Reply to ‘Correcting the ASCO position on phase I clinical trials in cancer’

I thank Dr. Burris for his letter. My article (Kimmelman, J. Phase I trials as therapeutic options: (usually) a betrayal of evidence-based medicine. Nat. Rev. Clin. Oncol. https://doi.org/10.1038/s41571-019-0264-7 (2019))¹ concerns an ethical question: are oncologists and others justified in presenting phase I trial participation as a therapeutic option for patients meeting the eligibility criteria? My view is that, generally speaking, they are not. For as long as physicians do not yet know how to use a drug, much less whether it works under near-optimum conditions, presenting phase I trials as a therapeutic option betrays the spirit of evidence-based medicine. It also fails to alert clinicians to the fact that their patients are, in fact, making personal sacrifices in the name of medical advances.

Dr Burris suggests (Burris, H.A. Correcting the ASCO position on phase I clinical trials in cancer. Nat. Rev. Clin. Oncol. https://doi.org/10.1038/s41571-019-0311-4 (2019))² that I confuse ASCO’s position (that phase I trials have ‘therapeutic intent’) with the position that phase I trials actually have ‘therapeutic benefit.’ Au contraire. My published response to ASCO’s phase I position statement makes clear that a declaration of ‘therapeutic intent’ has practically no bearing on the ethical question I am concerned with³. Intent is merely a mental state. Physicians who offer quack treatments such as caesium chloride outside of the context of a clinical trial might do so with therapeutic intent. This intent, however, does not justify recommending such treatments, much less asking patients to endure their disruptive effects. By contrast, claiming that something is therapeutic is a substantive claim about the relationship between the risks and benefits of a particular treatment in a specific population of patients. From reading ASCO’s and Dr Burris’s statements, it is nearly impossible to come away with the view that both regard phase I trials not merely as therapeutically intended, but substantively therapeutic.

Dr. Burris also faults my pharmacy analogy for failing to note the existence of oversight mechanisms, like institutional review boards (IRBs), in clinical research. I challenge this by asking for a single example where an IRB, or the FDA for that matter⁴, refused to allow a phase I trial to proceed owing to a lack of preclinical evidence of efficacy. Moreover, such bodies often acquiesce to the use of outdated study designs and evasive descriptions of the likelihood of clinical benefit, such as ‘we do not know if you will receive medical benefit from taking part in this study.’ Respectfully, I believe that ASCO’s phase I trial policy statement can do better for patients, researchers and oversight bodies. Firstly, ASCO might explicitly urge researchers to explain to prospective trial participants that phase I trials are primarily aimed at eliminating unsafe and ineffective treatments. Second, it might endorse the use of the more forthright language recommended by the US National Cancer Institute: “it is unlikely this intervention will help you live longer”⁵. Lastly, ASCO might seek to remind oncologists that phase I trials are, first and foremost, scientific undertakings. The ultimate moral justification for exposing patients to an unproven intervention within a trial is to evaluate the effects of the intervention. Accordingly, investigators and IRBs should address the continuing challenges regarding the preclinical rationale, feasibility and design, integrity of the analysis, and the completeness of reporting of data from phase I trials involving patients with cancer.