Ethics Forum: The Ethical Pitfalls of Clinical Trials

The Case

Your patient is deciding whether to enroll in a clinical trial at your institution and wants your advice about whether to participate. The study compares two drugs for acute pain following minor outpatient surgery. One drug is a newly developed analgesic that may soon be approved by the U.S. Food and Drug Administration. The other is an older, generic drug that has been around for decades. The protocol randomly assigns study subjects to one of several doses of the novel medication or a fixed dose of the older analgesic.

You are aware of review articles by pain experts describing how physicians commonly underdose the older drug. In fact, prior studies suggest that the “usual” or “standard” dose of the older drug—the one used in the study protocol—provides inadequate pain relief in nearly half of patients. That is why pain specialists carefully up-titrate the dose of this medication to achieve pain relief and routinely recommend that other clinicians do the same. The older drug’s package insert allows a wide range of doses; the study protocol’s fixed dose is at the lower end of this range.

You recognize that your patient may have her pain undertreated if she’s assigned to receive the older drug.

Do you advise your patient to enroll? Do you see any ethical concerns about the design of this study?

Discussion

This scenario raises an issue often faced by researchers designing a study to explore the effectiveness of a new therapy: What is the appropriate comparator? Although a placebo-controlled design represents the gold standard for clinical trials, assigning one group to receive a treatment expected to be ineffective is ethically problematic. Ideally, the control group should (at the very least) receive treatment considered to be the standard of care.

However, in the case described, the trial proposes a comparison group that receives a dose of pain medication known to undertreat pain in many patients. Even though the protocol employs a commonly used dose, the protocol does not allow optimal dosing. Thus, this protocol may be unethical because it assigns some patients to a suboptimal treatment regimen. In addition, the design may be biased toward the new drug because a range of doses of the new drug are being compared to a fixed, commonly ineffective dose of the older drug.

The counter-argument is that this study has a limited goal: to compare a new drug to a commonly prescribed dose of an old drug that would be considered by many clinicians to represent the standard of care. However, this justification raises ethical concerns as well.

What Constitutes an Ethical Clinical Trial?

The policies and recommendations contained in the Nuremberg Code, the Belmont Report, and the International Ethical Guidelines for Biomedical Research Involving Human Subjects and related consensus guidelines represent significant progress in the evolution of ethical clinical trials.^1-3 As institutional review boards are expected to take these well-established guidelines to heart, the question of ensuring an ethical trial may seem academic. However, when common or usual care is also suboptimal, deciding whether a study protocol is ethical may not be straightforward.

In fact, past trials of rheumatoid arthritis could be criticized for underdosing methotrexate in the comparator group.^4,5 More recently, trials comparing febuxostat with allopurinol required those in the allopurinol group to remain on a maximal dose of allopurinol known to inadequately suppress uric acid in a substantial proportion of patients with gout.^6-10

Of course, there may be situations when a treatment considered to be the standard of care is suboptimal but still an ethical choice for a clinical protocol. Concerns regarding cost, tolerability, and accessibility of a highly effective drug may make it a less useful comparator in a clinical trial. So, researchers may reasonably choose a more commonly used (but less effective) regimen. However, none of these considerations apply to the clinical scenario or trials mentioned above.

In thinking about the ethics of treatment trial design, an important paper by Emanuel and colleagues is worth consideration.¹¹ In it, the authors propose a number of principles by which the ethics of clinical trials should be judged. Among them, two are potentially violated by the study of pain medication described above:

• Value: The question this study can answer (How does a common dose of an old drug compare to a new drug?) is not the one clinicians need answered (How does an optimal dose of an old drug compare to a new drug?); and
• Favorable risk–benefit ratio: This study design does not maximize potential benefit to the study subjects as it could have by allowing up-titration of the older pain medication.

Another principle, informed consent, is also at risk if the study protocol did not clearly explain the known failure rate of the old drug at the designated, fixed dose. In fact, it is not clear that the consent could be written in a way that accurately discloses the discrepancy between optimal (rather than common) practice while also encouraging enrollment. Study recruits might decline enrollment if they know they might be assigned to a group with a good chance of having their pain undertreated.

