Good morning, Mr. Chairman and members of the Committee. I am Marc Manley, Acting
Associate Director of the Behavioral Research Program at the National Cancer Institute. I have
worked in NCI's tobacco control research program for more than 10 years. I am very pleased to
be able to testify before you today about one of the most serious public health problems facing
On September 19, 1997, the President called for comprehensive tobacco legislation with a goal
of reducing the smoking rate among young people by 50 percent within seven years. The
President stressed that the following five key elements must be at the heart of any national
- A comprehensive plan to reduce teen smoking, including a combination of penalties and
price increases that raise cigarette prices up to $1.50 per pack over the next 10 years as
necessary to meet youth smoking targets;
- Express reaffirmation that the FDA has full authority to regulate tobacco products;
- Changes in the way the tobacco industry does business;
- Progress toward other critical public health goals, such as the expansion of smoking
cessation and prevention programs and the reduction of secondhand smoke; and
- Protection for tobacco farmers and their communities.
The National Cancer Institute conducts and supports research, training, health information
dissemination, and other programs with respect to the cause, diagnosis, prevention, and treatment
of cancer, rehabilitation from cancer, and the continuing care of cancer patients and the families
of cancer patients. Tobacco use is responsible for approximately one-third of all cancers.
Foremost among the cancers caused by tobacco is lung cancer, which is the number one cancer
killer in both men and women. While certain other factors, such as radon and asbestos, have
been linked to lung cancer, far-and-away the most significant risk factor is cigarette smoking,
which has been linked to about 90% of lung cancer cases. Over 171,000 new cases of lung
cancer are anticipated in the US in 1998 (nearly 14% of all cancer diagnoses) with over 90,000
men and over 65,000 women expected to die of the disease in this year alone (28% of cancer
deaths). Lung cancer is among the most tragic of cancers in that it is largely preventable, and
once contracted, current interventions are of very limited efficacy with 5-year survival of all
stages combined being only 14%. In addition to lung cancer, mortality attributable to current
and past tobacco use has also been linked to cancers at other sites including lip, oral cavity,
pharynx, esophagus, pancreas, larynx, trachea, cervix uteri, urinary bladder, and kidney. Cancers
at several other sites have also been suggested to be linked to tobacco use and these putative
associations are currently under study.
The exposure to the cigarette smoke of others is also a significant issue. More than 3,000 lung
cancer deaths each year are attributable to exposure to the smoking of others (secondhand
NCI Tobacco-Related Research
The NCI has been a leader in tobacco-related research for three decades, and will remain
committed to reducing the burden of all cancers. The NCI currently supports research related to
tobacco totaling more than $85 million. Included in this portfolio is a broad range of research on
the molecular and genetic factors related to tobacco-related cancers; the mechanisms of cancer
causation and progression related to tobacco use; chemoprevention studies looking at the ability
of certain agents to prevent the onset or progression of tobacco-related cancers; population-based
studies of the genetic and environmental factors related to cancer, including behavior and other
lifestyle factors, and factors such as occupation, exposure to environmental tobacco smoke, and
differences in tobacco-related cancer rates among minority populations; clinical intervention
studies including smoking cessation, nicotine patch and self-help treatment, interactive video
interventions, and others; and community-based interventions in smoking cessation, particularly
those aimed at youth through mass media and schools.
More than a decade of research on youth and tobacco has made clear that a comprehensive
approach is what works best. Although we know that a price increase is one of the most effective
single interventions to reduce teen smoking, it should be augmented with other kinds of
interventions, such as educational programs, mass media programs, and restrictions on tobacco
advertising and sales to minors. The President's recommendations to Congress about the
essential elements of a tobacco control program are certainly consistent with the scientific
I would like to mention two specific initiatives that demonstrate NCI's commitment to reducing
tobacco use at the community level and especially among youth.
First is ASSIST, the American Stop Smoking Intervention Study for Cancer Prevention, a
community-based intervention developed by NCI to reduce mortality from cancer caused by
tobacco involving more than 6,200 organizations in 17 states. The ASSIST program may be the
most widely recognized of all of NCI's tobacco-related research projects. It is based on more
than a decade of research which identified the most effective interventions to reduce tobacco use.
A 10-year national demonstration project, ASSIST has made youth a major focus of its
interventions. Through media and educational campaigns, as well as efforts to strengthen
enforcement of local anti-tobacco laws targeted at youth, ASSIST will reach 91 million people -
approximately one third of the U.S. population, stop two million youths from becoming addicted
to tobacco products, and prevent nearly 1.2 million premature deaths. ASSIST, a partnership
between the NCI and the American Cancer Society, has put the most effective techniques for
reducing tobacco use into the hands of thousands of people and organizations in 17 states.
