American Journal of Law & Medicine

Making clinical trials safer for human subjects.


The pace of medical innovation is quickening, pushed by advances in biotechnology and pulled by growing demand for healthcare. Completion of the Human Genome Project has provided a multitude of data for determining genes and their functions, and stimulated the application of information technology and private capital for the development of new individualized drugs, gene therapies, and biomaterials for tissue and organ regeneration.(1) Concurrently, demand for healthcare continues to grow, expanding the market for medical innovations that are proven in clinical trials to be safe and effective.(2)

Clinical trials, in which new biotech and other medical products are tested on human subjects, provide much of the data used by the federal Food and Drug Administration (FDA) to determine whether the products are suitable for routine use in healthcare.(3) Thus, the trials are of obvious importance to medical progress and improvement of public health, and to those who have career and financial interests at stake.(4) But clinical trials are also important to the human subjects involved because the products being tested on them may remedy their illnesses, but may also pose risks since the products have usually not been previously tested on humans.(5)

The clinical trial is "a point at which research and the practice of medicine intersect"(6) because it is supposed to be designed and managed to achieve dual societal objectives: the generation of clinical evidence regarding the efficacy and safety of new products, information needed by the FDA to determine whether the products should be sold for medical use, and the responsible application of such products to selected human subjects for potential therapeutic benefit.(7)

In some respects, the clinical trial resembles the test procedures for introducing other new technologies into society. For example, the introduction of a new pesticide, aircraft, chemical manufacturing process, or method of producing energy, requires carefully designed field tests to discern gains, risks, and the adjustments needed to secure regulatory approval and societal acceptance. And like clinical trials, the test procedures must assure reasonable protections for workers and other persons exposed to these advances.(8)

Thus, the introduction of a new technology requires testing by conducting a carefully managed learning process with two main functions: an operational function to discern whether the new product or process can efficiently provide anticipated benefits, and a safety function to identify its risks, which may be latent and unknown at the outset, and to determine if these risks are manageable. Experience in many technological sectors indicates that the safety function is vulnerable to deliberate or inadvertent compromise when it is managed by proponents of the technology who are striving to achieve results which promise personal or organizational gain.(9)


Carrying out the safety function in clinical trials of new products containing genetically modified materials poses special challenges. Little is known or knowable beforehand about the risks posed by such advances, and many of the human subjects involved have been chosen because of their seriously impaired health, making them especially vulnerable to any risks. Compounding the challenge are contextual circumstances such as the multiplicity of individuals and organizations commonly involved in such a trial, making it a multi-party enterprise in which coordination and communication difficulties may arise and cause responsibilities to become diffuse and uncertain. In addition, there is urgency and pressure for success because of the substantial investment of capital, facilities and human resources over the long period of time usually involved in bringing a biomedical advance to market.(10)

The universe of biomedical advances now being tested includes many gene therapies, products containing specially created genetic material to be delivered into target cells of a person with the intention of curing the person's genetically-based illness. The strategy is to have this transference of new genetic material repair a mutated gene or inherited genetic condition, which is believed to be a contributing factor in causation of the patient's illness, or provide genetic material that adds missing functions or regulates the expression of other genes in order to defeat the illness.(11) Success, according to researchers, depends on effective delivery of the new genetic material into the target cells of the patient by using a vector, usually a disabled virus, in order to infect the target cells with the new genetic material and have the new genetic material thereafter perform as intended.(12)

Since the first clinical trial of a gene therapy in 1989, some 4000 human subjects have participated in over 500 gene therapy trials funded by the National Institutes of Health (NIH), with numerous other subjects enrolled in privately-financed trials approved by the FDA.(13) Among the illnesses being addressed in the studies are cancer, AIDS and other infectious diseases, cystic fibrosis, heart disease, arthritis and Alzheimer's. The trials increasingly involve delivery alternatives to viral vectors, and more diverse subjects, with most of the trials to date focused on safety rather than efficacy. Thus, NIH cautiously refers to this technology as "gene transfer" rather than "gene therapy" until evidence of efficacy in terms of therapeutic benefit is proven, and FDA reports that as of January 2000, it had not received any application to license a gene therapy product for use in healthcare.(14)

NIH has indicated its concern for several types of risks to the subjects: inadvertent transfer of the new genetic material to the subject's reproductive cells which could result in changes being passed on to offspring; transfer to the subject's non-target cells which could pose sudden or chronic health risks to the subject; and residual capability of a "disabled" viral vector to cause harm, such as infectious disease in the subject and their close contacts.(15)

