Skip Navigation
  • Text Size: A A A
  • Print
  • Email
  • Facebook
  • Tweet
  • Share

Blood Safety Transcripts

U.S. PUBLIC HEALTH SERVICE
ADVISORY COMMITTEE ON BLOOD
SAFETY AND AVAILABILITY

EIGHTH MEETING

THE RESERVE CAPACITY OF THE NATION'S

BLOOD SUPPLY

Day Two

Friday, April 30, 1999

8:08 a.m.

Holiday Inn Bethesda

8120 Wisconsin Avenue

Bethesda, Maryland 20814

C O N T E N T S

PAGE

Blood Donation by Individuals

with Hemochromatosis

An Overview of Hemochromatosis:

Dr. Vincent Felitti

Southern California Permanente

Medical Group 5

Statutory Requirements for Blood Donations

by Individuals with Hemochromatosis:

Ms. Mary Gustafson,

Food and Drug Administration 19

Perspectives on Blood Donation by Individuals

with Hemochromatosis:

Dr. Alfred Grindon

American Red Cross 28

Dr. Victor Herbert

Mount Sinai School of Medicine 45

Public Comment 67

Committee Discussion and Recommendations 87

Potential Contributions of Blood

Substitutes to the Reserve Capacity

of the Blood Supply

The Potential Therapeutic Benefits of

Blood Substitutes:

Dr. C. Everett Koop

Dartmouth College and Biopure, Inc. 160

Food and Drug Administration Oversight of

Blood Substitute Development:

Dr. Abdu Alayash

Food and Drug Administration 186

Dr. Toby Silverman

Food and Drug Administration 201

C O N T E N T S (continued)

PAGE

Potential Contributions of Blood

Substitutes to the Reserve Capacity

of the Blood Supply (continued)

Presentations by:

Dr. Peter Keipert

Alliance Pharmaceuticals 220

Dr. Timothy Estep

Baxter Hemoglobin Therapeutics Division 252

Dr. William Hoffman

Biopure, Inc. 281

Dr. F. Lou Carmichael

University of Toronto and Hemosol, Inc. 310

Dr. Stephen Gould

University of Illinois and Northfield

Laboratories, Inc. 323

Dr. Robert Winslow

Sangart, Inc. 348

Committee Discussion: Old and New Business 359

- - -

P R O C E E D I N G S

DR. CAPLAN: Good morning. As you will recall yesterday, we took an overview of some of the problems facing supply of blood for Americans and the need to preserve access and safety, as we start to think about meeting the challenge of shortage and what to do in that area.

This morning, we are going to look at one particular possible avenue for increase, blood, which is the use of blood from individuals with hemochromatosis, and this subject has been around in the literature a bit. People have been discussing it and talking about it. Despite the Chair's best efforts, there were even outbreaks of discussion about it yesterday, but now is the time to really have the panel reflect upon this and come to understand what some of the pros and cons are here.

Our first presenter is Dr. Vincent Felitti from Southern California Permanente Medical Group, who we asked to give us an overview on hemochromatosis.

DR. FELITTI: Ladies and gentlemen, I thank you for inviting me to describe the clinical picture of hemochromatosis, particularly as it relates to blood donation. The knowledge base from which I will be speaking is as an internist with Kaiser Permanente in San Diego, where we have been screening 50,000 adults for hemochromatosis each year since 1997 and where I personally treat and follow over 300 patients with this disorder.

Hemochromatosis is a disease that most people have never heard of and with which many physicians are still unfamiliar, but it is also recognized now to be the most common, potentially fatal, genetic disorder in the United States.

The essence of the disease is the extra absorption of about 1 milligram a day of iron more than would be absorbed by a person who did not have hemochromatosis.

Once absorbed, iron cannot be excreted by humans. Therefore, given sufficient time over the course of years and decades, excess iron builds up in various tissues causing the serious damage that is the well-known end result of hemochromatosis.

As a result of autopsy studies, biochemical tests, and now genetic analysis, hemochromatosis is now known to be present in 4 per 1,000 Americans. That is to say, there are approximately 1 million Americans who have the full-blown genetic machinery to develop iron overload disease during their lifetimes.

I would like to show a few slides to help fix this in your mind visually and to help understand the relevance of this most common of genetic disorders to the national blood supply.

It will be important for you to remember throughout that hemochromatosis can only be inherited. It cannot be transmitted.

The importance of hemochromatosis to the national blood supply lies in the fact that the goal of hemochromatosis, the goal of treatment of this condition, is to normalize the total body iron load, and the only practical way of accomplish this is by removing blood because of the iron contained in blood and because of the ability of the body to rapidly regenerate blood so long as iron is present.

This is another way of looking at the prevalence data. On average, any physician in the country, no matter what the specialty, knowingly or commonly unknowingly, sees a case of homozygous hemochromatosis, the full-blown genetic disorder, every 2 to 3 weeks.

Many of those cases, fortunately, as pre-symptomatic. Unfortunately, the large number who are symptomatic usually have their illness attributed to some other common disorder, and I would like to show you the way in which hemochromatosis presents, usually masquerading as other more common conditions.

The important thing about hemochromatosis is that it is a totally preventable condition if diagnosed early and if treated early by phlebotomy.

This is essentially the machinery underlying iron absorption, whether it occurs at a normal level or at an abnormal level, as in hemochromatosis. Iron is absorbed from the intestine. It is bound to a carrier protein transferrin, which moves it to blood, to muscle, to enzymes, which moves it into appropriate storage depots, or in the case of hemochromatosis, into excess storage and abnormal organ deposits, and also is responsible for minor loss in menstruation and child birth.

A key point to remember is that in hemochromatosis, there is no biological method of excreting or secreting iron once it gets into the body.

The way this condition presents itself most commonly is as skeletal manifestations, as joint pains, as true arthritis, which can mimic osteoarthritis, seronegative rheumatoid arthritis, or gout or pseudo-gout, even leading to the necessary for joint replacement. A surprising number of people with hemochromatosis have joint replacement at an early age.

Gastrointestinal manifestations of this condition are enlarged liver, abdominal pains, which appear to be not associated with stretching the liver capsule, diarrhea, cirrhosis, and primary liver cancer. All of these obviously have other causes and, in general, people with hemochromatosis are considered to have other causes, much to their misfortune.

There are cardiac presentations of hemochromatosis involving heart enlargement, slow heart rate, rhythm disturbances and congestive heart failure. The hematologic manifestation of hemochromatosis is anemia. We are certainly used to the concept of iron deficiency in anemia. That occurs because there is an inadequacy of a key building block for hemoglobin. At the other extreme, anemia occurs from an excess of iron because of the toxic suppressive effect that iron excess has on bone marrow function.

There are neuropsychiatric manifestations of iron overload of hemochromatosis. They include depression, dementia, profound fatigue, peripheral neuropathy, including deafness and tinnitus.

This is a photograph of two sections of a brain that have been stained with iron. The amount of iron in this brain is really quite extraordinary.

There are skin changes in hemochromatosis, bronzing of the skin, loss of bodily hair, and a rare condition called porphyria cutanea tarda.

The endocrine manifestations are numerous. Diabetes is a well-recognized complication and a common one of hemochromatosis. Sterility, infertility, testicular atrophy, and impotence are all the end result either of pituitary damage or of ovarian or testicular damage. Premature menopause is fairly common, and hypothyroidism is common.

So this is really an extraordinary array of presentations of this one condition. Diagnosis of the clinical case is precisely what one does not want to accomplish because it excludes the possibility of pre-symptomatic treatment.

Certainly, any internist is proud typically to make a diagnosis of clinical hemochromatosis. It is far better as a result of population screening to find pre-symptomatic cases.

I think an appropriate analogy would be to the pride one might have in diagnosing a case of tetanus versus the real social accomplishment that one might have by immunizing a population with tetanus toxoid.

So the goal in hemochromatosis, as in hypertension, is to find and treat the pre-symptomatic condition. Treatment of this pre-symptomatic condition involves the removal of blood repetitively over a lifetime, and that is what makes hemochromatosis a totally preventable condition, if it is diagnosed early and appropriately treated.

Treatment of hemochromatosis typically consists of removing 1 pint of blood per week for whatever number of months, sometimes beyond a year, that it takes to normalize the total body iron load. After that is accomplished, people subsequently must have blood withdrawn several times a year for the remainder of their lives in order to keep iron levels normal.

At present, this blood is drawn and discarded, even though no knowledgeable person believes that it is intrinsically dangerous to transfuse. Indeed, it has been transfused in Sweden for approximately 30 years without a problem, and in Canada for almost a decade.

I think in addition, anyone familiar with hemochromatosis patients will quickly realize that a large number of them have been super donors, having donated 40, 50, 80 pints of blood over the course of a lifetime to the Red Cross, but instantly becoming ineligible for further donation once the diagnosis is made, the diagnosis that explains their extraordinary ability to be super donors.

Lastly, because of the ability of people with hemochromatosis to regenerate blood so rapidly, these people are less likely than most ever to have needed a transfusion, and therefore, they are less likely to have previously contracted any transfusion-related disease that they might pass on.

Their need to recurrently donate blood, coupled with the observation that repeat donors are generally believed to be safer than one-time donors, makes a forceful argument that the United States should update its procedures for screening blood donors. A well-intentioned rule of thumb that excluded donors who might benefit from donation might well have served a good purpose in another area, but it is embarrassingly anachronistic in a time of advanced laboratory technology.

Because of all these factors, a situation exists that is truly a win-win situation, both for the Nation and for individuals, the 1 million Americans who carry two sets of abnormal genes for this condition, the Nation needs more pure blood. Approximately 1 million repeat donors need to give blood to prevent iron overload. That blood is acknowledged by all parties, and by the experience of several nations to be intrinsically safe. So a wonderful solution is at hand if you allow us to discard a much outdated rule of thumb that is inhibiting progress.

Is there anything that I can make more clear?

DR. PENNER: What is the incidence of the phenotype, the figure that you would have?

DR. FELITTI: The question was: What is the incidence of the phenotype?

DR. PENNER: Not the genotype.

DR. FELITTI: Correct.

The phenotype has a population prevalence for the most common mutation of approximately 1 in 8. You raise a very implementation question, namely how many of these people who are genetically homozygous will go on to develop the full-blown disease. That totally is a function of time.

I can tell you, we have looked at our own population of homozygotes, people with the full-blown disorder, and in those who were over 65 years of age, that is to say had adequate time to absorb iron if they were going to in their life, 90 percent of them had either signs or symptoms that logically were attributable to hemochromatosis.

Whatever the exact number is, most people feel that a clear majority of people who are homozygous for hemochromatosis will develop the phenotype, given sufficient time.

DR. PENNER: So that is the 1 out of 200, plus or minus?

DR. FELITTI: That is correct.

DR. CAPLAN: Other questions?

DR. HOOTS: I presume you are talking about, in terms of monitoring them phenotypically, you are following their ferritins. Is that correct, primarily, or are you using actually spec scans and other sophisticated means to follow?

DR. FELITTI: No. We follow their iron levels typically by ferritin.

DR. HOOTS: I know in secondary hemosiderosis in hematologic diseases like thalassemia that there is pretty good data that it correlates not 1 to 1, by any means.

So I guess what you are saying is it really should never get to the point where you would even have to do more sophisticated monitoring because, once you recognize the ferritin over 1,000 and you do the genotypes, then you would obviously preempt any of that morbidity, anyway.

DR. FELITTI: I think the basic point that you are raising, if I follow you correctly, is that ferritin is handy, but a less than 100-percent perfect measure of iron levels.

DR. HOOTS: Correct.

DR. FELITTI: That is absolutely correct.

So what we do initially to track iron levels is quantitative phlebotomy; in other words, keep track of the amount of weekly removal of blood to bring a person from their starting hemoglobin typically to the borderlines of anemia. Usually, that correlates well with ferritin, but as you are pointing out, it does not correlate perfectly with ferritin.

DR. CAPLAN: Did you have one more, John?

DR. PENNER: Yes.

Do you routinely do the genetic typing on your patients?

DR. FELITTI: We do. We do largely because it is available through the courtesy of Dr. Ernest Boitler at Squibbs Clinic at no charge.

Before the genetic test was available, it was certainly possible to diagnose and treat and follow people quite satisfactorily.

I think the main advantage of genetic typing is that it settles the argument with dubious physicians as to whether their patients have hemochromatosis or not.

DR. PENNER: Do you think any of the heterozygotes represent a problem?

DR. FELITTI: Oh, yes. There is no question about that. We have people who are currently genetically normal who are significantly iron overloaded. Presumably, there is another as-yet-to-be-discovered gene or more, and we certainly have heterozygotes who are significantly overloaded.

DR. CAPLAN: Let's do Jay, and then we have got to move on.

DR. EPSTEIN: Dr. Felitti, thank you for the very nice overview.

Since you are involved in surveillance, can you tell us what percent of persons at the time of their first diagnosis are discovered to be regular blood donors, and can you also comment whether in your opinion physicians are directing people to go to blood banks to have their phlebotomies, other than prescriptions for therapeutic phlebotomy?

DR. FELITTI: Yes. Let me answer the last half first. Yes, I believe that is true. I believe that there are physicians who direct patients to go to blood banks to donate blood, telling them not to mention that they have hemochromatosis. Yes, there is no question that that happens.

How often it happens, I do not know, but certainly often enough that I have seen it without any great search.

In terms of the first part of your question, could you restate that for me, please?

DR. EPSTEIN: Yes. At the time of their diagnosis, what percent of hemochromatosis patients are discovered to be already regular blood donors?

DR. FELITTI: I cannot give you a precise fix on that. I am impressed, however, from speaking with hemochromatosis patients that a surprisingly large number have been so-called super donors.

DR. CAPLAN: All right. We have got a lot of presentations. I want to thank you for that excellent overview.

I am going to ask Mary Gustafson from FDA to come and tell us a bit about the statutory requirements that influence the availability of blood from persons with hemochromatosis.

MS. GUSTAFSON: Thank you.

I will start with a couple of items about the agenda. I was given an honorary title of "Doctor," in which I appreciate, but I do not actually have one.

Also, the topic is entitled "Statutory Requirements for Blood Donations by Individuals with Hemochromatosis." The statutes governing blood and blood components, primarily the Food, Drug and Cosmetic Act and the Public Health Service Act, do not specifically address these issues. The laws provide broad governing principles about product quality and include prohibitions against adulteration of products and misbranding.

The Federal agencies that implement the laws passed by Congress are given the authority to interpret the laws, and they do so by promulgating regulations that are published in the Code of Federal Regulations.

FDA's regulations are codified in Title 21 of the Code of Federal Regulations, and what is published as a Final Rule, the regulation has the effect of the law under which it was written.

The next overhead.

It is in the Code of Federal Regulations that we find a rule that affects the collection and use of blood from donors with hemochromatosis. The regulation is found at 21 CFR 640.3(d). This section of the regulation addresses the suitability of donors for whole blood donation.

The specific citation pertains to therapeutic bleedings and reads: "Blood withdrawn in order to promote the health of a donor, otherwise qualified under the provisions of this section, shall not be used as a source of whole blood unless the container label conspicuously indicates the donor's disease that necessitated withdrawal of blood."

The intent of some of the older regulations is not very clear in our history, but I think the regulation was meant to be informative and not prohibitory. However, use of such labeled blood has fallen into disfavor over the past years.

From anecdotal reports, this appears to be related to concerns over blood safety in general. The consumer's greater attention to the blood they receive, that is, persons being transfused actually reading the blood container label as the blood is infused and the transfusing blood establishments implementing notification to physicians of patients receiving blood, is that the unit of blood has an additional therapeutic bleeding label.

Given the choice of transfusing a special-labeled unit or a unit without such labeling, the physicians would often choose the unit without the therapeutic bleeding designation.

Although this regulation is the only one directly related to today's topic, I would like to clarify another blood labeling requirement and its relationship to donations from persons with hemochromatosis.

The blood labeling regulations in 21 CFR 606.121 include provisions for labeling blood for transfusion according to the donor status; that is, whether the blood was donated from a volunteer donor or a paid donor.

The regulation was promulgated in the 1970's to reduce the risk of hepatitis B transmission by transfusable blood components. The regulations are very specific in the interpretation of who is a paid donor and who is a volunteer donor.

A paid donor is a person who receives monetary payment for a blood donation. A volunteer donor is a person who does not receive monetary payment for a blood donation.

The regulation continues by noting that benefits, such as time off from work, membership in blood assurance programs, and cancellation of non-replacement fees that are not readily convertible to cash do not constitute monetary payment within the meaning of this paragraph.

The preamble to the 1978 Final Rule contains more examples of benefits that do not constitute monetary payment for the purpose of labeling a donation that is from a paid donor. These include a large amount of time off from work, additional vacation time, rewards generally used as a motivation for donations such as lotteries, giveaways, a chance interest in a prize with significant dollar value, or non-monetary rewards associated with product promotion, reduction or cancellation of hospital charges that are unrelated to the transfusion and the cancellation of refund of non-replacement fees in blood assurance insurance benefits.

In essence, the only instances in which blood must be labeled as having come from a paid donor is when the donor receives direct monetary compensation or a gift that is readily converted to cash.

Under this rule, the donation from a person with hemochromatosis, even if there are benefits to the person associated with the use of the blood for transfusion versus bleeding the donor for strictly therapeutic purposes, would not require a "paid donor" label.

Now having provided a pretty black-and-white description of "paid" versus "volunteer," does that mean that FDA is not interested in the gray areas of incentives and what constitutes a good motivator to encourage donations from safe donors and what constitutes an undue enticement for the donation from an unsafe donor? The answer is no. We are very interested in this issue.

In the instance of hemochromatosis donors, the issue is not whether the donor will provide false information about having hemochromatosis, but whether being able to donate a unit of blood for transfusion as opposed to having to pay for a therapeutic phlebotomy might provide an undue incentive to not be candid in the donor interview about risk factors unrelated to hemochromatosis.

This discussion is not limited to hemochromatosis. Several years ago, we had the same discussion about autologous donors, do they truly meet the definition of a volunteer donor.

As noted yesterday in Miriam Sullivan's talk, this has resulted in, I think she said, less than 1 percent of all autologous bleeds being crossed over for general donation.

I think this all ties back into the discussions yesterday, particularly Dr. Williams' report on incentives for donation. We were most pleased with the inclusion of the incentive issue in the REDS survey and in terms of being able to collect data on a multi-center basis, and we will be most interested in seeing the final report of the 1998 data.

Are there questions?

DR. NIGHTINGALE: Captain Rutherford?

CAPTAIN RUTHERFORD: Mary, I have two questions. From your discussion, would it be safe to say that the FDA would not disapprove of the use of hemochromatosis donors as a raw product source for hemoglobin and hemoglobin-based oxygen carriers, provided they are biomarker-negative and meet the many physicals?

MS. GUSTAFSON: Where we have been on source material for any human blood-derived product is that the source material must meet the requirements in the regulations for transfusion or source material, and since we have not prohibited the use of blood from hemochromatosis donors, then meeting other donor suitability requirements, it would be suitable source material.

CAPTAIN RUTHERFORD: The second question is: Could they be collected more often or more frequently than the 56 days as a raw source provider, provided the physician sets the minimum hemoglobin levels?

MS. GUSTAFSON: We do have a section in the regulations that allows for collection more frequently if a physician examines the donor and finds that the donor is suitable for more frequent collection.

DR. NIGHTINGALE: Any other questions from the panel?

[No response.]

DR. NIGHTINGALE: If not, Dr. Gustafson--"doctor"--Captain Gustafson, I thank you very much, and I apologize for mislabeling an excellent talk.

MS. GUSTAFSON: Misbranding?

[Laughter.]

DR. NIGHTINGALE: Misbranding an excellent talk.

We will next have two perspectives. The first will be by Dr. Al Grindon and the second by Dr. Victor Herbert. They are scheduled for 30 minutes on the program because we anticipate substantial public comment.

I wish to give the two perspective speakers as much time as possible without impinging on the public comment. So could I have a show of hands, please, of the people in the audience who wish to give public comment?

I see four. If there are only four in the room, then I would hope that Dr. Grindon and Dr. Herbert could take the time that they need to express their perspectives, leaving time for both the comment and questions from the Advisory Committee. So we will play the time by ear, and we thank you all for your interest.

Dr. Grindon?

DR. GRINDON: Thank you very much. I appreciate the opportunity to visit with this group today.

I am employed by the Red Cross and on the faculty at Emory University, but I am here speaking as an individual and do not represent either of those groups.

I would like to spend a few minutes today addressing three questions: If patients with hemochromatosis were to be blood donors, how many units would be available? The second question, why is blood from patients with hemochromatosis not used today? And the third, can such patients be blood donors tomorrow?

I would like to begin with the data that you have already seen from Dr. Felitti that these are not Hispanic Caucasians. If you add Hispanics, that is another 30,000, and you end up at 5 percent of this group, a million. If it is .4 percent of a larger number, it still is a million, and that is about the same number that you have heard earlier today.

Note that these are homozygotes and not necessarily patients needing treatment throughout their course. So the question from the treatment of these folks is how much blood might be available.

I would like to share with you data from two publications that are in the press and will be published in June of 1999 in the Journal of Transfusion, and I will be going through a little bit of the data from each of these presentations.

The first one is McDonnell and coworkers, and these folks did an international convenience sample survey. What they did was to contact every patient they could find by whatever means they could find, including contacting patients directly, contacting organizations who might be able to have patient identity, contacting blood centers, and contacting practitioners. In each case, each of these groups would then contact the patient. They also used the Internet toll-free numbers, newsletters and other media sources. So they sent this survey to anyone expressing interest in having one.

This was an international survey. 2,300 of the 80 percent of respondents were from this country, and of those, 62 percent were men and 99 percent were white.

The data that I wanted to focus on, though, was these two numbers. The number of units withdrawn a month had a mean of about 2.6, and the maintenance, which is perhaps more germane to our discussion today, was .5 units per month for maintenance of the respondents in this survey. That is, of course, 6 units per year.

The other paper that I would like to discuss briefly was data from Barton and coworkers. This represents a group of patients who were treated by Dr. Barton's group in Alabama. They used probands as the subjects of this study rather than all their patients with hemochromatosis because they felt that probands were usually more severe than the relatives who might be subsequently diagnosed and, secondly, because probands may represent the general population more than relatives would because they are unselected. So these are probands.

They were diagnosed during routine medical care, and all probands treated by this group were included.

They were compared, then, to volunteer donors from the same part of the country. The testing throughout the course of these years was not identical, but reasonably comparable, and the eligibility also was comparable using the standards of the AABB. These standards were applied retrospectively.

Dr. Barton used what I hope many who are experts in treating this disease. He used what I think are standard approaches to the treatment of these patients and after induction sought to provide phlebotomy to keep the levels of less than 50 nanograms per Ml.

What is of interest here is that this produced a little more than 1 unit a year during the first year, and this is what is published or what will be published in June, but also with smaller numbers for subsequent years in Dr. Barton's experience.

The results, however, go on, and I would like to focus on two numbers here, the 59 percent and the 88 percent.

For the volunteer donors, the Red Cross donors, of all those who walked in to present, 88 percent were units that ended up being available for sending out to hospitals.

For the patient group, 59 percent during maintenance turned out to have units that would have gone through our system and been available for distribution. The number was a little lower during inducement, 44 percent, primarily because of the lower hemoglobin levels in these patients during the induction process.

So now we can get to a conclusion and say how many units would be available annually if you used phlebotomy during maintenance and used those units. If you have 50 percent who need treatment--and let me spend just a minute talking about that number--I am thinking about this not from Dr. Felitti's statement that 99 percent of people at age 65, or 90 percent, whatever it was, are going eventually to develop signs and symptoms and should be phlebotomized, but rather from the perspective of people being eligible as blood donors.

You begin by saying that three-quarters of the population are going to be greater than 17 years old. So it is not 100 percent that are eligible as blood donors, but only three-quarters of those.

Secondly, at any given time, women are less likely to be symptomatic. The obvious reason would be because of premenopausal blood loss or child-bearing. It is not quite that simple, but that is a reasonable kind of a statement.

There are some penetrants issues. The penetrants may not be 100 percent, and I think also the larger numbers that we hear assume perfect screening systems. Really, we may not be able to identify this 1 million people, at least right away.

So, when you go through this 50 percent, you start with 500,000. If you draw 6 units a year, you end up with 3 million units.

Dr. McDonnell's study did not address donor eligibility or unit suitability. So we have the assumption, which, of course, we recognize is not correct, that 100 percent would be eligible as donors and 100 percent of these units would be suitable. That is just unstated in the survey paper.

If you are drawing a little more than one a year and you start with 500,000 units, but only 60 percent of those end up being acceptable by our current techniques, then you end up with 300,000.

So the point of this is that it is not really clear how many units would be available were we to use blood from patients with hemochromatosis as blood donors. There is a log difference in these two numbers.

Could these numbers be off? Yes. Could it be more than 3 million? Certainly. I think that Dr. Felitti has told you that some heterozygotes become symptomatic, and if you include heterozygotes, you have a larger number.

There are also those who might be interested in having blood drawn from them because of concern about the so-called iron lesion and prevention of heart disease.

Understand also that these numbers deal only with maintenance, and if one were to address the use of units of blood obtained during the induction or de-ironing phase, you would have more units available.

Could the number be less than 300,000? Yes, I think so. I think that the biggest issue here is the issue of recognition of these patients with screening.

I might say also that there is a regulatory constraint. Some of these donors are going to need phlebotomy more often, even in maintenance, than several times a year, and might be constrained because of the ruling of the current CFR rule that Ms. Gustafson discussed that donors need to wait 56 days, unless, of course, they are seen personally by a physician on the day of donation.

So that is the number of units of blood that are available, and I think we do not know how many units there.

The next question is: Why is blood from these patients not used today? I have these three issues that I would like to address. The first of them, I think Ms. Gustafson had addressed really quite handsomely.

The first bullet is the CFR regulation. The second bullet, she discussed, and the third bullet is from the standards of the American Association of Blood Banks. Because of this confusion, the AABB has said this blood really should not be used.

The second is a financial motivation. These are data taken from the survey conducted by McDonnell and coworkers and show that the charge--not the cost now, but the charge--was ranging from $90 to $52, depending upon the place the patient went to get phlebotomy.

Most of those in the survey had some sort of insurance, and 39 percent had no out-of-pocket cost, but the mean out-of-pocket cost for the whole group was $45. So there really was for many of these folks a significant out-of-pocket cost.

As you have heard from Ms. Gustafson this morning, the FDA has been concerned about this. The FDA and the NHLBI convened a workshop in September of 1996 at which time Dr. Zoon, speaking for the agency, said what Ms. Gustafson has just said, that this is an area of great concern to the agency.

The AABB in a bulletin regarding what donor incentives would be unacceptable, dated 1994, and reaffirmed subsequently has said that an incentive is unacceptable if it is sufficient to entice someone who, acting solely on the basis of altruism, would not give blood.

I think the feeling of the people that I have talked to is that a $45 out-of-pocket cost is felt generally to be such an enticement. So there is this financial concern.

The third area I would like to discuss briefly is why this blood is not used. It is non-financial motivation. Patients with hemochromatosis are patients, and they are giving as a patient and not primarily as an altruistic donor.

Again, as Ms. Gustafson has said, almost as if she knew what I was going to say, she discussed the issue of autologous donations, and I think this is a helpful analogy to consider.

The issue is if an autologous donor meets all of our historical screening criteria and passes successfully all the blood tests that we do, should that blood be used for other patients, if not used by the autologous donor patient himself. That is to say, should that blood, if unused by the patient, be crossed over into the general population.

So do we do that? The answer is no, we do not. Why not? Well, there is an increased market frequency in donors who pass our regular screening criteria. There is an increased marker frequency in the autologous population.

In order to determine this, you have to do this very carefully. You have to match first-time donors with first-time donors. You have to match zip codes. You have to match age and sex. When you do all of that work very carefully, then you find that there still is about a three-times-higher marker frequency of anti-HBC, anti-core. This was before hepatitis C testing was available, much of this work.

As a result of this marker frequency, the assumption, which may be erroneous, is that there is a higher risk in marker-negative donors from the same population, and, therefore, the AABB standards in 1997 reflecting what has become common practice said that this blood should be used only for autologous transfusion.

I would like to move on to the third question, which is can these factors change. Well, yes, clearly, the CFR can be changed, and the labeling requirement removed. I would like to note here, as Ms. Gustafson noted, that the labeling requirement is not by itself prohibitive. Despite the label, that blood can be used. It is just that it adds such concern on the part from the blood center to the transfusion service, transfusion service to the clinician, and clinician to the patient, that it became unwieldy.

If you change the labeling requirement, though, I think there are concerns that need to be addressed, and that is the potential for the loss of patient autonomy, the right to know, the issues off informed consent. It is not that we feel that this blood intrinsically is any less safe, but that patients really have a right to know that this is blood from the patient rather than from a volunteer donor, at least there is that risk of losing that.

The second is the financial motivation. Clearly, we can change--"we," not me, but this group can consider reimbursement patterns and work to see those changed.

Some of the discussion yesterday that this group had was focussed in that area. So those things, i think, can be changed.

The non-financial motivation is a little more difficult. How do we demonstrate that patients with hemochromatosis are different than patients giving for themselves, the so-called autologous donors? I think because of the autologous issues that the blood banking community has gone through, it may be important to do such a demonstration.

Does it make any difference that we now have vastly improved testing, the NAP technology, for instance, for known agents that makes the risk of known transfusion transmitted disease very, very, very small?

What if we had viral inactivation technology for cellular components? That is not available today, but if it were available 2, 3, 5 years from now, would that make a difference? I think it would. If such technology were available, then paid donors even outside Rochester, Minnesota, and Iowa might be acceptable for blood donation generally.

Let me conclude, then, by stating that the number of units available, at least from maintenance of patients with hemochromatosis, is unclear. My feeling is that it may be far fewer than the numbers that have been discussed.

My personal feeling is that the 300,000 number may be closer than the 3 million number per year, and I say that to this group because I think the numbers of units is worth considering as you look at potential shortages of blood down the road.

The common constraints are fixable, with some difficulty, but they can be fixed. The non-altruistic motivation, if this is felt by this group or regulatory agencies to be a concern, that is going to be more difficult to address. It may not be a concern, but if it is, that is more difficult to address.

Finally, these concerns may be much less important and perhaps irrelevant with viral inactivation of cellular components.

Thank you. I would be happy to answer questions.

DR. NIGHTINGALE: Thank you very much, Dr. Grindon.

In the interest of time and making sure that everybody understands the ground rules in advance, Dr. Grindon, I would like to thank for keeping within his allotted time, which was 20 minutes.

I propose 5 minutes of questioning from the panel for Dr. Grindon, and propose that Dr. Herbert also have 20 minutes and 5 minutes for questioning. That would give each of the four public speakers 5 minutes to speak, plus some time for the panel to ask them questions, so we could have our break at 10:00 and the committee could have adequate discussion after that time.

So are there any questions from the panel?

Dr. Epstein?

DR. EPSTEIN: Are you aware of any actual studies on marker rates or window-period risks, that is to say, seroconversion rates, in this population?

DR. GRINDON: No, particularly if you consider this from the perspective that Dr. Busch spoke yesterday. Clearly, we have marker rates in the work of Dr. Barton, but that is not answering your question, which is what is the incidence and not the current prevalence, and the answer is no, we do not have such data.

Yes. Dr. Chamberland?

DR. CHAMBERLAND: I was just wondering if you could comment perhaps from the McDonnell paper on a question that Jay Epstein had asked earlier.

In that survey, I believe respondents were asked about previous history of blood donation?

DR. GRINDON: Yes. I am sorry. I did not write that down. I have the paper, and I can get it for you.

I do know in Dr. Barton's study, it was 29 percent. So a substantial number had been blood donors previously.

DR. CHAMBERLAND: I think my recall from the McDonnell paper was that it was around a third or a little bit over, about 35, 36 percent.

DR. GRINDON: Which is the same kind of range, yes.

DR. CHAMBERLAND: So in the same range?

DR. GRINDON: Right.

DR. NIGHTINGALE: Are there any other questions?

[No response.]

DR. NIGHTINGALE: If not, thank you very, very much, Dr. Grindon.

Dr. Herbert?

DR. HERBERT: Good morning.

I apologize for being physically handicapped, but no good deed goes understand punished, and when the Gulf War broke out, I rejoined my paracommando Green Beret unit, and my first Gulf War jump, which was my 153rd jump in 4 wars, resulted in damaging my spinal cord. So no good deed goes unpunished, and I am temporarily crippled.