From an interim analysis of the ASSIST program (Impact of the American Stop Smoking
Intervention Study on cigarette consumption, in press), we have evidence to suggest that
ASSIST is associated with a substantial difference in tobacco consumption in one-third of the
population of the United States. While these conclusions are only tentative, in combination with
data from some larger state-wide programs, they suggest that public health programs may be
effective in reducing tobacco use in large populations. This is very encouraging news for the
public health community, and we are looking forward to further analyses of this program as we
gather more information. We hope that ASSIST will be a model program that can be applied
throughout the United States.
Prevention and Cessation of Tobacco Use Among Children and Youth
One of NCI's top priorities for behavioral research is the reduction of tobacco use by youth. A
new research program was launched in 1997 specifically to address the gaps in our knowledge
about tobacco use by young people. A request for applications (RFA) was released in 1997 and
again this year to stimulate this critical area of research. More than $8 million in grants will be
awarded this year to support a wide spectrum of studies. Included in this portfolio are studies of
youth smoking cessation programs, determinants of teenage smoking behavior, the impact of
tobacco advertising, and the impact of reducing tobacco sales to minors.
Even with the research in place as described above, there is much more that can be done.
Research to Reduce Tobacco Use
There are many opportunities for research that will help us understand why children smoke and
how we can help them avoid tobacco use. A comprehensive tobacco control research portfolio
can contribute to real reductions in tobacco use by linking to ongoing and expanded intervention
programs. An innovative tobacco control research program is essential to guide new policies,
regulations, and programs that are being supported by public and private funds. Research must
address issues at the national, state, local, and individual level, and must include studies of adult
tobacco use and its impact on youth. Studies are needed that span the spectrum from basic
through applied research in both the biological and behavioral sciences. Of particular urgency is
the need for science-based information to inform federal and state policies and programs that are
now under development, including public education campaigns and smoking cessation programs.
A comprehensive program aimed at tobacco control research would include new research aimed
at understanding the initiation of tobacco use and nicotine addiction. Studies must also address
questions related to racial, cultural, and gender influences in youth tobacco use including the
impact of new, proposed, and future tobacco product regulations. It will be critical to assess via
research the impact of new tobacco products and promotions on initiation and cessation rates.
While many teenage smokers have tried to stop, most do not succeed. Research is desperately
needed to guide interventions in this important field. Clinical studies are needed to develop new
treatment regimens including pharmacological studies of nicotine delivery devices and other
drugs either separately or in combination with a variety of behavioral interventions. It will also
be critical to assess the success of community intervention programs and to develop new
statistical methodologies for this assessment. Using the methodologies developed as part of the
NCI's ASSIST program, the relative contribution of regulatory and program interventions to
changes in tobacco use can be assessed. As new interventions are developed, new assessment
tools will need to be developed in concert. As we have seen in the ASSIST program, it is
possible to conduct research programs that generate useful knowledge and improve the public's
Expanded participation in clinical trials relevant to tobacco-related cancers
For more than 30 years the NCI has supported a successful cooperative multicenter trials
program in treatment and, more recently, in prevention. Operating at hundreds of sites in North
America, NCI's cooperative groups have generated much of the evidence on which the practice
of oncology is now based. It is also currently the platform for two very large trials aimed at the
prevention of breast cancer and prostate cancer, respectively. The clinical trials program has
always paid close attention to tobacco-related cancers, and active programs for cancers of the
lung, bladder, head and neck, esophagus, kidney, and pancreas have been in place for years.
To meet these challenges, we must have a clinical research base that can bring the best of our
emerging knowledge about tobacco-related cancers along with the best ideas, best technologies,
and best people to the problems of prevention, detection, diagnosis, and treatment of these
diseases. The NCI would like to see all eligible cancer patients participating in clinical trials.
Accomplishing this very ambitious goal will involve utilizing new resources to lower the barriers
Prevention of tobacco-related cancers for 45-50 million former smokers
Premature deaths from smoking will continue for many years to come, even if all smokers were
to quit today. Heavy smokers retain some elevated risk for lung cancer when compared to
never-smokers, even after 20 years of cessation. More than 50% of lung cancers diagnosed in
the late 1990's occur in ex-smokers. Thus, the 45 to 50 million Americans who are former
smokers are a population that may benefit from chemoprevention research. They already have
taken a major step toward improved health, but require new medical options to gain full
advantage from this step.
The large population of former smokers in the Department of Defense, Veterans Administration
and managed care systems makes access to individuals for clinical trials in prevention possible.
Promising preventative interventions (e.g., vitamin A analogs, selenium, non-steroidals and
specific growth pathway inhibitors) are ready for testing. Prevention studies using aerosolized
delivery of new targeted therapies including gene therapies, antisense therapeutics, and targeted
small molecules will also need to be conducted. Key to these studies will be the identification
and validation of biological markers that will be used to define the activity of new preventative
Development of preclinical models for tobacco-related cancers
For clinical trials to have a good chance of success, the interventions being tested in them should
have to pass stringent tests in the laboratory that suggest they might work. For these laboratory
tests to be valid predictors of what will happen in people, the laboratory models must be accurate
mimics of human disease. For cancer this means that laboratory cancers should bear fairly close
resemblance to human cancer in their molecular characteristics.
