September 12, 2000

 

Stuart L. Nightingale, M.D.,

Office of the Assistant Secretary for Planning and Evaluation,

Department of Health and Human Services,

200 Independence Avenue, S.W.,

Washington, DC 20101, Fax:202 205 8835

 

Dear Dr. Nightingale,

 

The following document was composed to address certain questions that arose during the recent conference “Human Subjects Protection and Financial Conflicts of Interest” that I attended.  It includes a description of the point of view of the investigator as well as suggestions for improvements in the process.  I hope you find it interesting.

 

TRIALS AND TRIBULATIONS

 

Having attended the aforesaid meeting at the Natcher Center I decided to address one of the questions receiving little attention in the deliberations.  What are the views of actual academic clinical investigators on clinical trials and associated conflicts of interests?

 

            Why would academicians engage in clinical investigation anyhow?  The benefits  tobenefits to sponsors, clinical research corporations and hospitals are obvious.  The sponsor gets FDA approval for a monopoly, the institution gets overhead, the hospital gets to do procedures at the retail rate.  Participants often perceive that they receive additional or especially good health care, the latest therapy, and an opportunity to be altruistic. Additionally, for some research subjects this may be their predominant form of medical care.  For academic physicians what are the motivations?

 

            I have carried out clinical trials or have been involved in the design or write up of clinical studies for six companies, and am involved in a large-scale multi-site clinical study with the NIH.  Clinical trials and NIH clinical studies are two different organisms that must be discussed separately.

 

            Sponsor-initiated clinical trials  

 

Generally, such studies are designed by big or small Pharma after negotiations with the FDA, either to obtain approval for an agent, or in stage four trials to expand the database, or gain a new indication.  While some companies sometimes fund add-ons for individual sites, most of the time the goal is plain: to quickly accumulate enough patients to fulfill the study design.  This leaves the investigator without the creative pleasure of designing a study.  Academic institutions know that, and little academic credit is given to participation in a clinical trial.  I am asked to participate because I am already an expert in the field and my name will have worth to the sponsor, but that does me little good.  I get to go to one or two investigator’s meetings a year but I already eat too well and another hotel room in another city does not thrill.  In fact, those meetings are mostly bores and much too long for the purpose, which is to provide information about how to do the trial and to share information as it progresses.  For those of us who are active clinically, the honoraria do not cover lost clinical income or ingratiate us with our patients so, by the way, they should not necessarily be considered to create a COI but rather be considered compensation for time and effort.

 

            Signing up to do a clinical trial creates much trouble and stress.  Achieving a contract between a sponsor and the university is often agonizingly slow.  Part of the reason is that the provisions of the sponsor and the institution are often incompatible.  Master contracts have helped with this problem but the delay is palpable.

 

In our institution (and all of UC) there must be no profit derived from a clinical trial although there is often a little money left over that can be applied to educational and research purposes. Contrary to expressed opinions at the meeting, the academic IRB really acts as spokesperson and protector for the study participants, investigator and sponsor be damned! Good for them! But that means substantial delays prior to approval of proposals and consent forms.

 

Yet more delays are experienced. In an NIH study, the payment schedule begins when the study is initiated and much preliminary work can be done before enrollment of subjects begins.  In sponsor-directed clinical trials the clock begins when the first of many centers begins to accept patients. The longer the delay, the longer it takes to begin to accrue participants, causing much anxiety.  Study staff get paid regularly without regard to the trial start date, resulting in increasing costs as the trial start date gets pushed back.  If these time delays are significant, accrual times are so diminished that recruitment is insufficient and the study costs more than it earns.  While five thousand dollars gross per subject may seem like a lot of money, my guess is that except in the most strictly managed academic programs, trials often lose money.

 

 

            Sponsored trials also include long visits from the sponsor’s reviewers that eat up much coordinator time and sluggish payout, based on work already completed. 

 

When the data are collected at study completion, the frustration may build rather than diminish.  All the data are accumulated in the bowels of the sponsor and belched out according to the sponsor’s needs.  The speed of production of publications parallels the conformity of the data to the needs of the sponsor.  Negative studies are rarely completed for analysis.  I am still waiting for one after five years – and we thought the study to be a qualified success.  Repeated inquiry produces only excuses.

 

            I have volunteered to write up more than one study, but the sponsors demurred.  Reasons given included; “You academics take much too long.”  “ We also need papers for our careers.” “We have the statistical expertise in-house and they have to have something to do.” There is no recourse.  Information is power and total information is total power.  Eventually I got to be an author on a number of abstracts and papers describing the results of the research, some of it of moment.

