Barry Bonds and Roger Clemens are two world-class baseball players who played well enough to earn enduring respect and admiration. However, their reputations have been permanently tainted by their use of performance-enhancing drugs. Although they are both undeniably gifted, we’ll never know where their natural talent ended and their unfair advantage began.
The medical oncology version of this scenario played out in front of our eyes at the ASCO 2023 Plenary Session featuring the ADAURA trial. In this case, the unfair advantage was clear: Nearly two-thirds of patients in the control arm did not cross over to a standard-of-care treatment at relapse.
The ADAURA trial tested the value of the third-generation oral EGFR inhibitor osimertinib vs. placebo for up to 3 years in patients with a resected stage IB-IIIA non–small cell lung cancer (NSCLC) harboring an activating EGFR mutation. Initially presented in the ASCO 2020 Plenary Session, the preliminary analysis for ADAURA demonstrated a remarkably favorable improvement in disease-free survival (DFS) in the osimertinib arm, which led to its prompt Food and Drug Administration approval in this setting.
At the time, I was among the more vocal critics of the fanfare around this trial. The DFS improvement was impressive but, in this context, represented a low threshold for FDA approval. We already knew that prior trials of other, less effective adjuvant EGFR inhibitors routinely improved DFS but ultimately failed to improve overall survival.
Although in some cases, a DFS benefit can be sufficient to warrant changing practice, I would argue that overall survival is the most critical endpoint in a curative setting by a wide margin. I would also argue that cost should be a consideration for a drug priced at $440,000 over 3 years in the United States. At the very least, we – patients, oncologists, payers – should want to clarify what we get for $440,000 per patient, especially if that money could be better spent on other things.
Of note, we need to know whether the same overall survival may be achieved in this setting by treating only patients whose disease relapses, avoiding both the cost and toxicity of continuous treatment.
The relevance of this question became even more acute when an updated version of the ADAURA trial showed that the DFS benefit from osimertinib began to erode immediately after patients completed active treatment. Although the DFS benefit remained excellent,
The overall survival data – the critical test in my mind – were presented at the ASCO Plenary Session. The results were highly positive, with a hazard ratio for overall survival of 0.49 and a similar benefit observed across all eligible disease stages.
As with the presentation of the DFS data in 2020, these results were accompanied by adulation during the session and fawning in the media. However, a subset of people in the audience, and on Twitter, voiced a major concern: In the control arm, only 38.5% of the patients whose disease had relapsed (79 of 205) ever received osimertinib.
Post-protocol treatment included reassurances that patients in the control arm were offered crossover to osimertinib, but this only occurred starting in April 2020 and only when the treating investigator requested it.
The fact that only a minority of patients in the control arm ever received osimertinib means that ADAURA is not a trial that tests adjuvant osimertinib to osimertinib at relapse, the prevailing standard of care in the United States and the preferred treatment in this setting, based on NCCN guidelines for patients with relapsed/metastatic EGFR mutation–positive NSCLC since thein April 2018.
The change in standard of care in the United States and some other countries did not lead to an amendment in the trial, based on the argument that the trial was designed with DFS as the primary endpoint and that patients in the control arm are effectively off trial at relapse. That means patients would receive their country’s standard of care, which may be below, or different from, the standard in the United States or other parts of the world.
While defensible, others considered the low rate of osimertinib delivery in the control arm a serious flaw in the trial design and arguably an ethical problem.
Given this trial design, it’s important to question whether the trial magnified the difference in overall survival between the two arms compared with the standard of care in countries where adjuvant osimertinib will ultimately be marketed. Although I strongly suspect that the overall survival difference would have been significant without the disparity in access to osimertinib in the control arm, we will never know the magnitude of that difference.
With the current design, the trial demonstrates that osimertinib is associated with improved survival in 100% of patients with EGFR mutation–positive NSCLC who receive the agent vs. 38.5% of patients receiving it at some point after relapse.
At the end of the day, ADAURA followed conventional trial rules, which led to a windfall of accolades at the cost of suboptimal care for many of the patients randomly assigned to the control arm.
We need to decide as an oncology community whether we want to accept a system that confers rewards for stakeholders at the expense of patients enrolled in the trial.
We have developed a system that not only permits but depends on global trials enrolling many patients from countries with staging and off-protocol oncology care below standards of care in other parts of the world – effectively condoning and perpetuating disparities in optimal care globally. I am also saddened that some of the most respected leaders in our field have become upset when I question the ethics of accepting this system, one that I believe we have agency to change.
The question here is not whether osimertinib is a good drug; it is remarkably effective for patients with EGFR mutation–positive NSCLC. The question is how to design trials in a fair and balanced way that doesn’t compromise patient care or ethical standards.
If people think my assessment is too harsh, I openly welcome debate.
Bottom line: The ADAURA trial lays bare deeper problems with clinical trial design, and I would challenge readers to reflect on our clinical trial culture in oncology, which offers a system of rules that rewards denying the best identified care for patients on our clinical trials.
At the very least, should we celebrate these wins without fully acknowledging these problems?
Dr. West is clinical associate professor, department of medical oncology, City of Hope Comprehensive Cancer Care, Duarte, Calif. He reported conflicts of interest with Bristol-Myers Squibb, Boehringer Ingelheim, Spectrum, AstraZeneca, Celgene, Genentech/Roche, Pfizer, Merck, and Eli Lilly.
A version of this article first appeared on.