I would like you all to pick up the handout I left for you, which is our two-paged paper in last November's issue of the American Journal of Hematology, because what I am going to do in my allotted time is elaborate on it, including material presented this morning.

I will start first with Al Grindon's excellent presentation in which he stated indirectly that his organization was the culprit in creating the blood shortages twice a year and in creating not using good hemochromatosis blood.

You will recall he told us that his organization mandated in 1996, and I quote him, "This blood should not be used." I had spoken with Mort Spevak, and you will notice on my handout the first full paragraph on the full page of the written document that Mort Spevak, after pointing out that this hemochromatosis blood would solve all the problems in blood banking, both problems of contamination and problems of availability. What he did not say in that article when I asked him, "Mort, so why don't you use hemochromatosis blood at our blood bank?," he said, "Look, Victor, if I have one disaffected technician, all he has to do is call the AABB and tell them that I used hemochromatosis blood, and they will withdraw my blood bank approval by the AABB and I will be out of business. That is why I cannot do it." So the AABB is clearly the culprit in this matter.

As to the allegation that the Red Cross and the AABB will not use hemochromatosis blood, that is a lie. They use hemochromatosis blood all the time, and why do they use it? Because they do not know it is hemochromatosis blood, nor does the donor, because there are years of iron overload from hemochromatosis before there is any disease from it. So the donor does not know it, and the percentage of Americans who are heterozygous is 12 percent, which means about 12 percent of blood donors have at least 10 years of giving blood before they have any symptoms of iron overload, such as diabetes or liver disease or cardiac arrhythmias, all of which would preclude their giving blood anyway in the first place. The percentages vary according to ethnic groups in the United States.

For example, right here in Washington, D.C., Victor Guardia published in the New England Journal of Medicine, almost a decade ago, that 32 percent of all African Americans are heterozygous for hemochromatosis, and at least 1 in 100 African Americans are homozygous. That figure is now 1 in 80 who are homozygous for hemochromatosis.

There was a blood bank study by the Utah Group published in the New England Journal of Medicine in 1988 in which they took Salt Lake City blood donors, over 20,000 of them, and found that those blood donors who are given blood, 1 in 300 of that highly inbred Mormon population was homozygous. They had already given blood, and the blood banks did not know it and they used it, and 1 in 12 was heterozygous for hemochromatosis. That was published in 1988 in the New England Journal of Medicine.

We know that the frequency is much higher among Irishmen in Dublin, as Roberta Crawford's handout tells you and as you can read yourself in the 1999 issue of the Journal of Blood Cells, Molecules and Disease. A group in Dublin found that the frequency of homozygous hemochromatosis in Dublin is 1 in 100, and the frequency of heterozygous hemochromatosis is 1 in 5. So 1 in 5 Irishmen are heterozygous for hemochromatosis. They are all given blood until they get overt disease.

Their blood banks do not know it. Why? Because the blood banks do not bother to test iron status. They test hemoglobin because they lack the knowledge from the literature to know that you can have anemia from iron overload, just as you can have anemia from iron deficiency. So they turn these patients away who have low hemoglobins, and they tell them you are iron-deficient because they do not know, if you are iron-overloaded, you can be anemic from the iron deposited in your bone marrow. When you are repeatedly phlebotomized, your anemia goes away because the iron in your bone marrow, which is crushing your baby bone marrow cells, is gone. So now you make blood normally. So the blood banks are just kidding themselves.

After yesterday's session, Dr. Ronald Gilcher, who was the last speaker yesterday morning, came up to me, and he said, "Just between us."

[Laughter.]

DR. HERBERT: I said, "Is it all right if I repeat this tomorrow morning?," and he said, "Yes." He said, "Just between us," and I repeat he authorized me to tell you this morning.

We at my Oklahoma Blood Institute send our blood-banking vans all over our area, and we tell people, "Come, give a unit of blood. We will examine your blood for any disease that we might pick up in there, any viremia, hepatitis A, B, or C, iron deficiency, iron overload. Just give us a unit of your blood, and we will give you all of this knowledge in exchange and tell you if you have got some disease that we picked up and how to get it treated, how to get it promptly treated."

He tells me that by doing this, he never has a blood shortage. He uses the hemochromatosis blood, and those people then become regular donors because he tells them they have got iron overload and they have got to give blood regularly because that is the treatment.

You have got to get fixed in your mind the difference between a genotype and a phenotype. A phenotype is a person with iron overload disease. A genotype is a person who has a gene that predisposes to a disease. To have a genotype is not a disease. It is not an abnormality in the body. It is a warped gene which may or may not ever be expressed.

Fifteen percent of patients with genotypic hemochromatosis never express it as disease. They never get iron overload because we also have anti-hemochromatosis genes, which block iron absorption. If you are such a person, your gene for iron overload cannot be expressed.

Also, you can express a genotype where a person is just hemochromatosis into the disease. Why? Giving him iron, which blood banks, AABB blood banks, recommend to all patients who are turned away for a low hemoglobin, with the simplistic thinking that a low hemoglobin means low iron, means you should take iron, which is pure nonsense.

So I would like to go on with a few more notes, and then I will quit. If you read the title of our article, you will see it is a triple hematologic nightmare, "Under-diagnosing and Not Treating the Most Common U.S. Genetic Disorder: Iron Overload," and discarding each year tons of their good donor blood, creating artificial donor blood shortages in each of the past 30 years.

As Dr. Gilcher told me and allowed me to pass on to you today, that is not a problem for Oklahoma City because he does not let it become a problem for them.

I further wanted to draw to your attention the importance of using this blood and the importance of all blood banks starting to add to their blood testing, simply transferrin saturation. By taking a low hemoglobin and saying that is high in deficiency, that is nonsense and it is dangerous. You also must determine transferrin saturation. If it is over 45 percent, as it will be in 12 percent of your donors, you have got a patient with probable heterozygous hemochromatosis.

So you then go back and you get the serum ferritin, and if that is over 200, you have got a patient with hemochromatosis who needs phlebotomy regularly. You should phlebotomize him regularly for his and your benefit.

Thank you.

DR. NIGHTINGALE: Dr. Herbert, thank you very much, and I would specifically like to thank you in addition for your presentation, for keeping well within your time limit. This is of a great relief to the Acting Chair.

Are there questions for Dr. Herbert?

Dr. Gilcher, of course, gets to speak first.

DR. GILCHER: With all due respect to Dr. Herbert for whom I have a lot of admiration for over 30 years, I did say it a little bit differently, and I think I need to tell you what I did say to Dr. Herbert.

[Laughter.]

DR. GILCHER: What I told him was that in our community, we made the decision to remove the phlebotomy charge. So this is now done as a community service benefit. Therefore, there is no potential for secondary gain.

We did that over a year ago with the idea that we could move forward with the concept of utilizing these donors, but currently, we do not. I personally believe that these donors could be used, but that there do need to be changes made. So we have made a first step forward.

Dr. Grindon, if I can discuss what you and I talked about, when you think you have a good idea, what you usually find out is that somebody else thought about it first. So I think Dr. Grindon deserves that credit because he told me that he has actually done this about 7 or 8 years ago in his community, and that is remove the phlebotomy charge and made it a community service benefit.

So I think that what we have heard this morning has a lot of validity, and I think this is an issue that should be re-looked at, at this point in time, but certainly one of the steps I believe that would be necessary would be to remove the potential secondary gain that the patient or the person with hemochromatosis could get by getting "no charge," or that is elimination of the charge.

DR. NIGHTINGALE: Dr. Hoots?

DR. HOOTS: Just a question, Dr. Herbert, in terms of clarification or perhaps any of the other blood bankers who are here.

I presume in the screening, a routine hemoglobin is being done, not a CBC. Is that correct?

DR. HERBERT: Excuse me. Are you asking me?

DR. HOOTS: Yes, if you want to answer it, or if you would like to defer to somebody.

DR. HERBERT: All I can speak of in blood bank experience is our own blood bank experience.

DR. NIGHTINGALE: Could we ask Dr. AuBuchon or Dr. Davey to comment?

DR. DAVEY: Yes. It is either a routine hematocrit or a routine hemoglobin.

DR. HOOTS: The reason I am asking the question is just from a pure hematologic point of view, and the last thing I want to do is propose other things for blood banks to do, I think, under this aegis.

I was asking mainly in the case that if it were a CBC, rather than putting the blood bank in the unenviable position of kind of making diagnostic demands, that one could at least use an MCV as perhaps a screen, but since you do not have one, you can do that.

I think maybe one of the things that we could suggest, and maybe it is already routine, is if it is the practice to make therapeutic recommendations based on a low hemoglobin, which I would hope would not be true, that that not be done and that patients who are diagnosed with anemia based on one hemoglobin be recommended to be seen by their primary physician for further workup.

DR. AuBUCHON: In all of the blood centers that I have worked in, all the blood collection facilities, it has been always the recommendation that the patients see their physician for continued anemia and that a direct recommendation for therapy is not made by the nurse or other individual who is performing the hemoglobin, although I would note that in the population in general, the prevalence of iron deficiency anemia far, far outweighs the prevalence of anemia due to iron overload.

DR. HERBERT: But the number of patients with iron overload is twice the number of patients with iron deficiency, and those with iron deficiency are primarily infants, adolescents, and fertile females.

DR. NIGHTINGALE: Dr. Epstein?

DR. EPSTEIN: Yes. Dr. Herbert, I have a question for you. You have very clearly educated us that many of our donors do have hemochromatosis. They just do not know it and their doctors do not know, and we use the blood all the time and that is certainly true.

I am struck by the fact that when that happens, without anyone knowing it, it is an entirely different circumstance than whether it would happen knowingly. It troubles me that you seem not to focus on the issue of whether there is an undue incentive when the donor knows that he or she has hemochromatosis that the future health of the individual depends on phlebotomy and then is faced with a more complex choice when asked about behavioral risk.

I put this in the consent that, certainly, you could argue that the risk on the whole would be no greater than in the general population and, therefore, relatively low, but the problem is that in the arena of blood safety, we deal with extraordinarily low risks and an effort to eliminate them. I think it is the possible occurrence of a person that knows they have a need for phlebotomy and also knows that they have a risk factor and the is asked to be a truthful historian. I find it simply unhelpful to comment that it occurs unknowingly when the situation of whether it would occur knowingly is quite different.

So could you comment on the issue of knowing donation?

DR. HERBERT: Sure.

DR. EPSTEIN: How does life change after the donor knows?

DR. HERBERT: First of all, only 1 in 100 donors knows, and that 1 in 100 donors, who know, have all been told they have hemochromatosis by their primary care physician because they have started to develop symptoms of diabetes or cirrhosis or cardiac arrhythmias or sterility. That is how they may know, but that is only 1 out of 100 donors, because the vast majority of the donors or the people who would be donors who know, know because they have already been diagnosed as having diabetes, and the doctor was one of the rare physicians who did a transferrin saturation, as we require at Mount Sinai in every patient accessed to our diabetes clinic, but outside of Mount Sinai, I would guess that no patient knows he or she has hemochromatosis until they have severe diabetes, cardiac arrhythmias, et cetera, which would preclude them from being donors anyway.

So your question is a hypothetical, but unreal one.

DR. NIGHTINGALE: Dr. Epstein, for a follow-up?

DR. EPSTEIN: The world is going to change if there is mass screening for the genetic disorder.

DR. HERBERT: Is that bad?

DR. EPSTEIN: No, no. I think that is very good for public health, but we also have to consider the implications for the blood supply, and I do not think you have answered my question. It simply does not help to tell me that unknowingly hemochromatosis donors donate.

What I am asking you is, do you consider the issue of an undue incentive to be irrelevant, and if so, how do you explain that? The whole issue is how does it change once someone knows.

Once you know you have an obligate need for phlebotomy, do we expect that you will be an equally truthful altruistic donor?

DR. HERBERT: I would say that if you have an obligate need because you have iron overload, then it does not matter because if you sneak in somewhere else and give blood as a donor, you will be rejected anyway as a donor because of your diabetes, because of your abnormal liver function tests, which those who are no longer genotypes, but are already phenotypes have. They are rejected because of their abnormal liver function tests. All of them have abnormal liver function tests.

DR. NIGHTINGALE: Dr. Gomperts?

DR. GOMPERTS: Dr. Herbert, what is the utilization of deferoxamine in this patient population?

DR. HERBERT: Of what?

DR. GOMPERTS: Desferal.

DR. HERBERT: Deferoxamine?

DR. GOMPERTS: Yes.

DR. HERBERT: First of all, it is not good for genetic hemochromatosis. So it is not used because it has been shown not to be good.

However, it is the therapy of choice for iron overload due to repeated blood transfusions to counter a genetic hemolytic anemia, as in thalassemia, and there, as you know, thalassemics have to get blood frequently. They develop iron overload which would kill them and has killed many until they started using deferoxamine in all of them. So they are all getting regular deferoxamine therapy.

DR. GOMPERTS: I understand that, but in this particular disease process, especially in the symptomatic patient with diabetes, et cetera, Desferal is used or is not used?

DR. HERBERT: It is not used because it is too slow. Phlebotomy is much faster and, therefore, much more effective.

DR. GOMPERTS: It is not a financial issue?

DR. HERBERT: No. It has got nothing to do with money. It is just that when you do a phlebotomy, you take out 250 milligrams of iron with each phlebotomy.

DR. NIGHTINGALE: Dr. Gilcher?

DR. GILCHER: My comment actually is a question to you, Dr. Epstein.

After we made the decision that I told you about, removing the charge in our community, we looked at the idea of actually doing a similar study as to the one that was done and reported in the New England Journal of Medicine. However, we had not proceeded because in fact we are concerned about taking people who are actually blood donors and now identifying them. This is the situation that is in between really the question that you have asked Dr. Herbert.

How do you feel about taking a group of blood donors and then actually testing them to determine if they do have hemochromatosis and identifying them? Then, would you still consider them to be blood donors after that? This is a concern that we have had.

So we actually have not done the study for fear of actually losing donors, not actually increasing donors.

DR. EPSTEIN: Well, as Mary Gustafson stated, the FDA regulations do not preclude the donation. They only require the label. So we would require to label the blood as therapeutic phlebotomy.

DR. GILCHER: Of course, we know it would not be used by our physicians.

DR. EPSTEIN: I understand that, but if what you are asking is, are studies possible, I would say yes, studies are possible.

Are studies worth doing to try to resolve these issues? I would say yes. They are not going to be easy, though. Ultimately, what you are trying to do is look at seroconversion rates in individuals who are denying risk factors and continuing to donate in the face of a known diagnosis. That is really what you want to know here. Even that is a surrogate for ultimate transmission to recipients, but I think that is close enough.

DR. GILCHER: I agree with you.

DR. EPSTEIN: But those are not going to be easy studies.

DR. PENNER: As Mary Gustafson, the former Dr. Mary Gustafson, had mentioned, there are really two sticking points that we are running into on this situation. One is the identification of a disease state that would be benefitted by the phlebotomy or whatever procedure is done.

That issue was really, I think, developed to protect us from the blood donations from the polycythemic patients, a condition in which there is a malignancy rate that develops, and therefore, removing blood is beneficial to the patient, but one does not know when the patient is going to develop a form of an acute leukemia. Therefore, the question of whether you might have a transmissible something or another there has to be avoided.

I do not think those who put that issue together really reflected on the possibilities of hemochromatosis, and it would seem to me that it would be possible to address that very simply by asking for a change in the Act to allow hemochromatosis to be acceptable.

The second issue is the volunteer aspect and reward which has developed. In about an hour before I left to come to this meeting, my hemochromatosis patient was complaining, as usual, about the fact that he had to come in to our clinic to have his blood drawn, and that in the old days, he used to be able to go to the Red Cross and just have the blood drawn and there was no problem.

Then I explained to him again, "Well, you are not a volunteer when you go in, and therefore, your blood is going to be looked at differently." He said, "You already talked to me about that. That is foolish. Why can't I just go in, if the blood will be drawn as it used to for this condition by any of the Red Cross or any of the blood banks? Why can't I say after they draw it, I want to volunteer that blood or I do not want to volunteer that blood? And that will get me around the fact that I am just volunteering, and I am not doing this on a reward basis," but that would mean all of the blood would have to be drawn, as Dr. Gilcher has put forth, as they used to many, many years ago, just on the benefit of the fact that the patient was asked by his physician to go in and have his blood drawn for his hemochromatosis condition and that there would be no charge whether the patient donated the blood or did not donate the blood to the Red Cross.

So I said, "Well, that sounds like a very good idea. I do not know how that will fly, but I will bring it up at the meeting," and that is what I did.

DR. CAPLAN: I think we are to our public comment period. Did you have a list of names?

DR. NIGHTINGALE: Yes, I do.

I recognize two of those who wish to comment publicly. I see there are three. If we could move from left to right with Dr. Krikker and then Mr. Alexander, and then we will move along the front row.

DR. CAPLAN: I know we have got statements. So, if you could keep your comments relatively brief, that will let the panel ask questions, too.

DR. NIGHTINGALE: Dr. Krikker?

DR. KRIKKER: I think that before I make any comments about my statement, there are so many issues that were raised and so much should be commented upon that I think I would prefer to make some of the comments about some of the questions that were raised.

First of all, with reference to your question about what percentage of hemochromatotic patients have been donors, since 1982, I began to survey our patient members, and 39 percent had been blood donors anywhere between 5 and 46 years. Most of them, after diagnosis, in most cases, their bloods were no longer acceptable.

Since the hemochromatosis gene was discovered in 1996, it has raised so many more questions that the issue of hemochromatosis is not quite as simple as some may feel it is.

We now know that there are many transferrin saturations in ferritins which are not hemochromatosis, and there are many patients who never expressed the disorder. So that, the whole picture now is getting a lot more confusing than it had been.

First of all, there is the very important issue of the volunteer donor. I think that is unfair to compare the pre-1970 paid blood donors to hemochromatotics for this reason. The paid donors had a habit that they had to pay for, usually drugs, and we know the relationship of hepatitis to a drug habit. So that issue of volunteers as an altruism, I do not think it is quite applicable to the hemochromatotics who are very health-conscious. They take care of themselves, especially those who want to maintain their phlebotomies, and I think it is unfortunate that so very many families, where they have very many members who require phlebotomies, have no insurance or no jobs. I think it is really criminal that something is not done about this situation.

Recently, Senator Daschle's office contacted me, and I did send them some material, especially a position statement that our advisory board drew up. In 1997, I revised the 1983-1985 position statement that we had drawn up.

In my experience with the various blood bank executives, I was amazed at the variability of knowledge that each had. In 1994, I received a number of faxes and phone calls from some of the blood directors, especially out in the California area, out West, that President Clinton's recent Executive Order was a request to all the agencies to please review all their regulations and see which regulations could be revised, perhaps with the thought of rescinding, because they are now not in keeping with the times and should be changed.

With that, I began to get our new advisory board members, and we include, among our advisory board, people who have been involved with hemochromatosis, way back since the '30s and '40s, Drs. Bothwell and Charlton from South Africa, Dr. Olsen from Sweden, and in this the country, the Finch brothers.

It is significant that in Sweden and in South Africa and more recently in Canada, the bloods are being used, and in my conversations with the representatives of the transfusion organizations, they all admit that they have had no problems using hemochromatosis bloods.

First of all, the bloods are screened. If they have any overt disorder, certainly many of them are not going to be donating blood.

I have made the suggestion to several blood bank directors, why not do the usual screening techniques of people who want to give blood, even those who have hemochromatosis. We have to redefine what is "disease." Is a genotype individual diseased?

Dr. Herbert is correct that a phenotype does represent disorder, and first of all, they are not going to be donating. They do not want to.

I think that instead of the board making any snap decision at this time, as your Dr. Nightingale has provided you with some very excellent data, especially the recent journal, the Annals of Internal Medicine, you all were given copies of our position statement, the various blood communities, the FDA advisory board. Please re-read that. These are people who contributed to the position statement. They are knowledgeable. This is their work. They know every detail of it, and they recognize now that this is getting much more complex than it had been.

In all reason, do not disqualify someone because suddenly you find that he has hemochromatosis. He may never have expressed the disorder.

I did have a 5-minute talk prepared, but I will just forget about it.

DR. CAPLAN: Thank you.

DR. HERBERT: It should be noted that Dr. Krikker, who just spoke, is a physician on the staff of the State University of Albany Medical School, and she is exceptionally knowledgeable in this subject, more knowledgeable than anybody sitting around the table here and has been for many years.

She became interested when she, as an internist, did not recognize, until he died of iron overload at age 46, that her husband had iron overload. She then formed this organization. Well, maybe she did know it beforehand, but she then formed this very effective organization, and she is really the lead speaker in the world for this subject.

DR. KRIKKER: I would like to make one correction. I became a physician because as a young bride, my husband, who was a research chemist, was given a diagnosis of Mediterranean anemia by the professor of a teaching university. I always wanted to be a physician, and I was going to save my husband, go to med school, and diagnose him. By golly, my second year as a pathology student, there, setting the liver slides, was the diagnosis. I am biting my nails. How do I tell my medical professor, who is his physician, he may be all wet?

As a matter of fact, we said to him, "You know, Doctor, we want to get a second opinion on this." "Well, what do you think he has?," and I said, "Well, he thinks he has hemochromatosis." He says, "Nonsense. He does not have diabetes." So this is how I became a physician, and the story is a little more involved and you knew about it.

DR. CAPLAN: We have got to get these other people going.

DR. HERBERT: I just had to elaborate on the CFR.

DR. CAPLAN: No, no. Let me let the next person speak. Then we will get back to the discussion.

MR. ALEXANDER: Hello. My name is Randy Alexander, and I have hemochromatosis. I do appreciate everyone who is here, the committee members, all the present presenters. I appreciate very much their knowledge, expertise, and enthusiasm.

I have experienced firsthand the frustrations, the emotions, having to have blood drawn off and was very expensive, and I had few alternatives because I was unable to work for long periods of time. It was costing me over $800 a month to have blood drawn off, and it was going to save my life.

There are people out there who are going to blood clinics. They have no choice. The hemochromatosis patients do not understand why this blood cannot be used. I do not understand why this blood cannot be used. It is an emotional issue.

However, I do see all of the complications here. As Dr. Krikker pointed out, this is very complicated. What we have experienced in working on this issue for a number of years and working with Dr. Krikker, the Hemochromatosis Foundation, Roberta Crawford, the Iron Overload Disease Association, and my association, the Iron Disorders Institute, is that patients are screaming and clamoring for something to be done or to be heard. They feel that they are being ignored and they are not being listened to. This is a very good first step for people to be able to tell people that something is being looked at, that this issue is being addressed.

I think what is needed here and what has been demonstrated here, Dr. Grindon, they have opened the door, so to speak, in showing us the way of how to resolve some of these issues and in showing that people working together, sitting down, working together on a common ground, that no matter how difficult these studies or the information or what is before us is, anything worthwhile is worth working hard for. I have found that out.

I think we need to bring the CDC to be involved in this and looking at specifically what is the science that is needed, what is in question here, what data do we need, identify that. Let's work with the FDA. Let's work with the CDC, the NIH, the AABB, the Red Cross, the insurance industry. I do see the problem here with costs.

Running a non-profit organization and helping people is expensive, and blood is a good value, but we have to have consistency in the fact that if some people are going out and paying $300 for a pint of blood and it puts them in financial ruin, what options do they have?

We have an opportunity. We have a resource of blood here, people who have hemochromatosis, and when they find out, they feel they can contribute to help somebody else. Then, when they sit there and they have the blood drawn up and a biohazard sticker is put on that bag, that really does not set well with them and they do not understand. So those are some of the issues we face, and that is where the emotional aspect comes in.

I think working together, everyone coming together, stop the finger-pointing and everybody working in harmony in a goodwill environment, to resolve these tough issues, we need to do that.

I think today is the beginning of that, and I offer myself personally and my organization and the organizations represented here to work together. Safety is paramount because you see, even though I would like to have this blood being used and my blood being used, I know that any time that I may need blood and I want to be safe.

I appreciate your interest in this. Thank you very much.

DR. HERBERT: May I now comment?

DR. CAPLAN: No, no. Let me do these public comments. Then we will come back to the discussion.

DR. HERBERT: Okay.

DR. CRAWFORD: Hello. I am Roberta Crawford, Iron Overload Diseases. I would like to make three brief points.

Dr. Herbert has already made these points, but I want to make them again. The blood banking is a non-profit life-saving organization. I believe it has a moral obligation to test for the most common genetic disease in our society, and they do not know that they are not testing iron.

Maybe the medical directors and nurses do not do this, but we know, as we have heard from thousands of patients over 20 years and blood banks give out lists of high-iron foods to people with low hemoglobin. They say your iron is low and you need to eat these high-iron foods.

The problem is that some of these people with low hemoglobin, as Dr. Herbert pointed out, are already dying of iron overload, and they have not been diagnosed. They do not even know it. So that is reckless to tell people they have low iron just because they have low hemoglobin.

My last point is they do not know they are measuring iron, but if you are not blood testing, if you do not want to use this blood, you are using it if you do not test the iron because most of the people with iron overload are not diagnosed. They die undiagnosed, and they become blood donors.

Thank you.

DR. COLBURN: Good morning, everyone. Donald Colburn from the National Hemophilia Foundation.

We had a discussion with respect to this issue on our Blood Safety Working Group, and I want you to be aware that the group is against transfusion of this product.

However, I have been educated a great deal this morning to problems that this community has. How shall I say this? They are kind of not unsimilar to our community, and the problems revolve around costs, obviously, and then. where we are concerned, recipient safety, but they are two very separate issues.

We know that this is the most common genetic condition. That is what we have been told this morning.

One of the questions that would have to be asked is, is there any germ-line contamination and is this how we have made this the most common genetic disorder. Those are possibilities, but I think there is a solution here, though, that goes a little bit beyond, and depending upon whose numbers you want to believe that we saw today, it is something that I think a joint task force of the AABB, the RFC, and the ABC could really do something of a community service which would get back to something that Dr. Penner has been talking about and literally open up their collection sites at no charge for therapeutic phlebotomies. I think that might be one solution for this type of thing that the person comes in with a prescription for the draw, hands it to the nurse, and they get the draw done and there is no charge.

Instead of "biohazard" being marked on it, it could be whatever is done with "not to be used for transfusion" on the check-off. That would be a suggestion.

Thank you.

DR. CAPLAN: Let me ask you a question while I have got you up there. You said you were opposed, and safety, I assume was the--

DR. COLBURN: Yes. It sounds like it should not be a problem, but those are challenges that I am not sure all the information is in on. From where I sit and stand, there is a very common-sense thing that says: Why would a person want to receive a blood product from a known person with another problem? That is the simple context of it.

DR. CAPLAN: Any other comments?

Yes.

DR. SCHREIBER: George Schreiber from REDS.

Can I just throw out a couple of numbers from our REDS survey which might be of interest to people? We surveyed roughly 52,650 donors, and we have a survey ongoing now, a little over 90,000. So all of the returns are not in yet, but I am just reporting on what we have so far.

Out of that, we have 206 people, or a rate of 3.9 per 1,000, that identify themselves as being hemochromatosis.

This is an anonymous survey. So the donors are sampled. The survey goes out, and it comes back anonymously.

Just as a point of reference, we also have 1.5 per 1,000 that are polycythemic, and of those people who are hemochromatosis, 48 percent claim that a physician told them to donate blood for treatment versus 79 percent for the polycythemic patients. So you have another group that is creeping in that have probably much more of a health threat than hemochromatosis patients.

Forty-five percent of those with hemochromatosis donated primarily every donation or some donations because of the condition, again, versus 61 percent for those who have polycythemia.

Two other numbers, just as points of reference. 4.85 percent of those with hemochromatosis were repeat reactive on screening, on some screening tests. The way we do it, we do not have the confirmatory test back by the time we sample because it takes too long to get the confirmatory test, but we do mark the questionnaires so we know who is repeat reactive.

That is versus 4.32 percent of the general population. So it does not look like there is very much difference, particularly with the small number. You would not see very much.

Just in contrast, again, polycythemia has 16.25 percent that are repeat reactive. So, just as a point of reference for what we find in this survey, probably by the end of the summer we should have all the results in, in a more detailed analysis, breaking it down by every which way you can imagine.

DR. CAPLAN: Thank you.

Dr. Herbert?

DR. HERBERT: Yes.

Dr. Krikker mentioned this, and the discussion of it is in our Reference 6 in my handout. If you go to Reference 6, you will see it is the petition to the Food and Drug Administration, a docket number, filed September 5, 1996. This is the solution, action requested, that the FDA insert in 21 Code of Federal Regulations, et cetera, after the title "Therapeutic Bleedings" the following words, "Except for blood withdrawn from persons with iron overload disease whose blood should not be stigmatized, but should be labeled solely 'volunteer donor' in accordance with another CFR."

That is what I hope you folks on the committee will recommend approval of by the FDA. It solves the whole problem. It solves the problem that was brought up from the audience. Get rid of that "biohazard" sticker. You should get rid of the "biohazard" sticker, and the way you get rid of it is the FDA adopting our simple petition.

DR. CAPLAN: Good.

I think what we will do is let Jane say something.

[Laughter.]

DR. PILIAVIN: I am going to try one more time to get through to these people who are arguing that there is no problem.

I want to give you a little example of something that is a possibility. Say that there is a person who has been diagnosed with this disease, a fortunate person who is diagnosed early enough so that he has no symptoms, no other diseases, and is told that he has to pay $100 per draw for however long to get this iron out of his system. This person happens to be relatively poor and a closet homosexual. I would like you to think about the possibility that this person is going to decide instead to go to the local Red Cross and have his blood drawn and not bother to tell them that he has this risk.

That is in the most glaring terms what those of us who think about people's motivation are concerned with here. It seems to me that the only solution to this problem, if we want to be able to take advantage of this blood, most of which would be from people who apparently would otherwise be perfectly safe donors, is to do what Dr. Gilcher is more or less suggesting, and that is to have therapeutic draws for people with this illness be free at blood centers. That, therefore, removes the incentive for lying about dangerous health habits that we may have that would make our blood unsuitable for other people, but, otherwise, it is perfectly clear to me that this serves as a dangerous incentive, just like payment for blood did in the past.

DR. CAPLAN: I might have been out of the room when I missed this little pertinent fact, but in the phlebotomies and the blood draws, where is the money going? Who is making the money?

DR. PENNER: The staff. Essentially, you bring somebody in. You have to buy the bags. You have to have a nurse attend. You have to use a room, and that goes into it. So there is a fee tucked on--

DR. CAPLAN: For that.

DR. PENNER: --like everything else.

DR. HERBERT: According to this survey made by U.S. News and World Report in their story, "Throwing Away Good Blood," 2 years ago, the blood banks make $200 million a year from this.

DR. CAPLAN: All right. Let's take a 10-minute break and reassemble, and we will go at this issue with some discussion.

[Recess taken, 9:58 a.m. to 10:28 a.m.]

DR. CAPLAN: I am going to ask as we start this next session that we let the committee talk about this. I am not going to go into the audience unless there is a question of fact or something that one of the committee members wants to address, but then we will open it up a little bit for discussion towards the end of the session.

We have up until about 11:45 to talk about this. Let me set the stage with a couple of facts or opinions or statements that I think are interesting as we try to address the question of do we want to say anything about the use of blood derived from hemochromatosis individuals in the general blood supply, and this is me talking personally now, but hogging the podium again as Chair.

One, it is important to know that other countries have done this. Canada and Sweden, I guess Australia, and South Africa are all using this blood. So there must be some information available either in a general epidemiological way, or maybe there are some studies there. It would be useful for us to ask Steve or the FDA or anybody who might have information in CDC to see what could be found. I think it would be helpful, although I am actually not convinced that we have to await that information because, if there was a massive problem, we would probably know about. It would seem to me that we heard these countries had done this without any overt and obvious difficulty for their recipients.

The second thing we heard was that there is some presence of persons with hemochromatosis in the blood supply now. So what we are talking about is not necessarily allowing a new entry, but an increased entry into the blood supply. That is to say, there are people for various reasons who act as blood donors now.

Third, there is currently a fiscal issue. Jane's case is very striking about what sort of motivation somebody might have to disguise risk factors and take advantage of access.

A fourth question, to put it bluntly, is: If we trust our microbial and screening tests, are they adequate to pick up whatever risk factors would be introduced by these motivational things? Are we really going to put our trust in this safety screening system that we built?