Biological advances have changed the prospects for developing such models, and it is now
possible to engineer cells or even whole animals to have certain genetic properties that are
desirable in cancer models. Much more work is needed to develop and refine preclinical models
of the major tobacco-related cancers, so that clinical work in both prevention and treatment can
proceed with a higher probability of success.
Exploration of cancer susceptibility and gene-tobacco interactions
It has become increasingly clear that a person's genetic makeup influences susceptibility to
environmental agents such as tobacco smoke. Cancer susceptibility is not a simple function of
genes or environment but rather a function of the complex interplay of genes and environment.
We need to identify the genes that influence susceptibility to tobacco and to understand how
these genes function to modulate tobacco's carcinogenic potential. The identification of these
genes will enable us to predict who is at the highest risk for cancer development allowing for
better counseling, prognosis, and detection. This identification will also open research
opportunities to develop therapies to counteract the high susceptibility to cancer development.
Improved early detection and diagnosis of tobacco-related cancers
Tobacco-related cancers present attractive opportunities for the development of much more
accurate methods for earlier detection and better diagnosis. The population at risk is largely
defined (i.e., current and past smokers), and many classes of tobacco-related cancers occur in
anatomical locations accessible to biopsy or the obtaining of biological specimens (head and
neck, esophagus, bladder) or radiological images (lung). It is a reasonable expectation that more
sensitive diagnostic techniques will increase cure rates, since early cancers are, in general, more
curable than advanced ones. Molecular techniques can now pinpoint cancer-causing mutations in
the DNA of cells that appear in cells or cell debris from the mouth, aerodigestive tract, urine.
More sensitive digital imaging and image-processing techniques are in development that should
allow the detection of much smaller tumors in the lung, a notoriously difficult site for early
detection. Emerging technologies related to high-throughput, cost-effective, comprehensive
molecular analysis will soon allow a complete molecular profile of cancer cells and precancerous
lesions. The translation of these emerging technologies to applications that benefit large number
of patients is a daunting but necessary task. Further, the fostering of collaborations between
imaging scientists and investigators in other areas of biology and medicine will serve to hasten
the development of non-invasive methods of tumor characterization important to foster
improvements in diagnosis, prognosis, and treatment of tobacco-related cancers.
Improved National surveillance of tobacco-related cancers
There is an urgent need to enhance the Nation's ability to gather more extensive and more
intensive data on the burden of tobacco-related cancers throughout society, in order more readily
to develop and test hypotheses about trends in the incidence and outcome of these cancers as
patterns of tobacco usage change and as effective interventions are introduced on a large scale.
The effects of changes in utilization of tobacco products by the American people, together with
the results of implementing effective preventive or early diagnosis strategies nationally, when
these become available, will be difficult to discern without adequately sensitive surveillance
programs. NCI would like to expand its important databases such as the Surveillance
Epidemiology and End-Results Reporting (SEER), which tracks the impact of cancer on the
general population. The SEER database has proven to be of great utility to the scientific
community but the SEER database contains only about 13% of the US population and
representation of racial and ethnic groups within the database vary widely. There are also
limitations to the depth of data collection and entry that is currently possible.
We want to make sure that, to the maximum extent feasible, the SEER database not only tracks
changes in cancer incidence and survival accurately, but also contains enough of the right kind of
information to enable the generation of hypotheses about the basis for any observed changes in
trends over time. New initiatives will be developed to expand surveillance research to allow
more complete collection of information on risk factors, including exposure to tobacco products
and promotions. Emphasis on underserved or special populations at high risk (e.g. age,
race/ethnicity) will be a key component of the program.
Improved communication about tobacco-related cancers and interventions
The power of computer-based communications and the capabilities of the World Wide Web will
make possible new levels of cooperation in research on patients and populations. Slow and
cumbersome paper-based systems of data collection for multicenter studies will give way to
electronic communication, facilitated by enhanced links between sources of data at points of care
delivery (hospitals, offices, clinics) and the research database on which analysis of results is
based. In collaboration with other Federal agencies, and with the participation of many external
scientists and clinicians, NCI will modernize information links with its investigators, in a manner
that will be compatible with standards set by the International Committee on Harmonization for
North America, Europe, and Japan. We would also revise our criteria and standards for reporting
adverse drug reactions and treatment-related toxicity. The results will be a set of common
reporting requirements and terminological standards that will greatly increase speed and
efficiency in the reporting of clinical trials both to sponsors and to regulatory agencies.
Mr. Chairman, I hope this overview provides you and the members of the committee a sense of
the wide range of needs and opportunities in research on tobacco use and tobacco-related
cancers. This is a topic of critical importance to the health of the Nation. I would be pleased to
answer any questions you may have. Thank you.