| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1 Division of Pulmonary and Critical Care Medicine, Department of Medicine, The Johns Hopkins University School of Medicine, Baltimore, Maryland
Correspondence and requests for reprints should be addressed to Robert A. Wise, M.D., Johns Hopkins University School of Medicine, Johns Hopkins Asthma and Allergy Center, 5501 Hopkins Bayview Circle, Baltimore, MD 21224. E-mail: rwise{at}jhmi.edu
ABSTRACT
Pulmonary clinical research possesses no unique ethical issues compared with other areas of medicine, but the range of disease states and conditions in pulmonary clinical research are broad. The ethical values guiding clinical research in pulmonary medicine are defined by the Belmont Report (i.e., patient autonomy, justice, and beneficence). This article uses some of the most well-known clinical trials in pulmonary medicine to illustrate these issues and the numerous perspectives that must be considered. Ethical issues covered include introduction of new surgical procedures, placebo controls and drug withdrawal designs, smoking status in trials of patients with chronic obstructive pulmonary disease, and the tensions between community practice, clinical care guidelines, and evidence-based medicine.
Key Words: research ethics • chronic obstructive pulmonary disease • asthma • smoking cessation
There is nothing unique about clinical trials in pulmonary and critical care medicine that warrants special consideration regarding ethical standards. However, few specialties deal with as broad a range of severity and types of afflictions. Pulmonary and critical care medicine challenges the clinical trialist with diseases and treatments that span the entire spectrum of disease states, from screening asymptomatic people to the treatment of desperately ill individuals with little hope for effective treatment. Both symptomatic children and the critically ill elderly suffer from disorders treated by pulmonary and critical care physicians, thus posing widely disparate challenges and ethical dilemmas in pulmonary clinical research. Ultimately, the ethical values embodied in the Belmont report—patient autonomy, justice, and beneficence—are no different for pulmonary and critical care clinical research than any other field, but the richness of the research opportunities can provide the trialist with subtle and complex ethical dilemmas for which precedent may be lacking or conflictive. The purpose of this article is to review some of the ethical debates that have challenged pulmonary clinical trialists in recent years and, where possible, to indicate how they have been resolved. I am writing this article as a physician involved in clinical pulmonary research rather than as a professional ethicist, so my perspective tends toward the practical rather than the philosophical and toward the empirical rather than the theoretical. I have been associated personally, in one way or another, with most of the pulmonary trials described in this article, and the observations and interpretations are wholly my own and do not reflect any official position or opinion of other parties.
INTRODUCTION OF NEW PROCEDURES: THE NATIONAL EMPHYSEMA TREATMENT TRIAL AND LUNG VOLUME REDUCTION SURGERY
The National Emphysema Treatment Trial (NETT) emerged from published observations by Cooper and coworkers of the remarkable apparent success of lung volume reduction surgery (LVRS) in 20 patients. The results showed a dramatic 82% improvement in mean forced expiratory volume in 1 second (FEV1), with relief of dyspnea and improvement in exercise tolerance and quality of life. With up to 15 months of follow-up, no mortality was reported (1). Press releases and news stories cited these results as a major breakthrough in a hitherto untreatable disease. The response and enthusiasm among the medical community led to the very rapid implementation of this intervention in hospitals across the United States, in some cases before the case series was published. Patient demand was driven by claims in the lay press of dramatic benefit, and hospitals promoted the procedure directly to physicians as well as the public, emphasizing the benefits, minimizing the risks, and suggesting that LVRS was a breakthrough procedure rather than an experimental approach (2). In 1995, Medicare issued a special CPT code for LVRS. In the first 4 months after this code was available, 711 patients had this procedure (and probably many others also had it but were listed under other codes). The results from these first patients with codified LVRS were strikingly different from the original series: the mortality at 1 year was 26%, with prolonged hospitalization in 16% of patients, readmission within 15 months in 40% of patients, and an average hospital stay of 30 days in the first year (3). Not surprisingly, these early results caused great concern, especially when considered in conjunction with the cost of the procedure (approximately $35,000). The National Heart, Lung and Blood Institute (NHLBI) conducted a workshop in 1995, from which they concluded that the procedure was promising but a clinical trial was needed for proof of efficacy. The NHLBI recommended that a randomized trial with a nonsurgical comparison arm be conducted (4). The Health Care Finance Administration (HCFA) Technology Assessment Group also reviewed the relevant literature and concluded that the procedure lacked evidence of efficacy. HCFA therefore denied reimbursement and revoked the code for LVRS. The reversal of reimbursement for LVRS was greeted with a landslide of criticism within and outside the medical community, encouraged by tripartite pressures: from desperate patients hoping for a safe and beneficial treatment, hospitals who saw the procedure as a source of revenue to replace flagging income from cardiac surgery after widespread use of angioplasty, and political pressure exerted by the manufacturers of the stapled pericardial strips used for the procedure. Over $5 million was spent on the pericardial strips alone in 1995. The estimated losses just to the device manufacturers was in the tens of millions of dollars (5).
