Forensic Psychiatry & Medicine Managed Health Care

Reducing Clinical Research Risks


For Publication, NY Times, August 1, 2001

Dear Editor,

Drs. Shaywitz and Ausiello thoughtful essay(NYT Week in Review 7/29/01) reminds us that while all clinical research involves some risk, it is also possible to inform subjects of risks and to protect subjects from unnecessary risks. In order for subjects to be protected and to knowingly accept risks, not just clinical investigators and institutions but also the sponsoring pharmaceutical company must be considered as potentially jointly responsible. The requirement to meaningfully inform patients who are clinical research subjects is central not only by way of respecting their autonomy but also by way of protecting subjects from unnecessary risks and early intervention and damage control to reduce the level of necessary risk. A well informed research subject can not only make a meaningful choice as to clinical research trail participation but can also alert the clinical investigator earlier and more effectively as to emergent side effects. This is why an informed consent process including the assessment and enhancement of patient competency to communicate even unexpected side effects is important. However such an informed consent process involves sponsors allocating sufficient funds and to allow for the training of clinical investigators and disclosing sufficient information in a meaningful manner to allow for meaningful informed consent.

According to the Nuremberg Code and its progeny codes for clinical research, in addition to clinical investigators, sponsors are accountable for protecting human subjects from unnecessary risks. Support of such codes via a reasonable and fair policy delineating sponsoring pharmaceutical company responsibility is essential to avoid putting especially vulnerable patients at risk for sponsors shopping for the most compliant and cheapest clinical investigators and Institutional Review Boards. Without any sponsor accountability it is foreseeable that an atmosphere will be created where diligent clinical investigators and institutions who emphasize that informed consent is not a mere pro forma exercise but a process will be pressured by the clinical research market to lower their standards for informing and thus protecting vulnerable human subjects.

Yours very truly,

Harold J. Bursztajn, M.D.
Associate Professor of Psychiatry
Harvard Medical School
co-Director, Program in Psychiatry and the Law
Harvard Medical School @ Massachusetts Mental Health Center


(NYT Week in Review 7/29/01)

The Necessary Risks of Medical Research


By DAVID A. SHAYWITZ and DENNIS A. AUSIELLO

"FATAL Disease Cured by New Therapy" — it's the headline of researchers' dreams.

As modern biological science generates an ever-rising flood of promising discoveries, from the potential of embryonic stem cells to the nature of the human genome, the idea of one day being able to cure cancer, Parkinson's disease and other terrible illnesses no longer seems so fanciful. In fact, there is often an exhilarating sense these days that a cure for something — or for everything — is near.

It is not surprising, therefore, that following the recent deaths of Jesse Gelsinger and Ellen Roche in government-sponsored clinical trials at two major American universities, some have questioned why, in the era of transgenic mice and computer modeling of molecules, we still experiment on people. As one patient recently asked, "Why don't they just give us the good stuff?"

The answer is that it is terribly difficult to figure out what the good stuff is. And while the lab can identify promising new therapies, proof that they work can only come from clinical research, in which real people are carefully administered medicine and the results are meticulously documented. More often than not, even the most obvious and intuitive of treatments provide wholly unexpected results.

A classic example is the group of drugs called beta- blockers, medicines known to decrease the strength of heart muscle contraction. For years, they were not given to patients with weak hearts, until clinical trials demonstrated they actually helped patients live longer.

Or consider the use of bone marrow transplants to treat patients with advanced breast cancer. In part because other treatments were often only minimally effective, transplant seemed like a promising alternative. So promising, in fact, that it was difficult to set up clinical trials, as no one wanted to be randomly assigned to the non-transplant group. It took a decade until several studies concluded transplant was no better than conventional therapy.

Equally revealing are the raw statistics of new drug discovery: for every 5,000 compounds evaluated for treatment, only five will make it to clinical trials, of which just one will make it to market. How can the best scientists with the best labs money can buy be wrong 4,999 out of 5,000 times?

Ever since the discovery of DNA's structure in 1953, it has seemed only a matter of a few years before biology became a predictable process for understanding diseases at the most fundamental level, and then designing cures for them.

It was in this hope that special programs were established at the National Institutes for Health and other research institutions to encourage doctors to pursue this new science. As a result, many of the country's most prominent doctors — including J. Michael Bishop, chancellor of the University of California at San Francisco, and Harold Varmus, president of Memorial Sloan-Kettering Cancer Center — made their reputations through elegant laboratory investigation. Dr. Bishop and Dr. Varmus received the Nobel Prize for their discoveries.

Recently, however, doctors have begun to ask why the cures that molecular biology promised have not appeared, and to wonder if the laboratory is the best place to pursue them. Patients continue to suffer from cancer and heart disease, despite national crusades against both. And treatments for molecular diseases like sickle cell anemia and cystic fibrosis have evolved little, even though the genetic defects have been exquisitely defined.

The problem is that most biological systems remain too complex for science to predict how they will react to theoretically valuable treatments. And for that reason, the work that has actually changed how doctors treat the sick has come mostly from those researchers who focus on patients, not genes. For example, landmark population-based studies like the British Doctors; Study, the Framingham Heart Study and the Nurses' Health Study revealed the risks of smoking, hypertension and high cholesterol, and encouraged treating these problems aggressively.

Overall, it is becoming increasingly apparent that what really makes a difference for patients is not a doctor's detailed knowledge of the biology, but rather, his or her familiarity with the existing clinical literature, and particularly with current clinical trials. The best evidence of this is that the doctors other doctors turn to for care are usually not medical scientists, but scholarly clinicians fluent with the best data. Some have even had a role in generating it.

This is not to say that laboratory science has been neglected. Laboratory scientists are playing increasingly important roles in clinical research — for example, by suggesting disease genes for which patients might be screened. This year alone, such approaches have led to discovery of genes responsible for some forms of liver disease, lung disease, and diabetes in infants.

In addition, while drug discovery in the past relied heavily on trial and error, the medicines of the future will result from a careful study of the underlying biology. Already, the anti-cancer drug STI-571 and the protease inhibitor class of anti-H.I.V. medicines were developed with a specific drug target in mind.

Still, predicting how a new drug will work in the human body remains well beyond the reach of science. The clinical trial, not scientific theory, remains an absolute necessity.

The stakes are high. Clinical research is directly responsible for many advances in modern medicine, as well as for ensuring that patients do not waste time, money and hope on ineffective therapies. At the same time, as science generates ever more promising research leads, clinical investigators will feel pressed to work faster, with fewer safeguards.

When medical research causes the death of a human being, the wish to stop using people as test subjects is perfectly understandable. Such deaths should remind researchers of their absolute responsibility to explain experimental treatments to patients clearly and fully, and to do everything possible to minimize risks. But risk, in the end, defines why the research is being done. And even the most careful scientists and the most obsessive oversight cannot eliminate it entirely. David A. Shaywitz is a medical resident at Massachusetts General Hospital and a clinical fellow at Harvard. Dennis A. Ausiello is physician-in-chief of Massachusetts General Hospital and Jackson professor of clinical medicine at Harvard.


home * e-mail Dr. Bursztajn * more about Dr. Bursztajn * other web sites