Then, the last issue that might bear on this that I heard was there are different ways to allow people to indicate. This is John's patient's idea that you could do it or just put people in as donors. You could put them in under the system now, where they are paying for their phlebotomies, but designate the blood over, and we could certainly talk some about ideas about paying for phlebotomy, the real cost, so that blood centers could get the blood. People could designate it after it has been drawn, but that you minimize the financial incentive or barrier, depending on how you want to describe it, for people who have to get these bleedings done.

So there is a lot of interesting claims up on the table. Oh, let me add one last one.

There seems to be, which I heard, a fear still that the people providing the blood in this category of person are either diseased or abnormal or perhaps even, I guess I said to Steve, contagious. That, obviously, from a science point of view is just not true. So it is important to understand that this is not communicable, not infectious, not something you are going to give someone else, and the disorder, to the extent that it is seen as a disorder, is only a disorder if it builds up iron in you for a long period of time, and that brief exposure to this is not in any way risky or harmful. So that, even though there is concern, I think it is one that we can acknowledge that people have a concern, but I think it is one we can also address by saying this is not communicable and it is not a grounded concern.

It is obvious that people are right to wonder, but I do not think that should be a reason to provide a hurdle or an obstacle here in terms of whether or not we want to say something about trying to put this blood into the supply.

So, with that, what I would like to do is just open it up for preliminary thoughts or comments by committee members.

DR. GUERRA: Art, if I could just add a point to the last one that you made. Would children constitute a special case in terms of recipients of this blood? I am just not sure in terms of the load.

DR. HOOTS: Are you talking about from a risk standpoint?

DR. GUERRA: Right, in terms of the load.

DR. HOOTS: I do not really think so. If you are talking about either unbound iron or ferritin--

DR. GUERRA: Right.

DR. HOOTS: --and, again, I am speaking entirely off the top of my head and not with any calculated data.

DR. CAPLAN: That is my role. Do not take my role.

[Laughter.]

DR. HOOTS: If you look at ferritin in the modest range and also if you look at an elevation of serum iron, the amount of free iron that you are giving, the amount of bound iron you are giving is so negligible, relative to even what a child would do in the context of an upper respiratory infection where their ferritin as an acute-phase reactant goes up from 60 to 180, just because of the stimulation of the cytokines during an acute inflammatory event.

So I do not foresee that there would be even an infinitesimal risk of the extra amount of iron in there, and as far as I can ascertain from everything we have heard and also from what in dealing with iron states more on the iatrogenic side, like thalassemia, that this part of the threat issue would be virtually negligible.

DR. HAAS: I have just a quick follow-up.

Is that the same type of argument if there are patients receiving multiple units of these red cells?

DR. HOOTS: Yes, exactly.

DR. HAAS: I am thinking of the hemophilia model on the plasma side.

DR. HOOTS: I cannot conceive that even if you fractionated and it would all fractionate to one fraction, that there still would be enough extra unbound or bound iron or protein to cause any untoward events. You are talking about orders and orders of magnitude below the threshold of toxicity here.

DR. CAPLAN: Come up towards the mike. It looks like there is a technical answer or an additional answer.

DR. GRINDON: I am Al Grindon from Atlanta.

It is important to clarify that the blood that we are talking about being donated by these patients and used for recipients is not iron overloaded. Typically, this is during maintenance. So there is no risk of iron transmission.

DR. HOOTS: Even if it were during an overload state, I do not think it would be enough extra iron to add any morbidity whatsoever.

MR. WALSH: As a point of clarification, there seems to be three different groups of people affected here. One is the genotype group that is not symptomatic, has not been diagnosed, predisposed and may never present with disease. Two are those people who are phenotype and symptomatic, being treated for diabetes, cirrhosis, arrhythmia, that would be disqualified, anyway, from a center, I understand. Three would be those who would be perhaps phenotyped and not presenting with symptoms yet that would obviously benefit and that need phlebotomies.

Is that correct?

DR. CAPLAN: Right.

MR. WALSH: So, if we started out by just talking about whether all these groups would be included in a recommendation or one of the groups would be taken out, would that be helpful?

DR. CAPLAN: Sort of, although the first group, the undiagnosed is in. They are in there, anyway.

MR. WALSH: Right.

DR. CAPLAN: People who have overt symptomatology are out because they will go out on the screens. So we are probably talking in the third category for the most part.

DR. AuBUCHON: The blood-collecting agencies have invested an incredible amount of effort and resources to making the blood supply as safe as possible, and the FDA has helped to direct us to make sure that we are asking all the right questions of donors over the last decade, to the point where the blood supply is incredibly safe, as we heard yesterday.

I would find it antithetical that now this committee or anyone else in the Federal Government would then recommend that we take a step that would potentially decrease the safety of the blood supply, particularly when it appears that there is roughly little to be gained in terms of increased availability through that step.

On the other hand, I was very disturbed to hear the data from REDS this morning that was offered during the public comment session that not only are a fair number of hemochromatotics donating blood, but also polycythemics. That group is truly a concern, although no one really thinks that polycythemia or even acute leukemia is transmissible by transfusion. It clearly is a group that we would not want to have in our blood donor line.

I do not really know how we would approach that to "clean up" the blood supply, not that it truly is at any increased risk at the moment, but other than asking one more question, do you have polycythemia or hemochromatosis, which is kind of going in the direct opposite way that several of our speakers would have us go this morning, I am wondering if this is one thing we might consider.

I am offering this just off the top of my head and we probably need some additional thought and some additional calculations, but one approach would be a la the OBI approach, to allow the collection and transfusion of blood from patients with hemochromatosis on recommendation of their physician, as it were.

If the blood center offered free phlebotomy for anyone who needs it for therapeutic purposes, that would then allow us to possibly get the polycythemics to come out of the closet, identify themselves. We would not use their blood, but they would not be at any financial decrement, and it would take away the financial incentive of the situation that Jane was mentioning. At the moment, it bothers me very greatly.

DR. CAPLAN: Ed?

DR. GOMPERTS: The issue of labeling a unit of blood as hemochromatosis came up, clearly this is simply one way through this. Everything that I have heard and what I know about would indicate that there is no hazard from the point of view of transfusing a unit of blood from a patient with hemochromatosis.

Under those circumstances, what is the hesitation of actually putting that label on?

DR. AuBUCHON: Maybe someone transfuses directly could answer that, but I would think that anything that distinguishes a unit in any way would decrease a patient's willingness to receive that unit, and it would require the transfusing physician or nurse to spend a lot of time explaining to the patient exactly what hemochromatosis was and why this unit was not really at any increased risk.

I think we would have to be willing to accept those units into the blood supply without label or they will not get used. Logistically, it is too difficult.

DR. GOMPERTS: Is it just an education issue?

DR. AuBUCHON: Yes, but you would have to educate each and every recipient, and that is just too much time and effort in today's modern fast-paced health care, frankly.

DR. GOMPERTS: Maybe, maybe not.

DR. CAPLAN: One other issue about labeling is, do you label it because you are not sure that it is at risk or because you know it is at risk, and what I have not heard yet, and I wanted Jim to say something else or Jane, too, if she wanted to add to her story, if we are screening effectively and we heard the REDS data about what the incidence of infectious agents and risk factors were, if they were not different from the average population and we have got some number of people entering in, what is the risk?

I understand that there might be a risk or there could be a risk or we might attract in people who are not there yet, but before I throw a label on something, I want to know is it more risky or isn't it. I do not want to label it just because it might be. That would open us up to a lot more labels.

DR. AuBUCHON: The only risk that I see is the infectious disease risk associated with the monetary incentive to donate and not disclose the presence of hemochromatosis.

I agree with Keith. I do not think the blood itself is of any different quality.

DR. McCURDY: Independent Counsel think that there is an analogy here that might be useful. I do not think there is any question. Maybe there is, but there is very little question that so-called "paid donor blood" carries a higher marker rate and a higher seroconversion rate than so-called "volunteer donor blood," even with all of the incentives that have been put in it yesterday.

It has been assumed that this is because paid donors do not tell the truth in their history. It is my understanding that the amount of money that is provided to paid donors, mostly plasma phoresis donors these days, is relatively small compared to the overall economy at the present time.

We have no data that indicates that they do or do not tell the truth more often. There has never been, to my knowledge, a questionnaire sent out for anonymous return, like the REDS study, for paid donors.

I do not think we have really any good idea at all as to why paid donors are more risky, but they seem to be.

I am not aware of any study other than what George Schreiber presented earlier that gives marker rates in patients with hemochromatosis.

I have asked the question on one or two occasions about whether anybody screened either a random sample or consecutive sample for, say, hep C or one of the other or all of the other markers, and I do not know that such data exists.

It might not be too difficult to get that kind of data if it was desirable.

DR. CAPLAN: Other comments?

DR. HERBERT: Dr. Ray Gambino of Quest Laboratories reviewed 1.5 million sequential blood samples sent to his lab from physicians all over the metropolitan area for routine standard first testing for glucose, et cetera. He on his own on all of them did transferrin studies. He found in 1.5 million samples that 12 percent were heterozygous for hemochromatosis, and these are just normal healthy people going for routine exams in the New York metropolitan area. He published that in 1992.

DR. McCURDY: That is not the type of screening I was talking about, Victor. I was talking about screening for the infectious disease markers that are generally done in blood banks. They are done on a selected group of healthy normal individuals, who have passed a history screen for being blood donors. There is a small percentage of those. I do not have it. Rick Davey or somebody might have it, but there is a small proportion of those who have infectious disease markers without having any knowledge of it themselves. That is what I was talking about.

DR. CAPLAN: Let me put this point another way, and I am just tossing it up there for people to think about and answer. Then I will go back to John.

If you sampled, and there is no shortage of material to sample, people who have been through phlebotomy with a condition and saw what the different markers were in the pool, the fear is, though, that if you make this free or minimize the cost of phlebotomy, you will bring forward more at-risk people who will come in and use this, attracted by both the incentive than might have before. So you might get a skew in what the donation pool looks like from what the baseline pool is.

I guess my nuts-and-bolt question, to come back to it again, if you put an incentive forward and we cannot get the fiscal barrier removed, which I am not sure we cannot--that might be possible, but if we could not, would the testing, even with the behavioral distortions, still minimize the risk? That is what I have not heard anybody answer for me yet.

If you trust the testing, microbial level, that we have got now, then even if you had somebody stepping forward out of Jane's scenario, would you really raise the risk background incidence that much?

DR. PILIAVIN: Is everybody forgetting about the window period? Isn't that why we are so careful about the screening that we do? There is no reason to think that people with hemochromatosis are likely to have any cleaner lifestyles than all of the rest of us, especially since there seem to be so many of them.

[Laughter.]

DR. PILIAVIN: They are probably just like the rest of us. So it is the window period that the questioning is designed to solve.

We know it does not solve it completely, and partly, that is because other people lie, too, or simply have no clue that they are at risk because they did not know the person they were having sex with had used I.V. drugs, for example.

I just cannot imagine that an incentive, a major monetary incentive, is not going to have the same influence on these folks as it in the past has had on paid donors.

In regard to whether there is any evidence that paid donors are more dangerous, sure, nobody has done a study that asked them these same questions like the REDS study, but I would take the three-fold increase in markers as pretty good evidence.

DR. PENNER: To get back just to the labeling briefly, I agree with Dr. AuBuchon's comments that the minute we put a label on, then we have to go through the institutional review board. We have to then touch bases with the patient to see if he will accept it. They will not know about what hemochromatosis is, and the harangue to just get use of that blood would just not pay for it. That was, I think, our concerns all along.

Since we have already established, I think fairly well, that there is nothing wrong with the blood, it represents not a risk to the individual, why not address the first part of the problem and try to exclude hemochromatosis from that initial act of having to label? This would seem to me to be an approach where you exclude only that.

We do not add other labels because of other diseases. So why should we do for hemochromatosis? The only reason we are adding the label is because we want to exclude the polycythemics who are going to benefit from having their blood removed at this point.

DR. CAPLAN: Yes.

MS. O'CONNOR: Speaking as a consumer or somebody who has been transfused and so on, I think the patients have a right to know if this blood was collected from a therapeutic draw. So I would be opposed to putting a label on there that says this was obtained from a therapeutic draw.

When you guys test a unit of blood to see if it is C&B-negative or C&B-positive and you have actually tested it and found out that it is positive, the sticker goes on a unit of blood and goes on the floor. It does not go through an institutional review board or anything of that nature in the facility.

For the most part, patients do not even notice it, but for that astute patient who wants to know what they are receiving, I think it is important that that information be there.

DR. PENNER: If we addressed the other part of that by allowing it be volunteered by the patient who has given that blood, wouldn't that represent the same sort of response we get for any volunteer who walks in who says you can either use my blood or not? He is coming in because he is saying you can use my blood. The hemochromatosis patient can say the same thing and say, "I don't want you to use my blood. I want to take it home and put it on the tulips to keep the rabbits from eating them."

Enough said.

DR. CAPLAN: Let me go to Rich.

DR. DAVEY: If we are going to talk about labeling, it has to be linked with the incentive issue. We cannot talk about removing a label without removing the incentive, in my opinion. So, if we do not, we cannot unlink them. If the label goes, the incentive has to go, and that means we may have to think about somehow offering this as a service, or perhaps letting the donor, as you suggest, John, decide whether or not he or she has risk factors, but not involving the incentive issue to cloud their thinking. So labeling has to go with incentive.

DR. CAPLAN: Steve and I were just noodling along. Do you want to read this? This is something to think about.

DR. NIGHTINGALE: This would be a paraphrase of Dr. AuBuchon's previous statement: "A blood center should be able to use blood from patients with hemochromatosis without any restriction if that blood center offered free phlebotomy for all patients with hemochromatosis." That is not precisely what Dr. AuBuchon said. I believe that Dr. AuBuchon said "any therapeutic purpose," but that would be the notion that seems to be floating around the table at this time.

DR. HOOTS: I think one caveat that I thought Jim astute put in that that does not address is instead of just saying "with hemochromatosis," would you want to add "erythrocytosis" to that? That is what you want to screen out. So, if you use hemochromatosis or erythrocytosis, then it would give the opportunity and you would be able to screen that out.

DR. PENNER: For instance, for polycythemics?

DR. HOOTS: I was using "erythrocytosis" as a generic term to cover P-vera and everything else.

DR. PENNER: Oh, you wanted to include "erythrocytosis."

DR. HOOTS: Yes, because I thought I understood Jim to say that he would like those people to get free--

DR. PENNER: No. Just the hemochromatosis.

DR. AuBUCHON: I would think it would be essential that free phlebotomy be offered to any person who has it prescribed. The polycythemics would not have their blood transfused.

DR. HOOTS: Right, exactly.

DR. McCURDY: Could I ask for a point of clarification from something that Rick Davey said earlier?

Do all of the donors in the Red Cross system get either hemoglobins or hematocrits? They do not use the copper sulfate technique anymore.

DR. DAVEY: No, that is not correct. It varies from region to region in the Red Cross. Copper sulfate is used as a rough screen for hemoglobin.

DR. McCURDY: My thought was that if you really did a hemoglobin--and that measures density, not hemoglobin, but if you really did a hemoglobin or a hematocrit, then you would screen out those with erythrocytosis because they would be too high.

As far as I know, there are few, if any, blood banks who do that routinely. Maybe, Ron, you do.

DR. AuBUCHON: I am not sure that would solve the problem, Paul, because after these polycythemics have been in therapy for a while, their hemoglobin comes down to a range that is indistinguishable from mine.

DR. McCURDY: Yes, that is true.

DR. AuBUCHON: If I could return to the proposal that I guess I made, though I was not sure I was making one, I would like to hear more from the users on this committee. We heard Carolyn's comments. We have talked a lot in past meetings about patients' rights to know, and obviously, patients are very interested in having enough blood and they are very interested in having safe blood, but this is another patient-right-to-know issue.

In the system, as we are talking about it now, as a proposal, there would be no labeling of hemochromatosis units. They would just go into the inventory like any other unit.

Clearly, some of the units that we are transfusing right now already are from hemochromatosis patients, and we just do not know it. There would be a few more with this approach. What do the recipients think about this?

MR. WALSH: First of all, I think it is appropriate if you do a phlebotomy, it has to be on a script basis, and you would have to know why the physician was prescribing a phlebotomy, first and foremost.

If somebody was symptomatic, it would be red-flagged by the center, wouldn't it, for the indications we have already talked about? If somebody was not symptomatic, there would be no reason.

DR. DAVEY: They would have to go through a donor screen, like everyone else.

DR. CAPLAN: I take that as without reason to see that there is risk, you are saying what?

MR. WALSH: I would say in regards to free phlebotomy, absolutely, but with the physician's prescription and a reason for it to the center, which would theoretically red-flag a lot of the issues that we would be concerned with.

DR. CAPLAN: Yes, Larry.

MR. ALLEN: I am just wondering, if we are talking about labeling these units, would it also be necessary, then, at that point to label all blood to say that it might contain this?

DR. CAPLAN: I can only remind the committee that we went down this road a little bit with the CJD issue, talking about endemic and being in the supply and do we want to have special labels for something that might be there anyway. We talked about that a little bit.

I am personally moved quite a bit by the fact that I think there is some material in the blood supply now. So I take that question pretty seriously about trying to label. How would you label? This could be there, anyway. You could get big labels. It would be good for the label industry.

DR. PENNER: Use a broad label as opposed to a special label.

DR. HOOTS: I agree with you. I think it opens a can of worms. Here, I think the whole idea is to try to dis-incentivize the donation so that you can say that this donor population is no different from the regular population, and if you then label it and then later on we decide to label for other things, then essentially you have undone what you intended to do, which is you have put them in a labeled category in the same way that we might have done with CJD and chose not to do. I think that would be a mistake.

DR. HOOTS: I guess the question that I would like to ask is: What is labeled on blood right now?

DR. HERBERT: Hemochromatosis blood is labeled "biohazard."

MS. GUSTAFSON: By regulation, only the statement that I had on the overhead, which is that it was a therapeutic bleeding and the disease that caused the disease state for which the therapeutic bleeding was done.

Did you want general labeling?

DR. CAPLAN: Sure. We have time. Go ahead.

MS. GUSTAFSON: Do you want to go home this afternoon?

[Laughter.]

MS. GUSTAFSON: In the U.S. now, there is a uniform label that is used by the vast majority of blood centers, and it is accompanied by a circulate of information that gives more information about the blood component in terms of the transfusion state.

Actually, the infectious disease testing is in the circular of information. It is assumed that if a unit is labeled and released into an inventory that it would be negative for all of the infectious disease tests.

It has a donor number, and there is an identification of where the unit was collected. There is, obviously, the expiration date. There is the ABA and RH. There is the blood component's name and the statement that if it is for transfusion from a volunteer paid donor, and that is in the same prominence as the proper name of the product.

There are warning statements that are on now, which every unit of blood does tell anyone who reads the container that it may transmit an infectious agent. So that is on there now as a broad general statement.

Bruce, have I missed anything? Special typing. If there were special typings, like if it was chel-negative or any of those types, CMV negative, it is listed on the unit.

DR. EPSTEIN: Anticoagulant.

MS. GUSTAFSON: Oh, yes, the anticoagulant.

DR. EPSTEIN: The expiration date.

MS. GUSTAFSON: I mentioned the expiration date. Any additives that are added are listed on the container label.

DR. CAPLAN: Any unused blood, just to be clear about this, if you are going to toss it, there is a biohazard stick that is slapped on it, anyway, right?

MS. GUSTAFSON: No. A biohazard is only added if in fact there is an infectious agent.

DR. CAPLAN: No. I mean when you are about to throw it out.

MS. GUSTAFSON: No, no. If it is expired, you can tell it from the expiration date.

For some reason, if it was decided that it should be destroyed during processing, for whatever reason, generally an interim label "not for transfusion" would be added to it.

DR. EPSTEIN: I think it is important to clarify because there is misunderstanding.

The regulations do not require a biohazard label in the face of a therapeutic bleeding, and that would apply to hemochromatosis.

What is required is a statement that it was a therapeutic bleeding and the diagnosis that was the basis for the therapeutic bleeding. That is different than a biohazard label.

The problem, and I hear this very clearly, is that there is the implication in people's minds that that has the same meaning as a biohazard because it has been distinguished. I understand that, but, technically, it is not a biohazard label. That is not what is required or intended.

DR. HERBERT: Does your blood bank label "biohazard"? I have walked into 14 blood banks all over the U.S., and every one of them labels hemochromatosis blood "biohazard."

DR. GILCHER: No, we do not because we test every unit.

If the unit is untested, then it does have to have a "biohazard" label. We do test every unit. So we do not have to put the "biohazard" label on. I think that is the difference, where the blood center does not test it.

DR. HERBERT: You are one of the few.

MS. JONES: I think you need to be very clear when you are talking about a "biohazard" label on a unit of blood. Anything that is not going to be used for transfusion has "biohazard."

DR. HERBERT: No. It is labeled "biohazard."

MS. JONES: It is labeled "biohazard" because it is not going to go for anything further. It is not specifically directed at the hemochromatosis.

If somebody else comes in and does not want their blood used for transfusion, we put a "biohazard" label on, if they check that. It is not labeled. It is not tested.

DR. HERBERT: We d not require you to do that.

MS. JONES: I know, but it is just to keep it separate from everything else so that the techs do not make a mistake and put it into the transfusion.

DR. CAPLAN: That was my point that the labeling extended beyond hemochromatosis.

MS. JONES: Yes.

DR. CAPLAN: That is what I was trying to say. The trigger seems to be is it untested, not is it for disposal, although I think we slap "biohazard" stickers on everything at the University of Pennsylvania that we are about to throw out so that people do not take them home.

DR. HERBERT: The AABB solved this problem. It has not been mentioned because the AABB just added PCR, and when you add PCR, you pick up right away if it is infected with hepatitis C or infected with AIDS months before the test for AIDS comes back positive or the test for hepatitis C comes back positive. So there is no problem.

DR. CAPLAN: Jay?

DR. EPSTEIN: Two comments. First, I think that Jim AuBuchon hit the nail on the head, which is that we should not be talking about removing the label without removing the incentive.

At that point, if we think that hemochromatosis therapeutic phlebotomies are different than other therapeutic phlebotomies because the product has no known intrinsic risk, I think we are home-free. We have all been saying that the only issue here is the possibility of an undue incentive to a person who has obligate need for phlebotomy and knows it, above and beyond the fact that there can be a dramatic cost, especially in some cases an uninsured cost.

So, to the extent that it is a premise that there may be an undue incentive, one has to address the undue incentive in order to talk about changing the labeling. So we cannot put the cart before the horse. We understand labeling is part of the issue. There is actually no resistance by the agency to consider that as a remedy if we have a resolution to the incentive problem.

The second point that I would make is that we all realize that there are technologies that are making blood products safer and safer. These include improvements in donor screening and testing. They also will, we think at some point, include viral inactivation or pathogen inactivation strategies for cellular blood components, but we are not quite there yet, and we are still in an environment where we have residual risks and where we are very concerned about those residual risks, even though they are low.

We know from studies that have been presented and at these advisory committee meetings that the residual risks from infectious diseases are driven by window-period infections for which we do not have fully effective testing.

Yes, NAT further closes the window. There is a scientific debate whether it will completely close the window for HIV and HCV. Certainly, at this stage, it will not for HBV, and it is not clear that it will completely close the windows for HIV and HCV while we are doing mini-pool testing because when you pool samples prior to screening, there is a dilution factor.

Although we know that the viral titers ramp up very quickly, you have still got that dilution that you are dealing with when you do the test.

So there will still be a window period. There will still be a residual risk. It will be smaller, but all the more reason not to let through knowingly any additional window-period cases because you have got to realize that with these very low residual risks, the ones you let through may become the largest proportion of the ones that exist. The goal here is to lower that risk as much as possible.

I think at the present time, we still see donor screening, donor testing, and processing controls that reduce pathogens as a set of interacting safeguards, each of which is independently important. We do not want to compromise any of those safeguards, and that is the current outlook.

Perhaps in 10 years, if we think that we can truly sterilize pathogens in blood, it will be a different world, but we are not there yet.

DR. CAPLAN: Let's try another little thought. We are hedging about trying to separate the incentive from the actual blood product.

DR. NIGHTINGALE: The addendum was that the Health Care Financing Administration should consider therapeutic phlebotomy for individuals with hemochromatosis as a cost of ensuring and adequate and safe blood supply and reimburse providers accordingly.

DR. CAPLAN: In other words, what we are trying to do is get them to pay for phlebotomy. That should minimize the incentive.

DR. DAVEY: You may just want to make sure you include Jim's point that we want to include polycythemia in general or erythrocytosis in general, and make sure that P-vera is included in this.

DR. CAPLAN: Right.

I understand what the issue is about there is another group of people that are in there that maybe should not be.

I was actually just holding that on the side for a second to try and go at the issue of breaking financial incentive away from the hemochromatosis donor.

I do understand the other concern that is up on the table. That probably gets written as an amendment.

To go to Jane's point, if you want to shut down the window infections and you are afraid of incentivizing too strongly, if we are going to go in this direction, we need a recommendation that says we want the incentive minimized and then you can open the door.

MS. O'CONNOR: I missed who was going to pay for it.

DR. CAPLAN: We will, all of us.

MS. O'CONNOR: This committee?

DR. CAPLAN: Not individually.

MS. O'CONNOR: Did you specify?

DR. CAPLAN: This is a HCFA "ought to."

MS. O'CONNOR: HCFA, okay.

DR. SNYDER: Are you talking about for everybody? Are you talking Medicare, or are you talking Medicaid? Everybody that cannot do, you have got to look at what their congressional mandates are. You may be into legislation again. You really need to talk to somebody from HCFA about that.

DR. McCURDY: Art?

DR. CAPLAN: Yes, Paul.

DR. McCURDY: I do not feel that I have a good handle on what various insurance companies, including HCFA, will or will not pay for therapeutic phlebotomy. It is a legitimate treatment for a legitimate disease, and I have no understanding as to why they will not do that.

The uninsured or under-insured are a different problem than this. It could be taken care of by community service by the blood banks, but it is really a different problem.

I have heard from multiple sources through the last couple of years all sorts of different things about what the price of phlebotomy is and what insurance companies will and will not do, and I must confess, I am totally confused.

DR. CAPLAN: Again, I am not tossing out resolutions yet, but what I am trying to say is if we say that we recommend to the Surgeon General and the Department that the pool of donors is a valuable resource to deal with the supply issue and they must take steps to minimize the fiscal incentives, whether it is dealing with the managed care people or getting the blood banks to eat one more cost or talking to 11,000 HCFA bureaucrats about what they want to do, it seems to me our mission is to say we have to make sure that there is no incentive to disclose or hide any behavioral risk factor.

If the agency wants to go get this blood, which is pretty good quality blood, it must then find out why they do not pay for therapeutic phlebotomies or what to do about poor people. It has to minimize the fiscal hurdle or take away the incentive.

That is not bad because it does not make us solve it. It just tells them what they got to do if we think that that is blood that they ought to go get. There must be a coherent resolution in there somewhere.

Let me remind the committee. All we are trying to do, we do not have to micromanage the health care system. That is the good news. All we have to do is tell the Government to go minimize the financial incentive, take it to zero on why people come in for phlebotomy. You will then open up what we now believe to be a good additional source of blood to enter into the system.

That seems to be what I keep hearing. If you take the incentive out, then no one has objection to putting the blood in, even not labeled, so to speak, but the incentive must go out. Otherwise, there is risk from the window infection.

DR. AuBUCHON: You noted that several countries, Canada, Australia, Sweden I think you mentioned, used blood from hemochromatosis patients. I would note that all of those countries have socialized health care, and patients do not pay for things like phlebotomy.

DR. CAPLAN: I know. I know.

Then, another idea is they should finance tickets to Sweden for all people with hemochromatosis. They will bleed them there.

[Laughter.]

MR. ALEXANDER: Well, this is very interesting. I will go to Sweden. I would love to. I have never been there.

Randy Alexander, Iron Disorders Institute. I have hemochromatosis.

Something that is coming up here, and I thank you for the opportunity to speak, is that talking about costs in the insurance industry, talking about who is going to pay for this, the insurance industry does not understand this. The public does not understand hemochromatosis.

I respect your wanting to know if I am going to be transfused, what is hemochromatosis. Invariably, when we talk to patients or talk to people and they say what is hemochromatosis and then we educate them about it, it does not take forever to do it. We give them the very basics, and when they understand it, we tell them the treatment is blood. Then, they understand that they are not using the blood. The question is why. This is such a phenomenal resource that is being wasted.

We have had a tremendous record of success with the life insurance side of insurance industry. State Farm. We got to the head underwriter of State Farm. When that underwriter, that decision-maker, worked with us and understood what this was, they changed the whole underwriting procedure for State Farm. So has Indianapolis Life Insurance. So has TransAmerica.

So, when the public is aware of what this is, when the insurance industry is aware of what this is, and we are talking about costs, we all need to work together and educate the public and the insurance industry because, working together on this, it is going to help alleviate some of these questions and some of these problems we are faced with, especially when we talk about cost and talk about informed consent.

Also, if we work with HCFA and are able to make some positive changes in that area, as I understand it, does not the blood-banking industry follow the HCFA guidelines?

DR. CAPLAN: John?

DR. PENNER: This is going to put us into a very strange dichotomy in that the blood bank would then be billing the Government for doing a phlebotomy, which it then will use its blood and recover additional cost from the blood that it sends out.

In other words, there is something that would be very unique suddenly that all of the blood banks would be able to charge for doing their blood drawing.

We bill the insurance companies now in our clinic, and I am sure you do in clinics over in Sinai, and are reimbursed for that, but then the blood is disposed of, but now you have got a commodity. So you are kind of double-dipping in a way.

So there is something difficult here that has to be worked out. I think the blood banks could handle it.

DR. CAPLAN: Paul?

DR. HAAS: Maybe I am thinking about something that is painfully obvious, but, Art, when you were making your statement, you talked about the incentive problem and then that needs to be dealt with before we deal with the issue of the blood itself. I think that is really distinct. We ought to just talk about the incentive problem, period, and resolve that. Then, we still have the issue of should we use the blood, but I think the incentive problem crosses a lot of issues as I think Jim keeps saying.

Maybe that takes one item of the complication out of the issue. Just look at the question of whether or not we are bringing people into the system because we have the wrong incentives.

DR. CAPLAN: Before taking a few more comments, I was trying to do another on-the-spot pseudo-resolution or recommendation, and this one started off saying blood from persons with hemochromatosis is a potential valuable source to augment the diminishing supply. We could just say that. It is not the blood, but it cannot be used unless financial incentives to seek donation are eliminated. HHS should use this resource. It must, however, create policies that eliminate incentives to seek phlebotomies for the purposes of donation.

So that resolution is trying to drive for the blood is okay, it is your problem to cut the incentives out of the picture. It does not get to what about this other subpopulation of risky people who may be around, and I am not adverse to talking about that, but in the spirit of what Paul said, it seems to me we can say, if we want to say it, this source is okay. The blood is okay. It ought to be used to augment, but you have a problem, Government. You must device a policy to minimize the incentives to come in and be phlebotomized solely for donation purposes. That is your challenge.

We do not have to get into what John talked about. That is going to be for them to hassle out.

DR. DAVEY: I would just put it a little bit backwards. We want to minimize the incentive before we can be sure that the blood is okay.

DR. CAPLAN: Yes.

DR. DAVEY: Then, we can move on to the labeling issue, which is linked, but should follow the incentive issue.

MS. DUCA: My name is Anita Duca. I am from the American Red Cross.

The HCFA-proposed reimbursement regulation for outpatient services included a code for therapeutic phlebotomy, and the HCFA cap that was proposed was $22. I just wanted to give you that information.

DR. CAPLAN: I am sure business is rushing forward to capture that.

[Laughter.]

DR. GRINDON: I am Al Grindon from Atlanta. I wanted to make a couple of comments.

First of all, I would agree completely that eliminating the incentive is exactly what ought to be done, and that is what Dr. Gilcher has done. That is what we have done at the Red Cross in Atlanta.

I think before you move ahead from that, you need to ask now whether you feel that test-negative blood is okay. We have considered that for autologous donors and decided that it is not, that test-negative blood is not okay.

I think we need to make sure that we are comparing rates with donors and not with the normal population because the donor population has a much lower marker frequency than normal. So you have got to compare it to the other donor population.

You have to be careful to use Dr. Busch's concern and make sure you are looking not at prevalence. There is prevalence data. Other people have data on the prevalence of markers in patients with hemochromatosis.

