RIGHTS AND RESPONSIBILITIES
The philosophical debate over expanded access to investigational drugs takes many forms. People often try to reduce the standard arguments to simple dichotomies; for example, the ''mind versus heart" approach pits the scientific discipline of clinical trials against the compassionate use of experimental drugs for therapeutic purposes. The "beneficence versus autonomy" approach suggests that providing protection to people with HIV infection must conflict with respect for their individual rights.
In fact, most efforts to simplify the debate over expanded access do a disservice by diverting attention from a host of complex issues that must be considered in any discussion of increased access to AIDS drugs. These issues concern (1) the need for constraints on freedom of choice, (2) the capacity for informed consent in an environment characterized by restricted access to health care, (3) the potential for setting the rights of today's patients against the rights of future patients, (4) the shifts in concerns of institutional review boards, and (5) the problems that could arise from early and close collaboration between the FDA and drug sponsors.
This chapter is based on the presentations of J. Richard Crout, Dan Brock, Harold Edgar, Bernard Lo, Daniel Wikler, and Carol Levine.
FREEDOM OF CHOICE
The expedited review process, treatment IND regulations, and recent efforts to establish a parallel track system for AIDS drugs (all described in Chapter 1) have led some people to suggest that the FDA is moving away from its traditional role of consumer protection
This chapter is based on the presentations of J. Richard Crout, Dan Brock, Harold Edger, Bernard Lo, Daniel Wikler, and Carol Levine.
toward a new vision of patient autonomy. Scientists involved in these programs indicate, however, that they were never intended to promote freedom of choice as an independent value in drug development. Instead, they were designed to increase options for desperately ill patients with no therapeutic alternatives.
The extreme argument for freedom of choice has been that persons infected with HIV have "little or nothing to lose," so why limit their access to any drugs? The problem with this argument is that it fails to recognize the association between desperation and vulnerability. Exploitation of people with HIV infection by unscrupulous vendors with worthless products has been a significant problem. People with HIV infection do have something to lose: they can waste time, energy, and hope—or even become sicker—on substances that would never reach the marketplace through normal channels.
Totally free access to substances that may or may not be fraudulent is quite different from the opportunity to make informed, reasoned choices about products for which there is a reasonable expectation of effectiveness (based on preclinical or early clinical data). Clearly, persons with life-threatening illnesses are willing to assume greater risks in exchange for smaller potential benefits than other groups of patients. Expanded access programs recognize the right to assume such risks—with the advice and assistance of a personal physician—without abandoning the individual to the forces of the marketplace.
Early experiences with both treatment INDs and the parallel track approach have demonstrated that several external factors can limit a person's ability to make informed, reasoned choices about participation in experimental protocols. Two of the most troublesome factors are the lack of adequate information about drug products and the shortage of health care options—for some people with HIV disease, clinical trials may represent the only opportunity for access to a knowledgeable medical team.
For most persons with HIV infection and their physicians, the first encounter with a new drug or treatment alternative occurs through
the lay press. Traditional attitudes within the medical establishment about publishing first in the medical literature and then in the press have been tempered with regard to AIDS because of the recognized need to get information out "on the streets" as quickly as possible. News about successful studies fosters hope and gives patients a sense of control that they might not otherwise have.
But there are also disadvantages to this strategy. Physicians who treat HIV-infected patients complain that news reports do not have sufficient clinical detail to allow them or their patients to make informed decisions. For example, the patient may be especially concerned about one particular side effect of a drug, such as fatigue. News reports rarely present specific information about the incidence of an adverse effect unless the problem is severely debilitating.
The long delay between publicity about a new therapy and the publication of peer-reviewed journal articles limits the physician's ability to add substance to the decisionmaking process. Events surrounding the establishment of a treatment IND for aerosolized pentamidine illustrate this concern. The treatment IND was issued in February 1989, simultaneously with a press release announcing the effectiveness of the therapy as prophylaxis for Pneumocystis carinii pneumonia. San Francisco investigators presented a formal abstract describing community trials of the treatment at the international AIDS meeting in Montreal several months later, and by late June the FDA had approved the drug for marketing. As of March 1990, however, data from the community trial of aerosolized pentamidine still had not been published in a peer-reviewed journal.1
Another problem with depending on the media to disseminate information is the potential for bias. Reporters may have difficulty achieving objectivity in a news report based on a press conference called by a drug manufacturer, a funding agency, or an investigator who has devoted several years to a drug study. The need for eye-catching headlines also hinders efforts to place new discoveries in perspective.
