A Collusion of Misunderstanding
A dilemma confronts physician-investigators ... As physicians they are dedicated to caring for their patients ... As investigators they are dedicated to caring for their research ... These two commitments conflict whenever an individual physician-investigator comes face to face with an individual patient/subject.
In 1995, having just finished a fellowship in medical oncology, I was installed for a three-month stint as acting clinical director of Phase I trials at the Dana-Farber Cancer Institute. The Dana-Farber is a cancer center characteristic of elite academic cancer centers that engage both in clinical research, including high-volume Phase I trials, and direct patient care. My duties included caring for all patients already enrolled in Phase I trials and inviting new patients to participate in ongoing and newly sanctioned trials.
Phase I trials lack control groups, and unlike Phase II and III trials they do not test hypotheses (such as whether new drug X is at least y percent effective in disease Z, or whether new drug X is better than old drug Y). Rather, they address an estimation problem: drugs that have demonstrated promising anticancer effects in animals are administered to terminally ill cancer patients in an attempt to estimate the highest dose human beings can tolerate. Evaluating a drug's anticancer effect in human cancers, a primary objective of subsequent phase II and III trials, is not a defined goal in Phase I studies.
Having taken care of adults with terminal cancer before joining the Phase I team, I had witnessed the desperation that motivates many patients to participate in these trials and recognized in a vague but fateful way the peripheral regard most patients had for the sacrifices they were being asked to make on behalf of science, researchers, and future generations of patients. What I did not appreciate, prior to my direct involvement, was the moral difficulty inherent in navigating the tensions between patient care and research.
My experience taking care of, enrolling, and doing research on patients in Phase I cancer trials has forced me to question assumptions implicit in the actions I engaged in daily. Some of these assumptions reveal structural problems peculiar to Phase I trials, which in turn suggest design modifications that have moral consequences within these trials. Other assumptions underpin a more fundamental tension not amenable to technical solution, raising residual questions about human investigation generally, such as whether and under what conditions it is justifiable to trade on patients' hopes. (I should note that the difficulties I encountered would not, I believe, have been substantively different had my experience with Phase I trials taken place at any other academic teaching hospital.)
Subjects and Response Rate
In the United States, subjects in noncancer Phase I (dose-toxicity) studies are usually healthy volunteers. This choice is defended, by its proponents, on two grounds. First, sick persons are vulnerable and should thus be protected from harm, and second, results from healthy persons are free from the distorting effect that disease can have on drug metabolism and tissue tolerance. Critics of this practice argue that even though the risk of injury to healthy subjects is very small, because there are no health-related benefits the balance of risk and benefit is unacceptable. Others challenge this practice on scientific grounds, claiming that drug metabolism is so different in healthy from sick individuals that any transfer of results from healthy to sick may be dangerously misleading. Because Phase I cancer trials use drugs that often produce their effects by causing damage to DNA, with attendant concerns about long-term carcinogenic consequences, subjects in Phase I cancer trials are never healthy volunteers.
To be considered for Phase I cancer trials patients must have exhausted all known therapeutic options and have a life expectancy of at least three to four months. …