Issues of financial conflict are inherent in recommending any medical procedure for which physicians are paid for performing the procedure, or for which hospitals gain status and income for conducting advanced procedures. Financial pressures were particularly evident in the debate over whether LVRS was effective. The manufacturer of the costly stapled pericardial strips opposed the restriction on reimbursement for the procedure; they hired lobbying firms to influence congressional pressure on Medicare and the NIH. At one point, in opposition to the equipoise and ambivalence of the medical community, Congress passed a statute requiring that Medicare report how they planned to reinstitute payment for the procedure and took the unusual step of holding hearings on whether the surgery was indeed effective. At one point, the company that licensed the pericardial staple buttressing material sponsored a cocktail party at the American Association of Thoracic Surgery to influence the thoracic surgical community against the conduct of a clinical trial for the procedure. The invitation featured a picture of a fishnet, with the title "Trapped in the NETT" and inside announced that the proposed NETT was "HCFA's HMO model to control surgeons and restrict patient access." Thoracic surgeons were invited to meet the CEO of the pericardial strip manufacturer, and a former senator working for their lobbying firm, "to speak with the people who are working to return the power to make medical decisions to physicians and their patients." Surgeons who conducted the procedure, including one who reportedly received licensing fees for sale of the pericardial strips declared, "When done properly, this operation works. There is no question about it. Where there have been problems, they have come from...too many doctors doing it with little experience" (6). Thus, the lines of the debate were drawn between those who thought that the new procedure might do more harm than good and those who thought that patients were being deprived of a beneficial procedure. The ethics of an LVRS clinical trial were hotly debated at the 1996 American Thoracic Society Meetings, specifically the level of evidence needed to declare a surgical procedure nonexperimental; the risk versus benefit of the new procedure; the ethics of denying a costly but morbid experimental treatment to symptomatic patients; and the ethical mandate of whether to provide the surgery to patients and surgeons who thought it might help an individual patient.
In the debate, one of the procedure's pioneering surgeons stated that to deny patients access to the surgery was a Tuskegee study of the elderly—forcing victims of emphysema into an untreated control group. The analogy was that elderly Americans were being treated like the poor African Americans who went untreated for syphilis without consent. This was echoed in testimony to congress by a former President of the American Lung Association. The Tuskegee argument was rebutted by the historical truth: the ethical breech of the Tuskegee study was that the study patients thought they were receiving routine care while they were enrolled in research, thus not giving adequate informed consent. No proposal had been put forward that participants in the NETT should not know that they were involved in a clinical trial. Ironically, the procedure's originators, who included physicians receiving payment from the pericardial strip device manufacturer, admitted that the original series of patients who had undergone LVRS did not provide informed consent under IRB review because it was a standard surgical practice to remove lung tissue from patients with emphysema. Informed consent was only given by patients afterward to undergo follow-up testing after the surgery had been completed to document its success. Proponents of an LVRS clinical trial argued that they were not proposing that patients should be enrolled without providing informed consent, but questioned the ethics of providing an unproven and potentially morbid treatment to hopeful and symptomatic patients absent a strong level of evidence.
Ultimately, and despite congressional political pressure opposing it, HCFA (later renamed Center for Medicare Services [CMS]) and NIH-NHLBI jointly sponsored a randomized, controlled trial to study LVRS (the NETT), with the first patient randomized in 1998. HCFA supported the clinical costs of the trial, while NHLBI paid for the research components. The NETT underwent intense public scrutiny and debate. CMS was ordered to report the trial's progress and any interim scientific findings to Congress, and an external board of ethics experts was asked to review the ethical propriety of the trial. An independent Data and Safety Monitoring Board (DSMB) was created and was required to review the trial's interim results every 6 months and to conduct a comprehensive literature review every 3 months, to judge whether there was compelling evidence that the treatment was effective so that the trial would not be prolonged excessively.