The Barton data I showed you has some of that data, but that is not enough. What you really want is the incidence data, and that is something we just do not know, whether the incidence is any different. I would guess that it is not, but that is data that is not available yet before you say blindly, "Well, we know this blood is just as safe." We think it is, but we do not know that.

I might say, finally, that a study design would be very difficult. It is easy if there is a difference. You can do that with relatively small numbers of people. If there is not a difference, and that is what we think is going to be the case with hemochromatosis, then the question for people like Dr. Epstein is how low a Type 2 error are you willing to accept. That is a very complex issue to do that.

DR. CAPLAN: It does seem to me, by the way, following up that point, if the committee decided to go toward steps to minimize incentives and access the blood, then another thing we might suggest is to have a study done of just this type before we go "whole hog," so to speak. That would not be terribly onerous to do.

DR. AuBUCHON: If I may object to your last statement--

DR. CAPLAN: I mean not onerous in setting up the study.

DR. AuBUCHON: --that would be a huge study. You would be looking at multiple years, multiple millions of dollars. I would love to have it, but my guess is we are going to have to fly blind on this one.

DR. CAPLAN: For a while.

I mean, we could at least do surveillance for a while.

How are you doing over there on that?

DR. McMURTRY: I am coming along.

DR. CAPLAN: Struggling? Okay. Then, we will keep talking for a while.

[Laughter.]

DR. CAPLAN: Yes, Keith.

DR. HOOTS: In terms of juste that whole statistical thing, in light of what Mike said yesterday about the upswing and the incidence, would it be shortened a little bit if since these might be frequent donors, we get a lot more incidence data more quickly than we would, say, in a routine donor population?

I know it still would be expensive. It would still be protractive, but it might shorten it relative to what we are used to for just the general population.

DR. CAPLAN: Other comments?

I think what Mac is going to put up there is the statement that I started. By the way, sometimes it sounds to the committee perhaps like I am trying to push or force a position here. I am not. I am just trying to phrase up what I thought I heard said about the blood being of value and trying to see whether we want to move in this direction toward recommending something about the minimization of the financial incentives.

If we do not want to do it today, we do not have to, and we do not have to do it at all. I am just tossing something out for your deliberation, not anything else.

We do not get higher grades if we answer this today than if we did the next time or some other time.

DR. HERBERT: The sharp difference is that polycythemia is a malignant disorder. Every polycythemia patient eventually dies of acute myelogenous leukemia, and as you know, at Sinai, we for 35 years have been running an NIH study at a number of institutions, including our own, showing that every polycythemia patient eventually dies of acute myelogenous leukemia. So we are dealing with a virus.

DR. CAPLAN: All right. So what does that, then, say? Blood for persons with hemochromatosis is a potentially valuable--

DR. NIGHTINGALE: Resource.

DR. CAPLAN: --resource--I wrote it, and I cannot even figure it out--to augment the diminishing supply, but it cannot be used unless financial incentives to seek donations are eliminated. DHHS should use this resource. It must create policies that eliminate incentives to seek phlebotomies for purposes of donation.

So we want to move into a motion mode?

DR. AuBUCHON: Should it say "eliminate incentives to seek donation for purpose of phlebotomy"?

DR. CAPLAN: Oh, yes. You are right. "Seek donation." In that last sentence, yes.

DR. DAVEY: I would add that DHHS should consider using this resource, at this point.

DR. HOOTS: That is what I was going to say, too.

DR. HERBERT: May I mention that hemochromatosis patients who do not know it, which is 99 out of 100 hemochromatosis patients, are more frequent blood donors because every hemochromatosis patient who does not know it gets a perceptible lift when he/she gives a unit of blood. They say, "Hey, this is a great thing." So they become frequent blood donors because, after every donation, they get a perceptible lift, a perceptible euphoria, perceptibly feeling better.

DR. CAPLAN: Let's try "should consider using this resource," and then we flipped around the "donations for phlebotomy." It was right. That is what I meant to write down.

MR. ALLEN: Art, would it make sense to add that we acknowledge that these donors' blood is already in the blood supply?

DR. CAPLAN: No. It will be in the record of this discussion as to how we got to that resolution, the point that Dr. Herbert made. They are in there as already there. So I do not think we need it as a preamble. They will be in the text.

DR. HERBERT: Oh, go ahead and put it in.

DR. SNYDER: Art, I have got a question for the blood-banking people. Out of patients who have donated in the past or donors who have donated in the past, is there any data that you have or anecdotal data that shows subsequently they are diagnosed? Does that make sense?

DR. DAVEY: I do not think we have those data at hand.

Do you know of any?

DR. AuBUCHON: No, I do not know of any such data.

DR. SNYDER: The only reason I am asking is that would give you a heck of a marker as to how much is in the supply right now.

MR. ALLEN: There was not even acknowledgement that it was in there until someone brought it up. We hear from the blood centers.

DR. SNYDER: I'm sorry?

MR. ALLEN: We did not hear that these donations were already in the blood supply from the blood centers. We heard this from one of the speakers.

DR. SNYDER: Right, I understand that, but former donors who are subsequently diagnosed, do they have come back and tell you they cannot donate anymore?

DR. AuBUCHON: To my knowledge, that information is not known, and George Schreiber's data from Westat comes the closest to that today, but that does not directly address your question.

DR. CAPLAN: Mary?

DR. CHAMBERLAND: Perhaps we need to modify the statement that begins "DHHS should consider using this resource."

I mean, technically speaking, the Department allows use of this resource. It is linked to the labeling requirement, but I am not sure it is actually correct as written because right now we are allowing use of blood collected from persons with hemochromatosis.

I know what the intent is, but I just do not think it is technically correct.

Would you consider Jane's suggestion of "removing barriers"?

DR. HERBERT: She is destigmatizing it.

DR. CHAMBERLAND: "Should consider removing barriers"? "Removing barriers for the use of this resource"?

DR. PENNER: Yes. That would give them the flexibility to alter that. Yes, that is good.

DR. CHAMBERLAND: Yes. It is a little more general.

DR. HERBERT: I do not understand why you do not use the direct word and beat around the bush. The direct word is "destigmatizing."

DR. CAPLAN: If you are going to be a subset, stigma is one problem. There are some others. There are other barriers. That is why.

DR. KUHN: Just a definition, is barrier also considering labeling?

DR. CAPLAN: Yes, sure.

DR. AuBUCHON: Mr. Chairman, maybe I was not listening, but I never really heard a yes or a no from our consumer representatives on this committee.

I guess by them not saying no, they are saying this is okay, but is this really all right with you guys?

DR. KUHN: Going back to your question, Karen was so nice to draw me a nice label h ere, and on it, I see that there are words such as "disclaimer," "it may contain infectious agents," even to the point of saying "CMV" on it at times.

I am wondering if as a consumer who would be receiving this that hemochromatosis should be mentioned.

I know I heard the argument saying that we have to explain that to a patient. If I am looking at this and I say it may contain infectious agents, the first question I am going to have is what are those infectious agents and how are they going to affect me.

Again, an educated consumer is going to say how is it going to affect my immune system, what kind of reactions may I have to this, and again, I guess my concern is I am still not comfortable with the point of removing as a label, a barrier meaning a label. I am not really comfortable yet of removing a label.

It gives that patient the right to know information, and I know it might take another 5 minutes to explain to a patient what does this mean. I know if you were to explain it to me and if I was going to receive this, I do not think I would have any problems at this point, especially after I had some of my other questions answered graciously by Keith Hoots. So it would not affect someone who is immune-compromised.

I still think that we have a responsibility to give patients the right to know, especially if we are going to make a step such as this a policy issue, which we are going to enter something else into the blood supply.

MS. O'CONNOR: I think it would be appropriate to look at other countries' experience, in this case, because we are talking about removing the incentive.

If Sweden has in fact used this successfully for 30 years, I do not know how long the other countries have, but that would certainly make me as a patient feel fairly comfortable that this is an okay thing to do.

DR. KUHN: Do we know what is on the labels?

DR. CAPLAN: We know the Swedish would not say something they cannot defend.

[Laughter.]

DR. CAPLAN: That is an old Minnesota-ism.

DR. HAAS: I am one family member removed from being a consumer, but I guess the reason why I brought up my earlier point is that I think there is an important issue about the incentives, and although as I sit here today and listen to a lot of the information, I am becoming more comfortable with the quality of this, I still think there are other issues out there that need to be addressed on that level.

I am a little uncomfortable dealing with the quality issue today. I am more comfortable dealing with this incentive stuff, and if the question to me is would I be willing to vote for something that talks about taking out this negative incentive, the answer is yes.

Am I willing to say that this is a product that ought to get into the bloodstream? I am leaning in that direction, but I would probably be uncomfortable voting for it at this stage.

DR. DAVEY: Just to comment a little bit more on that point, I think the idea is if we can remove incentives for this condition that we will be able to enter these people as volunteer blood donors. So there really would not be any particular right to know because these people would be considered as totally qualified volunteer blood donors. If we get rid of the incentives and approach it that way, then there will be no specific right to know about this illness in my view.

DR. CAPLAN: I will just add, my worry about the label issue is that you do not want to label blood from a group if you do not have any reason to think that there is risk because labeling is tricky.

The second issue with labeling is do you want to label for comparative risk. There may be other groups in there now who have a greater risk than this that we are not labeling saying this blood came from X, Y, or Z group.

So, again, I am not against informed consent, but I want to make sure that if you are going to label it, you are going to label it with something that tells you there is risk there.

MR. ALLEN: It is my understanding that all blood now has that generic label on it that you and Carolyn are talking about. That is my understanding. All blood already has that label on it.

So, unless we are going to get specific with each possible disorder that comes with that label, then I do not see the need to go any further with this.

DR. CAPLAN: Jim is right, however, to ask if there is anyone else from the consumer end who wants to talk about this labeling issue. Is there anyone else on our committee that did not say something?

[No response.]

DR. CAPLAN: Okay. Please identify yourself again.

DR. GRINDON: Grindon from Atlanta.

I do not mean to be presumptuous, but suggesting a pathway for this group, it seems to me the first thing is to remove the financial incentive from all donors because there is risk for those with polycythemias we talked about. That is clear. That is the first thing. The second thing is to collect data, and the third thing is to deal with the labeling issue.

I think you have got to get the data before you deal with the labeling, and we do not have that data, but you can remove the financial incentive today and everyone will be better off.

MR. SUBIAN: Sal Subian, America's Blood Centers, New York Blood Center.

On the issue that you just raised--and I agree entirely, and I think it is so important that we should at least state about labels--we know that certain groups, for instance people from Asia, have a higher incidence of hepatitis. Should we label units with the ethnic origin? I think this would be unacceptable.

So I think that the label has to be the last one, and I agree with what Al Grindon just said. Removal of the financial incentive would probably be the first very solid step towards that.

DR. HERBERT: I disagree with the statement that we have to have yet another study before we can destigmatize hemochromatosis blood.

We have already had hundreds of studies which have been published which show it should be destigmatized. All you have to do is read your own copy which you all have of the January 1998 issue of Annals of Internal Medicine, with over 100 pages of articles on the fact that hemochromatosis blood should be destigmatized. I mean, how many more times do we have to put it in the literature from how many more thousand laboratories before you guys accept it as real?

DR. CAPLAN: Mary?

DR. CHAMBERLAND: Just a couple of thoughts. I agree with many others that an important first step is the removal of financial incentives, financial barriers.

In comments that have come out about the "need for studies" or the ability to actually study this, a couple of things. One is, in follow-up to Tricia's comments about other countries have done this, unless they have done this in a way in which they have identified the units as coming from individuals with hemochromatosis and then followed up the recipients who specifically got such identified units, you are not really going to have a good way to specifically study, I think, what the experience is in other countries.

I could be wrong and would be most welcome to hear if anybody has more specific information about other countries' experience.

I think, empirically, it would be theoretically possible to study the issue, but I have become increasingly pessimistic without putting down any facts, figures, power calculations, whatever, about our ability at least in the short term to study this epidemiologically beyond just prevalence of marker rates.

The comments that have been made by Mike Busch and Al Grindon earlier about do we need to really look at incidence, window periods, that is what Jay said is what driving is residual risk, and I think this, epidemiologically, could be quite a large undertaking, multi-centers, many years, whatever, but perhaps there is a need to at least work out some estimates of sample size, power calculations, whatever. That might be a first step, and we could sort of see what might lie ahead.

I think the last point, something which Paul and Al Grindon have talked about on a couple of occasions, even if you remove financial incentives, do we have a level of comfort that blood that is given for therapeutic reasons for reasons similar to the autologous donor, is that in any way different from the volunteer donor who does not have a medical condition or whatever to do that?

I do not have a good answer for that. I am not even sure if we have a good way to measure that, to be perfectly honest, but if I understand Paul and Al Grindon, that is what they have sort of raised on the table.

DR. CAPLAN: Let me make a suggestion here and then go back to the discussion again.

I think if you talk about considering removing barriers to using this resource, one of the happy things that happens sometimes is if it is necessary to do design or figure out certain studies, be they surveillance or whatever, that actually fits under the barrier description. We do not have to design the studies here. All we have to do is say remove it on the barriers and, just so I can get it on the record, that could include the commissioning of studies at vast cost or cheap, depending, to answer some of these questions either by epidemiologic surveillance or by trying to track recipients back to where the blood source was.

So the reason I like language about removing the barriers is that it opens the door to this kind of information, and it might even open the door to pilot use and that sort of thing. It is basically not for us to design it here. It is for us to say go do it, if that is what we want to do.

Get the financial barriers out of the way, then remove the barriers by coming up with enough information so that consumers can say that the quality is there, do not label it, we are okay with this. Whatever other studies, either literature reviews or surveillance studies and so on, that is the barrier removal part. We would then get back something, I hope, from the agencies that would say this is what we have done, we are going to use this and we are going to do that and this is why it is okay.

So, being now in love with my resolution, more than I ever was fond of something, I want to see if I have got a motion for it. Then, I would like to put it in play, and if we do not want to do this today, we do not have to.

If anybody wants to move it, since we are going toward lunch, either we will kind of roll along here and do something or maybe we will table it for the future.

DR. EPSTEIN: I would like to suggest a rewording of the very same concepts. I will just read this out, and if it sounds good, we can write it down and put it up.

I would start with the issue of destigmatization and say: "The committee finds that blood products obtained from persons with hemochromatosis carry no known increased risk to recipients attributable to hemochromatosis, per se, and therefore may be a valuable resource to augment the diminishing blood supply. However, the committee also recognizes that the obligate need for phlebotomy can constitute an undue incentive for blood donation due primarily to financial considerations. For this reason, DHHS should create policies that eliminate incentives to seek donation for purposes of phlebotomy. As such undue incentives are removed, DHHS should create policies and eliminate barriers to using this resource."

[Applause.]

DR. CAPLAN: Good. I like that.

DR. HOOTS: I move we put Dr. Epstein's proposal on the table for a vote.

DR. GILCHER: I will second that.

DR. CAPLAN: Discussion?

DR. NIGHTINGALE: The record should note that the FDA got applause.

[Laughter, applause.]

DR. CAPLAN: I might even more this one along without slapping it up on the board. We can read it again if you want to hear it again.

Any other discussion?

[No response.]

DR. CAPLAN: Would you like to have the question called? Should we read it one more time?

DR. EPSTEIN: "The committee recognizes that blood products obtained from persons with hemochromatosis carry no known increased risk to recipients attributable to hemochromatosis, per se, and therefore may be a valuable resource to augment the diminishing blood supply. The committee also recognizes that the obligate need for phlebotomy can constitute an undue incentive for blood donation due primarily to financial considerations. For this reason, DHHS should create policies that eliminate incentives to seek donation for purposes of phlebotomy. As such undue incentives are removed, DHHS should create policies that eliminate barriers to using this resource."

DR. CAPLAN: All in favor?

[Show of hands.]

DR. CAPLAN: Against?

[No response.]

DR. CAPLAN: That is unanimous.

Any abstentions?

[No response.]

DR. CAPLAN: Okay, so we solved that.

[Laughter.]

DR. CAPLAN: I think what we will do at this point, then, is reward everyone for their attentiveness to this important issue by taking our lunch break a little bit early, but I will ask that you try to come back close to 12:45. We can actually end a bit earlier. We will move up the schedule.

Who is up first? Dr. Koop is. When we come back, if he is here by then, we will let him go first. If not, I will flip the schedule around and we still start with somebody else, but we will try to get going close to 12:45.

Thanks.

[Whereupon, at 11:45 a.m., a luncheon recess was taken, to reconvene at 12:53 p.m., this same day.]

A F T E R N O O N S E S S I O N

[12:53 p.m.]

DR. CAPLAN: If we could get the audience to take their seats, please.

I am very happy to report that the consensus that we achieved this morning on hemochromatosis survived through lunch. It seems to have held up. That must be a pretty table point of consensus. So that is a very good thing for the committee to have reached resolution on a thorny issue.

I meant to make sure that Steve noted in the record that obviously we would like some feedback from the Secretary about the recommendation as appropriate, and it would be helpful to get that in a timely way.

We heard a lot during the past day and a half about the problems in the availability of blood, blood products, supply issues that exist, and supply issues that are emerging. We thought that it would be then useful as we begin to explore this supply issue not to resolve all dimensions of it, but to educate ourselves more about blood substitutes as another alternative for handling issues of reserves and handling issues of shortage.

So we have speakers this afternoon that will address that subject and open it up for us. We have a very distinguished first speaker, C. Everett Koop of Dartmouth College and Biopure, former Surgeon General and person who I think will be well known to the committee as our first person to provide testimony.

I am going to ask him to come up to the lectern there, and we will begin with Dr. Koop who is going to talk about the potential therapeutic benefits of blood substitutes.

Welcome.

DR. KOOP: Thank you, sir.

The expedited development of blood substitutes or, more accurately, alternatives for red blood cells is a key public health issue that I think we must address today.

We have the opportunity in this arena to gather and focus key recommendations to guide the Public Health Service and the Food and Drug Administration to support new initiatives in this field because I believe that alternatives to red blood cells will have a tremendous impact on public health, not only in the United States, but significantly worldwide, which can be realized through available technology in the near term, which can address immediate needs due to shortages or safety concerns with the existing supply of red blood cells, and which can offer additional advantages regarding shelf life and storage conditions that significantly improve logistics so that they are more readily available where and when they are needed.

The quest for an alternative to red blood cells has been narrowed to the search for agents that can best be described as "oxygen therapeutics." This is not just a shift in terminology, but it has to do with the actual function of what red cells and their alternatives do, which is to deliver oxygen therapeutically to tissues and to organs.

Let me use an analogy for progress in telecommunication. I can remember the day when we had no telephone, and then wonder of wonders, we had one we could crank on the wall, but the only person who could talk to us was my grandfather who lived two streets away.

Then we had a real telephone, as did many other people. First, it was a party line, and then it was a private line. Then we went from no dials to rotary dials, finally to touchtone, then automatic dial telephones, and finally to cell phones. Each was considered at the time to be a great improvement over the previous one, but each had its limitations which we corrected in part by the development of the next generation.

Some of the newer cell phones today can do everything except make decisions for you, but most significantly, they do not require the pervasive infrastructure that land-based phone lines do.

So it has been with the development of oxygen therapeutics. They have the capability of providing an unprecedented availability to patients in need. Not only could they eliminate the need for infrastructure required by type-specific perishable blood, but also they can allow for widespread pre-positioning without inventory management or refrigeration.

As we consider the possible role of oxygen therapeutics, we need to look to the need of the patient who benefit from their use; in particular, what are their requirements for safety, efficacy, and most important, availability of an oxygen therapeutic.

As to safety, we have recognized for some time that a blood transfusion is not, and never will be, totally safe. As we all know, the safety of therapeutic transfusions has been greatly enhanced by improvements in donor screening and testing for infectious agents.

Even this month, I have seen ads in the newspapers endorsed by the Red Cross that tout a new process for inactivation of viruses caused by AIDS and hepatitis C in a pooled plasma product, but for red cells, there has been no alternative. Since the present use of over 12 million units in the United States alone demonstrates a truly major need, we have learned to accept the risks associated with the administration of allogeneic blood.

These risks, due to a variety of factors including both known and unknown infectious agents, immunosuppression, processing errors, and others, have been considered minimal and management when weighed against the loss of benefit of not using blood at all, but now, with the potential for use of the new oxygen therapeutics, we have the opportunity to provide a consistent pharmaceutical quality and a safe alternative that addresses many of our safety concerns, both present and future.

In clinical settings--and I remind you that I was in them for almost 50 years--there are many reactions to blood transfusions, but even health professionals think, "After all, blood is a natural substance, and anything that happens with its use is normal and, therefore, okay."

Yet, if blood were to be looked at as a pharmaceutical product, many of these same reactions would be classified as adverse events or serious adverse events, which would be carefully listed in the product's package insert.

With the consideration of oxygen therapeutics, we and, most important, our patients may no longer have to accept many of the safety concerns with the use of blood, and the potential benefits, which oxygen therapeutics offer in terms of efficacy, may be even more important than that.

As to efficacy, we know that patients have a critical need for timely and efficient delivery of oxygen to the tissues under a variety of circumstances. Oxygen therapeutics should affect the prevention and/or reversal of tissue hypoxia and its consequences when there is significant blood loss, either local or systemic, inadequate vascular bed support, or tissue injury such as myocardial infarct or stroke.

This means that an oxygen therapeutic, such as a hemoglobin-based oxygen carrier, which I shall refer to now as HBOC, should provide quick and efficacious oxygen delivery in the microcirculation.

It is in this area where HBOCs seem to work differently and more efficiently than do red blood cells. Other oxygen therapeutics, the perflourocarbons, or PFCs, are also in development. Without prejudice, let me say I will focus on HBOCs since they have an oxygen transport more akin to red blood cells and appear to have greater advantage in dose-ranging, colloidal effects, and convenience.

Let me mention in passing, although this audience appreciates it full well, that under transfusion circumstances, RBCs have a reduced effectiveness in the immediate hours post-transfusion. Furthermore, many times RBCs very poorly serve the purpose for which they are truly given, which may include inadequate oxygen supply to the tissues, a diminished cardiac or respiratory function, or circulatory impairment because of vascular obstruction, tissue injury or both.

We know that normally 98 percent of the oxygen in blood is contained in the red cells in contrast to plasma which contains only 2 percent, but we also know that red blood cells do not uniformly and consistently flow through all of the capillaries in contrast to plasma that generally does flow without impediment through all of the capillaries.

One major advantage of HBOCs is they increase the distribution and release of oxygen via molecular hemoglobin dispersed in plasma to more effectively perfuse hypoxic tissues, especially in places where red blood cells cannot or temporarily are not going because of the pathology under treatment. This occurs because HBOCs are almost 1,000 times smaller than the bulky red blood cell which cannot travel through stenotic or restricted flow regions.

Therefore, the use of HBOCs should be superior when measured in terms of clinical outcomes to either allogeneic or autologous red cells. There is enhanced tissue oxygenation by utilizing the plasma space more effectively.

Simply put, the circulatory system, as a result of HBOCs in the plasma, becomes more efficient. Even at lower doses, HBOCs have the ability to deliver oxygen and can actually enhance the nature of oxygen offload from the remaining red blood cells. More oxygen is taken up in the lungs and diffused to vital organs with each pass of blood through the tissues.

There is evidence that this occurs at low blood pressure or when there is restricted red blood cell flow. In these situations, blood is typically not expected to be beneficial, but HBOCs are.

So we have come to set our sights higher when we think of the possible benefits of an HBOC that truly affects the microcirculation rather than just considering the old blood substitute model. I say that because the proper HBOC can provide oxygen to the tissues when, immediately after infusion, and where transfused red blood cells are ineffective. When I say where, I am referring to areas of impaired or restricted blood flow or where red blood simply are not available.

When HBOCs are finally available and broad-scale clinical experience develops, I expect that textbooks will need to be rewritten because everything we know about oxygen transport is dependent upon their being delivered in packages known as red blood cells.

Now we have the opportunity to see patient management enhanced significantly. In the future, patients with anemia and sub-optimal oxygenation will not necessarily be given units of red cells to raise their hematocrits, but will be dosed with HBOCs to achieve a specific effect. The effect is patient stabilization reflecting improved oxygen distribution.

The ideal oxygen therapeutic should: one, be a significant oxygen-carrying and delivery agent with physiologic control; two, be convenient, ready to administer, and require no reconstitution; three, should not require the use of an oxygen mask or supplemental oxygen; four, be universal, meaning it requires no typing or cross-matching; five, be volume expanding; six, assist in the generation of new red cells; seven, be stable at room temperature and for long periods of time so that it can be pre-positioned and stockpiled; eight, be available in large quantities from a readily available, cost-effective, safe source, and be produced to uniform consistent standards in large quantities according to pharmaceutical Good Manufacturing Practices; and, of course, finally, must be a safe product.

In addressing the approval process, I feel the need to recount some history and enumerate some precepts and remind you of some truisms that can easily stand in the way of acceptance of a new product, whether it be by the Food and Drug Administration, by physicians, or by the patients themselves.

Let me start with a little personal history. Shortly after Pearl Harbor, I found myself working at the University of Pennsylvania under the aegis of the National Research Council in one of about 10 laboratories assigned the task of developing a substitute for plasma.

The concern, of course, was the treatment of wounded in World War II, and I was assigned to work with gelatin made from bone. Others were given gelatin made from pork skin. Some were given colloids like gum acacia. Even hemoglobin solutions were used back then, but they were uniformly fatal. That is a fact over the years that has prejudiced some against later-generation polymerized hemoglobin in solution which is really a safe therapeutic.

History certainly repeats itself, and some of the experiences I had more than 50 years ago will make clear the same pitfalls exist today with HBOCs. Several possible sources of gelatin were considered for a substrate: first, the bone gelatin I have mentioned; and secondly, pork skin gelatin which is derived from dirty pig hides from American slaughterhouses, as compared to the bone, which was sun-dried from India.

Bone gelatin, being clean, required very little in the way of heat sterilization. As a result, we achieved a high molecular weight gelatin, and that was excreted by the kidney slowly. It sustained the blood pressure and enabled a wounded soldier to be transported from a field hospital to a base hospital, stabilized and not drifting into shock.

Pork gelatin, on the other hand, being dirty as it was, required repeated heat sterilization, which resulted in the degradation of the gelatin molecule so that it was so small, it left the circulation via the kidney so rapidly that it would not support the blood pressure in a wounded soldier being transported, and he arrived at the base hospital in shock and was at high risk for death.

To give a comparison, if an albumen molecule is the size of a football, bone gelatin that we made had the diameter of a cigar, but was as long as a football field. Pork gelatin, on the other hand, because of its degradation with heat sterilization was the same diameter of a cigar, but it was only as long as a football.

Therefore, all gelatin solutions were not the same, and I want you to remember that all oxygen therapeutics are not the same.

This prompts me to say, in retrospect, a lot of the trouble we had in World War II in investigating substitutes for plasma came not from the materials we used as the substitute, but impurities in the vehicle solutions which we were not able at that time to eliminate with techniques then available. This, of course, has all changed, and fortunately for the better.

So, to recapitulate what I have just said, all plasma substitutes were not the same. All blood substitutes are not the same. All HBOCs are not the same, and experience derived from the use of one cannot necessarily be carried over to the use of another.

HBOCs have an evolutionary history of their own through research and clinical trial, and they are continually being improved.

I said before that HBOCs not only delivery oxygen where red cells cannot go, but they also enhance or facilitate the uptake of oxygen in the lungs and its discharge in the tissues. Also, quite remarkably, there is evidence that HBOCs appear to have a hematopoietic effect; that is, they stimulate the growth of new red blood cells, and it may even serve as an iron substrate.

By ensuring that tissues are well perfused, HBOCs can provide an oxygen bridge, if you will, after blood loss, and foster the more rapid restoration of a patient's own red blood cells.

The beauty of all this is that an HBOC could result in an effect similar to erythropoietin and, at the same time, supply oxygen and iron to the patient. An HBOC also acts as an effective plasma expander, such as does albumin.

My former colleagues at the FDA face many challenges in addressing novel products, and the challenge of assessing oxygen therapeutics as an alternative to red blood cells has been significant.

Because the FDA is responsible for regulating blood, a product that by its nature has no efficacy criteria for approval, it has been a challenge for them to establish acceptable efficacy criteria for alternatives to red blood cells.

When an approved comparator product has no regulatory efficacy criteria, what criteria, then, do you use to judge an alternative? We have the opportunity today to provide FDA with guidance and recommendations that can only expedite the availability of this needed class of therapeutic products.

Also, comparing the processing of blood for transfusion and that of possible alternatives, blood donations for transfusions, that is, red cells, because they are obtained from individual donors, are minimally proceed and quality is essentially tested into them on an individual basis from over 12 million donations in the United States alone. Each unit of blood is also subject to variations related to the donor, who certainly is far from being a controlled uniform source.

HBOCs, on the other hand, are true biopharmaceutical products, and they are manufactured in large quantities under pharmaceutical controls to yield reproducible consistent products.

Today, we must consider what are the possible alternatives to red cells, when red cells may not be available or when we want to reduce the risks associated with the ordinary use of red cells. I would strongly recommend that you direct our colleagues at FDA to judge alternatives to red cells on their own risk/benefit profile as biological products. After all, if we were to truly do a risk/benefit profile of blood today, I am not sure that it would come out. I personally think that it would be difficult for blood to be approved by the standards that are currently being applied to blood substitutes.

We also do not need to fear that your recommendations to expedite implementation of alternatives to red cells would lead to some major public health crisis. Red cell alternatives or oxygen therapeutics, in a manner similar to that of cell phones that I talked about, could be implemented in a step-wise fashion. Certain patient populations would immediately benefit from their use, and as more data and greater acceptance and understanding occurs, broader applications could be approved for use later on.

I think an HBOC red blood cell alternative that meets the criteria I set before you earlier should have expedited consideration for approval by FDA. Such an agent can meet several unmet needs for many serious critical care conditions: for example, helping to conserve the use of RBCs and avoiding any lessening of maintaining the highest standard of safety in the red blood cell supply, or providing improved oxygenation of tissues via the microcirculation.

For example, this could potentially prevent the creeping damage to heart and brain after infarct and stroke. There is the potential relief of pain and sickle-cell crisis. There is the oxygenation of areas of low flow and low pressure; for example, in shock or multiple organ failure or dysfunction. There is also the time when it could substitute in some cases for some of the indications for hyperbaric oxygen. Certainly, it would simplify blood logistics and enable the treatment of blood loss in remote areas where RBCs are usually not available.

As to availability, one should not forget the humanitarian benefit of a safe and effective HBOC. The advantage of an HBOC in America is enormous, but it is ever much more so in a developing country. It could mean delivery of supportive critical care for all, no matter where they are positioned any place in this world.

What about lesser-developed countries or many troubled spots around the world? Suppose that the group "Doctors Without Borders" were able to use an HBOC for the countless kids that step on land mines? What can be said about the Balkans can be said about Vietnam, Cambodia, and other areas of disturbance. These are countries without the infrastructure to supply safe blood for transfusion.

To illustrate what is possible, consider the implementation of cellular telephone systems in developing countries that do not have hard-wired telecommunication networks. Cellular phones in those countries have already enabled those countries to leapfrog forward in a cost-effective way without the need for that traditional infrastructure, so it could be with HBOCs.

So, too, could developing nations have the advantage of a mass-produced pharmaceutical which could be stockpiled, transported, and utilized without the need for refrigeration, typing and cross-matching, screening for HIV and other transmissible agents, all of which require an infrastructure of specialized equipment, disposables, and, of course, most important, trained personnel.

A transfusion of blood in many developing countries gives the added burden of the possible transmission of malaria, hepatitis, AIDS, or all three, and the availability of an HBOC could stop that form of transmission immediately.

The benefits of an HBOC to the military are really incalculable. Besides the obvious improvements in more effective care, which could be rendered in the proximity of injury during combat, a stable, typeless, oxygen-carrying plasma expander which could be pre-positioned, even buried int eh ground in advance, would decrease the need for expensive handling and restocking of a perishable blood supply. In this respect, military preparedness for blood generates a huge cost whether it is used or not.

As a case in point, it cost well over $100 million to set up and maintain a blood transfusion service to support our troops in the Desert Storm engagement. Military personnel all over the world were recruited as blood donors, and the shelf life for that large amount of blood collected for that huge undertaking was 28 days.

We should also consider the overall logistics of the blood supply. Allocating alternatives to red cells to some defined patient uses will ease the growing burden and cost of obtaining more blood in the face of a swindling donor population, and of the blood collected, we need to assure that there is minimal waste. We should strive to optimize the use of blood for its intended purpose and obtain alternatives from sources that will not be competitive.

