An Idea for Modifying Phase I Clinical Trials: A Patient's View

Human participants in a protocol differ in two ways from the rats and dogs that preceded them. First, because human participants are not distinguished by breed, they do not have identical or blanket reactions to an experiment; in contrast, animals that are bred or selected for laboratory research react the same. Second, human patients have differing mental and personality characteristics. In experimental trials, doses can be adjusted in advance to accommodate the size and other physical aspects of human participants, but the trials cannot compensate for the thoughts and emotions of individual participants. These important facets of the whole person are largely ignored when trial programs are designed.

An experiment that uses brown, gray, and other rats in addition to carefully bred white rats and that adds wolfhounds and Chihuahuas to the laboratory population would approximate the varied human population that serves as protocol participants. Because different breeds would probably have different reactions, a single success would not be applicable to the general rat or dog population. Considering that, in human trials, some participants may be inclined to act at cross-purposes to the experiment, it is a wonder that any success can be validated for use.

Although physical differentiation may be impossible to overcome in most human protocols, the nonphysical differences may present a source of great utility. Because human temperament ranges quite randomly through all body types, successful results with protocol participants who share the same temperament could apply to all body sizes and shapes. The solution, it would seem, therefore rests in the recognition and use of personality traits.

Most patients do not ask to participate in protocols but rather are content to either let the disease follow its natural course or adhere to the advice of their health care providers (who in turn may be somewhat fatalistic). Patients who do seek to participate in protocols are those who question the status quo and who are most eager to alter it. These useful traits often make them the despair of those who care for them because they question the program in detail: They want to redesign the protocol to fit self-perceived needs. To a minor extent, they often do succeed in having the protocol adjusted, thereby rendering trial results still more suspect.

What would happen if a control group of more or less uncontrolled participants were permitted to change a protocol in radical ways? For example, what if participants were allowed to combine substances as long as the combination was not known to cause death? Or, what if participants were allowed to increase drug doses as long as the amount was not known to cause permanent crippling? A hundred such patients might improve the chances of finding a cure a hundredfold when the new combinations of chemicals or altered drug doses are tested. One breakthrough by a single participant could be followed by other participants who are only too eager to achieve a like cure. Gradually, the successful dose schedule could be adjusted to lessen dreaded side effects.

That some of these 100 participants would die of the effects of their medication would be unavoidable-but better that a few fall in the storming of the bastion than no storming be attempted. Without direct assault, cancer becomes a battle of attrition, with the tumors digging in ever more securely. I have come to know from experience that tumors do learn how to resist agents over time. Tumors must be kept off-balance and under constant attack by a variety of weapons. To use weapons singly or desultorily is a design for toughening tumors. They must be punished constantly, even if the price is constant pain and discomfort for the patient. Letting a patient choose the poisons (under professional guidance) adds something to the will to struggle. We who are struggling to escape cancer do not, obviously, want to die of it. We do prefer death in the struggle to life under cancer's untender rule. The enemy is not pain or even death, which will come for us in any eventuality. The enemy is cancer, and we want it defeated and destroyed.

Just before assaults on fortified positions, U.S. Civil War soldiers would pin their names and addresses to their uniforms to make it easier for the body-sorters to do their work after the battle. Patients going into these modified protocols could likewise place their names on specific protocol adjustments. Survivors could then proclaim: This is how I wanted to die-not a suicide and not passively accepting, but eagerly in the struggle.

Commentary

Top Commentary Author & Article Info References

The Man behind the Words

Drs. Christopher K. Daugherty, Mark Siegler, and Mark J. Ratain: Professor Zimmer, a patient who had a refractory malignancy, participated in several phase I cancer clinical trials at the University of Chicago. With the consent of his widow, his essay is published in this issue of Annals. Professor Zimmer's perspective has helped us to better understand the motivation and vulnerability of persons who have life-threatening incurable diseases and who choose to become involved in investigational therapies. Further, his views have encouraged us to reevaluate our attitudes toward phase I cancer trials and to initiate an ongoing research project that was designed to examine the motivations and behavior of the participants and physicians involved in clinical trials.

Although Professor Zimmer describes personal experiences encountered during his battle against cancer, his views on how some patients respond to incurable illnesses certainly apply to many other terminal diseases. Zimmer raises serious questions about basic aspects of clinical trials, including informed consent and risk-to-benefit calculations, and challenges what he considers the paternalistic attitude of clinical investigators, institutional review boards, and regulatory agencies. At one point in his essay, Professor Zimmer makes a radical suggestion: Patients who participate in phase I cancer trials should be permitted to alter a research protocol in hopes of benefiting from additional therapeutic options as well as to advance scientific knowledge in the field of cancer care. Zimmer urges us to reconsider the 50-year-old Nuremberg paradigm of clinical research. This paradigm suggests that human research participants are ignorant and vulnerable and must be protected by regulatory mechanisms from unscrupulous clinical investigators who are more concerned about advancing science and their careers than they are about providing benefits to individual patients [1].