One of the contentious issues was that participating institutions were initially enjoined from providing LVRS to patients not participating in the trial, as such participation would be a conflict of commitment by the institutions. Because the participating institutions were chosen on the basis of their demonstrated expertise in conducting the surgery, this rule would have effectively removed access to the surgery by a large number of the most experienced institutions. After consideration, it was concluded that enforcing this rule would not be feasible for surgeons who were not trial investigators. Therefore, it was modified such that participating NETT surgeons were not permitted to conduct LVRS outside of the trial.
The trial was approved by all of the participating IRBs, except for Washington University in St. Louis (WUSL); the principal investigator at WUSL thought that NETT was unethical. Originally, he had participated in the trial design in the hope that there would not be a randomized control group or that the trial would be limited to include people who did not meet "conventional" indications for the procedure. Ultimately, however, the WUSL IRB concluded that there was not sufficient equipoise for the trial to be conducted at their institution and questioned whether there was sufficient equipoise to conduct the trial at all. WUSL took the unusual step of disseminating their rejection letter to all of the other participating IRBs, though it was not clear whether their intention was to influence other IRBs or simply to inform them of their action. No mention was made in their rejection of the trial that faculty at their institution might have had a financial conflict of interest through licensing fees for the pericardial strips used in the procedure, so it is not known whether this contributed to their rejection of the trial.
An interesting ethical and legal question was raised regarding Medicare's funding the payment for the surgery and follow-up medical care for trial participants. Medicare, by statute, covers only 80% of the charges for these procedures with the additional 20% co-pay, paid by the patient or their co-insurance. For the most part, the investigators and participating NETT hospitals were willing to forgo the charges for the 20% co-pay, so study participants did not receive bills for participating in research. It was surprising to many that this would be illegal under the current statutes, which contain provisions designed to prevent "kick-backs" from medical providers to patients. Ultimately, this was partially resolved by an advisory opinion from the Office of the Inspector General that, except under very specific circumstances, it would be unlikely that participating physicians would be prosecuted for such behavior. Nonetheless, the physicians and hospitals involved in the NETT who did not take diligent action to collect co-payments were in violation of the law. To date, none of the NETT physicians or hospitals has been prosecuted under this statute, but they still remain caught in a gray area between the law and their compassionate desire not to charge subjects for participating in research.
As the NETT progressed, under very close observation by the DSMB, subgroup analyses were conducted to identify characteristics of patients who would particularly benefit from the surgery, so those patients could be excluded from the trial and be permitted to have the surgery at the earliest possible time. As it turned out, the opposite was true. The DSMB identified early a subgroup of patients with certain radiographic and physiologic criteria who were highly susceptible to death after this surgery, which was rapidly reported in the medical literature (7). Within 48 hours of the Board's findings, all such patients were excluded from the trial, and the results were put into the public domain within 6 weeks (8).
The publication's title, "Patients at High Risk of Death After Lung Volume Reduction Surgery," may have affected the degree to which this procedure was not accepted into the wider medical community. It perhaps should have been titled "Some Patients at High Risk of Death After Lung Volume Reduction Surgery" (emphasis added). Headlines in public news media cast doubt on the surgery, and the patient request rate for LVRS dropped dramatically. Because of the high demand for LVRS, and the desire to accommodate as many patients as possible, it was initially projected that more than 5,000 patients would be enrolled in the trial. With the reduced demand for LVRS, the actual number of enrollees was considerably lower, ultimately only 1,218.
However, the initial projection that 1 out of 3 participants randomized to the medical control arm would seek LVRS at another institution also proved incorrect, with only 1 out of 20 crossing over. Therefore, the study was completed with preserved power to detect a survival benefit. The final results showed that the overall benefit of LVRS was quite modest, with only 15% of patients demonstrating an improvement in exercise capacity of 10 watts or more compared with 3% in the medical group. There was no difference in mortality at 2 years, but 8% of the surgical patients died within 90 days of surgery. Although it was conventionally thought that patients who had the best exercise capacity after pre-surgical rehabilitation were the best candidates for LVRS, subgroup analyses showed that such patients had higher mortality after the surgery compared with those who received medical treatment, particularly if they also had diffuse emphysema. Patients without a beneficial response to exercise training had the greatest benefit from surgery in terms of both exercise capacity and survival if they also showed predominant upper lobe emphysema.
