Complex Clinical Trial Issues Require an Ethical Balancing Act

October 13, 2020
Maurie Markman, MD

OncologyLive, Vol. 21/No. 20, Volume 21, Issue 20

The fundamental objectivity of major scientifically oriented public health agencies has been called into question, while the issue of what constitutes sufficient evidence for a therapeutic strategy to be considered “standard” is also being challenged.

These are complex times for the clinical research community. The fundamental objectivity of major scientifically oriented public health agencies has been called into question, while the issue of what constitutes sufficient evidence for a therapeutic strategy to be considered “standard” is also being challenged.1

These dilemmas highlight a related question faced by oncologists who participate in clinical trials and, most importantly, who are required to explain to patients why participation in a given cancer study may be important for them to consider.

The current discussion highlighted in news reports regarding the validity of retrospective or prospective observational studies versus therapy specifically determined through the process of randomization is at the core of the debate regarding the clinical utility of convalescent plasma in the management of coronavirus disease 2019.1

In fact, the argument is similar to a discussion that may occur commonly in the practice of an oncologist who elects to participate in clinical trial research. Although the decision to become research subjects in investigative efforts is—and must always be—made by the individual patient with cancer themselves, when physicians consider asking patients whether they might be interested in participating in a therapeutic study, these professionals are surely also asking themselves whether a trial would be in the patient’s best interest.

It is here where we directly confront the complex but critically relevant issue of equipoise. As magnificently articulated in the landmark essay by Samuel and Deborah Hellman in the New England Journal of Medicine almost 30 years ago, the fundamental roles of the (a) physician and (b) the physician–clinical scientist differ—at times substantially2: “The physician, by entering into a relationship with an individual patient, assumes certain obligations, including the commitment always to act in the patient’s best interests….The clinical scientist is concerned with answering questions—ie, determining the validity of formally constructed hypotheses. Such scientific information, it is presumed, will benefit humanity in general.”

The clinician’s dilemma occurs in the potential conflict between these 2 most often complementary but, unfortunately, surely not identical goals. The authors wrote: “The ethical obligation created by the covenant between doctor and patient requires the doctor to see the interests of the individual patient as primary and compelling. In essence, the doctor-patient relationship requires doctors to see their patients as bearers of rights who cannot be merely used for the greater good of humanity.”2

If we accept the primacy of this clearly articulated role of an oncologist, under what circumstances would it be ethically acceptable for that clinician to permit a patient with cancer to be randomized to one therapeutic approach versus another?

It is here that the concept of equipoise comes into play. As noted by the Hellmans, if participation by a specific patient in a given randomized clinical trial is to be considered an acceptable therapeutic option, it is essential that “the physician believes that the severity and likelihood of harm and good are evenly balanced.”2 Conversely, when considering possible alternative strategies and potentially available novel approaches in disease management “if, however, he or she believes that the new treatment may be either more or less successful or more or less toxic, the use of randomization is not consistent with fidelity to the patient.”2

Clearly these are very strong statements, but assuming we accept the basic foundational argument differentiating the role of the physician from that of the clinical-scientist, obvious questions follow. For example, would it be ethical for an oncologist to enter a patient into a randomized clinical trial if one of the arms of the study included a standard-of-care antineoplastic drug delivered at what is known to be a potentially excessive toxic dose/schedule, such that the clinician would not administer this specific regimen to patients who were not participants in this randomized study?

In fact, this was precisely the situation in the oncology regulatory clinical trials arena for a number of years when single-agent pegylated liposomal doxorubicin was delivered at a dose of 50 mg/m2 (once every 4 weeks) as a control arm in a number of ovarian cancer studies in platinum-resistant disease when it was well recognized that this dose level was rarely—if ever—administered in noninvestigative clinical practice.3 Considering the issue of equipoise, the question is whether a physician who was also serving as a clinical investigator on such a study was truly acting in their patients’ best interests if they suggested or encouraged patients to become research subjects in these studies.

Fortunately, the highly disturbing ethical concerns associated with these historical ovarian cancer studies are most uncommon today in the realm of oncology clinical trials. However, clinical cancer investigators are frequently called on to consider recommending possible entry of patients into an investigative effort and they must weigh whether participation is truly the optimal therapeutic option for an individual patient at a specific point in their cancer journey.

Further, oncologists must seek to navigate carefully the ongoing intense debate within the cancer community regarding the required level of evidence to permit novel alternative strategies to become an acceptable standard of care. For example, how does an individual oncologist view the clinically impressive results of a nonrandomized experience from a major academic center published in a high-impact medical journal in evaluating the issue of equipoise?

Additionally, many situations exist in which promising anticancer agents may be available only through participation in FDA-sanctioned randomized trials, where in the opinion of the patient’s oncologist the decision to deliver the superior therapeutic strategy may have to be left to the luck of the draw.

Despite these complexities, or perhaps because of them, clinicians should carefully consider the issue of equipoise, as viewed by individual oncologists based on their clinical knowledge and experience as well as their intimate understanding of the patient’s medical history and personal goals, when any individual is to be offered participation in an investigative effort.

The fact that a study is open at the oncologist’s institution and a patient meets eligibility criteria should only be the starting point in thoughtful consideration and subsequent discussion of the appropriateness of this possible therapeutic option.


  1. Kupferschmidt K, Cohen J. In plasma OK, critics see politics, not science. Science. 2020;369(6507):1038-1039. doi:10.1126/science.369.6507.1038
  2. Hellman S, Hellman DS. Of mice but not men—problems of the randomized clinical trial. N Engl J Med. 1991;324(22)585-1589. doi:10.1056/NEJM199105303242208
  3. Markman M. Serious ethical dilemma of single-agent pegylated liposomal doxorubicin employed as a control arm in ovarian cancer chemotherapy trials. J Clin Oncol. 2010;28(19):e-319-e320. doi:10.1200/JCO.2010.28.9934