One way to deal with Zimmer's challenges is simply to respond that because no patient can really understand the complexities of clinical trial research design and methodology, these matters are best left to experts. But this discounts the fact that some patients, such as Professor Zimmer, have a high degree of knowledge, motivation, intelligence, and comprehension. For these patients, a paternalistic response may be facile. Another response could acknowledge that although some patients may understand the foundation of clinical trials, participants should not have a role in designing or modifying research protocols. From a scientific standpoint, Zimmer's notion of allowing a "control group of more or less uncontrolled patients ... to change a protocol in radical ways" seems preposterous. His suggestions, if adopted, could generate research anarchy with little, if any, meaningful information emerging from clinical trials. In addition, if patients with cancer and other persons with incurable diseases are regarded as vulnerable individuals who are susceptible to inducements and misplaced hopes, there would be a great risk for harm to patients who volunteer to participate in loosely regulated trials that might not yield any meaningful scientific information.

Research Participants as Patient Advocates: The Lessons of AIDS

During the 1960s, 1970s, and much of the 1980s, the foremost concerns of the U.S. Food and Drug Administration (FDA) and the general public regarding clinical research and drug development were to minimize risk and protect potentially vulnerable patients from ambitious clinical investigators and from a profit-motivated pharmaceutical industry [1, 2]. A social trade-off that permitted slower progress in the conquest of disease [3] was accepted to provide a safer, more closely monitored environment that protected research participants from harm. Thus, for the past three decades, the FDA and institutional review boards have been permitted to decide whether a clinical trial is scientifically worth conducting and whether the risk-to-benefit ratio is one that participants should be permitted to consider. Traditionally, patients and participants did not become involved in decisions about whether clinical research should be conducted. It is interesting to note that this paternalistic system developed and became established in clinical trials as the rest of society and other health care institutions increasingly recognized the importance of autonomy and the right of patients to decide what is in their own best interests [4].

Today, however, the medical community cannot ignore the claims of dying patients on the grounds that they are ill-informed, naive, vulnerable, or misled. With the development of the acquired immunodeficiency syndrome (AIDS) epidemic, activists have demanded modifications in the research and regulatory process, thereby challenging the Nuremberg paradigm. Such demands have resulted in major changes in clinical research, drug development, and the clinical trial process, at least with respect to drugs for the treatment of AIDS [5, 6]. According to the literature, advocates emerged with fury [5] and demanded to be able to make their own risk assessments regarding investigational drugs. One mission of such organizations as ACT-UP (AIDS Coalition to Unleash Power) was to fight for the right of patients with AIDS to have access to experimental therapies. Activists claimed that persons who have life-threatening illnesses do not need protection from the potential harms of research trials. They argued that the role of the federal government and of such regulatory bodies as the FDA should be to maximize patient opportunity and choice to participate in clinical research rather than to maximize patient safety. Reforms in the research process were achieved because many patients with AIDS had a high level of motivation and knowledge; a strong political community was willing to fight policy and public relations battles; social attitudes shifted away from paternalism, especially with regard to life-threatening or terminal illness; and pharmaceutical companies were eager to develop, market, and profit from new and effective drugs for an incurable condition for which effective therapy was not available [6, 7].

During the past decade, the FDA has responded to the pressures and criticisms of AIDS activists and patients with AIDS [5-7]. Beginning in 1987, the agency developed a new set of regulations that heightened respect for patient autonomy by allowing patients and their physicians to assess the risks of experimental therapies. To be made available for further studies, drugs and therapies only had to show some evidence of substantial benefit, even if safety measures remained unknown after relatively small phase II trials that involved only a few hundred patients. The FDA has increasingly recognized the importance of surrogate markers of survival in AIDS and has created a less adversarial relationship between drug developers and representatives of regulatory agencies. The U.S. Congress subsequently enacted legislation that obliges the government to become involved in testing drugs that had previously been available only through an underground system and whose safety had not met U.S. standards [5]. Other notable changes have included parallel drug research (that is, fast tracking of drugs), whereby patients could take a drug even without availability of phase II data if the agent had been proven safe and continued to be tested clinically [8]. Thus, because of pressures from AIDS activists, the FDA and other branches of the U.S. government have become willing to reassess the balance between efficacy and safety. This reassessment has led to modifying the Nuremberg paradigm into the AIDS paradigm: The potential efficacy of clinical research is more important than the potential risk. The specific outcomes from changes in the policies of the FDA have included rapid approval of didanosine and, more recently, rapid approval and release of several new protease and reverse transcriptase inhibitors [9-12].