Thus, the NETT showed that the initial claims of proven benefit from LVRS were overstated, but it also provided a basis for limited indications for the surgery that subsequently have been incorporated into Medicare payment policies. It can be argued whether NETT ultimately saved many patients in the high-risk groups from nonindicated surgery and risk of death, but it is clear that the trial conduct has provided a strong evidence base for assessing the relative risks and benefits of the procedure and permitting patients and physicians to make an informed decision regarding whether to undergo the surgery. Now that the procedure is paid for by Medicare for patients likely to benefit, the enthusiasm for it has markedly diminished, and there is little promotion for the surgery by hospitals and physicians. In the 5 years since the NETT results were published, only about 400 patients have undergone the procedure, compared with the more than 1,000 patients in the first 4 months after the initial results were published in the lay press (9).
ISSUES RAISED BY THE NETT
Ultimately, the NETT experience is one of the major success stories of clinical research, having a meaningful impact on healthcare policy and standards of care (10). However, the NETT story raises many issues that are pertinent to the ethical conduct of clinical research. Surgical procedures, in particular, are inherently problematic, insofar as they are conducted within the context of clinical care and are not subject to the same scrutiny as new drugs or devices. What is the level of proof needed for new surgical procedures, or procedures in general, before they are no longer considered to be experimental? It would be unthinkable that a new drug would be as passionately promoted and widely adopted as LVRS after an uncontrolled exposure in only 20 patients. When a surgeon performs a new procedure, what is his or her obligation to seek informed consent and IRB oversight? What is the obligation of a physician to disclose publicly and to patients financial conflicts from the invention of devices that are inherent to the performance of the procedure? Is it ethical to deny coverage for experimental treatments outside of the context of a clinical trial? Should physicians who participate in research be enjoined from providing the same treatment to other nonresearch patients as evidence of equipoise?
Surgical trials also raise the vexing question of whether sham surgery should be used in the control group to minimize crossover and surgical placebo effects. Although the NETT did not have a large crossover from medical to surgical treatment, a recent trial of surgical approaches to spinal disease was essentially uninterpretable because of the high level of crossover from medical to surgical therapy (11). The accompanying editorial suggested that this might have been avoided by a sham surgical control (12). In this regard, some recent airway device trials for COPD have proposed employing sham control procedures, including anesthesia, mindful of the attendant risks. Thus, the risk of subjecting volunteers to a scientifically invalid study needs to be weighed against the risk of subjecting volunteers to potentially harmful placebo interventions.
In retrospect, much of the acrimonious debate over LVRS could have been avoided by: forgoing pre-publication press releases, which fueled demand by desperate patients; by open disclosure of financial conflicts of physicians, hospitals, and device manufacturers; by eliminating congressional pressure on government agencies to fund or to not investigate experimental procedures that bring revenue to their home states; by review of early cases of experimental surgery by an IRB or similar body to provide a rational and measured introduction of new methods; by limitation of hospitals in broadly promoting and advertising unproven experimental procedures under the ruse of advanced technology; and by encouragement of agencies and insurance companies who pay for new procedures to evaluate outcomes data before developing coverage policies.
PLACEBO CONTROLS AND DRUG WITHDRAWAL: THE SOCS TRIAL
In 2000, the Declaration of Helsinki was updated to define the conditions in which use of a placebo comparison group is considered to be ethical. The new guidelines stated, without condition, "A new method should be tested against the best current methods" and a placebo is acceptable only in the circumstances "where no proven prophylactic diagnostic or therapeutic method exists" (http://www.wma.net/e/policy/pdf/17c.pdf, accessed Dec 8, 2006). The U.S. Food and Drug Administration (FDA) considered this new update with unease, because they felt that approving drugs for noninferiority compared with current therapy was different from proving efficacy against placebo, except for cases in which a placebo was clearly harmful for patients, such as antibiotics for serious infections or chemotherapy for advanced cancer. Robert Temple, the Policy Director for FDA's drug divisions, considered the ethical implications of the revised Helsinki Declaration, which barred use of placebos, to be "bizarre" (13). In 2002, the World Medical Association (WMA) modified their position by issuing a clarification to the 2000 declaration, indicating that placebos could be used in some limited circumstances: "Where for compelling and scientifically sound methodological reasons its use is necessary to determine the efficacy or safety of a prophylactic, diagnostic or therapeutic method; or where a prophylactic, diagnostic or therapeutic method is being investigated for a minor condition and the patients who receive placebo will not be subject to any additional risk of serious or irreversible harm." Nonetheless, this guidance still leaves considerable room for interpretation. There is not a bright line showing whether an incremental improvement in treatment should be considered "compelling," or whether disorders such as COPD and asthma should be considered "minor conditions." Most investigators and IRBs accept some expanded interpretation of the Helsinki Declaration under certain circumstances: when the benefit of the knowledge gained for the community might outweigh the potential for harm; when the period of risk is not long; when adequate provisions are in place for rescue therapy; and when trial monitoring for safety and efficacy is instituted (14). Thus, in practice, the stringent limitations placed on placebo-controlled trials by the Helsinki Declaration are often ignored by the medical and scientific community. However, the debate over use of placebo-controlled trials has spurred the more critical review of the unquestioned necessity for placebo controls when the benefit of the treatment seems obvious (15).