In conclusion, I have not covered many of the facets of this developing field, but from my perspective, I strongly recommend that you consider how the development process of oxygen therapeutics could be expedited.

It seems to me that in the face of the diminished reserve of the domestic transfusion capability and the likelihood of other conflagrations around the world like Kosovo that there are several options.

One, FDA could take a second critical look at the accumulated data on safety and effectiveness and reassess whether or not an HBOC could be approved now or soon. Second, consider approval of an HBOC for a specified limited set of indications, or, three, a combination of one and two.

I wish you well in your deliberations, and I think we all are fortunate to be on the cutting edge of such an advance in therapy. Today, you have a unique opportunity to recommend solutions to a pressing public health issue that could allow for tremendous new medical advances in the future.

Thank you.

DR. CAPLAN: Thank you.

[Applause.]

DR. CAPLAN: The floor is open for questions.

DR. PILIAVIN: I read your paper while I was coming here on the plane, and I was very impressed with the possibilities for third-world countries and so on.

I am just wondering what the cost of such a product, given I understand how much development has to go into any kind of new product. Will it be at all soon that these products would be actually economically available to developing countries?

DR. KOOP: I think that as it is produced in larger quantities and time goes on, that is true, but I think it is safe to say that this product can be produced cheaper than you could procure a unit of blood for transfusion by the usual method.

DR. PILIAVIN: Eventually, after development costs?

DR. KOOP: No, right now.

DR. PILIAVIN: Right now?

DR. KOOP: In this country, right now, you can produce this as a unit cheaper than you could procure a unit by our regular infrastructure method.

DR. CAPLAN: Dr. Koop, what is the situation with respect to this and other blood substitutes in the approval process in other countries?

DR. KOOP: I can only speak about South Africa and tell you an anecdote that the FDA already knows. So it is not letting any cats out of the bag.

We have done three clinical trials in South Africa with our human material, and you do know that we have a veterinary material that is very much like the human material, which has been approved by FDA and is now being used by multiple species of animals, even though it was approved for use just in dogs.

Our South Africa experience showed that we were able to progressively save more and more blood in the three successive clinical trials, and as I understand it, these were civilian trials. When the military saw what the results were and were able to read the protocols, they moved on their government and said it was too important an advance for South African civilians and or their possible military use in the future that there had to be some kind of fast track used to approve this product for human use.

DR. CAPLAN: So there have been clinical trials at least there?

DR. KOOP: There and several places in Europe, and we had some early Phase I stuff in Central America.

DR. CAPLAN: Any other questions?

DR. PENNER: No evidence of any renal damage over time and all of this?

DR. KOOP: No. We have had no evidence of renal damage by physiologic test and nothing by histologic section thereafter.

DR. CAPLAN: Maybe one final question, what is this situation with respect to application to the FDA? I know you said that the evidentiary standard issue is in place, so to speak, what standard to hold the blood substitute to.

There are protocols submitted, protocols under review. What is the situation here?

DR. KOOP: Other people here know this better than I perhaps, but my understanding of it is that we were approaching a point in Phase III trials where we thought that the one we were then doing would be the one that would satisfy FDA, and when that was finished, they asked that we do a study on 600 patients, orthopedic patients that had expected large blood loss, and we are now 45 patients into that study, but that is a huge study.

A study like that is an exorbitant expense, when we feel in our hearts and on the basis of data that we have already proven just about all we can prove about that study.

MS. JONES: FDA has asked you to do additional studies. How many patients were used in your original studies?

DR. KOOP: In the what?

MS. JONES: IN the original studies before.

DR. KOOP: 320.

DR. CAPLAN: I want to thank you for getting us off to an understanding of some of the potential therapeutic benefits of blood substitutes.

I am going to next turn to the presentation by the Food and Drug Administration on oversight of blood substitute development. In light of the first presentation, that seems like a good sequence.

We have got Drs. Alayash and Silverman.

DR. ALAYASH: My name is Abdu Alayash. I am with the FDA. The focus of my presentation today will be basically to deal with the safety aspects of hemoglobin-based product with some biochemical emphasis, primarily because of the fact I am a biochemist and secondly the hemoglobin is a rather complex and unique product. We need a little bit of biochemical educational aspect today.

Let me start with the outline which basically summarized the current approaches by industry and many in the research community as far as the modification of these products, but before I do so, look at the top of the picture. This is the hemoglobin within its natural environment, the red cells.

As you all know, we have a huge amount of hemoglobin within the red cells. This high concentration of hemoglobin favors the stabilization of the hemoglobin in its natural tetrameric form.

The red cell also contributes its metabolites, 23DBG, which basically cross-link two of these sub-units, and additionally modulate the affinity of hemoglobin towards oxygen.

If you take the hemoglobin out of the red cells in a dilute medium, the hemoglobin tend to break down, and if you were to infuse the hemoglobin, it would be cleared by the kidney and it would be extremely dangerous.

The other aspects, of course, the ability to release oxygen will be impaired. Sulfur hemoglobin solution in theory would release far less oxygen than within the red cell because of the 23DBG, and you can see the difference between the P50 and the red cells and outside the red cells.

So industry and the community approach the two basic problems by artificially cross-linking the tetramere with agents. They cross-link it either between the two alpha sub-units or between the two beta sub-units. Sometimes they add additional micro-molecules to increase the size and also to enhance the oxygen delivery.

Another approach is to actually polarize the hemoglobin with a protein polymerizing agent such as glutaraldehyde. The result is huge clusters of these hemoglobin. You can see two tetrameres cross-linked, three, four, and so on and so forth. They are also able to reduce the amounts of tetramer in the final solution.

Another approach that has not been used by industry, but it is still at the developmental stages by the Navy primarily, it is to actually encapsulate the hemoglobin. So all of these issues are easily resolved, and, of course, there is one simple issue that has been overlooked by everybody in the field, the fact that the red cells besides all of these features that I have provided has also provided a very efficient system that keeps the hemoglobin in its ferrous functional form.

Having hemoglobin outside the red cells, of course, the hemoglobin will be vulnerable. Of course, this is an issue that has sort of bogged the field for some time.

If you look at the literature now--and this is basically what I have abstracted from two recent reviews in the field--what are the preclinical experiences? So far, I remember this is published largely by the community and, to a certain extent, by industry itself.

The first thing that will strike you is the fact that the first point that sulfur hemoglobin is able to scavage nitric oxide. That leads to hypertension. The fact that the hemoglobin is very small, unlike the red cell, it can reach the parts of the vascular system where nitric oxide is produced. Nitric oxide, we now understand, is a very important vasodilator. So removing nitric oxide results in vasoconstriction and hypertension.

Some effects have been reported on the platelets. It affects only the microphage activation and even vasculitis. This was reported, by the way, by the Dutch group, the Dutch army, and in this case, they seem to think that is the polymerizing way in glutaraldehyde use to polymerize the hemoglobin that was responsible for the transient lesion seen in small vessels.

If you add hemoglobin to neurons, a report suggests that hemoglobin actually kills the neurons but leaves the glial cells intact, adding iron chelators or reducing agent will ameliorate the conditions.

Oxidation of hemoglobin is an issue, and free radical injury is also an issue. Only until very recently, people started to pay attention to these two very important issues.

The enhancement of endotoxin effect, more than one report in the literature suggests that hemoglobin and the endotoxin interact. Hemoglobin affects the clearance of LPS from the circulation. Hemoglobin increased the lethality of the LPS.

Clinical experiences. This is, of course, very difficult to review the literature. There are very few reports, but, again, there are a couple of recent reviews that came out, which suggest the common theme.

These are the two major reports in the literature about the human experience. Again, one of them, the hypertension on vasoconstriction has been seen with most of the hemoglobin that we are familiar with. G.I. distress, this varies from minor to moderate effect in the G.I. tact, which again most people seem to agree that it is because of local spasm throughout the G.I. and due also to the fact that hemoglobin interferes with an nitric oxide physiology.

One report that was found was published to show that there was some drop in the platelet count in recent human clinical trials.

The pressor effect seems to be the dominant effect here, and there are a number of theories and suggestions in the small community of ours to explain the effect.

One of them, of course, is the nitric oxide binding, as I have alluded to this fact. Hemoglobin actually reaches the vascular system, the area where nitric oxide is produced, and there is a tremendous affinity between the two. That would lead to the vasoconstriction and hypertension.

The issue where you have a small product versus a larger product, this is a debate, very active in the community. I am sure you are going to hear that for the rest of the day.

DR. CAPLAN: Can I just ask you one question? On the effects, are these single-dose administration effects? What is the range, depending on the studies?

DR. ALAYASH: There are a variety of studies. There are those with single does, and there are those with escalating doses. The hypertension had been seen in almost all of these conditions or under all of these circumstances.

The other hypothesis that has been put forward recently is a natural response to the fact that these products deliver too much oxygen. Although I have read recently that somehow the nitric oxide is actually involved in that.

There are suggestions that endothelia and natural vasoconstriction is produced. It may turn out to be a number of these factors that I have explained, but I think the nitric oxide is the front runner so far.

So what is nitric oxide? These are the things that I have listed that we need to know, and I am sure you have heard about the general interest in the scientific community about this small molecular that would produce.

Nitric oxide, we know it is of the EDRF, produced by the vascular system, produced by a very efficient enzymatic system. It is short-lived in seconds. It is very reactive. It reacts with oxygen, and, of course, it reacts with a number of biological molecules, including the hemoglobin, even within the red cells.

There is a long list of functions. The two functions that are relevant to us is really vasodilator function, an additional function that started to be appreciated by the scientific community, the fact that nitric oxide, besides its vasodilator effect, is actually an antioxidant.

By removing nitric oxide, you remove that vasodilator function, and also, you are removing an antioxidant property or function.

I have made this finding as simple as I can, based on 10 years of research in this area and, of course, reading the literature. What we are dealing with, as far as I am concerned, are two problems. The first problem lies with the product itself, hemoglobin, and the second problem is, of course, the area where the hemoglobin finds itself in.

Let me start with the product. This simply shows you that hemoglobin is really different from any other plasma or blood-derived product. This is a product that we saw in the ferrous functional form, and this is the form that we would like to keep the hemoglobin at, if you would like.

The ferrous form carries oxygen, and also, the ferrous form, if it reaches a production side, it will scavage nitric oxide very rapidly. As it a reversibly binding oxygen, of course, it turns into another form, which is the ferryl nonfunctional form. This action, you will remember, occurs on a daily basis within our red cells, but because of the enzymatic system that I have described, we, you and I, will have about 1 to 3 percent of your total hemoglobin that is in the nonfunctional form.

Incidently, if the hemoglobin were to encounter nitric oxide, you would also produce ferric nonfunctional form. So, if the hemoglobin reaches that, it not only loses the ability to carry oxygen, but it will turn into a nonfunctional form.

The third form of the hemoglobin is the ferryl, which is less popular in our community as yet, but there is some evidence which suggests that if the hemoglobin were to reach certain parts and certain conditions with very little hydrogen peroxide that is produced by the tissue, by the vascular system, by the platelets, by microphages, hemoglobin can turn into a toxic species which is known as the ferryl species. This is the point I have tried to make.

This is a product that if you leave it alone would simply turn into different--each step of the way. You are decreasing its functionality, and you are increasing its toxicity, but the good news here, as far as I am concerned, all of these reactions, now we know more about them than 5 years ago or 4 years ago. Now we know actually, and there are, as you probably will hear today, by means of genetic engineering or chemical engineering--you can actually restrict some of these reactions.

The other problems, of course, is the locality where hemoglobin finds itself. One of them is the vascular system. What you are looking at here, in simple terms, is the vascular system. This is the inside of the interior of blood vessels. These are the endothelia cells. These are the target cells of the smooth muscle.

What we understand is, of course, there are two scenarios under normal circumstances, of course, and the compromised circumstances.

Under the normal circumstance is what we understand now. This is the important element I was talking about. Usually, there is a redox-equilibrium. There is always more nitric oxide than the harmful oxidant in the vascular system. This is how nature kept it, and this is why we now recognize nitric oxide as an antioxidant.

Because of this dynamism, if you like, the other harmful oxidant, the chances of them being formed, is reduced.

In a compromised vasculature--and this could be somebody with trauma, somebody with sepsis, somebody with diabetes, somebody with sickle cell anemia, the situation is different. You normally have less nitric oxide than these oxidants, and to have hemoglobin here, of course, hemoglobin will tip the balance toward this scenario, which is an unfavorable scenario.

The question, of course, if hemoglobin is here, which we know this is the point of contact, it can easily reach vasoconstriction, hypertension, and that could lead to some other consequences which I have described.

So what are the questions from a biochemical point of view or from a regulatory point of view are really these three set of questions. There is some overlap in there.

Now that we recognize the nature of the product and the locality where the product sort of impacts the vascular system, the question clearly will become what are the short- and long-term effects of having hemoglobin in the vasculature and the possibility of hemoglobin disturbing the balance and also interaction as a cellular component.

If what I have described to you is a potential aim, but yet a toxicity, it is going to affect the development of a successful blood substitute.

How is it going to affect the criteria of evaluation? This is something Toby will address.

To what extent can we actually tolerate or control this redox potential toxicity and/or vaso-reactivity? In my humble opinion, I think these actions could be understood. They are beginning to be understood. The next thing is we need to control them if we really want to see the promise of a blood substitute to be fully realized.

Thank you very much.

DR. CAPLAN: Questions?

DR. HOOTS: What do the animal toxicity studies show? If you inoculate a rodent with LPS and then give him the blood substitute where you have got a vascular injury state, do you accentuate the vascular injury state?

DR. ALAYASH: Within the community, there are two schools of thought. Some people particularly in Europe have shown that in a septic shock animal model, hemoglobin does aggravate the situation. As I said earlier, it could affect the clearance of LPS.

The same experiments were done by a different group and gave a slightly different thing. It is not really fully understood, but if you take it to a simpler system in vitro, we ourselves have done that and many other people. Yes, the interaction between the two, LPS and hemoglobin, somehow that interaction ultimately affects the vascular system.

DR. CAPLAN: All right. Thank you.

DR. ALAYASH: Thank you.

DR. CAPLAN: We are going to hear next from Dr. Silverman of FDA.

DR. SILVERMAN: Thank you.

My name is Toby Silverman. I am a medical reviewer in the Office of Blood at the Food and Drug Administration.

I wanted to start off by saying that I truly appreciate the inventiveness of those people in the telecommunications industry and the truly rapid pace at which they have brought really these wonderful products to market, but I also understand that for the products that we are concerned with today, we are in a position of trying to imitate what nature has so very elegantly evolved over a very long period of time, millions of years.

As Dr. Alayash has said, we are only now beginning to scratch the surface of understanding, and we are only beginning to appreciate the nuances of oxygen transport physiology in red blood cell biology.

Could I have the first slide, please?

In November, I presented an overview of medical and regulatory considerations for the development of so-called blood substitutes for combat casualty care in San Antonio.

My talk came out of a review of discussion points at a meeting held by the Institute of Medicine in September of 1998 on the use of resuscitation fluids for combat casualty care.

The Institute of Medicine conference, which was convened in September, was charged with reviewing the state-of-the-art of fluid resuscitation, identifying targets for therapy, and then was further charged with making recommendations for future research directed at acute treatment of massive blood loss on the battlefield.

A subsequent talk later in November at a conference here in Washington, D.C., was designed to expand the discussion to instances of civilian trauma and to open the discussion about the design of clinical trials primarily in elective surgery.

I would like to announce that in September of this year, probably everybody here already knows the Food and Drug Administration, the Department of Defense, and NIH will cosponsor a 1-1/2-day conference regarding clinical trial design issues for this class of products. After that conference, it is FDA's intent to publish a guidance document.

Blood substitutes and oxygen therapeutics are biological drugs or drugs. I want to clarify that when I use the term "blood substitutes," I do not mean to imply that any of the products in development can actually substitute for all of the properties or activities of whole blood. Rather, I mean that these products have been designed to substitute for or imitate the oxygen-carrying and delivery capabilities of blood.

As I noted earlier, nature has evolved a very elegant transport and delivery system whose nuances we are only beginning to understand.

Having said that, the ability of products in development to perform these tasks effectively and safely is not assumed and will be the subject of ongoing clinical trials.

It has been said that trials now are larger than typical for biologic products, and most biologics approved to date have been for relatively small patient populations, but there have been some indications for some biologic products where studies have included several hundred to thousands of patients per cohort. There is no fixed rule about sample size. Sample size is heavily dependent on the anticipated risk benefit profile.

General efficacy considerations for drugs include such considerations as increase in survival, the prevention or slowing of disease progression, a decrease in morbidity, or measurable symptomatic relief, and in that order in terms of importance. These endpoints are to be distinguished from drug activity endpoints.

So what do I mean by drug activity endpoints? Drug activity is measured as the result obtained in a biological or chemical or physical assay, either in vitro or in vivo. On occasion, such activity endpoints have been used as surrogates for efficacy, and this brings us, of course, to the definition of a surrogate marker or endpoint.

Surrogate markers are used to diagnose disease or evaluate patient response to treatment. A surrogate marker should reflect what is happening in the underlying disease. The relationship between the surrogate and the true endpoint of interest should be such that an effect on the surrogate marker reflects an equivalent effect on the disease or the true clinical endpoint of interest.

Use of any surrogate endpoint or endpoints, such as blood pressure, lactate levels, base deficit, oxygen consumption, tissue oxygenation, organ functional assessments, and a long list of others must be validated as correlating with survival in hemorrhagic shock or exsanguinating hemorrhage before use in lieu of a mortality endpoint.

Evaluation of so-called blood substitutes in cases of blunt and penetrating civilian trauma, FDA anticipates that mortality will be the endpoint of choice for clinical trials in this clinical situation, particularly hemorrhagic shock or exsanguinating hemorrhage.

The reasons for this are as follows. If administration of a resuscitative solution results in worsened mortality, I think everybody would agree that efficacy would not have been demonstrated.

If a resuscitative solution neither improves or worsens survival and also does not improve a major morbidity, I think that most people would agree that efficacy would not have been demonstrated.

If a resuscitative fluid does not worsen mortality, but does result in a major irreversible morbidity in those who do survive, then most people, I think, would agree that efficacy would not have been demonstrated.

Now we get to the very difficult situation. If a resuscitative solution improves survival, but results in a major morbidity that impacts permanently on a recipient's ability to function, then I think that people would agree that efficacy would have been demonstrated for the mortality endpoint, but there will be a larger social question of the quality of the life saved, and that will require a lot of discussion. That is not an FDA issue, per se.

I think it is important to remember that in many situations, particularly field settings, many more people will be exposed to product than the population potentially helped by administration of the product.

I would like to emphasize that the ability of the EMT or the combat medic and military trauma to triage those who might benefit from those unlikely to benefit will probably be quite limited.

If a resuscitative solution is not anticipated to improve mortality associated with trauma, then the ability of such a product to improve a major morbidity could be used to demonstrate efficacy of the product for use in trauma. That is to say, patients who do survive and have an improvement in a major morbidity endpoint would be assumed to have showed efficacy for the product.

Having said that, the product should have an effect on a serious morbidity that has substantial impact on day-to-day functioning. An impact on a short-lived or self-limiting morbidity will usually not be sufficient, but the morbidity need not be reversible provided it is persistent or recurrent.

As with the mortality endpoint, use of any surrogate endpoints must be validated as correlating with improvement in a major serious morbidity before use in lieu of the morbidity endpoint.

These various scenarios speak to very different risk benefit assessments. Where blood is routinely available, use of one of these products certainly would not worsen mortality. Morbidity associated with product use will require careful assessment and quantitation.

Where could one evaluate the trauma situation? Field use, studies and circumstances where blood is not routinely available as in ambulance, hospitals lacking a blood bank or ready access to a local blood center.

One could also study the use of these products in situations where blood is available, but with randomization of study subjects to drug or to blood. We will return to some of these points when we talk about perioperative use.

It is not clear whether the results of studies under relatively controlled situations, as in the ER, could be extrapolated to field situations, either civilian or military. Nor is it clear if efficacy in cases of civilian trauma could be extrapolated directly to efficacy in combat situations where there may be prolonged delay to definitive care. This may also pertain in the civilian setting under certain circumstances, but not all, where there is care under adverse conditions, both environmental and physical, under uncontrolled circumstances, and where there is limited monitoring or therapeutic resource availability.

With regard to the military, worldwide, approximately 20 percent of soldiers who are wounded in action die. This is data that was presented at the IOM meeting in September. Ninety percent of combat mortalities occur before entry into the medical system, 80 percent within 30 minutes of the injury. Fifty percent die as a result of massive blood loss, 25 percent due to surgically uncorrectable torso injuries, 10 percent to surgically correctable torso injuries, and 9 percent to peripheral injury.

Penetrating trauma is the major cause of combat casualties both in the past and at the present. Increasing use of more effective body armor has resulted in an increase in the percent of casualties suffering from blunt trauma.

About 10 percent of those mortally wounded survive to enter the medical system, and they die from results of hemorrhagic shock, head injury, or contamination from the G.I. tract. They have been the main focus of military trauma care during the 20th century, and of this, 24 percent die of the hemorrhagic shock, about half from head injuries, and about 10 percent from septic shock.

So how does one go about developing a product, given all of these constraints? Before embarking on an evaluation of any of these products in trauma settings, FDA believes that products should be evaluated in Phase II studies under controlled conditions such as elective surgery. Such studies would also provide a basis for evaluation of products for perioperative use in Phase III.

In the Phase II use of the product under controlled conditions, such as elective surgery, it would permit evaluation of hemodynamic effects in toxicity, a preliminary estimation of maximum tolerated dose and a preliminary evaluation of toxicity at that does, an evaluation of drug activity for temporary reversal of physiologic transfusion triggers.

Up to this point, I have talked about circumstances where use of one of these products may actually save the life of the recipient. Under such conditions, the risk-benefit paradigm shifts very heavily towards efficacy. It is a truism to say that the better the product at saving lives, the more obvious the clinical benefit. While it is true that the efficacy of blood has never been demonstrated in a rigorous clinical trial, the utility of blood in treating life-threatening anemia, I think, is not in question.

The historical database from the period prior to availability of blood, I think, answers this question resoundingly. However, there are many considerations to keep in mind, and these include the known risks associated with the use of blood, which have been elaborated earlier, and the unknown risks associated with the use of blood. I am thinking particularly of emerging infectious diseases here. These are risks that cannot at this point really be quantitated.

Because of this recognition, FDA has agreed to accept reduction in or avoidance of allogeneic red blood cell usage as a surrogate for reduction in the risk of allogeneic red blood cell transfusion.

I think it needs to be said that avoidance of allogeneic red blood cell transfusion does not equate to avoidance of all allogeneic risk. FDA is not asking companies to measure the number of permanent adverse outcomes attributable to blood usage in a clinical trial, as such a demand would necessitate really enormous studies.

It is anticipated, at least early on, that traditional transfusion triggers will be used for licensure of early-stage products.

FDA is also not asking companies to measure oxygen delivery capabilities of their products at this time or directly in the efficacy endpoint, as evaluation of oxygen delivery capabilities would require several things, including the development of a new potency assay to reflect the oxygen delivery capabilities of the product in vivo and the development of new transfusion triggers known as dosing guidelines.

FDA does ask that sponsors evaluate the safety profile of their products. Again, more patients are likely to be exposed to product and blood than are anticipated to benefit from avoidance of an allogeneic red blood cell transfusion, and I want to repeat that avoidance of allogeneic red blood cell transfusion does not equate to avoidance of all allogeneic risk.

Adverse events may be either new and unanticipated or be of the type reported to be associated with the different forms, in this case, of the hemoglobin-based oxygen carriers, but I do not want to exclude perfluor-chemicals here.

Adverse events reportedly associated with the use of hemoglobin-based oxygen carriers overlap substantially with adverse events known to occur perioperatively.

FDA believes that contrary to clinical trials for most other products, clinical trials for these products capture efficacy data in the safety endpoint. Many of the adverse events for the hemoglobin-based oxygen carriers in particular have been thought to have occurred as a result of the vasoactivity of the product.

They may also have occurred as the result of inadequate or inappropriate offloading of oxygen resulting in tissue ischemia.

Therefore, FDA believes that studies should be powered for safety as well as efficacy, and that the safety endpoints should be defined prospectively.

Since adverse events are likely to increase with increasing dose of the product administered, FDA will ask that the number of oxygen-carrying units of both the product and blood be reported.

It is anticipated that adverse events leading to permanent morbidities might be the primary safety focus of clinical trials for perioperative use, and this will be heavily dependent on the extent to which these types of adverse events occur in the comparator group. If they are very rare, then evaluation of serious adverse events may suffice.

For purposes of data analysis, FDA suggests blinded review of all new and novel adverse events in predefined categories of adverse events, and the Data Safety Monitoring Board should be blinded to treatment allocation, and a blinded determination of serious adverse events leading to permanent sequelae if they occur at a sufficient rate. Again, the Data Monitoring Board should be blinded to treatment allocation.

FDA also recommends prospectively defined safety stopping rules. FDA anticipates that clinical trials for perioperative use would be stopped early and unblinded for safety considerations only, particularly permanent morbidities. There will be no stopping of the trials early for the efficacy endpoint.

Sample size calculations, safety boundaries and statistical analyses will be the subject of negotiation between manufacturers and FDA as the clinical situations will be different from one company to another.

In closing, I would like to say that the questions of benefit and risk raised here will be answered at one time or another. They will either be answered as part of ongoing clinical trials or if they are not studied adequately before licensure, they will surely be answered, perhaps not as clearly as we might like, after licensure as product use increases.

As you have heard and as represented today by Dr. Alayash, our academic and scientific colleagues have raised a number of very important questions regarding the safety, and I might add even the efficacy of products in development. I would like to venture an opinion here that for elective surgery, where patients may have other options, including postponement of the elective procedure, autologous pre-donation, other options, we have a responsibility to provide information that will allow those patients and their health care providers to make an informed choice.

While the clinical trials performed before licensure will not provide all of the answers, I believe that we need to have studied enough patients to gain a reasonable assessment of potential risks, as well as benefits to patients.

The sample size for any given clinical trial will depend on the underlying rate of adverse events of interest in the comparator arm.

I would like to add that while each of these products is unique and has individual specific characteristics, I believe that we can learn some common lessons with regard to clinical trial design from the outcome of the trauma trial that was performed by Baxter and will be presented, I believe, here today.

Had the only trial prior to licensure been a study in surgical patients, we might not have learned that such a product was associated with a negative outcome in a more urgent care setting. As these products are likely to be viewed as interchangeable with blood in situations where blood either is now or could be used, I believe that product labeling for a surgical indication alone may need to include also a prominent warning or even a contraindication for use in trauma given that the product will not have been studied in such settings.

I would like to go back to the analogy with the telecommunications industry, a manmade enterprise. I think it behooves all of us to remember that a very funny commercial with some important points that ran on television a long time ago, longer than I care to remember. If you fool with Mother Nature, you should do so with humility. Well, sometimes you might be lucky. You might also get burned.

I would also like to correct the record on some statements that were made about what was and was not p resented to the Food and Drug Administration.

We are not aware of the three clinical trials from South Africa having been submitted to FDA in a form that would have permitted us to identify them as having been done in South Africa.

I would also like to correct the record that with regard to a Phase II or III clinical trial, the agreement there--I am not aware and we are not aware that such a trial was agreed to as a Phase III clinical trial.

DR. CAPLAN: Why don't you stay there one second, Dr. Silverman, and see if we have any quick questions.

DR. SILVERMAN: Sorry. I am walking away with the microphone, too.

DR. CAPLAN: No?

DR. SILVERMAN: Perfectly clear.

DR. CAPLAN: Thank you for that thorough overview of the requirements of clinical research design trials.

We have next a presentation from Dr. Peter Keipert of Alliance Pharmaceuticals.

DR. KEIPERT: I would like to thank the organizers for this opportunity to present today. Looking back historically, I have been in this field for over 15 years, and certainly, 10 years ago, the blood banking industry viewed our group of companies in this development effort more as a threat than an aid. I think it speaks volumes that we are now here being invited to this meeting to present a potential relief, if you will, to aid in some of these availability issues that we are talking about.

If I could have the first slide.

There has not been a lot said about perfluorocarbons. So this will be my chance to perhaps point out some of the unique differences between fluorocarbons and hemoglobin.

Hemoglobin has sometimes been described as being the more preferable route because it is closer to Mother Nature. Perhaps after this talk, we can decide whether fluorocarbons are closer to mother's milk and therefore perfectly acceptable as a transient oxygen carrier.

The main purpose of these products, as has been alluded in the previous two presentations, is for delivery of oxygen. This is a temporary oxygen carrier intended for intravenous use, and we are really targeting this product for use in elective surgery. I think the rationale for that will become clear during the talk. It is based on where the majority of blood transfusions occur.

While trauma may be a very dramatic event when it happens in terms of transfusion needs, when we talk to blood bankers, most of them do not even track what percentage of their blood goes to trauma because greater than 50 percent is used in elective surgery and the rest if used for medical transfusions.

The unit does for our product is shown here. It is packaged in a ready-for-use container, containing about 110 mL of the emulsion. This is a product which is stable enough to withstand terminal heat sterilization, according to the normal industry standards.

Active drug in the container is about 65 grams of perfluorocarbon, and based on our Phase II studies where we have looked at drug activity and made sophisticated oxygenation measurements, we now know that this is equivalent to at least a unit of red cells in terms of its ability to delivery oxygen to tissues. We anticipate a shelf life for this product of approximately 2 years.

The first major distinction between fluorocarbons and hemoglobins is the way they transport oxygen. This is your more traditional sigmoidal hemoglobin association curve.

You can see that under normal conditions, hemoglobin is designed to be fully saturated, and it unloads anywhere from 25 to 30 percent on average in the resting state as the blood goes from the arterial side through to the tissues and comes back on the mixed venous side.

Fluorocarbons on the other hand have a straight line dissolubility curve. So you can increase the efficiency and the carrying capacity of these products simply by increasing your arterial PO2. So, during surgery when the patient is being ventilated with higher concentrations of oxygen, we can get in excess of 80 or 90 percent extraction of our oxygen as this blood travels from the arterial side of the circulation down to the mixed venous side.

This is physically dissolved oxygen which then dissolves and flows into the tissues according to the local PO2 gradients.

This is a little cartoon to help illustrate two points, and that is that in the larger vessels, of course, you have a huge concentration of red cells flowing, but even in these vessels, you have a small annulus, a small plasma ring where small particles tend to flow. This is where you find a lot of the platelets flowing near the endothelium.

So, already in these vessels, we have a concentration of emulsion particles near the vessel wall which can then start to diffuse oxygen into the tissues.

However, the reason I point this out is that often when people think about fluorocarbons, they think about the experiment where you draw an arterial blood sample before and after dosing with the fluorocarbon, and the simply look at oxygen content. They do not see a big difference, and then they jump to the conclusion that how can such a small difference in content have such a large apparent benefit in terms of tissue oxygenation.

The answer to that really comes when you look at the microcirculation in the tissues. As most of you know, the microcirculation is a huge surface area. The red cells spread out. You end up with large plasma gaps between the red cells, and in some cases, you have vessels that are only intermittently profused. However, they are almost always profused with plasma.

Therefore, these emulsion particles which are 40 times smaller than a red cell can provide oxygen delivery to these regions, and particularly to an ischemic area in places where red cells may not flow.

So we have always seen a significant benefit in terms of tissue oxygenation, which you would not a priori predict based on simply looking at a sample of arterial blood.

In the interest of time, I am not going to summarize actual data from a vast set of preclinical studies that we have done to look at efficacy. These have been done in multiple animal species and multiple labs, academic centers, and at Alliance Pharmaceuticals itself.

This schematic, however, summarizes the findings that we have observed. When the PFC is administered into the circulation, of course, you increase the oxygen content in the plasma compartment, since this is where the emulsion particles flow, and you therefore increase oxygen delivery to the issues.

If you start going into the tissues and you look at what happens, as I mentioned, this is dissolved oxygen. It does not have to dissolve through a red-cell membrane. So you have preferential or facilitation of uptake of this dissolved oxygen as it goes through the tissues, and if you get into the tissues, and we have done this in several animals now where we put electrodes in various organs and we isolate venous drainage from a particular site, we can demonstrate an increase in the venous drainage from that particular site, as well as an increase in tissue oxygenation.