Physicians who treat HIV-infected patients have suggested several ways to improve communication about promising new drugs. One suggestion has been to encourage a standard format for press releases
that includes basic information about study subjects, methodology, endpoints, and results (the same type of information that would be included in a formal abstract). Another option is to speed up journal publication. Alternatively, an independent group could be designated to review data and present key results in an informational letter to physicians, perhaps in the FDA Drug Bulletin. Other options include a mechanism similar to the "Clinical Alert System" adopted by the National Cancer Institute, or a federally sponsored on-line data base that would provide relevant data from clinical trials to HIV-infected patients and their physicians.
The proposed policy statement for parallel track protocols emphasizes the development of appropriate mechanisms to educate potential drug recipients and their physicians. Initially, the sponsor must provide patients with enough information to compare the potential risks and benefits of a new drug with the risks and benefits of other treatment options. The sponsor also has the more difficult task of ensuring that information acquired during the course of a parallel track protocol—especially with regard to adverse effects—is relayed to participants as quickly as possible. Physicians involved in the early days of the parallel track protocol for ddI report that lack of such information sometimes hampered their efforts to provide appropriate care for their patients.
Two different access problems may distort decisions about participation in conventional clinical trials, as well as in expanded access protocols. The first involves access to health care in general (see Chapter 7). The second involves access to specific drugs.
The AIDS epidemic has had a disproportionate impact on the urban poor in the United States. Socioeconomic factors associated with high rates of intravenous drug abuse, such as poverty, unemployment, and inadequate education, are also associated with higher rates of HIV infection, especially in the major population centers of the Northeast. Repeated studies have shown that access to primary medical care is inadequate for the impoverished men, women, and children who are most likely to become infected with HIV. Contributing to access difficulties is the fact that the proportion of physicians
practicing primary care specialties has dropped precipitously in this country over the past two decades. The number of physicians who are willing to provide primary care for Medicaid patients with HIV infection is extremely small.
One dilemma created by this situation is that patients may have to enter drug trials to obtain basic health care. All of the efforts by institutional review boards and others to ensure that participation in randomized clinical trials is voluntary may mean very little if the patient has no alternative form of care. Similarly, the patient's right to withdraw from a trial at any point is jeopardized if dropping out means losing touch with essential health care providers.
Parallel track protocols are unlikely to ease this situation because the primary care provider is the fundamental link between the patient and the drug sponsor. A patient cannot participate in a parallel track protocol unless his or her primary care physician certifies that the patient meets the requirements of the protocol and that all efforts have been made to use standard therapies. The physician also must agree to participate fully in patient education and to monitor the patient closely for adverse effects of the investigational drug.
Patients who are fortunate enough to have a primary care physician have an advocate in the search for an appropriate experimental therapy. Sometimes, however, a physician will learn that a patient is not eligible to receive a desired drug because he or she does not meet the entry criteria for relevant protocols. A San Francisco physician who encountered this situation recently polled his colleagues about how they would handle it. A surprisingly large number replied that they would ignore the entry criteria and, if necessary, falsify laboratory data rather than deny the patient access to a potentially beneficial drug.
Discussions about misrepresentation in clinical trials have focused mainly on patients who knowingly take drugs that are not authorized by a study protocol or who surreptitiously have their drugs analyzed to determine whether they have received the investigational drug or a placebo in a blinded study. The informal poll described above indicates that the problem could go much deeper. Open-label studies, such as the parallel track protocol, are especially vulnerable to misrepresentation because regulators and sponsors keep reporting requirements to a minimum.
Every time a patient or physician misrepresents a patient's clinical status to enroll the patient in a drug trial, the quality of data collected in that trial diminishes. For example, a study protocol might be designed to assess whether patients who have failed to improve on standard therapies benefit from a new investigational drug. If a patient gained entry to such a study without trying standard therapies, information related to that patient would distort conclusions drawn from the entire trial. Improvement in the patient's condition would be regarded as evidence that the drug had the potential to help a certain group of patients—those who had failed to respond to other measures—when in fact the evidence did not pertain to that group at all.