Should Active Therapy be Withdrawn from Asymptomatic Patients?
Use of placebos, particularly when they are given to patients with stable disease in place of effective treatment, to cause destabilization, raises particular concerns. In asthma research, treatment withdrawal in asymptomatic or well-controlled patients is an efficient study design and is commonly done, but it presents several ethical dilemmas (16). For example, the SOCS Trial was designed to examine the effectiveness of a long-acting ß-agonist (LABA) as replacement therapy in patients whose asthma was well controlled by an inhaled corticosteroid (ICS). Patients with asthma who were stable on ICS were randomly assigned to 4 months of treatment with placebo, a LABA, or continued ICS. After the 4-month treatment period, all three groups received placebo to determine whether there was a prolonged protective effect of ICS compared with LABA (17).
Results from the SOCS trial showed no significant differences in conventional outcomes between the ICS and LABA groups, but a threefold increase in treatment failures in those who switched to LABA (24% versus 6%, p = 0.004). There was no prolonged protective effect or rebound effect from discontinuing any one of the active treatments compared with placebo. The findings from this study have proven immensely important insofar as it changed treatment practice for asthma to avoid LABA monotherapy for persistent asthma. However, the SOCS trial was strongly criticized because of the use of a placebo treatment arm:
The authors did not address whether there was clinical equipoise between the safety or efficacy of either of these agents compared with placebo. If no equipoise exists between the study arms, then we question whether it was justifiable to expose the placebo group to the risks of under-treatment, and all patients to this risk during the placebo run-out phase....This concern is particularly relevant to children enrolled in the study. Since 97% of all participants who were randomized were already being treated with ICS, this protocol appears not to offer prospects of direct benefit, particularly for those who received placebo. (18)
The criticism of the placebo group in this study was echoed by others (19).
The SOCS study highlights the ethicality of using placebos for patients with asthma. There is no question that the knowledge gained from this study has had a major impact on the approach to managing patients with persistent asthma. There is also little doubt that the patients with asthma participating in the trial were subjected to some increased risk of asthma destabilization during the drug withdrawal to placebo treatment. The SOCS trial raised the issue of whether adequate controls were in place to monitor and rescue those patients who developed symptoms. Very likely this was the case, because no serious harm was reported for any of the participants. Moreover, nearly two-thirds of the patients on placebo had no ill effects over 4 months, suggesting that many patients are using ICS either longer than needed or without immediate need—a study result that has not been emphasized. Thus, one has to consider this trial within the daily practice of clinical medicine where it is common to attempt treatment withdrawal in patients with well-controlled disease and monitor the results. If one accepts the dictum that, in clinical research, withdrawing treatments from patients with well-controlled disease is unethical, it seems logical that one should never ethically withdraw treatment from stable patients in clinical practice except for intolerance. Nonetheless, this trial, and the criticism that it received, has chilled investigators' interest in withdrawing active treatment in patients with stable lung disease, even under close observation. As a result, we will be less likely to know whether certain medicines, once started during the acute phase of a chronic illness, can be safely withdrawn.
OBLIGATIONS TO FOLLOW ESTABLISHED CARE PRACTICES: THE CHILDHOOD ASTHMA MANAGEMENT PROGRAM
When Are Trials Obligated to Follow Care Guidelines?