The net result, then, is an increase in mixed venous PO2, and this has been demonstrated repeatedly in animal studies and in human studies, but when this was our initial only piece of efficacy for drug activity, the question arose what happens at the tissue level. What if you have metabolically disturbed the system? You could see increases in mixed venous PO2 due to blood shunting, poisoning of a particular organ. So it forced us to go back and do additional studies to look at tissue oxygenation, and we have in fact now demonstrated that there is no detrimental effect, and in fact, there is a very positive effect at the tissue level.

Just to summarize all of these studies in general terms, we observed what I have described here as positive oxygenation signals, and what I mean by positive is that they go in the direction that you would expect them to go in.

If you put an oxygen carrier into the circulation and you do not metabolically disturb the system, you would expect an increase in mixed venous gases. This is in contrast to some of the work that has been published with hemoglobin solutions where vasoconstriction actually decreases blood flow to the tissue, and you do not see this increase.

Once again, in contrast to some of the vasoactive hemoglobin solutions, we see completely stable hemodynamics with this product. We have no disturbance in cardiac output, pulmonary pressure, and systemic vascular resistance.

I have alluded to this already. We have been able to demonstrate enhancement of tissue oxygenation using a variety of invasive PO2 electrodes, both needle electrodes and surface electrodes in at least five tissues, three of which could be considered fairly critical, the heart, the brain, and the gut, in conditions of acute anemia.

In some of these studies, we have also been able to look at some index of organ function. For the heart, we have looked at left ventricular diastolic pressure or stroke work or other indices of myocardial function. In the brain, we had a study where we looked at auditory potential regeneration after an ischemic insult, and in muscle, we can look at ATP levels. We have actually shown a benefit and an improvement in these indices of function.

So let me now switch and say a few words about where this product is in terms of clinical experience. Looking at Phase I and Phase II of clinical studies combined, we have a total of 17 different studies in over 500 subjects of which 340 have actually received the active drug.

In terms of dose ranging, we have approximately a 10-fold range here in doses. If we look at the actual grams of active drug that have been administered, anywhere from 30 to 315 grams, and if you will recall the first slide showing the unit dose, which contains about 65 grams, you can appreciate that this is anywhere from a half unit to almost 5 minutes of product.

The breakdown is shown here. We have over 200 subjects in the early healthy volunteer and Phase I studies in both cancer and surgery patients.

The safety profile that was generated in these studies was very clean. We in fact did two supplemental healthy volunteer studies to look very carefully at immune function and platelet function and coagulation, since these are effects that have been seen previously in animal studies with PFC emulsions, and we are able to determine that with our current formulation, we have a very clean profile which then allowed us to move into the large multi-center Phase II studies shown here.

I will show you a couple of pieces of data from these studies. There were multi-center studies that were run in the U.S. and in Europe, and I will also touch on one of our cardiac surgery Phase II studies.

How do we want to use this drug? The whole purpose of this meeting is to see how these products are comparable or different from blood and what we could do to help alleviate some of these shortages.

While in terms of safety there are some significant differences between these products which should be fairly obvious, blood, of course, requires typing, has the risk of contamination, potential for immune suppression, and the well-known risk of disease transmission, our product, does not have any of these restrictions associated with it. It can be administered to anybody, and it is essentially based of synthetic clean ingredients.

In terms of efficacy, you do have some differences. Our product, and we have seen this in our animal and our Phase II studies, provides an instantaneous burst of oxygen delivery. We can see that systemically and at the tissue level.

With red cells now, if they come from the blood bank and if they have been in the fridge for more than a few days or a week, there is a progressive storage lesion that occurs. The oxygen affinity of these red cells increase. The membranes stiffen. PH changes. Cytokines are released from any residual white cells. You basically have a product which initially does not give you the oxygenation burst that you are looking for, and this has actually been demonstrated in critically ill patients in clinical studies.

The other big difference, of course, is the circulating time of these products. Red cells, of course, will circulate for several weeks. These temporary oxygen carriers are generally clear within 1 to 2 days. So then the question arises how do we optimally use this shorter circulation time to benefit the patient.

We have always felt that the best approach is to allow the patient to be their own donor and to couple the use of these products with autologous conservation techniques.

What I have listed here are the three common techniques currently available, preoperative autologous donation, autologous blood salvage which is done interoperatively, and acute normal volemic hemodilution.

On the left, I have described some desirable benefits that you could think of that you might want for an autologous blood product. You would want it to avoid risk of disease, which, of course, it will do. You would like it to be fresh, high-quality blood, with platelets and coagulation factors.

You want to eliminate clerical errors, prevent contamination, and be able to use it in an emergency at a fairly lost cost.

Pre-donation, as you can see, really only fulfills the first of these criteria, and yet, that has been an important enough driving force to make this fairly standard practice in many orthopedic centers and in some neurologic surgery centers, although there has been more and more literature now really questioning the logistics and the costs and the complexity of doing this and questioning the cost effectiveness of it. And as was mentioned yesterday, close to half of this blood continues to be discarded and not used.

Autologous blood salvage has some additional benefits, but it is really only when you get to acute normal volemic hemodilution which is done at the time that the patient comes in for surgery that you can start to appreciate all of these benefits.

This is fresh blood drawn from the patient. It never leaves the operating room. It stays with the patient. At the end of the procedure, the blood is given back.

So, conceptually, hemodilution is the most attractive of the three.

Let's look at how frequently or how common these techniques are used. We tend to look at the blood transfusion market in terms of the MSBOS schedule. MSBOS stands for the maximum surgical blood order schedule.

Hospitals have these schedules. They know often by surgeon how many units of blood this particular surgical procedure will require, and if you look at those procedures in the 2- to 4-unit category, there are approximately 4 million of these procedures per year in the U.S.

Of those 4 million procedures, at least 2 million of them actually get transfused with allogeneic blood, and that is the most common and routine mode of therapy.

A half-a-million about these patients will pre-donate their own blood during the 6 weeks leading up to surgery. On average, they will pre-donate anywhere from 1 to 3 units. This is mainly done in orthopedics, but it is also done, to some degree, in other surgical procedures.

Autologous blood salvage tends to be reserved for the higher blood loss procedures. You generally have to salvage at least 3 units of blood in order to make it cost effective to fire up the machinery and have the technician running the equipment. So that is used in only about a quarter-of-a-million cases.

Then you have hemodilution, which in this country in particular is still quite under-utilized. If you do surveys, it averages about 100,000 or less patients per year.

So then the question comes up: Why is it so under-utilized if it has these potentially attractive features? I think it is based on sort of the up-front, the a priori limitations of this technique.

In order to make hemodilution efficacious, you have to be a little more aggressive. You cannot simply pull off 1 unit or 2 units and then expect that to make a big difference in terms of transfusion outcome.

By being more aggressive, you have to be willing to tolerate lower interoperative hemoglobin. In some patients that are young and healthy, this is generally not a big issue, but that leads you directly into the safety concern in the elderly patient with compromised cardiac function due to cardiovascular disease, where the clinician now has a fear of taking away some of that oxygen reserve from the patient.

You know the patient is about to lose blood. So it seems a little counter-intuitive to pull off 3, 4, 5 units of that patient's blood.

So those are the limitations. There was recently a metaanalysis published by Bryson in 1998 looking at those studies that have been published in the literature, and the general take-home message is that some studies appear to show a benefit, some, despite significant efforts, did not show a benefit, and you really have a lack of very good clinical proof in the literature.

Part of the reason for that is there never has been a company or a product or some marketing effort to make this happen. So we feel that the approach should be to allow the patient to be their own donor and to combine these two modalities, combine the use of an oxygen carrier with acute volemic hemodilution, and we have now coined an expression for this. We call it "augmented ANH," as a new autologous method, where we use the oxygen carrier to offset the limitations of the hemodilution procedure itself.

The use of the product can decrease the safety concern because you are now providing an extra margin of safety with respect to oxygen delivery, and the clinician can see an improvement in oxygenation based on the physiologic status of the patient and the oxygenation of the blood.

So, in the simplest terms, our drug now becomes an enabling technology. It enables the clinician to finally do hemodilution properly. They can collect more units. Now the hemoglobin level that that patient actually bleeds at during surgical blood loss is much lower. The patient tolerates that lower level, and you, therefore, lose less red cells during the surgical procedure.

This is illustrated a little more clearly in this cartoon, where the patient comes in on the day of surgery and you initially collect anywhere from 2 to 4 units. The number of units you collect, of course, depends on the patient's hemoglobin level and their body mass, but you collect several units of blood.

You take them down to a hemoglobin level of 8 or 9, which is still perfectly safe with respect to oxygen delivery. Blood is now less viscous, less red cells. Now you start surgical bleeding, and during the time that the patient is losing blood, instead of immediately re-transfusing these very valuable fresh units of blood, which is what historically they have been forced to do, now you administer your synthetic oxygen carrier and you allow that drug to transport oxygen during this acute anemic phase. Then, once you have achieved hemostasis or you get to truly critical levels of anemia, you have to start giving back your fresh autologous blood.

So you lose less red cells during surgery, and you have a fresh product to give back, which has all the platelets and coagulation factors still intact because this blood never goes into the fridge. It stays with the patient.

This is essentially also a nice cartoon depicting the two large Phase II studies that we have done. So, if you keep this picture in mind, it will help to describe the protocol for you. These were two studies run in parallel. One was run in the U.S., focussing primarily on urologic procedures, radical cystectomy, radical prostatectomy type surgeries.

Then we had a parallel study in Europe in orthopedic surgery looking at primary hip revision and spine surgery. In Europe, we added 2 doses to single-unit does and twice that the U.S. we only had the higher does.

Briefly, the design. All patients were fully instrumented with PA catheters. The reason we mandated this for the Phase II study is because this was a drug activity study. We wanted to document what we had seen in our previous animal studies with respect to oxygenation status and the mixed venous blood gasses.

All subjects underwent acute normal volemic hemodilution to a target hemoglobin of 9. This was also a unique feature of this study. WE assembled over 30 or 40 anesthesiologists and surgeons from all of these countries to come up with a prospectively defined list of physiologic transfusion triggers.

We ended up defining both absolute values and changes from baseline that the clinicians would agree if they encountered those values, they would be justified in administering a transfusion. So we look at increases in heart rate, decreases in blood pressure, changes in cardiac output, hemoglobin levels, EKG changes, and drops in mixed venous blood gasses.

When the patients reached one of these triggers, they were randomized to either treatment or one of two controls, and then we looked at the activity of the drug in terms of was the trigger reversed, and if more than one trigger occurred, they all had to be reversed and then we looked at duration.

I will very briefly just show you those two findings from the studies summarized here. This is the percent reversal of all the triggers that occurred at that first trigger time point. In both studies, this is the 1.8-gram dose. These are equivalent doses.

We have a significantly higher reversal of triggers, and this is any or all of the triggers that occurred, versus the control group that received the unit of autologous blood at the time of that first trigger.

Then, if we look at the duration of the effect, we can see that, once again, we achieved the primary endpoint in both studies who had a significant prolongation versus the duration in the control group.

You will note that there is a difference in the magnitude of the effect here. Here, you can see the single unit being approximately equivalent to a unit of blood.

The main difference between these two studies is really due to the rate of bleeding in these different surgical procedures. In the U.S., you have much more rapid blood loss in the urologic-type surgery. Here, you have much slower, more prolonged oozing type of bleeding in the orthopedic surgery. Therefore, we were able to see these longer durations of physiologic stability.

This is the Phase II cardiac study that I wanted to also show to you to illustrate now this concept of augmented hemodilution collecting more autologous blood with the drug and having an impact on transfusion.

This is the amount of blood that we deliberately harvested from the patients just prior to going on to bypass, to protect those units from the circuit. A liter was collected in the electrolyte control and in the lower-dose group, and in the higher-dose group, we collected about 1.6 liters.

You can see the impact on transfusion, on average the low dose and the control, where we pulled the same amount of blood. We did not see a difference in transfusion outcome, although we did see a reduction, somewhat of a reduction in transfusion compared to historical controls at that site where no blood harvesting is done.

However, when we combined more aggressive harvesting with the higher dose the drug, we were able to see an 83-percent avoidance, and this is through discharge of allogeneic blood and a very low transfusion rate per patient.

So this is an illustration of the concept. Let me now bring you up to date in terms of where we are in our clinical development. We have a Phase III transfusion avoidance and reduction study that is currently underway in Europe. It is a multi-center study in 30 clinical sites across 8 European countries.

We have a Phase III cardiac surgery study that will look at the use of the product in combination with augmenting interoperative autologous donation, this concept of collecting more blood. This protocol is in the process of being developed, and it will be discussed with the FDA shortly.

The endpoint here will be the traditional endpoint that the FDA has indicated is an approvable endpoint, which is the avoidance of allogeneic red cells. This will be primarily a North American study, although we may include Canada and the European site in this as well.

In parallel, we will also run a dose escalation study in surgical patients, and this was at the request of the FDA, as Dr. Silverman just explained. There is this potential for off-label use. These are oxygen-carrying drugs that may be very popular once they become approved, and then the question comes up if they are used and if a clinician gets near the end of the case and the patient may still be unstable, they may be tempted to give an extra couple of units of the product. So we wanted to be able to demonstrate that even if we push this drug to much higher doses that we still would not encounter any unexpected serious adverse events.

This study is in the position now of having a full protocol ready and which will be finalized with the FDA in the very near future. So these two studies will run in parallel.

One of the questions we were asked to address when we came to the meeting was what potential impact can this product or our clinical indication have on the blood supply.

So I have put together this analysis, going back again to the maximum surgical blood order schedule. You will recall that there are approximately 4 million of these procedures in the U.S. Of these procedures, at least half of them, or 2 million patients, are getting transfused, and on average, they are using approximately 2-1/2 units each.

So the total number of red cells actually transfused in surgery is approximately or at least 5 million units per year in the U.S.

Based on our data and our modeling that we have done, we believe that using this technique of augmenting ANH with the product that on average across these MSBOS, 2 to 4 units, patients that we are targeting, we could see an average reduction per patient of about 1-1/2 units. So, on average, they are getting 2-1/2. If we reduce by 1-1/2, multiply that by 2 million patients, you have the potential to spare up to about 3 million units.

Keep in mind this is across all patients. This will translate into complete avoidance in those patients that are currently only getting 1 or 2 units. It will amount to a significant reduction in those patients that are currently getting 2 to 3 units, and it will amount to some significant reduction in patients that are currently getting greater than 3 to 4 units.

This is the total potential pool, assuming we could get into all of these surgical procedures. One then has to consider sort of a market penetration rate, and we have kind of laid out three different numbers here, a 20-percent penetration that could be achieved in a short amount of time, going anywhere up to a 50-percent penetration, and you can then see anywhere from 600,000 to 1.5 million units that could be spared that would not be used in these patients that would then be available to address other patients. This gets up into the range of 10 to 12 percent of the total blood collected in the U.S.

So let me just end with a couple of words about where we are in terms of being able to meet this need from the manufacturing point. Raw materials, we have a big advantage here that they are already available at industrial-scale quantities, and at exceedingly high purity levels.

Our fluorocarbon is greater than 99.99 percent pure and is available in metric tons.

We have a production facility shown here, and the nice thing about emulsion technology is it is a simple additive process. Every reagent you put into the tank gets mixed, homogenized, and ends up in your final bottle. Emulsion technology has been validated many times over because of other products that have preceded ours. Fluosol, the first-generation emulsion, oxygen carrier developed by Green Cross, which to date represents the only oxygen carrier ever approved by the regulatory authorities. You have Intralipid and Liposyn. These are I.V. fat emulsions.

So there is a previous regulatory history in terms of our commercial scale. We have a market-launch facility completed in our facility in San Diego. Currently, the capacity of this facility could generate enough units of product in excess of 800- to 900,000 units. That would allow us to treat approximately 400,000 patients per year, and with a modest investment of approximately 20- to 25 million, we can triple our capacity. We can increase to 1.5 million patients per year, and the biggest advantage here is that our product in no way competes with the existing blood supply in terms of drawing on a raw material.

Thank you for your attention and the opportunity to address the panel.

DR. CAPLAN: Questions? The floor is open.

DR. HOOTS: Just one technical question. Obviously, oxy-hemoglobin dissociation is irrelevant here, and you showed that first slide. When you are at an FIO2 of about room air or 21 percent, what kind of pulse occs can you maintain on somebody's gut and normal lung capacity?

DR. KEIPERT: Under room air, if a patient has normal oxygenation, you are going to see a very marginal effect. The carrying capacity under room air is very low.

However, we have done studies, if not designed by chance because of a protocol error, in dogs that were hemodiluted and hemorrhaged to critical levels, and when you get to critical levels of anemia, the experiment was done under room-air breathing. Then we actually saw a benefit in terms of reversing the myocardial dysfunction, but under the normal surgical patient that is perhaps not that critical, we would benefit and will advocate the use of higher-inspired oxygen while the patient is being ventilated to maximize the drug efficacy.

DR. PENNER: I am sorry. I missed the half-life.

DR. KEIPERT: The intravascular half-life is dose-dependent. It ranges anywhere from 6 to 12 hours in the top-load studies. At the current doses we are using, we would expect it to be more than adequate for the intraoperative surgical use.

DR. PENNER: And excretion?

DR. KEIPERT: Excretion from the body, you mean?

DR. PENNER: Yes.

DR. KEIPERT: Yes. The fluorocarbon is eliminated very much along the same lines that an anesthetic gas is eliminated. It is ultimately blown off through the lungs. The fluorocarbon particles are taken up by the phagocytic cells, the Kupffer cells in the liver and the macrophages in the spleen, and then the fluorocarbon slowly leaves the body by expiration via the lungs.

DR. PENNER: This does not have to be kept frozen anymore?

DR. KEIPERT: No. In contrast to Fluosol, this product is liquid, ready for use in the bottle. We recommend refrigerated storage to get the maximum shelf life, but it is stable enough that it can tolerate days or even weeks at room temperature.

DR. PENNER: The fluorocarbons, you really receive no benefit until the hemoglobins got down to around 4 grams or less. Apparently, that is not so with this newer procedure.

DR. KEIPERT: It really depends on what the endpoint is that you are looking at. This emulsion has a much higher oxygen-carrying capacity than the first-generation Fluosol product did, but the benefit and the efficacy of the drug really depends on the type of animal model or the clinical model that you are studying.

You will get more and more benefit of the product the more anemic the patient is. That is true. You will get a higher percentage of oxygen consumption coming from the oxygen carrier that you have put in.

DR. CAPLAN: All right. I think we will thank you, and we will move on now to Dr. Estep from Baxter Hemoglobin Therapeutics Division.

DR. ESTEP: Thank you, Dr. Caplan and ladies and gentlemen of the committee and also the audience.

As I think most everyone is aware, Baxter Health Care Corporation is not currently in clinical trials of any so-called blood substitute. However, until recently, we have been very heavily involved in the testing of these sorts of products. In fact, collectively, we have enrolled over 1,300 patients in either treatment or control groups for these kinds of solutions. I think it is on that basis that Dr. Nightingale invited us to participate here.

I would have to say that on the basis of that, we have gained quite a few learning and experiences on the use of these products. They have not all exactly been pleasant, but I think that one often learns as much from the things that do not work out very well as those that do, and it is certainly in that spirit that I am speaking to the committee today.

In that vein, basically what I wanted to do today was, first of all, talk about our summary experience which is shown here in the first slide, and then digress a little bit into an issue that was raised by Dr. Alayash, the hemoglobin vasoactivity, since that is a very important and controversial characteristic of hemoglobin-based oxygen carriers, and then move on to talk about our view of how these solutions may in fact impact blood product usage and then summarize with where we are today.

Although as a company, Baxter worked fairly extensively on the development of hemoglobin-based products in the '70s, there has been a continuous effort since 1982 to develop products based on human hemoglobin.

In the course of that effort, we have performed over 100--really closer to 200 pre-clinical studies and over 1 dozen clinical trials.

On the basis of the events that occurred primarily last year and a bit in the previous year, we concluded that we should not continue development of what we now think of as the first-generation product, either those based on human hemoglobin or the so-called first-generation products to which we acquired rights by virtue of acquisition of Cematogen, which was the leading company in development of those technologies.

The reasons for this decision are many-fold, but basically, they are summarized in this particular slide.

First of all, in our lead clinical trial, which has been well publicized in the lay press, that is, the U.S. Trauma Study, we certainly did not see efficacy in that trial. In the analysis of those data and the data from our other studies collectively, we identified a side-effect profile that gave us some concern and also the regulatory authority some concern, specifically with the occurrence of some fairly rare, but important side effects. These basically fell into the category of organ failure or multi-organ failure. This suggested to us that there was additional work that needed to be done both pre-clinically and clinically to better understand the role of hemoglobin potentially in exacerbating or maybe even initiating these sorts of side effects.

This, of course, implied that the time required to, in essence, fix the first-generation product would be substantially longer than we had originally anticipated.

In addition, really in parallel with that, we, as I already mentioned, did acquire the capability to develop and assess recombinant hemoglobin-based products, and this really plays into the previously expressed concerns about vasoactivity, which may in fact be very closely related to the side-effect profile, the fact that the Cematogen effort had been focussed for a couple of years, specifically on mitigating vasoactivity, by introducing changes in the hemoglobin molecule using the techniques of molecular biology.

So, on the one hand, we had a problem or a set of problems that we thought may well be due to hemoglobin vasoactivity, and on the other hand had acquired a new technology which appeared to be able to mitigate this particular problem. So we really decided to refocus our efforts into the recombinant area primarily because it offered the opportunity to substantially improve the safety profile of these products.

I am going to add a little bit of additional information on the subject of vasoactivity, and I will show a little bit of data and partial answer to a question that Dr. Caplan raised earlier about what this really looks like.

As has been mentioned, it is observed most readily as a pharmacologic increase in blood pressure, and what that means is an increase in blood pressure that does not appear to be related to the ability of hemoglobin to expand volume or by virtue of its oxygen-carrying capacity, but rather to another effect.

Although a number of systems are probably involved in this, the most important one does in fact appear to be the interaction of hemoglobin with nitric oxide. Hemoglobin binds it very avidly, and it also can catalyze the metabolism of nitric oxide in probably several different ways.

So, in essence, the issue is that hemoglobins will tend to diffuse at varying extents out of the vascular space and, if you will, intercept the nitric oxide on its way to the smooth muscle cells, and since N.O. is a vaso-relaxant, this results in a net increase in blood pressure.

We really have spent the last 10 years characterizing this effect. We observed it in the late 1980's and wondered about whether this was good, bad, or indifferent. In fact, we had accumulated a lot of preclinical data that suggested it might be useful, and this is an example of the kind of response we are talking about.

This happens to be in rats. I will show other data in a moment in another species and in mankind. This is a plot of the increase in mean arterial pressure versus dose of our first-generation product, which is DCLHb, or trade name Hemasist.

There are a couple of features that are of interest. One is that this effect is observed at relatively low doses, just 100 or 200 milligrams per kilogram. It does represent about a 30- to 40-percent increase in mean arterial pressure, although as one increases the dose further, the maximum increase in mean arterial pressure does not change, but the duration changes. It gets longer at higher dose.

This range is fairly relevant because the range of clinical doses that are being evaluated these days range from about 500 to in excess of several thousands of milligrams per kilogram. So this is a phenomena that occurs at the doses that we are talking about with regard to these kinds of products.

As I mentioned, though, this is not necessarily all bad because several of the contemplated indications involve, in fact, states exhibiting low blood pressure.

Whether you are talking about hemorrhagic shock or septic shock, one might thing a priori that increasing blood pressure to a certain extent in those states would be helpful. Indeed, that certainly appeared to be the case in our preclinical testing and some of our early clinical testing.

We have accumulated data that suggests in models of stroke that exchange transfusion with hemoglobin solution substantially mitigates the infarct size associated with stroke, and the pressure response does contribute to that.

In surgical patients in perioperative settings, hypotension is an issue that anesthesiologists tell us they worry about, and there was thinking that at least this side effect would be mitigated by the infusion of hemoglobin solutions, and that in fact appeared to be the case in our perioperative studies.

We also have some clinical evidence that actually rather small doses of hemoglobin solutions can substantially reduce the incidence of hypotension during hemodialysis.

On the other hand, there has always been an ongoing concern that if too much vasoconstriction is occurring in the wrong place, i.e., in some critical organ, then blood flow may be compromised to an extent that that organ would be damaged.

In particular, there has been concern about pulmonary vessels because they seemed to be particularly dependent on N.O. secretion as a mechanism for maintaining appropriate blood pressure, and the third area of concern was that increases in blood pressure may exacerbate bleeding either in actively bleeding patients or patients that have a newly formed clot.

So there has really always been a question of is this a net benefit, is it a problem, is it benign, and I think it is informative to show just a little bit of representative clinical data.

This is a rather typical example of the kind of response that we have seen in perioperative patients. This happens to be a study in anesthetized patients, that is, anesthetized when this was infused, undergoing abdominal aorta repair with a fairly low does. This is about 2 mL's per kilogram of our particular product.

You can see, although there is some variability here in the data, that in the patients infused with the hemoglobin-based product, there is a rapid and statistically significant increase in blood pressure compared to those patients receiving control solution.

This is actually very typical. We have seen this in a number of other perioperative patient populations. So there is no question that the vasoactivity is manifested in these patients. However, I should note that our preclinical studies suggested that this response can be mitigated by a number of maneuvers, if you will, anti-hypertensive agents, manipulating fluid volumes, certain other pathological states.

So the kind of response you observe very much depends on the patient population, and one population in which we consistently have not seen a significant difference in blood pressure is in the treatment of hemorrhagic shock patients.

This was a result from one of our Phase II studies. This is about the same dose as the perioperative patients you saw in the previous slide, a total of about 2 mL's of DCLHb solution.

You can see that the difference between the treated patients which are in blue and the control patients in red is not statistically significant nor clinically significant, and we have also seen this in both of the Phase III studies we did, both the U.S. Trauma Study and the European Study.

So, for some reason, there is some difference about the pathology or the condition or the state of these patients so that you do not see the significant difference in increase in blood pressure that we see in other patient populations.

So where do we stand today in terms of what we believe the manifestations are? Certainly, there are some side effects that have been observed that are almost certainly due to this particular property. Increases in systemic blood pressure and pulmonary vessel resistance are almost certainly due to this phenomena.

The G.I. effects that have been alluded to earlier we think are probably due to this phenomena, and it may be that the rare but important organ failures that have been observed may be related to this phenomena as well.

So, recognizing this, we and others have really looked at how one could mitigate this particular response, and there are really two basic strategies that have been pursued.

One is to polymerize the hemoglobin; that is, incorporate it into a higher molecular weight entity which would reduce the propensity to extravasate into the interstitial space.

The second approach is to alter the heme pocket through molecular biology to selectively inhibit the intrinsic ability of hemoglobin to interact with N.O.

I will show you some data that we have accumulated in a swine model that we particularly favor because the cardiovascular system in a swine is very similar to that of humans. In fact, we found the pressor response in swine parallels that in human volunteers in our Phase I study very closely. So we think it is very predictive of the response you would see in conscious humans.

This shows the effect of polymerization of our first-generation product. We happened to do this study using glutaraldehyde as a polymerizing agent, but we have seen the same type of results with other polymerizing agents.

What you see here with the filled circles is the response to the DCLHb, somewhat similar to the data that I showed in rats, although this is as a function of time with a fairly significant dose, 2 grams per kilogram, certainly within the clinical range being contemplated. You can see a very consistent vasopressor response of about 40 percent.

If we make the molecules larger, we still see a response very much in the same degree, even when we take care to remove the tetrameric material.

So, in this species, certainly, we did not see a significant mitigation of the pressor response simply by enlarging the molecule.

The bottom line is the response to albumin. So there may be a little bit of activity due to volume expansion, but not much.

On the other hand, as we have started to evaluate heme pocket mutants, we have gotten different results. The top two curves are the response, again, to DCLHb and to the first-generation recombinant product, which is also a stabilized tetramer. You can see they are very similar, but when one modulates the heme pocket to substantially reduce N.O. binding, you see the pressor response has been significantly attenuated.

So this is one of the reasons that we think this is really the optimal approach to use if one wishes to address this particular problem.

So where we stand basically with regard to the mitigation strategies is that while one can observe some reduction in this response in rodent models which are commonly used by polymerization, that really does not seem to affect what we believe to be the more relevant model in the swine, although heme pocket modifications can do so.

With regard to the clinical significance, we really have undergone a sea change in our thinking. I think it would be fair to say that 2 years ago, we were probably the champions of vasoactivity as probably "good for you" school of thought. There are some patients that still may benefit from this kind of activity. Our distribution, if you will, in terms of patients for whom this would be beneficial is probably less considerably then we thought it was before.

The other sobering thought is that identifying which patients fall into which category may well be difficult than we originally conceived. Therefore, we are trying to walk on the other side of the road, so to speak, and try to develop products that do not have this particular approach.

Let me shift gears just a little bit and talk about more of the efficacy side of the equation, specifically with regard to blood sparing. Again, with hemoglobin-based oxygen carriers, there are two basic approaches that have been taken. One is simply to use the HBOCs as a direct red cell replacement, that is, when a patient is assessed as needing a blood transfusion, they would receive the hemoglobin-based solution instead or the use of hemoglobin-based solutions in a hemodilution type of protocol as Dr. Keipert just described for the fluorocarbon solutions. The same sort of logic would apply for hemoglobin-based oxygen carriers.

In two pivotal clinical studies that Baxter has performed, basically using HBOC solutions as direct red cell replacements, we did see a statistically and, in the minds of the investigators at least, clinically significant reduction in blood avoidance, 19 percent in one case and 24 percent in another, using relatively low doses of DCHLb. The average dose in these studies was on the order of about 60 grams which is roughly equivalent to the amount of hemoglobin in one unit of blood.

We also think, although we have not personally done this kind of study, but some of my colleagues and the other companies will probably be talking about this more later, the hemodilution strategy should also work in blood sparing.

However, I should note that in both cases, there are some limitations that need to be appreciated. First of all, the half-life of these products is on the order of anywhere from 12 hours to perhaps a maximum of 48 hours, depending on dose and the particular entity that is being evaluated, and this is far less than the approximately 60-day half-life of fresh red cells. So that, as you might expect, is going to limit somewhat the degree of blood sparing that can be achieved, and I should emphasize it is probably not going to be linear with dose, although there is not enough data to really say that for sure empirically.

The modelling that we have done and I think others have done would suggest that if you get one-for-one replacement, sparing if you will, low doses, that is not necessarily going to extrapolate up to high doses.

However, for those patients who do typically require a few units, we believe that a reasonably high percentage of those patients could be spared blood transfusion, a red cell transfusion completely.

With regard to the use of these products in hemorrhagic shock, I would note that a number of the inherent characteristics do make them attractive for these kinds of indications. The longer stability, the fact that you do not need typing and cross-matching, low pathogen risk, and the fact that during their storage, they maintain their ability to offload oxygen of tissues as compared to blood which rapidly loses that capability during storage.

Indeed, in the extensive preclinical testing that we did with DCLHb, basically every time we evaluated in a different model, we seemed to get a positive result. So there was a suggestion of strong possible efficacy for HBOCs in this indication.

On the other hand, as already mentioned, we did not see that in the clinical studies that were recently completed, and we think part of that is due to the dose limitations. Basically, the average dose was on the order of half-a-liter, and another realization that fell out of our analysis of those studies was that the concurrent therapies probably mitigated the effectiveness of the hemoglobin-based oxygen carrier.

One thing that we, in the wisdom of hindsight, did not adequately appreciate was the huge amount of additional fluids that patients typically get.

I should say that in our study design, basically we were adding the HBOC to the current therapy. So all of the patients got current therapy in the treatment group. They had the HBOC.

However, the average amount of fluid that the treated patients got was on the order of 10 liters. So, literally, the 10 liters may have diluted out the effect of the half-liter.

I think the assumption that you can merely add this to concurrent therapy may not have been a valid one. You may need to alter concurrent therapy.

Also, while we really appreciated the fact these were going to be difficult trials, they actually turned out to be a lot more difficult than we imagined.

There was a tremendous heterogeneity of the patient population, and the distribution of mortality expectations was very different than what we had hoped.

I think this is one of the more important learnings from this study. When you are using mortality as the primary endpoint, and that was the defined endpoint in this particular study, you need to be operating in a patient population that has a relatively high mortality.

What is shown in red here is the distribution we had hoped to access in our study population with the average mortality being about 40 to 60 percent. That is, we wanted to get patients who had a fair likelihood of dying on the basis of their severity of injuries in the current therapy, but who were not so severely injured that they would die no matter what.