Before the emergence of AIDS and its effect on clinical trials and drug development, clinical research was perceived as a process of human experimentation that sometimes involved cruel, even inhumane, treatment [13]. Today, clinical trials and access to the drug development process are viewed as valuable societal goods and rights that should be guaranteed. Previous regulatory policies that were designed to protect research participants from harm are now viewed by many as having created unnecessary paternalistic barriers to potentially lifesaving therapies. Clinical investigators and others who are involved in the process of developing drugs that are effective against AIDS have clearly learned the importance of obtaining support from the patient community, even to the extent of welcoming the involvement of patients in the design and planning of clinical trials [6, 7, 13].

Fast Tracking and Advocacy in Cancer

All patients with life-threatening diseases have certainly learned from the experience of AIDS activists and patients with AIDS. Most notably, many survivors of cancer and family members of patients who did not survive have developed strong activist and lobbying groups for their causes [7, 14, 15]. Such organizations as the Breast Cancer Coalition have become involved in efforts to influence public policy and government funding that specifically targets breast cancer research and therapy [16]. In addition, the regulatory changes that resulted from the AIDS movement have been extended and now apply to the approval process for anticancer drugs. The FDA has made a commitment to the same fast-track approach for anticancer drugs as has been used for the development of drugs effective against AIDS [17]. The FDA has also indicated a willingness to use surrogate markers (for example, partial responses) rather than survival rates as a measure of drug effectiveness when considering the approval of anticancer agents. These policy changes have markedly reduced drug-approval processing times. Many of these drugs are now being approved in less than 6 months from the time a new drug application is submitted; in contrast, previous approval times often exceeded 1 year. An excellent example of the effect of changing regulatory policy is the recent approval by the FDA of irinotecan as second-line treatment for advanced colorectal carcinoma without data from phase III randomized clinical trials [18, 19].

Policy changes have come as a result of intense societal pressure from highly motivated groups, many of which have strong political support. In light of these recent changes, Professor Zimmer's essay is especially poignant. On one level, Zimmer eloquently raises the question of whether some individual patients who have a fatal disease could push investigators to move faster than they otherwise might prefer in conducting phase I clinical trials of potentially beneficial new drugs. On a more general level, Zimmer is asking fundamental questions about the relation between individual patients like himself-those who have incurable diseases, such as cancer or AIDS, and who are willing to become participants in clinical research-and the social and political aspects of clinical research trials.

Clinical Research: An Eloquent Request To Rethink the Traditional Approach

Clinical investigators in cancer research are aware that such differences as age, sex, race, genetics, previous chemotherapy regimens, disease, performance status, concomitant disease, and individual patient variations in drug metabolism create a heterogeneous population of patients. These differences can complicate attempts at identifying a maximum tolerated dose (the goal of a phase I trial) that can be applied to a broader population of patients with cancer [20, 21]. Professor Zimmer realizes that the problems associated with physical heterogeneity may be insurmountable. He argues, however, that researchers could compensate for the problems by recognizing and constructively incorporating certain personal and psychological characteristics and goals of patients, including how aggressively they wish to pursue potential therapies, into phase I research trials. Zimmer's suggestion may be radical, but it highlights the importance of a patient's motivation to participate in a phase I clinical trial despite minimal chances for benefit.

Recognizing the effect of these personality traits, Zimmer points out that some patients, because of their assertiveness and awareness, "... seek to participate in protocols ... question the status quo and are most eager to alter it." Professor Zimmer argues that when such patients participate in a conventional phase I trial, investigators should consider the advantages of evaluating the suggestions of these patients on implementing modifications in the phase I study design. Zimmer describes a "control group of more or less uncontrolled patients" who are allowed "... to change a protocol in radical ways"-a suggestion that, if adopted as policy, could have alarming repercussions for clinical investigators and regulatory groups. On a more reasonable note, Zimmer encourages medical professionals to change the focus from minimizing potential risk to maximizing potential benefit if a patient has a terminal disease and is involved in a phase I trial. Zimmer's words clearly echo those of AIDS activists. Equally challenging is his suggestion that the design of such an unusual trial might be feasible and ethically acceptable if it were conducted with patients who had special characteristics, including the proper motivation, ability to understand the substantial risks involved in a radical trial, and willingness to accept such risks.

Zimmer suggests that society has persons who are adverse to risk and others who are risk takers. Perhaps the crux of his essay is to ask whether risk-taking behaviors in the scientific battle against cancer should be denied to individual patients who have incurable cancer and therefore have little to lose. Zimmer wonders why we deny the request of dying patients that they be permitted to approach death with a heroic struggle.