Public concern over the safety of clinical trials is periodically aroused by revelations of harm and deaths among the human subjects involved. For example, recent review of many clinical trials of new conventional drugs, involving some 45,000 children, has found eight deaths and numerous other adverse events.(16) In the case of gene therapy trials, such concern was first aroused in September 1999 by the death of a young adult subject, 18-year-old Jesse Gelsinger. Gelsinger had a relatively mild form of a rare metabolic disorder which was manageable through diet and drug regimens. But, he voluntarily enrolled in a clinical trial of a gene therapy at the University of Pennsylvania in order to test the therapy for the benefit of others suffering from similar diseases related to liver function.(17)

Within one day of receiving the therapy, Gelsinger developed systemic blood clotting. Over the next three days, he suffered respiratory disease, liver and kidney failure, and then died. This tragic event, caused by overreaction of Gelsinger's immune system to the adenovirus used to deliver the genetic material, was the first death proven to be caused by a gene therapy trial, and prompted investigation of the trial by NIH and the FDA. Investigators found that the researchers violated the trial's protocol in numerous ways.(18) As a result, FDA has initiated an administrative proceeding to disqualify the principal investigator from further clinical studies.(19) Gelsinger's family has also sued the researchers and trial managers for damages.(20)

FDA and NIH thereafter investigated other gene therapy trials and found that many subjects in these trials had suffered "adverse events," many of which had not been reported by the researchers or organizations involved.(21) For example, medical researchers at Cornell and Tufts Universities had failed to report to NIH six deaths among their subjects in gene therapy trials for regenerating blood vessels.(22) Although some if not all the adverse events may have been due to the subjects' underlying illnesses, the failures to report to the oversight agencies constituted violations of their obligations regarding such events.(23)

The investigations also illuminated a disturbing feature of many of these reporting violations; namely that many of the researchers and organizations involved held significant financial interests in the companies which made the genetically modified materials being tested.(24) The influence of such financial interests in compromising reporting duties and other safeguards has become a major issue for the agencies. Heretofore, the agencies have not regulated the financial ties of researchers, relegating this matter to the universities and other organizations where trials are conducted. However, studies now show that the universities have sought to hire and retain biotech researchers in the face of competitive pressures by relaxing traditional strictures on financial conflicts of interest.(25) As a result, many reforms aimed at lessening financial conflicts of interest have been proposed in order to reinforce the safety function in clinical trials' management.(26)

Can clinical trials of gene therapies be made safer for the human subjects involved? This article addresses this question by examining basic ethical principles for safeguarding human subjects in biomedical research; FDA and NIH safety management systems for gene therapy trials; factors which can undermine the safety-management systems during performance of gene therapy trials; various reform initiatives; and the potential value of safety management experience with other new technologies and its transferability to the context of gene therapy trials.


Over several decades, esteemed organizations have sought to provide ethical guidance for protecting human subjects in medical experimentation and biomedical research. Shocking revelations of sadistic experimentation by doctors in Nazi Germany led to enunciation of the Nuremberg Code in 1949.(27) The Code calls for fully-informed, voluntary consent by human subjects as the essential requisite for their enrollment in medical experimentation, and for prohibitions on experimentation which is "random," "likely to cause unnecessary suffering or death," or which poses risks which exceed "the humanitarian importance of the problem to be solved."(28)

These principles were amplified in 1964 by the World Medical Association in its "Helsinki Declaration," a moral code of conduct for medical researchers.(29) Recognizing that medical research on humans may be done for various beneficent purposes, the Declaration divides such activities into two broad categories: research for the diagnostic or therapeutic benefit of a patient, and research done solely for scientific purposes "without implications of direct diagnostic or therapeutic value" for the human subjects involved.(30) Informed consent and other relatively conventional principles of due care for patients are enunciated for the former case, but new precautionary principles are set forth for safeguarding human subjects in the second category.(31) With regard to research solely for scientific purposes, the Declaration provides that "it is the duty of the physician to remain the protector of the life and health" of human subjects involved, to discontinue research which, if continued, would be harmful to the subjects, and not to allow scientific or societal interests to ever take precedence "over the well-being of the subject."(32)