Unfortunately, although the average was indeed around the 40 to 60 percent the distribution of patients we got in terms of expected mortality, and this was on the basis of the TRIS scoring system, which is actually not a perfect system, but is the most widely used system for estimation of this mortality, we got a very different profile. In fact, the patient population tended to be made up of those patients who would probably survive with current therapy or who would die no matter what. So this made it very, very difficult to demonstrate a mortality endpoint in this particular clinical trial.

So what are the lessons learned? Well, I have already mentioned that mortality is a very difficult endpoint, and I will come back to that in our recommendation section. I noted that concurrent therapy may in fact need to be modified if you are contemplating HBOC treatment, and we probably need to stratify the patients a lot more highly than we did in the study. We accepted patients that had penetrating trauma. We accepted patients that had blunt trauma, and a number of people have subsequently pointed out, these are really very different patient populations.

So we would probably need to define that even more tightly than we did before, and also, although we did a lot of preclinical testing, I think we should endeavor and will endeavor in the future to incorporate even more features of the actual human clinical situation into those models as much as practical.

There are a couple of other aspects of HBOC that also will probably impact on blood product usage. One is that there is a suggestion, and I would say this is only a suggestion and has not been clinically demonstrated, that erythropoiesis may be stimulated or enhanced because hemoglobin turns out to be fairly adept at delivering iron in the right place. If that is true, then that might impact the efficacy in blood sparing.

In addition, as you have already heard, these products are being contemplated for a number of indications for which blood is not currently utilized, and one could imagine that that will indirectly impact the amount of blood products that are required for these patients, but I think it is rather difficult to predict which way it will go because you can imagine if you start saving patients that historically have expired, they may go on to require more blood products.

On the other hand, if you reduce morbidity substantially, patients leave the hospital sooner and they may require less. So I think at this point in time, it is a little difficult to predict which way it is going to go, other than it is probably going to change significantly.

As I mentioned, we had two recommendations with regard to facilitating approval, as was solicited, and they really fall into two different categories.

One topic that has already been raised is the comparison of hemoglobin-based oxygen carriers to blood, and that is a very natural and appropriate comparison if you are talking about blood sparing.

The problem is that we are appropriately characterizing the risk and benefit of the hemoglobin-based oxygen carriers very systematically, but that has never been performed for blood and red cell transfusion. I think now that some of these kinds of products are on the horizon, that is basically mandatory and not only to assess whether to approve the products, but which product to use in a given patient population.

We would suggest this be done both in a preclinical toxicity testing and as well as in clinical trial assessment of transfusion efficacy.

There is a little bit of literature in both of these areas, which suggest that the issues may be a lot more complicated than I think are assumed, and I think it would be very valuable to expand that database.

The second recommendation really falls out of our experience in treating hemorrhagic shock patients, and that is whether mortality must be the primary end point for these kinds of studies.

I believe this is in fact being reassessed by the agency on the basis of Dr. Silverman's comments, and we would certainly endorse this because I am not sure that is a practical or appropriate endpoint given our experience in trying to assess that in this kind of patient population.

The current status of our particular program is that basically we are in preclinical development of what we think of as a second-generation product. It will be recombinant-based. It will be modified to intrinsically reduce the vasoactivity, and we hope to initiate clinical trials within several years. The approval time frame will obviously depend on how successful we are in going through our preclinical development process and also how well the regulatory pathway os defined at that particular point in time. That may well change over the next few years.

I would just like to summarize our overall opinion, again, to reiterate that while we have had several bumps in the road, if you will, in our particular program, we do remain optimistic about the long-term capability of these solutions to be useful in the practice of medicine. We do think they will reduce and/or eliminate the need for blood or red cell transfusions in many patients, but that the treatments of patients not traditionally transfused may indirectly impact blood product usage as well.

So that, the availability of these solutions will have a very complicated effect potentially on blood product usage and in fact the practice of transfusion medicine.

Thank you very much.

DR. CAPLAN: Thank you, and thanks for that review of perhaps not a triumphant experience in research so far, but still an interesting set of lessons there.

Comments or questions?

DR. PENNER: I have one question. On the hemorrhagic shock patients, how did you get approval for the saline controls?

DR. ESTEP: If you want to know the entire process, I guess it must be understood that blood sparing was not the indication. We were looking at mortality as an endpoint, and so after a lot of deliberations with many trauma care professionals and a number of discussions with the agency, basically we felt that the appropriate control was simply a volume control, almost a procedural control for that study, because we were simply adding the HBOC treatment to standard therapy.

DR. PENNER: In that red cell replacement?

DR. ESTEP: No. It was not intended to be a red cell replacement type of protocol.

DR. PENNER: For hemorrhagic shock, they did not believe that red cell replacement would be efficacious?

DR. ESTEP: That was something we obviously measured, but that was not determined to be the appropriate efficacy endpoint.

At that point in time, we believed we had a product that would reduce mortality, and the agency believed that was the only appropriate endpoint to use in that particular study. So we mutually agreed to go forward on that basis.

DR. FEIGAL: We are talking about replacement in the field before you got to the hospital?

DR. ESTEP: No. Actually, there were two studies that we ran. In the U.S. study, this was in the emergency room. In a similar European study, we were treating the patients in the field prior to the hospital.

DR. CAPLAN: Keith?

DR. HOOTS: Did the trauma study include head trauma?

DR. ESTEP: No. We did exclude those patients.

DR. CAPLAN: So, as a final thing, I just want to be sure. One of the things you thought was inadequately understood was outcome studies on the baseline transfusion practice interventions. You did not think they were adequate as a baseline for you when you went into the mortality as outcome requirement? It was the third-to-last slide. I was trying to see what you were asking for about studies there.

DR. ESTEP: On the recommendations part?

DR. CAPLAN: Yes.

DR. ESTEP: The outcomes really were related more in the blood sparing indication. The logic that I was trying to articulate was the fact that if you are sparing blood, the implication is that the risk benefit of doing that is better than the risk benefit of using blood, if you will, but that risk benefit is really not well defined.

In fact, the other aspect of this, I will just mention, because I attended the session yesterday, which was very interesting. As we start thinking about cost effectiveness, it is not clear if we even know what the cost of blood transfusion is, and knowing what I know now from yesterday's session, I will probably add, it would be very useful to define what the real cost of blood transfusion is as well.

DR. CAPLAN: Thank you.

Next is Dr. William Hoffman from Biopure.

DR. HOFFMAN: Good afternoon. I am Bill Hoffman from the Biopure Corporation, and I am here today to talk to you about applications for Biopure's oxygen therapeutics, which are hemoglobin, Hemopure and Oxyglobin.

I am going to share the podium this afternoon with Dr. Bernadette Alfred, who will briefly discuss our manufacturing processes when I finish.

Until one year ago, I was director of the Surgical Intensive Care Unit at the Cleveland Clinic where I administered Hemopure to patients in clinical trials. So I am going to describe our clinical development programs, as well as some of my personal experiences with Hemopure.

Biopure Corporation in Cambridge, Massachusetts, was founded in 1984 in order to pursue a primary scientific interest in developing oxygen therapeutics for both the human and veterinary use.

The company has 180 employees. It has divisions in clinical and basic research, as well as marketing and sales. The company also has large-scale operating/manufacturing facilities.

Two of Biopure's products are shown on this slide. On the left is Oxyglobin, which was approved in January of 1998 for the treatment of anemia in dogs. The product on the right is Hemopure, which is the hemoglobin-based oxygen-carrying solution that is now in advanced clinical trials in humans in the United States, Europe, and South Africa.

This is a cartoon actually from yesterday's newspaper. It is a raccoon that was injured, a gunshot wound, and he is in the hospital observing himself from above, obviously a near-death experience. The veterinarian is pushing Oxyglobin wide open.

I do not know if anyone has seen the paper today. I do not know what the outcome of this experiment was, but, of course, this would be off-label use.

Hemopure is polymerized hemoglobin with an average molecular weight of 250 kD. In contrast, native hemoglobin has a molecular weight of 64 kD. It is applied in a solution of 13 grams per dL. A major logistic feature of Hemopure is that it is room temperature-stable, specifically stable for greater than 2 years at 2 to 40 degrees Centigrade.

Hemopure requires no reconstitution and is ready to administer without preparation or warming. It is manufactured by extraction and purification of hemoglobin derived from bovine red blood cells.

When you look at Hemopure, the color is similar to that of blood, but it has liquid properties that are more like lactated ringers. It has a low viscosity of 1.3 centipoise at 37 degrees Centigrade, which is approximately one-third the viscosity of blood.

Hemopure is isosmotic and isoncotic. It has a half-life in the circulation of about 24 hours. Hemopure is right-shifted hemoglobin, characterized by a P50 of 38 torr. This means that the release of oxygen is enhanced relative to red blood cells. Hemopure utilized the abundant chloride ion for release of oxygen. In contrast, human red blood cells are dependent on 2,3-DPG for the release of oxygen.

2,3-DPG is depleted in bank blood and is not restored for several hours after transfusion. Therefore, Hemopure may release more oxygen per-unit time than transfused red blood cells.

The clinical development of hemoglobin has followed a logical progression leading toward a variety of clinical applications. Knowledge gain from safety and preliminary efficacy findings from early studies form the basis for subsequent trials.

The perioperative trials have been based on an "oxygen bridge" concept which was first observed in some of the early animal studies using Hemopure. I will describe the oxygen bridge in more detail later.

We have conducted Phase I, II, and III studies where data were obtained on pharmacokinetics, safety, and efficacy, with doses up to 244 grams of hemoglobin. For the current Phase III orthopedic surgery trial in the U.S., the maximum dose is 300 grams of Hemopure.

Clinical indications studies include use as a blood or red cell substitute. In these studies, the endpoint has been the elimination for the need of allogeneic red blood cells. Allogeneic means another person's red blood cells.

In addition, Hemopure has been investigated as an oxygen therapeutic in clinical situations where blood is not normally used, but where enhanced oxygen transport may be desirable.

Indications that are being developed include the ability to oxygenate tumors and make them more susceptible to killing by chemotherapy or radiation therapy.

Three major features of the oxygen bridge are listed on this slide. An oxygen bridge is used to treat a severe anemia. A severe anemia occurs in patients who have a low hemoglobin level and symptoms severe enough to require the infusion of bank blood, specifically someone else's bank blood.

Hemopure is able to increase oxygen-carrying capacity through sequential dosing until the new red blood cell production in hemoconcentration restore the patient's old red cell concentration, or hematocrit.

Finally, this concept has been validated in multiple clinical studies, which I will show you on the next slide.

The oxygen bridge is schematically illustrated by the figure at the top. On the Y axis is the patient's hematocrit, and on the X axis is the days postoperatively. Shown in solid red, the hematocrit declines after surgery as a result of blood loss and hemodilution. Hemodilution occurs when patients get a lot of fluid that dilutes their own red blood cells.

Some patients can get through this period of anemia without receiving a transfusion. In order to do so, their cardiovascular system compensated for the low level of hemoglobin by pumping the few cells they have around faster. Many patients cannot compensate adequately, and they become fatigued. Their wounds do not heal well, or they develop more serious cardiac or respiratory problems.

Normally, these patients require a transfusion of one or more units of red blood cells so that normal function can be restored. In this situation, Hemopure can be given instead of red blood cells. After 4 or 5 days, when the patient's own hematocrit is adequate, Hemopure is no longer required.

Below the figure are six arrows which illustrate six of the important clinical trials of Hemopure, utilizing the oxygen bridge concept.

In the first studies, shown in blue, a single dose of Hemopure was administered in the operating room after the patient had lost a half-liter of blood. The maximum does administered in these trials shown on the left was 244 grams.

The second line describes a study where Hemopure was administered in the operating room, also after 500 mL's of blood loss, but in this trial, an additional dose was given on the following day. This study demonstrated the safety of Hemopure being given on consecutive days.

The results of these two investigations were very promising because they showed that Hemopure could reduce the need for red blood cell use.

The next three trials shown in pink were major clinical trials where Hemopure was given to patients undergoing elective operations in cardiac surgery on the top, abdominal aortic aneurism surgery in the middle, and finally general surgery, non-cardiac surgery, with the bottom pink arrow.

In these trials, the ability of Hemopure to substitute for red blood cells was the primary endpoint. Hemopure was given at the time the patient's physician decided that the patient required an allogeneic red blood cell unit. At that time, patients in these studies were randomized to receive either Hemopure or the red cells.

As you can see, as these studies progressed, the maximum dose allowed also progressed. In addition, the time that the patients were allowed to be treated with Hemopure was also increased up to 6 days.

In these studies, Hemopure was consistently effective in eliminating the need for even a single allogeneic red blood cell transfusion in patients who were randomized to receive it.

Finally, the last arrow in the slide represents our ongoing pivotal trial in the United States which is orthopedic surgery patients. The maximum dose in this trial is 300 grams, and we are currently enrolling patients.

In summary, Hemopure has been given to over 450 subjects. The maximum dose in the clinical trials to date has been 244 grams, but up to 300 grams or 10 units is allowed in the ongoing studies.

The maximum plasma hemoglobin concentration achieved has been in the range of 4 to 5 grams per dL. We have accumulated safety, pharmacokinetics, physiologic, and efficacy data on these subjects, and we have an ongoing pivotal trial.

I am going to review some of the results from the individual trials. This first slide shows the Phase I studies which were done on normal volunteers. These studies evaluated escalating doses and increasing rates of administration.

Exercise physiology was also examined in normal volunteers who were anemic and given Hemopure. Finally, immunology studies were done in 8 normal volunteers looking at serial exposures over a 1-year period.

The results of these studies are summarized very briefly on this slide. Increasing infusion rates and increasing doses were well tolerated in the normal volunteers.

Hemopure produced statistically significant enhancement of pulmonary diffusing capacity, which is an important measure of respiratory function.

Hemopure groups had lower heart rates and lower non-invasive cardiac indexes, particularly when they exercised. In other words, Hemopure was successful in lowering the stress response to the anemia that was produced.

Finally, infusion of 45 grams of Hemopure was equivalent to infusing 110 to 150 grams of hemoglobin as autologous blood and restored normal exercise tolerance in these normal volunteers.

This slide shows two studies that were done with Hemopure. I would just remind you that these "n's" over here are patients exposed to Hemopure, and these studies, there are roughly equal numbers of controls.

There were four studies in the United States looking at safety and tolerability in a variety of patient populations, including those undergoing major urologic procedures and emergency gynecological procedures.

There were three European safety intolerant studies conducted looking at acute normal volemic hemodilution which is a fairly common practice in Europe for blood conservation.

The two-dose escalation studies in the U.S. that I have already described are shown here. They enrolled a total of 80 percent in the Hemopure group, again, roughly as many controls.

Finally, there were two studies looking at the feasibility of use of Hemopure as an oxygenating agents in patients with sickle cell anemia. One study was in sickle cell patients in crisis, and one study was in sickle cell patients undergoing exercise.

Clinical studies with Hemopure have occurred at multiple institutions in the United States, Europe, and South Africa. This slide shows a partial list of the U.S. investigative sites. I think these represent some of the best community and university-affiliated hospitals in America.

This slide shows the advanced clinical trials with Hemopure. The cardiopulmonary bypass trial, I will soon describe in more detail.

In the abdominal aortic aneurism trial, 46 patients were exposed to Hemopure, and there were 23 controls.

In the European and South African non-cardiac surgery study, 83 patients were exposed to Hemopure. There were 77 patients who were controls.

In all three of these studies, Hemopure was effective at eliminating even a single allogeneic transfusion at the long-term follow-up time point.

The U.S orthopedic trial listed here is an ongoing pivotal trial with plan sites in the United States, South Africa, Europe, and Canada. The total number of Hemopure exposures to date is approximately 450. Again, we have approximately equal numbers of control patients.

These subsequent slides review in more detail regarding the cardiopulmonary bypass trial. This trial was a randomized double-blind trial. The control group received allogeneic red blood cells. It took place at 14 centers, a combination of university and community hospitals, including the Cleveland Clinic, Massachusetts General, and Emory University.

There were 50 patients randomized to receive Hemopure and 48 randomized to receive allogeneic red cell units on a one-to-one randomization scheme.

Patients were randomized at the first postoperative allogeneic transfusion decision, which occurred before intensive care unit discharge.

This slide summarizes the efficacy endpoint and shows by postoperative day the number and percent of patients within the Hemopure group who had no postoperative allogeneic red cell transfusions.

The first line in yellow represents postoperative day zero, which includes the day of surgery up until midnight that night. On the day of surgery, you can see that Hemopure was 100-percent effective at eliminating the need for red cell transfusion. That means that none of the patients within the Hemopure group who otherwise would have gotten at least one red cell transfusion received a transfusion during the first approximately 16 to 18 hours after surgery.

If you can extrapolate these data to broader surgical populations, this suggests that if emergency surgery is required and red cells are not available or there is a delay in the availability of red cells, it is highly probable that a relatively low dose of Hemopure which was used in this study will bridge all patients until blood is available.

On postoperative day one, which includes approximately 36 hours after the time of surgery, 80 percent of patients randomized to Hemopure still had not yet received a single allogeneic unit.

The table goes on until day 28, which is the long-term follow-up time point, which is approximately 4 weeks later, and at the end of the trial, there were 34 percent of the patients who had been randomized to the Hemopure group had not received a single unit of allogeneic blood.

In this trial, investigators were asked why the patients were transfused. In all cases, patients were transfused for a low hemoglobin level, but there may have been additional reasons. One of those additional reasons is a vital sign abnormality, and what we are talking about here are hypotension which is a low blood pressure or arrhythmias which is an abnormal change in the heart rhythm.

It turned out that prior to the first infusion, there were a total of 18 patients who were randomized to receive either Hemopure red cells because they had a vital sign abnormality.

Eleven of these patients were randomized to receive Hemopure. The envelopes were opened. Eleven of them were randomized to the Hemopure group, and seven were randomized to receive red cells.

In the second row, you can see the effective treatment on the vital sign abnormality. In only 1 out of 11 cases was the vital sign abnormality still present after the Hemopure infusion. However, it was still present in four out of seven patients who had received red blood cells. This result suggests that red blood cells are relatively ineffective at correcting vital sign abnormalities associated with anemia. On the other hand, Hemopure was quite effective in this trial.

The third row is rather interesting because this row shows the number of patients that developed new vital sign abnormalities after the first infusion. There were no such patients in the Hemopure group. However, two patients in the red cell group developed new vital sign abnormalities after the first infusion of red cells.

An important part of the oxygen bridge is restoration of the hematocrit, or red cell concentration by the time of discharge. This ensures that patients will no longer require transfusion.

The slide shows serial hematocrits in the patients in the cardiac surgery trial randomized to receive red cells shown in black or Hemopure shown with the open circles.

You can see that its screening, which is up here, the hematocrits in the two groups are approximately 38, and, of course, they are not statistically significantly different.

At baseline down here, which is prior to randomization, that is after cardiopulmonary bypass--remember, this is a postoperative treatment study--the hematocrits in the two groups are also no different, approximately 24 percent.

The red cell group shown in black is having an increase in hematocrit on postoperative days one, two, and three, as a result of red cell transfusion.

On postoperative days one, two, and three in the Hemopure group, the hematocrit remains low. This is because the oxygen transport needs of these patients are being met in large part by the Hemopure solution.

By postoperative day six, which is the discharge day, the hematocrits in the two groups are equal, and, of course, they are equal again on day 28.

Therefore, despite the reduced amount of transfusion in the Hemopure group, relative to the red cell group, the hematocrits are the same by the discharge date. This increase in hematocrit in the Hemopure group may be the result of better hemo-concentration, better mobilization of fluid, or an erythropoietic effect due to the iron that was alluded to earlier, or direct stimulation of erythropoiesis by Hemopure.

Historically, hemoglobin solutions have been associated with blood pressure increases. One way to examine blood pressure is to compare the biggest increase in the patient's blood pressure from baseline. This gives you an idea of the vasoactivity in the Hemopure group relative to the red cell group.

In this slide is the mean maximum change from baseline and diastolic blood pressure. The red cell group is shown with the blue bar and the Hemopure group is shown with the red bar.

The mean maximum change in diastolic pressure from baseline in the two groups is 20 and 22 millimeters of mercury, and these are not statistically significantly different.

For systolic blood pressure, which I have not known, the mean maximum increase in the two groups was 30 and 36 meters of mercury, respectively, also not statistically significantly different. Therefore, Hemopure was not associated with greater extreme elevations in blood pressure in this trial.

This slide summarizes briefly some of the safety variables in the study. Serious adverse events in the cardiopulmonary bypass trial are equally common and of similar frequency in both groups.

Maximum increases in blood pressure were not statistically different between groups. That is the slide I just showed you.

The only non-serious adverse events that were more common in the Hemopure group were transient fever that lasted less than 12 hours and jaundice. Jaundice tended to occur more frequently in patients in the Hemopure group who had received 120 grams.

This jaundice results from the physiologic processing of hemoglobin to bilirubin. It appeared in the trial to be benign. It was not associated with any worsening of liver function tests.

Interestingly, the increased report of fevers were not associated with increased reports of infections. Moreover, in the other two major clinical trials done, the aortic aneurism trial and the European and South African noncardiac trials, fevers were actually reported more often in the red cell group than in the Hemopure group.

To summarize, Hemopure successfully eliminates red cell transfusion in a significant proportion of cardiac surgery patients, as well as vascular and general surgery patients, as demonstrated by the three clinical trials that have bene completed to date.

Hemopure can also reduce the total number of red cell transfusions. Restoration of red cell concentration or hematocrit occurs so that the patients are effectively bridged by Hemopure to completely avoid allogeneic transfusion.

Furthermore, we have shown that in cardiac surgery patients, vital sign abnormalities resolve more often with Hemopure than with red blood cells.

In the major clinical trials to date, Hemopure has shown equivalent safety to the controls.

Finally, an independent data monitoring committee reviewing the results of the cardiopulmonary bypass trial stated that the trial was adequately conducted and controlled to be sufficient as a pivotal trial for efficacy.

Ongoing and plan studies for Hemopure are listed on this slide. In the specialty of oncology, we are investigating the ability of Hemopure to oxygenate hypoxic tumor cells and make them more sensitive to radiation therapy and chemotherapy.

We have an ongoing orthopedic surgery trial in the United States, Canada, Europe, and South Africa that will serve as a pivotal trial for license application in the United States.

We are also planning Phase II studies with the military in stable trauma patients and non-cardiac surgery patients as a means to develop further trauma indications and further military applications of Hemopure.

Finally, to close, I would like to talk about four rather remarkable cases. In the first case, the patient was taken through an 8-1/2-liter blood loss with a relatively low dose of Hemopure. This was in an early Phase II study.

The Hemopure maintained hemodynamic stability until cell salvage blood was available. This patient completed the surgery, had an ICU stay of 12 hours, a hospital stay of 5 days, and never required transfusion with his own pre-donated autologous blood.

In the second case, we treated a compassionate use patient who had refused allogeneic blood, but consented to receive our product. This patient had blood loss down to a hematocrit of 7 and developed chest pain and ischemic EKG changes.

The woman had been bleeding from a uterine fibroid that was stopped with hormonal therapy. Hemopure was administered in relatively low doses that were titrated to alleviate any symptoms of myocardial ischemia.

After the first does of Hemopure, she was able to get out of bed with assistance and use a bedside commode, and she had no ischemic systems.

After approximately 1 week of treatment, treated with approximately 30 grams every day, her hematocrit had increased to 18 percent, which was a level adequate to keep her symptom-free, and she was discharged from the hospital.

In an additional case, a patient who was randomized in a clinical trial to receive Hemopure became anemic intraoperatively and developed ischemic EKG changes while under general anesthesia. These were successfully reversed with the application of Hemopure.

The fourth case was a very recent case. It occurred in the last week. This was a South African patient who was allo-immunized, had severe sepsis, with disseminated intravascular coagulation, and required surgery for removal of the infected focus.

Her anesthesiologist would not allow the surgery to proceed because there was no blood available, and we supplied the hospital with several units of Hemopure. About 90 grams were infused in that patient, total. Blood eventually did become available for her, but she was discharged from the hospital after about 5 days, her sepsis completely resolved.

In summary, I hope I have given you an understanding of the logical progression of Biopure's clinical trial program. The oxygen bridge concept has been tested and validated in multiple surgical populations, including cardiopulmonary bypass, aortic reconstruction, and multiple general and orthopedic surgery patients.

Hemopure has been safe to date. We have rather remarkable evidence of elimination of exposure to red cells, red cell recovery, hemodynamic stability, and reversal of severe anemic symptoms in our compassionate use patients.

I want to thank you very much for your time and attention.

DR. CAPLAN: Dr. Hoffman, before you go in that direction, you can use the time you have got however you want, but it is probably about 5 minutes by my count. So you may forego some questions if we go down the road of, let's say, manufacturing capacity display.

DR. HOFFMAN: Let's have questions.

[Laughter.]

DR. CAPLAN: John?

DR. PENNER: Just a question. Is there any sign of reticuloendothelial cell blockade as this product breaks down? In other words, a situation where you might have enhanced problems with infection.

DR. HOFFMAN: No. We have not seen in our clinical trials any increased infection. We did a preclinical study in rats, which was published in the Journal of Clinical Investigation, where cardiovascular function was actually improved. These are rats infused with endotoxin, and we just treated this compassionate use patient who had severe sepsis and organ dysfunction, as evidenced by the DIC, who recovered completely. So we have not seen that at all.

DR. PENNER: But none of the other patients have demonstrated increased susceptibility to infection?

DR. HOFFMAN: No.

DR. CAPLAN: Does the product have any reason to cause concern about zoonoses?

DR. HOFFMAN: The product is sterile, and probably the best person to answer that is the manufacturing people, but the product is sterile.

DR. CAPLAN: A related question. I am just curious. When you give this product to animals, dogs or whoever else, have they seen performance enhancement?

DR. HOFFMAN: They do, and there have been some 20,000 units sold in the veterinary market. There have been a lot of uses, anemia, sepsis, and whatnot, and the veterinarians are reporting performance enhancement. Yes.

DR. CAPLAN: Paul?

DR. McCURDY: The need for transfusion, like beauty, is often in the eyes of the beholder. Were any of these randomized trials blinded? Did you make an attempt to blind?

DR. HOFFMAN: The trial I described in detail was double-blind. The investigators were blinded.

The other trials have transfusion criteria, but we do allow the investigators a lot of leeway.

CAPTAIN RUTHERFORD: Captain Rutherford, DOD.

DR. HOFFMAN: Yes.

CAPTAIN RUTHERFORD: In your autologous, you said you had three that got better and four that really didn't, and two of them had adverse events. Did you track those autologous units over the date of the shelf life of the units to determine if they were past the 14 days of 2,3-DPG?

DR. HOFFMAN: I don't know the answer to that. That is a very good question. One would hypothesize that the reason the red cells were not effective in reversing vital sign abnormalities were that they were not unloading oxygen effectively, but I do not know the age of those red cell units.

In general, it is the hospital's practice to release the oldest blood because it is going to expire. I expect the blood was near to its limits of storage, but I honestly do not know. That is a very good question.

Those are not autologous units. Those were allogeneic.

DR. PENNER: Just a comment. I am concerned about the bridge data because, as we followed the Jehovah Witness group, oftentimes you get by surprisingly with 7 grams and 6 grams and so on and we get them through the surgery. Although we all are very fearful of them getting into cardiovascular problems, they do oftentimes survive with these very low levels, and we cannot tell that there is a distinct difference if we have other individuals that we have transfused.

One would need, I would think, a lot of patients to be able to feel comfortable that you had really accomplished something with the bridge approach, even though it may save blood because we will be less nervous about it.

DR. CAPLAN: I think that is probably going to be it. Thank you.

I think we have got Dr. Carmichael next from the University of Toronto and Hemosol.

DR. CARMICHAEL: I would like to thank the organizers for inviting me to this meeting, to be able to present to you some of the thoughts and positions of Hemosol to the advisory committee. I would also like to thank you for the opportunity to stand up. Those seats are incredibly uncomfortable.

[Laughter.]

DR. CARMICHAEL: The concern about shortfalls of the blood supply is what brings us here today. Part of the solution may be to better manage the use of allogeneic or donor blood during elective surgery using strategies based on blood substitutes.

I feel that the blood substitute industry can help in this area since a large portion of blood use occurs perioperatively during elective surgery, approximately 50 percent or 60 percent of the blood consumed. Many patients receive only 1 to 2, perhaps 3 units of packed cells.

A number of clinical transfusion strategies have been developed and adapted by industry as presented today that will help to alleviate part of the concerns with respect to the blood supply.

Our goal is to reduce the use of allogeneic blood in the surgical setting. With registration, we will develop the products' broad potentials.

I was going to give you a quick overview of how the product is made.

Is there a pointer here? I do not have a pointer.

All right. You probably do not even have to see the slide. I will just go through it.

We start with outdated bank blood from a U.S. source. This is blood that has already been approved for patient use.

This blood is brought to us up in Toronto. We are the token Canadian company, and it is extensively washed and the cells are lysed. The hemoglobin is collected, converted to carboxyl hemoglobin, and then pasteurized at 62 degrees for 10 hours. This is followed by another step which is not shown in this slide of a viral filtration, using a viral guard filter. These two steps are to ensure the removal of viruses and bacteria from the hemoglobin.

We then take the product through two steps of chromatography. We end up with hemoglobin, greater than 99 percent pure. This is subsequently cross-linked with the trisaccharide raffinose, which has been oxidized to give you a bifunctional agent, which actually cross-links between the two beta chains into the 2,3-DPG pocket. It also gives us intramolecular cross-linking which gives us oligomers up to a molecular weight of approximately 500. The overall molecular weight of the product is approximately 150 daltons.

The product is then reconstituted in Ringer's lactate and is ready for infusion into patients.

The advantages of these products is similar to what people have said previously. The product is immediately available for use. It is a universal donor, essentially equivalent to O-negative blood. There is no cross-matching required. It is essentially free of virus and bacteria. It has a prolonged shelf life, up to a year or greater. Stored at minus-70 degrees, it has probably an indefinite shelf life.

Once you bring the product out of the fridge, it has sufficient oxygen delivery, the same as the day it was made, and as was mentioned earlier, much more efficiently than blood that has been stored for 10 days or more.

The low viscosity and small molecular size makes this product ideal for going into areas where red cells have difficulty penetrating. We think that it will help in the management of the red blood cell supplies.

I would like very briefly to go through some of our Phase II studies, just to give you an idea of what we have done, also.

Our first Phase II trial was an orthopedic surgery, looking at patients undergoing total hip and total knee arthroplasties. This was a multi-center trial carried out in Canada.

The design was to use interoperative autologous donation, IAD as we call it or hemodilution as Dr. Keipert called it earlier.

In this trial, we removed just 500 mL's of blood harvested from the patients, and it was replaced with increasing doses of Hemolink.

The patients were hemodynamically stable. There was no discernible evidence of clinical or biochemical abnormality to end-organ function, and there were no serious adverse events.

The G.I. upsets that we saw were primarily nausea and vomiting, and along with the G.U. hesitancy, they were probably related more to the dose of narcotic that was given for postoperative pain.

In a second Phase II trial that was carried out in the United Kingdom, again in orthopedic patients, we used a different design. This is the design where the Hemolink was used as a direct blood replacement or as a bridge to transfusion.

The transfusion trigger was 9 grams per dL, and as the patients bled down to that transfuse and trigger, they would then be randomized to either receive Hemolink or packed red cells.

Again, there were no limiting adverse events related to Hemolink, and patients were hemodynamically stable and end-organ function remained normal.

We have undertaken two trials in cardiac surgery, both in patients undergoing coronary artery bypass grafting. We have one study that is currently running in the U.S. and one that has just been completed in Canada and the United Kingdom.

Again, we used the interoperative autologous donation where 500 to 1,500 mL's were harvested from these patients prior to going onto bypass and being replaced with increasing doses of Hemolink.

This procedure has the particular advantage in cardiac surgery in that it protects that blood that is harvested from exposure to non-epithelial surfaces of the bypass pump, which would activate platelets, white cells, complement, et cetera. Then, when these patients required a transfusion, they would receive back their own fresh blood.

In conclusion, we felt that Hemolink could safety be used in cardiac surgery patients as part of the blood conservation strategy. We have now just embarked on our initial Phase III trial, which was based on the design of the trial I just talked about, and this, again, will be in patients undergoing cardiac artery grafting, interoperative autologous donation. Blood will be harvested prior to going on to pump and replaced with Hemolink. We anticipate the completion of this study early in 2000.