What if investigators could find a way to incorporate Zimmer's challenging ideas into rigorous clinical research? Such a merger might be accomplished by first studying the complex ethical and personal issues that affect the motivations of patients and their decision to participate in investigational phase I trials. A more informed and meaningful dialogue between investigators and patients who have terminal illnesses would recognize patients' vulnerability but also their autonomous decision-making ability. Innovative ways to design clinical trials, including phase I trials that consider patient motivation, understanding, and risk-taking behavior, could emerge from such a dialogue.

The process of drug development currently begins with a phase I study. Overall, the chances of response in patients in phase I cancer trials have been low [22, 23]. Response rates of only 4% to 6% have been shown, and complete responses are extremely rare. Of significance, and perhaps contrary to what many persons believe, the overall chance of fatal reactions to drugs in these trials is also extremely low, approximately 0.5% to 1%. Also of importance is that most responses in these studies have occurred at 80% to 120% of the subsequently recommended dose for phase II trials, which is at or near the maximum tolerated dose. If patients were participating in phase I trials solely for altruistic reasons, that is, to advance cancer research and to help future patients, phase I trials would probably carry less ethical conflict. Professor Zimmer's essay reminds us that the ethical dilemmas presented by phase I trials and other clinical research are numerous and complex and extend beyond issues related simply to altruism and human experimentation.

Zimmer's Impact on Research Methodology

Thus, challenged and encouraged by Professor Zimmer's ideas, we have initiated a research program that has been designed to explore the motivation, understanding, and rights of patients who choose to participate in phase I cancer trials. The research began with a pilot study of patients with cancer who had given informed consent to participate in phase I trials at our institution [24]. Objectives of the study were to identify and quantify factors in the ethical decision-making process of patients who had refractory malignant conditions and chose to participate in phase I trials. Oncologists who consented patients for the trials were also surveyed. Our findings led to the conclusion that patients are motivated to participate in phase I trials almost exclusively by the hope of therapeutic benefit. Although no patient in the pilot study mentioned altruism as a major motivating factor, we found evidence that some patients may have altruistic feelings toward other patients with cancer. Other conclusions from this study were that patients perceive themselves as adequately informed about the investigational nature of phase I trials. Despite this perception is the particularly concerning fact that most patients in our pilot study apparently did not know that the purpose of phase I trials is to study dose schedules, toxicity levels, or both. From information gathered on the survey that had been administered to oncologists, we concluded that many of them used phase I agents with therapeutic intent and that they believed patients derived psychological benefit by participating in clinical trials.

We have expanded our initial examination of patients and physicians involved in phase I studies by attempting to validate the findings from our pilot study [25]. In addition, we have conducted a feasibility study that attempts to incorporate the personal decision-making process of patients involved in phase I trials and that allows patients to determine the amount of risk they are willing to accept in a clinical trial by giving them the option of selecting (within defined limits) their own doses of investigational agents [26]. The goals of this study were to increase patient understanding and autonomy through direct participation in research decisions and to increase patient satisfaction with clinical research.

The focus of our examination of research ethics and informed consent in early clinical trials of cancer would seem extremely relevant and timely in light of the recently published report from the President's Advisory Committee on Human Radiation Experiments [27]. The Committee's report, although commissioned to study previous abuses in research, includes a study of patients who were participating in clinical research. The findings echo our published results [24], as do the concluding recommendations, which call for further empirical study of the current process of informed consent by participants in human research.

Conclusion

Deficiencies in the current process of clinical trials are apparent. In cancer research, such problems include accrual, the ability of patients and physicians to make autonomous decisions, and arguments from patients and advocacy groups in favor of adequate reimbursement [27, 28]. By conducting empirical research on the clinical trial process, we may begin to address some of these problems. We believe that our studies can affect not only the care of patients with cancer but also the care of patients who have other life-threatening illnesses. Our hope is that this research effort may lead to increased freedom of choice for patients with cancer who are participating in phase I trials and that this might result in greater patient satisfaction, improved patient outcome, and an increased number of patients who choose to participate in cancer clinical trials. Finally, we hope that this research initiative can shed some light on issues relating to informed consent in clinical research and human experimentation in general and on the decision-making processes of potentially vulnerable patients. Even if our research demonstrates that clinical outcomes do not change and that patients do not benefit from increased freedom of choice in research trials, we must remember that there are some patients like Professor Zimmer for whom the most important benefit may be having their personal views respected and perhaps put to use. We are also confident that clinical investigators can develop a better understanding of what motivates patients who willingly choose to participate in clinical research.

Dr. Siegler: The MacLean Center for Clinical Medical Ethics, University of Chicago, 5841 South Maryland Avenue, Chicago, IL 60637.

Dr. Ratain: Section of Hematology and Oncology, the Committee on Clinical Pharmacology, Cancer Research Center, University of Chicago, 5841 South Maryland Avenue, Chicago, IL 60637.