The Childhood Asthma Management Program (CAMP) evaluated the effect of anti-inflammatory drugs on childhood lung growth in asthma. A total of 1,041 children ages 5 to 12 years with mild-to-moderate persistent asthma received an ICS (budesonide), an inhaled nonsteroidal anti-inflammatory drug (nedocromil), or placebo, and all three groups received albuterol as rescue medication. The primary outcome was lung growth as measured by FEV1 over 5 to 6 years of follow up (20). During the development of the CAMP trial design, questions were raised as to whether it was ethical to treat children with asthma outside of expert panel guidelines, even with the admission that evidence to support the guidelines was not compelling. In addition, one of the participating IRBs questioned the ethicality of submitting children to the risk, albeit minor, of methacholine inhalation during follow-up visits. Although methacholine challenge is a routine clinical procedure for asthma diagnosis, all of the participating children presumably had their asthma diagnoses established clinically and confirmed with a baseline methacholine challenge test. Thus, the accrued benefit for the children participating in the study was thought to be small. However, because the risk was considered low and the monitoring was close, methacholine challenges were permitted by the review committees and the monitoring board.
The CAMP results showed that ICS are associated with less axial growth (by about 0.5 inches) and lower FEV1 (but the same percent predicted values because the children were shorter), fewer emergency department visits, and fewer steroid courses. Nedocromil was no different from placebo for most of the efficacy outcomes.
The CAMP trial posed several interesting ethical issues. Is it ethical to not treat with ICS or nedocromil when they are considered by expert panels to be part of the "standard of care"? In this case, the standard of care was not supported by rigorous evidence, though the subtlety of this is not always apparent to review committees. As it turned out, the treatment with nedocromil for children like those enrolled in CAMP was little better than placebo and was an unnecessary treatment. The ICS proved to have complex effects with some benefit and some harm, but overall were effective for asthma treatment. The lesson from CAMP, therefore, is that guidelines and standard of care ought to be thought of as starting points for clinical research rather than the end point, particularly when the evidence base is less than convincing. The ethicality of conducting research involving methacholine challenge tests in children with asthma is a more vexing issue insofar as there was little accrued benefit from such testing except in those with airways reactivity regression and thus might be taken off active treatment.
When Are Trials Obligated to Follow Community Practices?
The converse debate occurred during the grant review process of a trial of nocturnal oxygen therapy in patients with COPD with nocturnal oxygen desaturation (NOD). Although Medicare covers therapy with nocturnal oxygen for patients with NOD, the common practice of treating these patients is supported by neither current guidelines nor the evidence base. Nonetheless, it is common practice in the community to provide nocturnal oxygen for patients known to have NOD, despite lack of evidence. The ATS/ERS guidelines state that testing and treatment for NOD is not recommended for COPD except in special circumstances, such as sleep disturbance or the presence of complications from hypoxemia (e.g., polycythemia or cor pulmonale) that are not explained by the awake PaO2 (21). The GOLD guidelines state that, "A decision about the use of long-term oxygen should be based on the waking PaO2 values" (22). Some have suggested that nocturnal oxygen might be harmful to patients with isolated NOD because of induction of hypercapnia (23). Because of this controversy, the NHLBI convened an expert workshop on this topic, from which a trial of NOD during sleep was recommended as a priority. Accordingly, a small, randomized, double-blind, proof-of-principle trial was proposed to study the biological effects of treating patients with COPD with satisfactory daytime oxygen tensions and newly diagnosed NOD using oxygen compared with sham treatment for 6 weeks. The trial was soundly rejected by the NIH review panel evaluating the trial on the basis of ethical impropriety, and an ethical concern was registered with the NIH: "... the risk associated with oxygen therapy removal from subjects once NOD [is detected, sic] is unethical and unacceptable even for short periods of time" (personal communication). Obviously, the trial was not approved, the investigators were chastened, and the study will not be conducted. The lessons learned are that community practice may deviate from expert opinion or scientific evidence, and once a therapeutic practice becomes established, it may be considered "unethical" to test whether it is correct, no matter how weak the evidence.
SMOKING STATUS IN COPD TRIALS: WHAT ARE THE OBLIGATIONS OF INVESTIGATORS?