If many patients and physicians choose to follow this course, future patients will not have accurate information on which to base their own choices about treatment alternatives. The tension between the needs and rights of today's patients and the needs and rights of future patients is an unfortunate corollary of the drug evaluation process. The only way to obtain accurate information about the risks and benefits of an unknown agent is to introduce it through a series of careful, methodical clinical trials. Understandably, however, today's patients often view immediate access to a potentially beneficial drug as a higher priority than the gathering of information for future patient populations. (Sometimes, however, participants in trials are also direct beneficiaries of the results. For example, when the recent trial of AZT in patients with asymptomatic HIV infection was terminated, more than 90 percent of the participants had not yet developed AIDS and were immediately offered AZT.)
Organized expanded access may help resolve this conflict, but only if parallel track protocols do not interfere with enrollment in conventional clinical trials. Patient advocates suggest that one way to increase patient accrual in conventional trials and to decrease the risk of misrepresentation is to include patients, their advocates, and their primary care physicians in the planning of each protocol. Patient representatives can provide important insights into the factors that make a particular trial more or less appealing to the target population.
INSTITUTIONAL REVIEW BOARDS
Expanded access reflects some underlying changes in the philosophy of drug regulation, and these changes will be mirrored in the func-
tioning of institutional review boards, the local organizations charged with protecting the rights of individuals who participate in clinical trials. These changes are subtle and do not call for eliminating the traditional concerns of IRBs; they may, however, challenge IRBs to modify and revise some commonly held principles.
First, there has been a shift in emphasis from nonmaleficence to beneficence; that is, from preventing harm by protecting patients from risk to actively promoting patients' welfare by providing them earlier and broader access to experimental drugs.
Second, a different aspect of patient autonomy is receiving more emphasis. IRBs have always been concerned with patients' rights—to be free from subtle or overt coercion in making decisions about whether or not to participate in a research protocol, to have full information to make informed choices, and to take the risks associated with participation in protocols, as long as those risks are understood. Recently IRBs have had to recognize that many patients are more concerned with their right to take serious risks than with their right to be free from coercion.
Finally, IRBs will have to adjust their perspective on the selection of subjects for participation in trials. Traditionally, IRBs have acted to protect individuals and groups from being included in trials simply because they were easily accessible—for example, prisoners—or because they were vulnerable—for instance, drug users or pregnant women. Today, when patients believe that participation in clinical trials offers substantial promise of benefit, these restrictions may be viewed as discriminatory rather than as protection from harm.
The effect of the AIDS epidemic on drug regulators has not received a great deal of attention, but some observers suggest that calls for expedited development and early expanded access could have a major impact on the way regulators view their own responsibilities. For decades, regulators and drug sponsors have had an almost adversarial relationship. Sponsors produced and organized huge amounts of data and presented it to the FDA for evaluation. The FDA played the devil's advocate, often focusing more on the potential for adverse effects than on the potential benefits of a candidate drug. Regulators were rewarded for refusing marketing privileges to a drug or device that later turned out to be harmful; there were no comparable rewards for making a beneficial drug available quickly.
Advocates for patients with many different diseases have complained that the traditional system placed too much emphasis on caution. But some scientists and legislators worry that new procedures could err in the other direction. If FDA officials become deeply involved in the design of drug protocols, will they still be objective when the time comes to evaluate the data generated by those protocols? If a protocol does not exactly answer the questions that must be addressed to assess safety and efficacy, will the regulator who helped shape the protocol be willing to turn to sponsors and suggest starting over?
Identifying promising drugs and getting them to the marketplace as quickly as possible are extremely worthwhile goals, but everyone should be aware of the potential costs as well as the potential benefits. For example, the new emphasis on expedited development greatly increases the demands on postmarketing testing and surveillance systems. Drugs could be approved for marketing even if some questions remained about the most effective ways to use them. Physicians who treat patients with drugs approved through expedited review must understand the importance of responding quickly to all requests for information and of relaying all questions and concerns to drug sponsors as they arise.