Over the next decade, marked by enormous growth in biomedical and behavioral research activities in the United States, the need to clarify ethical guidelines for protecting human subjects in projects for scientific purpose became apparent, and a National Commission was created to accomplish this task. The National Commission for the Protection of Human Subjects of Biomedical and Behavioral Research issued its report in 1979, the "Belmont Report," which provides the foundation for the protections now afforded human subjects in gene therapy and other biomedical trials by government agencies.(33)

According to the Belmont Report, research programs and projects must adhere to three basic ethical principles: "respect for persons" through full implementation of informed consent procedures; "beneficence" in research on humans by "maximizing possible benefits" while "minimizing possible harms;" and "justice" in the distribution of research benefits and burdens across society.(34) But despite thoughtful discussion of the three principles, the Report concludes with permissive recommendations.(35) Prescriptions, prohibitions and other strict limitations are avoided, and an ethically-informed but flexible decision-process is propounded for researchers to follow in designing and conducting activities with human subjects.(36)

In striving to avoid restriction, the Report even attempts to refute the Hippocratic maxim "do no harm" and replace it with a qualitative balancing analysis:

   ... avoiding harm requires learning what is harmful; and in the process of 
   obtaining this information, persons may be exposed to risk of harm ... 
   Learning what will in fact benefit may require exposing persons to risk. 
   The problem posed ... is to decide when it is justifiable to seek certain 
   benefits despite the risks involved, and when the benefits should be 
   foregone because of the risks.(37) 

Thus, the Belmont Report offers a morally-informed but ultimately permissive guidance to biomedical researchers--namely some sort of qualitative cost-benefit analysis--for determining the protections afforded to human subjects. Only a few unavoidable limits on researcher discretion are expressed: e.g. "brutal or inhumane treatment of human subjects is never morally justified," a higher level of justification is needed for enlisting a "vulnerable population" (e.g. children, prisoners) as human subjects, and "relevant risks and benefits must be thoroughly arrayed in ... the informed consent process.(38)

Encouraged by the Belmont Report and other permissive rationales, including those articulated by the National Bioethics Advisory Commission,(39) government regulators and grant providers, and individual researchers and their organizations, are now engaged in authorized clinical trials for new biotech products despite their potential for harming the human subjects involved. According to Jesse Gelsinger's father, "I have read that my son's death has been called by one of the leaders in this field as a pothole on the road to gene therapy. His death was no pothole. It was an avoidable tragedy from which I will never fully recover."(40)


FDA and NIH require safeguards for human subjects in clinical trials of gene therapy and other biomedical advances, but from different jurisdictional and cultural stances.(41) FDA is mandated by several federal statutes to regulate drugs, medical devices and "biologic products,"(42) including gene therapy product.(43) It thereby regulates clinical trials for such products if the findings are to be considered by the FDA when it is subsequently called upon to determine whether the products are sufficiently safe and effective to be sold and used in commerce. Thus, FDA regulations govern gene therapy clinical trials, whether privately or publicly funded, so long as the results will figure in subsequent marketing applications to the agency for regulatory approvals.(44)

NIH provides federal grants to researchers in academic and medical research centers for scientific purposes.(45) Unlike the FDA, it is not a regulatory agency. Nevertheless, it has responsibility to ensure that the research work it supports is appropriately conducted. Thus, its policies, grant terms and conditions and generic agreements with the researchers and organizations it supports obligate these parties to adhere to approved research plans and protocols, requirements for avoiding fraud and other forms of "scientific misconduct," various procedures for protecting human subjects, and informed consent requisites.(46)

Both organizations view protection of human subjects as an integral feature of their responsibilities. However, FDA's regulatory culture emphasizes the agency's needs for useful findings on efficacy and safety which will enable it to make regulatory decisions which are factually supportable and otherwise credible. Thus, protections for human subjects are implemented to serve this purpose. Similarly, in NIH's research-promoting culture, safety of the subjects in trials it funds is subordinate to the agency's drive for scientific progress, a condition which suits the career and financial motives of researchers.(47)

Subordinating the safety of the few persons in a trial to the "productivity objectives" of clinical research, such as generating data useful to regulators or making scientific progress, comports with the Belmont Report because emphasis on productivity facilitates clinical research which may benefit many person--a gain which conceptually outweighs the risks borne by the few subjects in each trial. Conversely, elevating safety of the few subjects to primacy would have the effect of burdening and obstructing clinical research and thereby retard the rate at which biomedical advances are gained for the benefit of the many. …

Log in to your account to read this article – and millions more.