Smoking cessation is a central tenet of COPD care. Unfortunately, most smoking cessation programs are only partially effective, which leads to complex issues for clinical trial investigators. In most studies, the scientific objective depends upon the participants' stable background condition, meaning that current smokers should continue to smoke and that quitters should continue to refrain from smoking. This may conflict, however, with the investigator's obligation to provide adequate treatment and information to patients within the context of clinical research. Thus, judgments are required to determine what the study investigator's obligation is to induce smoking cessation in therapeutic trials of patients with COPD. Is it equitable to enroll only smokers or only nonsmokers in trials in which it is not known which group might benefit? The several large COPD clinical trials that evaluated the efficacy of ICS to modify the course of COPD enrolled mainly active smokers or those who had recently quit (24, 25). Although not explicitly framed in this way, these trials could have been interpreted as testing the hypothesis of whether treatments can make it safer for high-risk patients to smoke. This is similar to the issues raised regarding the propriety of testing AIDS vaccines in those who have high-risk behaviors (26).
SUMMARY
The case studies presented in this article illustrate the ethical dilemmas, conflicts, and uncertainties that investigators engaged in pulmonary clinical research face. Often these issues appear differently in hindsight, after the research is completed – sometimes reflecting well on the investigators, and sometimes not. In virtually all such situations, clinical investigators are dedicated to conducting rigorous, scientifically valid, and ethically sound trials. The difficult job of the clinical trials is to balance the sometimes conflicting ethical values using judgment and wisdom.
DISCUSSION HIGHLIGHTS
DR. ROBERT R. FENICHEL (Independent Consultant, Washington, DC): There are really 2 kinds of benefits from a drug—sometimes you get one, sometimes you get the other, and sometimes you get both. One effect is to make patients feel better, and the other is to avert irreversible harm. Patients frequently relinquish the chance of feeling better (for example, by not taking their medicine). That being so, it should be acceptable for the same patients to accept randomization into a trial in which symptoms are likely (active treatment) or unlikely (placebo) to be relieved. A patient who is uncomfortable with his or her treatment can always drop out, having lost nothing but time. If the expected effect of active treatment is to avert irreversible harm, or if an expected side effect of symptomatic therapy is to increase irreversible harm, then the issues of informed consent are more difficult, but the choices involved are again choices that patients make in or out of trials. Some patients choose surgery over radiation therapy for laryngeal cancer because of improved survival, but others choose radiation therapy because of improved preservation of speech. (26) Some patients go skydiving, and some don't. Many women still take hormone replacement therapy, knowing that the effect is to increase irreversible harm, but symptoms are reduced.
FOOTNOTES
This article was funded by Boehringer Ingelheim and by Pfizer, Inc.
Conflict of Interest Statement: R.A.W. has served as a paid consultant to Boehringer Ingelheim and Pfizer and received payment for the preparation of this article and attendance at the symposium at which it was presented. He also receives research funding from Boehringer Ingelheim which provided support for the symposium that generated this article. R.R.F. has served as a consultant to AVEO Pharmaceuticals, Abbott Laboratories, BioCryst Pharmaceuticals, Boehringer-Ingelheim Pharmaceuticals, Bracco Diagnostics, Bristol-Myers Squibb, CV Therapeutics, CoTherix, Corgentech, Daiichi Sankyo Pharma, Eli Lilly & Company, Eyetech Pharmaceuticals, Forest, H. Lundbeck A/S, Hope Medical Enterprises, Idun Pharmaceuticals, Janssen Pharmaceuticals, Johnson & Johnson, KAI Pharmaceuticals, King Pharmaceuticals, Laboratoires Fournier, Lifescape BioSciences, Luitpold Pharmaceuticals, LungRx, MDS Pharma, MedImmune, Medtronic, Merck, Mylan-Bertek Pharmaceuticals, Navitas Pharma, Neurochem, NicOx, NovaCardia, Orion Pharma, Otsuka America Pharma, Prestwick Pharmaceuticals, Reliant Pharmaceuticals, Roxane Laboratories, Sanofi-Aventis, Schwarz BioSciences, Scios, Servier Canada, Shire Development, Solvay, TAP Pharmaceutical Products, Teijin America, Vanda Pharmaceuticals, Watson Pharma, Wyeth-Ayerst, and Yamanouchi Pharma America.
(Received in original form January 3, 2007; accepted in final form January 19, 2007)
REFERENCES
This article has been cited by other articles:
|
|
 |
|
  R. A. Wise and M. B. Drummond The Role of NETT in Emphysema Research Proceedings of the ATS, May 1, 2008; 5(4): 385 - 392. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
B. T. Thompson and D. Schoenfeld Usual Care as the Control Group in Clinical Trials of Nonpharmacologic Interventions Proceedings of the ATS, October 1, 2007; 4(7): 577 - 582. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
