Although narrow-entry, randomized, placebo-controlled trials based on preplanned deaths and overall survival may be the gold standard in the laboratory today, human beings are not laboratory rats. Sound scientific research can be done without lethality as a planned endpoint.
—Donald M. Gleason, MD, and Ryan G.N. Seltzer, PhD
The requirements for sound evidence of a drug’s therapeutic benefit have translated laboratory experience to human testing. In the laboratory, experimental animals give their lives to lethal testing of drugs and scientific analysis. LD50, the terminology denoting an anticipated 50% death rate of laboratory rats in a drug study, reflects standard procedure in this setting.
In human drug studies as well, this translates to preplanned deaths when a drug is tested for treatment of lethal disease. The current concept of overall survival when testing drugs in humans demands a preplanned number of deaths so that the test drug can be shown to be more effective than placebo or a placebo equivalent. Overall survival amounts to nontreatment of lethal or debilitating disease in the study patient. We tend to overlook the human cost of these studies.
Ignoring Drastic Effects
A classic double-blind study performed by Rammelkamp and colleagues in the 1950s in the Great Lakes Naval Training facility demonstrated that penicillin, properly administered, completely eradicated beta-hemolytic streptococcic infection in young recruits with pharyngitis and prevented the sequel of rheumatic heart disease.1 No mention is made of the known, lifelong debilitating effect of rheumatic heart disease occurring in 17 of the young sailors in the placebo cohort.
This pattern of focus on the success of a treated group, while ignoring drastic effects on the placebo group continues in today’s studies. In the AFFIRM trial of enzalutamide (Xtandi), there were over 700 preplanned fatalities.2 In the trials of abiraterone (Zytiga), > 500 preplanned fatalities were documented.3,4 Is overall survival, a holdover from the animal lab, an ethical determination? Is it the only valid determination of drug effectiveness?
A comparison between early studies of lethal infectious disease—renal tuberculosis and HIV-AIDS—addresses this question. Lattimer demonstrated that triple-drug therapy arrested potentially lethal renal tuberculosis at every stage of the disease. There was no placebo cohort and no deaths, preplanned or otherwise.5 On the other hand, in early trials of zidovudine, the drug was administered to patients with lethal HIV-AIDS in a 1:1 ratio to placebo despite the known outcome of untreated HIV-AIDS.6 One can legitimately ask if anything was accomplished by mortalities in the placebo group.
The insistence on placebo-controlled preplanned deaths in lethal disease has additional ramifications. Enrollment of patients into a study is usually the rate-limiting factor in clinical trials. When death is a planned outcome in 50% of study patients, physicians are not anxious to enroll their patients in such studies.
Furthermore, the informed consent process seldom clearly expresses this reality to the volunteer subject, a violation of one of several tenets of the Belmont Report guidelines for ethical treatment of patients. Patients offering their lives and well-being to a study expect “beneficence”—that is, they expect the study will not do them harm. Moreover, patients with lethal disease are a very vulnerable population, willing to take a chance for survival.
Taking advantage of a vulnerable population and ignoring the concept of beneficence are not within the concept of justice according to the Belmont Report, yet are integral parts of the overall survival concept pervading current human research. It is not surprising that human research studies of lethal disease take years to accumulate patients, and at such financial expense that the price of new drugs reflecting the protracted studies is usually prohibitive.
More Humane Options
There are alternatives. If a drug to be tested is known to be superior to its comparator, the concept of “equipoise” suggests that the comparator study need not be done. Extending this concept, if during a study there is clear superiority of one drug to its comparator, “clinical equipoise” suggests that logically and humanely, the effective drug should be available to all participants.
Furthermore, surrogate markers may be used as primary endpoints according to the International Conference on Harmonisation, “when the surrogate is reasonably likely to predict clinical outcome.” When predictive surrogate markers are available, there is no need for a study to require planned deaths. Surrogate markers may include imaging studies, biochemical markers, circulating tumor cells, etc.
Yet another approach is to abandon comparator, placebo-controlled trials entirely, in favor of registries. Registries are currently employed in postmarketing surveillance and are responsible for uncovering recondite side effects and drug interactions. Introducing registries in the premarketing space, replacing phase III trials, would facilitate the evaluation of a drug at different stages of disease simultaneously at greatly reduced cost.
Such investigative registries exist for the study of rare diseases (such as Gaucher’s disease), where the luxury of a population of 1,000 volunteers and 500 deaths are just not available. Disease-specific registries would feature wide entry criteria, no placebo, and retention of patients who might otherwise leave and bias a study because of placebo concerns. Statistical techniques exist to deal with confounding variables, simulate the benefits of randomized controlled trials, and offer a humane way to meld scientific and humanitarian concerns.
Although narrow-entry, randomized, placebo-controlled trials based on preplanned deaths and overall survival may be the gold standard in the laboratory today, human beings are not laboratory rats. Sound scientific research can be done without lethality as a planned endpoint. ■
—Donald M. Gleason, MD
Clinical Professor, Urology (retired)
College of Medicine
The University of Arizona
—Ryan G.N. Seltzer, PhD
Mel and Enid Zuckerman
College of Public Health
The University of Arizona, Tucson
Disclosure: Drs. Gleason and Seltzer reported no potential conflicts of interest.
1. Denny FW, Wannamaker LW, Rammelkamp CH, et al: Landmark article May 13, 1950: Prevention of rheumatic fever. Treatment of the preceding streptococcic infection. JAMA 254:534-537, 1985.
2. Scher HI, Saad F, Tapin Me, et al: Increased survival with enzalutamide in prostate cancer after chemotherapy. N Engl J Med 367:1187-1197, 2012.
3. de Bono JS, Logothetis CJ, Molina A, et al: Abiraterone and increased survival in metastatic prostate cancer. N Engl J Med 364:1995-2005, 2011.
4. Ryan CJ, Smith MR, de Bono JS, et al: Abiraterone in metastatic prostate cancer without previous chemotherapy. N Engl J Med 368:138-148, 2012.
5. Lattimer JK: Renal tuberculosis. N Engl J Med 273:208-211, 1965.
6. Levine C: Has AIDS changed the ethics of human subjects research? Law, Med Health Care 16:167-173, 1988.