In our Phase III program, Hemosol focused on areas where there is frequent need for transfusion of blood. The two areas, of course, are cardiovascular surgery and orthopedic surgery. Although the amount of blood cells normally transfused per patient is usually small, being only 1 to 2 units, the overall impact on the blood supply could be great due to the large number of cases that are performed annually.

We use the approach of interoperative autologous donation, where the patients' own blood are removed or harvested just prior to surgery and replaced with an oxygen carrier and volume-expanding solution, namely Hemolink.

If the patient loses blood during surgery and requires transfusion, their own blood, complete with coagulation components, is immediately available to be returned to them.

The advantage of the blood substitute use during harvesting is that it allows for a greater amount of blood to be harvested or sequestered without incurring a severe degree of anemia.

With this IAD approach, the use of allogeneic packed blood cells for transfusion will be reduced, such that a smaller fraction of the blood supply will be used, making more units available for use in other areas and taking pressure off the current system.

The major determinants of Hemosol's reserve capacity in the next 1, 2, and 5 years will depend upon the availability of raw materials, our manufacturing capacity at Hemosol, and, of course, regulatory approval of the product.

The raw material requirements will require a collaborative an cooperative relationship between the blood collectives and the HBOC industry. In this slide, I have listed a few of the areas where we may cooperate.

We are currently using outdated bank blood which ranges from 1 to 2 of the total blood collections, amounting to approximately 250- to 400,000 units annually.

A coordinated effort with the collectors to get this material to industry would increase utilization of the current donations.

A number of collections are deemed to be not suitable for transfusion, such as underfills and overfills. If the collectives could make these units also available to the HBOC industry, it, too, would ensure a better use of donation.

A very rough estimation from these initial two points suggests that the blood collectors could supply approximately a half-a-million units annually, enough to produce 2- to 300,000 units of HBOC, which could significantly impact the supply of red cells.

Red cells obtained through therapeutic bleeds, such as those from patients with hemochromatosis that we heard this morning, with appropriate testing could be directed to further processing by the HBOC industry.

Currently, autologous and directed donations are dramatically underused, as was previously mentioned, and following appropriate testing again, these could also be directed to industry, even if outdated.

Blood collectors also normally target blood types that are in high demand, for example, Type O, and may defer other blood types, such as Type A and Type B that are used less commonly.

If the blood collection centers would recruit all blood types more aggressively and then direct the excess inventories to the HBOC manufacturers, this would allow for better resource utilization, and the collectors would not lose as a result of maintaining high, less usable, inventories.

This would require, again, a careful coordinated effort between the HBOC industry and the blood collectors, but could greatly enhance utilization of current inventories and expand further the quantity of donor blood that is available for use.

We could extend our blood collections to other countries where validity or validation testing for viruses is deemed adequate.

At Hemosol, we currently have our grow-your-own studies under way. We are looking at a long-term program to grow red cells ex vivo starting from stem cells. The results are promising so far. However, the project is not thought to come into fruition for another 3 to 5 years.

With the above considerations in mind, the projected reserve capacity for Hemolink is there. In the first year, this will be minimal. However, as our manufacturing facility comes on board in the next couple of years, we should be able to supply approximately 200,000 units. As we go forward into 5 years, we would project that anywhere from 400 to possibly a million units could be produced from our facility.

In summary, then, we are studying the efficacy of Hemolink in orthopedic and cardiac surgery, and using it to facilitate interoperative autologous donation as part of a blood conservation strategy.

We have learned considerable amount about the safety of our product in our Phase I and II studies over the past 2 years, and now we are moving rapidly into our Phase III trials.

We are enlarging our production facility and are developing a two-tiered system for manufacture in North America and Europe.

We look forward to working with the red blood cell suppliers and regulatory agencies to facilitate further contribution to the reserve capacity of the blood supply.

Thank you very much.

DR. CAPLAN: Thank you. That was a model of conciseness.

DR. CARMICHAEL: Thank you.

DR. CAPLAN: Questions?

DR. CARMICHAEL: I should have spoke longer because I do not want to go back and sit down.

DR. McCURDY: If you grow your own red cells from stem cells for this process, why don't you do it for direct transfusion? You could grow only O's and never have a problem.

DR. CARMICHAEL: That is a good point, and then we would not have the problem of the short half-life of the product. That is a very good suggestion. I will work on that one.

CAPTAIN RUTHERFORD: In your manufacturing process, you say you use Ringer's lactate as a solution. Are you looking at any other newer solutions that are coming out that are being FDA-approved?

DR. CARMICHAEL: We have not at this point, no.

DR. CAPLAN: All right. Thank you.

DR. CARMICHAEL: Thank you.

DR. CAPLAN: Next, we have Dr. Steven Gould from the University of Illinois and Northfield Laboratories. I am going to just let you know that at the end of that talk, I am going to have Steve chair the meeting until the end, for the very simple reason that I have to go watch my son get an award at a music thing. If I drive like crazy and they store enough Hemolink and Biopure up the road, I probably will survive the trip.

DR. GOULD: Thank you. We appreciate the opportunity to present the state of our work today.

I am going to share with you some new data that has not been previously presented publicly, which will demonstrate an increase in survival that occurs in patients treated with our product, following life-threatening blood loss due to trauma.

In the way of background, our material is human polymerized hemoglobin. Our starting material is human red cells, as shown here. The active ingredient, as we have heard today, is the four-part hemoglobin molecule known as the tetramer.

We lyse the cell and extract this protein, as we have heard. All human hemoglobin is essentially the same immunologically. That is the basis of the universal compatibility.

The problem has been historically not oxygen-carrying capacity. Any hemoglobin outside of the cell effectively carries oxygen. The problem as we have heard several times is that toxicity of this small molecular weight species of hemoglobin.

Our approach, based on nearly a quarter of a century of development now, has been to polymerize or clump these tetrameres using glutaraldehyde as the cross-linking agent into larger molecules of 2, 3, or 4 tetrameres locked together. There is both, in turn, in-trauma molecular cross-links that occur.

An important second step in our process is the purification to remove virtually all unreacted tetramere. Our final release specs call for well under 1 percent tetramere in the final preparation.

One unit of our material is shown here, and it is different. One thing that the committee should understand, a unit of blood has a distinct characterization. One unit of a blood substitute varies from sponsor to sponsor.

One unit of our material delivers 50 grams of hemoglobin and 500 mL volume at 10-gram concentration, and I believe it is the only blood substitute being developed that actually delivers 50 grams.

There is a reason for that. Our team has been surgically based, and we feel it is important to maintain the constant approach to a bag of red fluid. This is in vitro when in fused. In vivo, the increment is 1 gram per decaliter in the patient, the same as occurs with a unit of red cells. We think that is really crucial in an urgent setting.

The P-50 in our case is modified using pyridoxal phosphate. It is slightly right-shifted compared to red cells. It has an intravascular half-life in patients of about a day, and it has a shelf life in excess of one year.

So we consider this to represent the ideal resuscitative fluid. It allows simultaneous replacement of the lost volume in a hemoglobin that occurs following hemorrhage. It is immediately available due to its convenient storage properties. It is universally compatible. It is effective an ambient oxygen tensions. It does not require any supplemental oxygen, and I am going to share the data today that demonstrates it is safe during rapid and massive infusion, and is capable of supporting life in man in the virtual absence of any circulating red cells.

This slide indicates how the material will be used. On the left is the normal two components of our blood volume, the red cells making up 40 to 45 percent of the total volume, which we characterize by the hemoglobin concentration, to balance the plasma, normally amber.

The middle of the slide represents a 30-percent bleed which I have chosen because it is sufficient to cause a fall in blood pressure, and clearly, one bleeds both plasma and red cells. Upon arrival in an emergency room, the first step that occurs is an obligatory replacement of this lost volume to restore hemodynamics.

Currently, that is done with salt water to allow for the 45 minutes to 1 hour it takes to obtain fully compatible cross-matched cells. There are options, if necessary.

On the right is a characterization of how resuscitation occurs with PolyHeme. A single solution allows for the rapid restoration of the lost volume, but simultaneous restoration of the lost hemoglobin. The plasma turns red, and the red cell mass remains reduced. It has the potential to fundamentally alter the early care of the injured patient in the current settings, and because of its ease of use and convenient storage, it has the potential to be moved out of the hospital into settings where blood is currently not used for resuscitation.

This is a summary of our current active clinical trials. There are three. We are doing a Phase III trial on elective surgery, vascular surgery at a dose of 6 units. This includes using the product during both hemodilution and blood replacement.

The protocol calls this an aggressive hemodilution because of the significant concentration of hemoglobin being delivered. We obtain 6-unit hemodiluted blood from every recipient in the experimental arm. That goes without saying. We can obtain very large quantities of blood. That trial is underway.

The Phase II trial in trauma is ongoing at doses up to 20 units, as required, and is being given as blood replacement on an urgent setting following traumatic blood loss.

We are enrolling patients on an ongoing basis in a compassionate use protocol as life-saving therapy on a case-by-case basis when blood is unavailable due to primarily immunologic incompatibility.

We continue to publish our work in peer-reviewed scientific journals. Our two most recent publications are shown here. They document the absence of any serious safety issues.

In addition, they demonstrate the ability of the use of PolyHeme in this setting to reduce the use of allogeneic blood on essentially a unit-per-unit basis.

This slide summarizes our experience, excluding the Phase III data, which we are not privy to as the trial is underway.

From the outset, we have focused on the large volume use of PolyHeme. That is how blood is used. That is how we believe PolyHeme will be used, and we have been interested in obtaining this experience.

There are 151 recipients here. This is ongoing when I prepared the slide. We are impressed at the 53 patients that have received between 6 and 20 units.

For those not familiar with this, the normal volume of blood in an adult is 10 units. So there are 19 recipients who have received the equivalent of one complete blood volume exchange, and seven additional patients who have had up to two times their blood volume replaced with PolyHeme.

There has been an absence of serious adverse events related to the product. Based on that experience, we believe that the most compelling use for PolyHeme is to address a critical unmet medical need by providing oxygen therapy as support during a life-threatening blood loss when blood is potentially unavailable. The important clinical benefit will be a reduction in mortality in this setting.

I want to clarify a couple of points on the prior slide. Some might question that and the existence of that need. We believe the primary scenario for unavailability would be in remote locations, either in the civilian or the military setting. However, hospital settings in the urban environment occur on a daily basis. As a vascular surgeon, I can assure you that every day, unplanned surgical hemorrhage occurs in the operating room.

Issues of incompatibility, we have heard about over and over today. They occur in both acute blood loss and in chronically transfused patients for whom blood simply cannot be found.

Lastly, there are the inevitable issues of inventory and balance that occur, due to the vagaries of supply and demand.

It is difficult to precisely quantify. The point to emphasize, the incidence of unavailability of blood in this country is not zero.

With regard to the concept of a life-threatening red cell hemoglobin, I agree with the earlier comment about Jehovah's Witnesses. We have been interested in the response to progressive blood loss for the last 25 years. We have published a number of very, very useful, in our opinion, studies documenting the adaptation that occurs to progressive blood loss. We have also published clinical trials in patients with religious objection to blood, demonstrating that although patients can indeed adapt, there is a point below which that adaptation is no longer adequate and life is not sustainable.

The literature supports that in bleeding surgical patients, the mortality exceeds 80 percent when the red cell hemoglobin is below 3 gram per decaliter or a hematocrit of 10 percent. In some instances, in some of the reports, the mortality goes as high as 95 percent. That is a healthy patient.

With compromised cardiopulmonary dysfunction, we would expect this life-threatening level to occur at an even higher level.

Until now, the challenge has been how to design and conduct the story to address this critical need, specifically how could one answer the question of whether PolyHeme could reduce mortality due to insufficient red cell hemoglobin levels when blood is unavailable. We have conducted that study, and I am going to share those results with you now.

This assessment was done in a non-randomized protocol design that some might consider a simulated setting or a surrogate population based on the fact that blood was available, but not used.

We included patients who sustained substantial blood loss due to trauma, but did not receive blood replacement. Instead, their initial infusion of an oxygen carrier consisted of 6 units of PolyHeme or more, and I am going to report the mortality and compare that to the historical data.

This slide is worth a couple of moments because it explains how we are going to assess efficacy. Normally, the total concentration of hemoglobin in an individual is only the hemoglobin carried by the red cell. That is the only source we have.

When PolyHeme is infused, as we saw in an earlier slide, there are now two sources of hemoglobin. We have always felt that the physiology of oxygen transport by PolyHeme is equivalent to that of the red cell, and that there are in fact additive. I believe that the data we are going to share will emphasize that.

After the infusion of PolyHeme, a sample of blood can be obtained, the plasma and red cell separated, and the contribution of hemoglobin and oxygen due to each of those can be precisely quantified.

In essence, the equation explains our protocol. We are enrolling patients who are bleeding and losing red cells, but we are not transfusing them with blood. This, of course, would be unethical were we not replacing them with an alternative form of hemoglobin in an effort to obtain and maintain a satisfactory total concentration of hemoglobin.

So the question is what is an adequate total concentration of hemoglobin. These are the guidelines that were adopted in a 1988 consensus conference on the perioperative use of red cells that my colleagues and I were part of, and again, for those who are not familiar, I call this the short course in transfusion medicine.

The normal range of hemoglobin in adult men and women is between 12 and 15 grams per decaliter. As one bleeds, that falls. It is rare to give blood if the hemoglobin is greater than 10. It is rate not to give blood when the hemoglobin is below seven. There is a raging debate. I am going to circumvent the debate and define the therapeutically desirable range as somewhere between 7 and 10. As I have indicated, even in this consensus statement, it is clear that hemoglobins below 3 are life-threatening.

Now we can look at the data. We have enrolled 53 patients who sustained blood loss following trauma, did not receive blood, and were infused with PolyHeme, 53 who received 6 units or more. At the end of infusion, when we made the measurements, 27 of them had red cell hemoglobins less than 3 grams per decaliter.

I am going to start with a specific illustration to help you understand how we analyze the data. This is the young individual who received a high-velocity gunshot wound to the abdomen and arrived in the emergency room with a presenting total hemoglobin of 5.2 grams per decaliter, equivalent to an hematocrit of 15 percent, well below the acceptable level and approaching the lethal level. All of this hemoglobin was carried in the form of the red cell.

Our investigators were sufficiently confident based on their experience to take him to the OR to control his injuries and rapidly infuse him with 10 units of PolyHeme in 24 minutes. It is of interest that this is in fact more rapid than blood can be infused because of the flow properties of the material.

There are two observations I want to focus on. The first is that his total hemoglobin was increased to 7.5, back to the acceptable range, and essentially all of the hemoglobin was carried in the form of the PolyHeme.

His red cell hemoglobin was 0.7 grams per decaliter, virtually unmeasurable, clearly incompatible with life.

If I show you the group data now, this slide shows the concentration of hemoglobin on the Y axis. The blue band is the 7 to 10 therapeutically desirable range, and this is the mean point just prior to infusion in these 27 recipients. The mean value was 8.6 grams, precisely correct in terms of when surgeons should start to infuse.

The dotted line here is really a first-time observation. It represents what I might call the natural history of untreated blood loss, untreated in that these patients did not receive red cells. This is the study that one could not do previously.

It represents the red cell hemoglobin that would occur were these patients to have their lost blood volume replaced with salt water.

The final red cell hemoglobin is 1.6 gram per decaliter as a mean for the group. In contrast, the total hemoglobin is maintained in the therapeutically desirable range. It is this delta or this increment that is the benefit of the PolyHeme in this setting.

Twenty-seven bleeding patients did not receive blood and had an adequate total hemoglobin concentration.

DR. CAPLAN: Dr. Gould, just for curiosity, why didn't they receive blood?

DR. GOULD: It was part of the protocol.

DR. CAPLAN: No, no, but I mean they were randomized with--

DR. GOULD: No. Let me go over it again. This was a non-randomized trial design. This was done in trauma settings after IRB approval when informed consent was obtained. The patients were told that they were to receive a developmental product in lieu of red cells. They made a conscious choice or a family member made a conscious choice, but the point to emphasize, Dr. Caplan, is that consent was obtained in every individual.

This is the most important slide I am going to show you today. This is the mortality outcome. Remember, we are talking about 27 individuals who had a red cell hemoglobin below 3 grams per decaliter in whom the expected mortality is greater than 80 percent.

Four of these 27 individuals died from a mortality rate of 14.8 percent. This figure is compatible and consistent with the mortality rate reported in the literature from major trauma series when blood is used and represents death due to the injury.

I want to break it down by subset. Of the 27, 20 of the individuals had red cell hemoglobins below 2. There were three deaths from mortality of 15 percent, and perhaps most impressive, five individuals, including the sample I showed you, had red cell hemoglobins below 1 and not a single patient died.

There is no magic here. We believe that the mechanism of this increase in survival is the ability of the PolyHeme to maintain an adequate total concentration of hemoglobin in the absence of red cells. So total hemoglobin never got below 3.

We think this evidence definitively makes the case that the PolyHeme is effectively delivering oxygen in this setting.

For a comparison, this is the outcome for the entire 53 recipients. There were nine deaths overall for a mortality of 17 percent. In the 26 recipients whose hemoglobin was above 3, there were five deaths for a mortality of 19.2 percent. The point to emphasize is that there is no increase in mortality above this otherwise critical level of red cell hemoglobin.

I am also going to do a comparison to the literature based on a paper that we published in the New England Journal in 1986 where we studied seven individuals with religious objection to red cells, all of whom had red cell hemoglobins below 3. Six of the seven died for a mortality of 85.7 percent. Our observed mortality here was 14.8 percent for an absolute risk reduction of 70.9 percent, with a P-value of .001. We all understand P-values.

I want to talk about the concept of clinical significance and use a paper written by Sackett in the British Medical Journal in 1995 which talks about the number-needed-to-treat concept. The number needed to treat to achieve a benefit is the reciprocal of the absolute risk reduction. It is a way of describing the relative benefit of an active therapy compared to a control group.

In this instance, the number is 1.41. What this means is for every 1.41 patients treated, the benefit of the active therapy will be achieved. Specifically, every 1.41 patients in this setting treated with PolyHeme, one life will be saved.

Importantly, there were no serious adverse events related to the infusion of PolyHeme in this trial; most importantly, the traditional adverse events that have occurred with other hemoglobin solutions.

In summary, we think this data illustrates that PolyHeme reduces mortality for red cell hemoglobins less than 3. By maintaining an adequate total concentration of hemoglobin, I think the data points out that the product is safe during rapid massive infusion and essentially resolves the issue of unavailability of blood.

We believe it is fair to conclude that PolyHeme successfully addresses a critical unmet medical need.

Lastly, we were all asked today to address the potential impact on the blood supply. I would like to break it down into two discussions, one, on conservation of the present supply and, two, on the expansion of the blood supply.

Conservation, I think based on the data I have shown with the doses I have shared with you, we should have a significant impact on reducing the amount of blood that is used in patients.

Previously, most of the attention has been on the benefits that accrue to individual patients. While we agree with that, we are glad to have the opportunity to talk about the benefits to the system. We expect to eliminate the use of uncrossed match blood based on the availability of this product.

We think that because of the volumes, we are able to safely and rapidly infuse. It will have a substantial impact on reducing the use of blood in the urgent setting, and I have not even begun to address today the use in the elective setting.

Expansion is actually a bit more complicated. It is clear that we use blood as our starting material. So some might question how we are actually going to help the blood supply. This requires, I think, an understanding of the way the current blood suppliers, collectors actually work. I make the analogy that blood is a highly perishable commodity. Up to day 42, it is great. On day 43, it has to be discarded. It is not like a supermarket where you can drop the price and put it on sale.

So the blood collectors today use the concept of minimal outdate to do their collection. In fact, although we talk about blood shortages, we know that we collect close to 2 million units in excess of the number of units transfused every year because it is based on the concept of shelf life and inventory balance.

We propose to convert in collaboration with the collectors to the concept of maximal collection. We can use blood that is outdated, and there is a precedent for this over-collection in place already.

The incidence of O-negative blood in the population is about 6 percent. The incidence of O-negative blood in collections today approaches 12 percent. The reason for that is that all of it is used.

The collectors have estimated with us that overall collections, if they do not need to worry about outdating, could be increased anywhere between 10 and 50 percent. So we think there is an opportunity in collaboration to enhance overall collections.

In closing, just to summarize, I think the clinical development of PolyHeme has the potential to impact in a number of ways. I think most important to us is the ability to address this critical unmet medical need, but overall, we feel we can reduce waste, enhance economies of scale for the reasons I have outlined, facilitate inventory management for the collectors.

We think PolyHeme will expand the availability of oxygen carriers to new settings and resolve the issues of inadequacy or unavailability of oxygen carriers in the present setting.

Thank you.

DR. CAPLAN: Thank you.

Questions?

DR. CHAMBERLAND: I may have missed it, but either in your current trials that are ongoing or trials that are planned in the future, do you plan to use a design which would involve randomization in which one group of patients would get your product and others would be transfused with red blood cells?

DR. GOULD: We have done a randomized trial. One of the references I showed you was in a randomized trial.

Our Phase III trial is randomized where a group will receive, if they need it, allogeneic red cells. The point of this trial is to do the trial in the setting when blood is unavailable.

We feel that scientifically comparing this outcome to a group receiving red cells is not appropriate. The appropriate scientific control is patients who would receive salt water and not red cells. So, for efficacy, this is the way to do the trial.

DR. CHAMBERLAND: I am also not certain, though, that it is fair to compare your observed mortality in this trial with historical controls of some 15 years ago because a lot of other things have changed in treatment of trauma patients and whatever. So there is not a good bench mark here.

DR. GOULD: I agree with your comment.

I put that slide up to give you an example, and of the studies in the literature, I selected that one for several reasons. First of all, I was the principal investigator. So I agree with what you said. I can tell you that the approach of taking care of the patients was the same.

I fully, fully acknowledge the importance of simultaneous controls. I do not think that I can design a study--I know I cannot--where I take a group of patients today and treat them without giving them blood unless there are religious objections. The majority of patients with religious objections will not take this product. So I use it for that purpose. It was in our hands in a group that was carefully studied with full instrumentation.

The value is that it supports all of the trials in the literature documenting that hemoglobins below 3 are incompatible with life. If I were talking about hemoglobins of 7, I agree with your earlier comment. That is not a sufficient level to talk about increasing survival. So I share your comments. I put that up to help put this in context of people who may not be familiar with the literature.

DR. CAPLAN: Thank you.

Our final speaker of the day is Robert Winslow, and he is from Sangart, Incorporated.

DR. WINSLOW: Thank you very much.

I am going to be mercifully short, and so I am going to hold this microphone because it just slipped off.

I want to thank Steve Nightingale for the opportunity to make a few remarks at the end of this very interesting afternoon, and the reason I asked to be at the end is because I hope that the comments I have will be directed more at the future toward products that are coming long and are in the academic, shall we say, pipeline.

I am going to talk about some work in very short summary that has been carried out in a continuous publicly funded program that began in 1985, the group of my colleagues at the Letterman Army Institute of Research in San Francisco, and continued from 1985 through about 1993 when that facility was shut down.

The program continued with substantially, that is to say, some of the same personnel, myself and a couple of others at the University of California-San Diego. From 1993 to 1998, we worked collaboratively between the Department of Medicine and the Department of Bioengineering. That work was primarily supported by a program project grant from the NIH.

It is impossible in a short time to summarize all of that work and what it mean, but I hope to make a comment on a couple of salient points, if I can find the pointer.

In order to make this summary, I want to direct my comments toward three model molecules represented here in scale model drawings. On the left is cross-linked hemoglobin with a molecular weight of about 64,000, a molecular radius of about 2.7 nanometers, a viscosity of 1 centipoise, and a molecular volume of 860 cubic nanometers.

Polymerized hemoglobin, this particular model polymerized with Ring-open raffinose, has a molecular radius twice that big, almost 4.9 nanometers, molecular weight of about 160, viscosity a little higher, 1.36. By the way, the viscosity of water is 1, by comparison. It had a volume of 4,000 cubic nanometers.

Then, this very large molecule, which is a molecular at surface decorated with polyethylene glycol, or PEG, has a very interesting characteristic. Each molecule of hemoglobin is surrounded by an immobilized shell of water, which gives the molecule a very, very large diameter of 14.1 nanometers, even though the molecular weight, 123 kilodaltans is smaller than the polymerized hemoglobin, and its viscosity is very high, 3.39, and its molecular volume is 93,000.

This high viscosity you might think is a bad quality for a blood substitute. In fact, it is not. The viscosity of whole blood under the same conditions is a little bit less than 4. So this has about the same viscosity as blood.

Studies in the microcirculation that were carried out in the Bioengineering Department at UCSD have been, I would say, the most interesting and revealing to us, and I show this summary slide which shows in a hamster, a skin-fold model, a capillary blood flow reducing from 100 down to 50 percent on the Y axis, as animals are progressively hemodiluted from a hematocrit of around about 50 which is normal for a hamster down to around about 10.

As you hemodilute with dextran as a control, the white circles, we see the capillary blood flow drops, as is well known to occur in shock. However, if you hemodilute with the low molecular weight or small hemoglobin, in this case, the alpha-alpha hemoglobin developed by the Army, similar to the Baxter product, we see a tremendous drop in capillary blood flow.

The larger molecule, polymerized hemoglobin has much less of an effect, and the largest molecular has even less effect than hemodilution with dextran.

These results correlate very well with blood pressure results that you have heard a number of speakers talk about. This is a series of experiments done in rodents in which exchange transfusions are carried out to the tune of 50 percent of blood volume, followed by a rather severe 60 percent of blood volume hemorrhage.

This model is chosen because it is about 50-percent lethal in the hour past transfusion which makes it a very good model to test overall efficacy in animals.

As you can see, the blood pressure drops immediately on hemorrhage in control animals. Their hemoglobin at the beginning of this hemorrhage is about 14. With cross-linked hemoglobin, that is to say the low molecular weight alpha-alpha, there is the marked rise in blood pressure follow by a moderate drop during hemorrhage, although the blood pressure does not really drop below baseline. You heard Tim Estep talk about this property earlier.

These animals have a hemoglobin of about 10, but a plasma hemoglobin of about 3.9. With polymerized hemoglobin, as you have heard other speakers refer to, the rise in blood pressure is significantly less, but on hemorrhage, they also drop their blood pressure. As in both of these cases, the larger standard errors here occur because animals are dying.

In the polymerized group, initial hemoglobins are 10.8, with a plasma hemoglobin of 4.8.

The most impressive part of this slide is this last panel. This is the PEG hemoglobin, the very large molecules, and you see this very, very impressive hemodynamic stability.

There is no rise in blood pressure to speak of, just a very, very small bump, virtually no mortality, as I will show you in a subsequent slide, and tremendous hemodynamic stability, despite the fact that these animals have a total hemoglobin at the beginning of hemorrhage of only 7.6 grams and a plasma hemoglobin of 1.9.

If you look at the survival curves in animals, this is a very short-term experiment, but similar to what one might expect in a hemodilution hemorrhage kind of protocol. The animal surviving, again for normal controls, is about 50 percent. A dramatic loss, all animals die in the alpha-alpha or small hemoglobin group. Polymerized hemoglobin does better initially, but eventually all those animals die as well, with the remarkable survival of animals transfused with PEG hemoglobin.

If I had another hour, I could explain to you all of the theoretical reasons for this, but I think we understand this now quite well. To summarize, I made this bullet slide which gives you what we believe are the salient points in design of an efficacious hemoglobin-based blood substitute.

First of all, we believe that maintenance of capillary blood flow is the most significant parameter. It correlates best with survival. It correlates best with acid-based balance, and also results like that have been seen in human trauma studies which relate survival to acid-based balance.

Maintenance of capillary blood flow has bene shown now quite clearly to be dependent upon the limitation of facilitated oxygen diffusion, as well as facilitated HbNO. diffusion, or NO scavaging. These can be limited by two physical means. One is increasing molecular radius according to Stokes' law and the other is increasing the viscosity according to Einstein's law of viscosity.

Viscosity is apparently much more important than we would have expected previously based on simple hemodilution experiments and is a key property having to do with maintenance of sheer stress in the microcirculation.

Finally, we believe the second major component that is necessary is maintenance of vascular volume by providing oncotically active material in hemoglobin.

Using those principles, we now go back and say can we design an oxygen transport solution that would have superior properties, actually superior to blood in certain circumstances. I am going to show you one such formulation which is simply a mixture of hemoglobin with a very low hemoglobin concentration of 2.5 grams percent with an equal volume of Pentastarch. This is a hydro-zyethyl starch, a molecular weight of about 250,000, 5 grams percent.

This solution has the same viscosity as blood, about 3 centipoise compared with 3.5 for blood. It has a very low P-50, which again I am not going to go into in detail, but we believe this is an essential component, a low P-50 relative to red cell P-50 and a high oncotic pressure.

This solution, containing a very small amount of hemoglobin now, when given to animals, results in very impressive survival curves. Again, this is survival fraction remaining alive in rats after several hemorrhage, and we see, again, about 50 percent of animals dying in the control group who begin with a hemoglobin of about 14, compared with animals exchange-transfused with a product I just described which we call "hemospan" because it is a combination of hemoglobin and Pentaspan.

These animals begin their hemorrhage with a hemoglobin of only 5.6 grams, a hematocrit of 14.8 percent, and a plasma hemoglobin of only 1 gram per decaliter.

As you can see, the survival is actually slightly better than the controls, despite this very, very low level of hemoglobin.

The control I show in this slide is Pentastarch alone because it has the same oncotic pressure and the same viscosity as this combination solution, but without hemoglobin.

What this slide is telling us is that the simple addition of only 1 gram per decaliter of sulfur hemoglobin in a solution that is properly formulated with regard to viscosity and oncotic pressure is extremely effective in this particular model of severe hemorrhage.

What does this mean? The reason I came to show you these results is the following. A product like this with a very small amount of hemoglobin, these actually contain only about 10 grams of hemoglobin per 500-mL unit, which means that you can produce 3 to 5 units of this product from every unit of donor blood, either outdated blood or whatever source, which is a very, very efficient process. It means that the cost of the product will be very low because the low cost of raw materials. The dosage of hemoglobin is low, which means lower toxicity. Of course, it can be sterilized, and this particular product can be stored lyophilized.

I think the bottom line is that this means that products like this, when they find their way into clinical use, it will mean a much more efficient management scheme for blood in general.

Finally, I know you cannot read this literature, but I want to just say that every logical step that I have described to you has been documented in the peer-reviewed literature, and if anybody is interested in this information, I would be happy to provide it.

Thank you very much, and thanks again, Steve.

DR. NIGHTINGALE: Thank you, Dr. Winslow.

Questions for Dr. Winslow from the Advisory Committee?

Captain Rutherford?

CAPTAIN RUTHERFORD: The FDA recently approved Hexstan as a new plasma volume expander. Have you considered trying diluting Hexstan with your hemoglobin solution?

DR. WINSLOW: Actually, we have looked at a number of starches, and the critical property here is viscosity. Oncotic pressure is important as well.

Headstarch, which is the primary component of that product, does not work in this formulation because of those two properties.

DR. NIGHTINGALE: Are there any other questions?

[No response.]

DR. NIGHTINGALE: I do not see any.

On behalf of Dr. Caplan and the entire membership of the Advisory Committee and the Department of Health and Human Services, I really sincerely want to thank all of the presenters of the last 2 days of the meeting.

For the public, the unedited transcript of the meeting is posted on our web site, www.hhs.gov. There is a forward slash to "partner," then a forward slash to "Blood Safety," one word. We hope that will be posted by close of business on Monday.

There will be hopefully an approved summary of the meeting in accordance with the CFR for public meetings, which we plan to post by close of business on Wednesday.

As we had alluded to earlier, this meeting was designed to prepare us for a briefing of the Surgeon General who is the Blood Safety Director of the Department of the Blood Safety Committee on the Reserve Capacity of the Blood Supply. However, it is obvious that it would appeal to a broader audience, and we will be exploring with the presenters the possibility of various ways of presenting this to a wider audience, including publication in print or on the web.

In advance, I will thank all the presenters for their continued cooperation as best as they can with that plan.

With that, I would entertain as the Acting Chair a motion to adjourn.

DR. FEIGAL: So moved.

DR. KUHN: Second.

DR. NIGHTINGALE: And seconded.

All in favor?

[Chorus of ayes.]

DR. NIGHTINGALE: We are adjourned, and thank you, once again.

[Whereupon, at 4:36 p.m., the Advisory Committee was adjourned.]

- - -