 

So the investigator’s motivation is to obtain a modicum of academic support and give one’s patients access to promising technologies. Participation rarely pays off.  Clinician academicians could do it for the visit fees, which exceed those obtained from practice but usually there are not many of these, in comparison to coordinator visits, so the financial reward is limited. Some investigators become lecturers, which is more lucrative, but academicians do not have to carry out clinical trials to join the lecture circuit. To demonstrate the limited benefits, UCLA School of Medicine developed a clinical trials center a couple of years ago, assuming that faculty would flock to trials offered by sponsors.  No such thing.  Only when the investigator has been sought out and the trial was in her area of specific scientific interest were contracts concluded.  The Clinical Trials Center has brought a high degree of fiscal responsibility to the trials enterprise and has protected investigators from many of the hassles associated with moving a proposed trial through the system.  The study coordinators working for the Center constitute an independent group whose income does not depend on an individual protocol. They are instructed to consider the participants first and to execute studies entirely according to protocol. 

 

I doubt that I will do another sponsor-developed clinical trial.

 

NIH-funded clinical studies:

 

            An NIH-based study is another matter entirely.  I chair the Ethics Committee of SWAN, the Study of Women’s Health Across the Nation. The rights of participants have been carefully considered at each step of protocol development and execution.  This includes such matters as how to communicate intelligibly, what is fair to ask, what to require in the consent, what information to distribute to participants for transmission to their health care provider and how to couch the notifications.  We studied how to provide research results to the participants in a comprehensible manner and how to anonymize or functionally anonymize data relating to stored samples to be tested in the future.  Investigators are compensated generally in proportion to their time and effort.  Most important, the investigator group has control over the date and the manner of publishing it.  SWAN’s commitment to the science and the joint responsibility of investigators for the integrity of the results, ensures an effort to tell the whole scientific story very well.

 

            So what would I suggest that the government and the scientific community do to foster the clinical trials necessary to translate scientific advances:

 

  1. It is the academic investigators who provide probity to such studies.  They require a modicum of inducement to conduct a trial.  Compensation for work, time, and meetings should not be construed as a COI.

 

  1. Consultations associated with clinical trial study design constitute a service to the public and compensation for this service is well deserved.  Each institution should have the freedom to decide how to deal with this purposeful compensation.

 

  1. Investigators with general consultancies to or any equity in a sponsor should not personally conduct a clinical trial of the sponsor’s drug or appliance. Not only would their involvement tend to diminish the perceived quality of the results, but it also would serve little purpose.  Let others do the grunt work and avoid a serious COI.

 

  1. The scientific community should request of its journals that they require full disclosure in the reporting of clinical trials results, including who owns the data, who did the analysis, who wrote it up and the contributions of the named authors.  This degree of disclosure is consistent with the “culture of disclosure” we agreed upon at the meeting. 

 

Many journals are owned by scientific societies.  Writing disclosures into the rules should be simple and if the NEJM and the JAMA do it, the others will follow suit.

 

       5.  A more radical (better but less likely solution) is to require that all multi-site Phase 2 and 3 studies be analyzed and written up by an impartial group.

 

6.            Regarding the term “COI” I agree with Maria Angell that it resides in the circumstances and not in any effect on scientific objectivity. That is why disclosure is so important – because the judge is the reader or the participant not the institution.

 

7.      We should explicitly indicate that the investigator responsibility include personal evaluation of ethical status and scientific validity of the proposed study and their perception of research equipoise.

 

8.   Adverse event reporting, a critical issue that was not discussed at the meeting, can create a very serious C O I for investigators conducting clinical trials.  Adverse events are either minor or serious, related (to the trial drug) or unrelated.  Events must be recorded quickly and accurately, and transmitted both to the sponsor and to the IRB. The sponsor must notify the FDA within set timelines. Hospitalizations, for whatever reason (related or unrelated), are always serious adverse events.  Sponsors do not like adverse events especially related ones.  When informed of such an event, they may inquire as to the investigator’s degree of certainty that the event is related.  They may even ask the investigator to get permission from the participants to have their medical records scrutinized by the sponsor or designee.  These additional exercises may influence the investigator to classify the adverse event as unrelated.  Furthermore, reporting too many related adverse events may brand an investigator as overzealous and reduce access to additional studies.

 

      This COI directly interferes with the integrity of the research record and puts participants at undue risk.  It is clear that for the commercial clinical trial organizations, distortion of the research record may be systematic and pervasive. The system may not work well for academic investigators either. Empirical studies could readily shed light on this problem.  I propose that the FDA carefully examine its adverse event reporting system to see whether a method could be developed to ensure objectivity.  For example, all adverse event reports could be sent via the Web directly to the FDA Medwatch system where they could be compiled and stored for analysis and aggregation.  The reports could be sent anonymously to the Sponsor and to all the involved IRBs electronically.  This system would eliminate huge amounts of paper work, improve the detection of serious problems and make it easier for the investigators to give objective reports.

 

I hope that this is interesting to agencies involved in regulating clinical trials.

 

Sincerely yours,

 

 

 

 

 

Stanley G. Korenman, M.D.

Professor of Medicine

      Associate Dean, Ethics and Medical Scientist Training

      UCLA School of Medicine