The Importance of Equipoise

Ideally, a study should address treatment groups from the starting point of equipoise—that is, the investigators should be unsure which therapeutic arm will be better treated.¹² While the results of clinical trials are rarely certain before completion, those assigned to an undertreated comparator group are at a distinct disadvantage. This raises the potential criticism that the study design was “rigged” to favor the new drug at the expense of the study subjects in the comparator group.

Would You Enroll?

When faced with a clinical decision, many patients ask their physicians, “What would you do?” A similar question could be asked here: Would you want to enroll as a subject knowing that you might be assigned to the older drug at a dose that is commonly ineffective when dose adjustment could provide relief? If the answer is no (or even I’m not sure), should you encourage your patients to participate in such a trial?

I would not enroll in the pain trial described in the vignette above. If I were a patient with gout, I would not have enrolled in the febuxostat/allopurinol trials knowing I might receive allopurinol at a suboptimal dose. Accordingly, I would not encourage my patients to enroll in such trials.

The Bottom Line

Choosing an appropriate comparator in a clinical trial can be tricky.12 Ethical issues may arise when the comparator group does not receive what we know or believe to be the best available treatment. We should set a high bar for clinical trials, and that means that a treatment that is common but suboptimal may not be good enough. Indeed, it may even be unethical.

Dr. Shmerling is clinical chief of the division of rheumatology and program director of the BIDMC rheumatology fellowship at Beth Israel Deaconess Medical Center in Boston.

References

1. The Nuremberg Code. JAMA. 1996;276:1691.
2. National Commission for the Protection of Human Subjects of Biomedical and Behavioral Research. The Belmont Report. Washington, DC: US Government Printing Office; 1979.
3. Council for International Organizations of Medical Sciences. International Ethical Guidelines for Biomedical Research Involving Human Subjects. Geneva, Switzerland: CIOMS; 1993.
4. Strand V, Cohen S, Schiff M, et al. Treatment of active rheumatoid arthritis with leflunomide compared with placebo and methotrexate. Arch Intern Med. 1999;159:2542-2550.
5. Lipsky PE, van der Heijde DM, St Clair EW, et al. Infliximab and methotrexate in the treatment of rheumatoid arthritis. Anti-tumor necrosis factor trial in rheumatoid arthritis with concomitant therapy study group. N Engl J Med. 2000;343:1594-1602.
6. Becker MA, Schumacher HR Jr, Wortmann RL, et al. Febuxostat compared with allopurinol in patients with hyperuricemia and gout. N Engl J Med. 2005; 353:2450-2461.
7. Schumacher HR Jr, Becker MA, Wortmann RL, et al. Effects of febuxostat versus allopurinol and placebo in reducing serum urate in subjects with hyperuricemia and gout: A 28-week, phase III, randomized, double-blind, parallel-group trial. Arthritis Rheum. 2008, 59:1540-1548.
8. Ogryzlo MA, Urowitz MB, Weber HM, Houpt JB. Effects of allopurinol on gouty and non-gouty uric acid nephropathy. Ann Rheum Dis. 1966;25:673-680.
9. Perez-Ruiz F, Alonso-Ruiz A, Calabozo M, Herrero-Beites A, Garcia-Erauskin G, Ruiz-Lucea E. Efficacy of allopurinol and benzbromarone for the control of hyperuricaemia: A pathogenic approach to the treatment of primary chronic gout. Ann Rheum Dis. 1998;57:545-549.
10. US National Library of Medicine DailyMed: FDA information: Allopurinol tablet. Available at http://dailymed. nlm.nih.gov/dailymed/search.cfm?startswith=allopurinol. Accessed February 2012.
11. Emanuel EJ, Wendler D, Grady C. What Makes Clinical Research Ethical? JAMA. 2000;283:2701-2711.
12. MacKenzie CM, Paget SA. Ethics in Clinical Trials In: Hochberg MC, Silman AJ, Smolen JS, et al. eds. Rheumatoid Arthritis, 1st ed. Philadelphia, PA: Mosby Elsevier; 2009: 413-418.