At this juncture where all the data is not yet available, who really knows what happened? I have heard many theories developing. Was it post-progression receipt of PD-1 or PD-L1 therapy for the control arm? Doubtful, as most sites for this trial were in areas where these agents are not yet available. Was it the difference between PD-L1 therapy compared to PD-1 therapy? After all, pembrolizumab has already been proven to offer a survival benefit in this setting in a randomized phase 3 trial.1 There is no clear evidence of such, as it has been difficult to discern clinical efficacy or safety differences from multiple trials with different PD-1 and PD-L1 antibodies. Most likely, trial specific patient populations probably contributed to the negative results. Previous evidence of such is no more difficult to identify than the dramatic differences in response rates between the phase 12 and phase 23 trials of atezolizumab in patients with advanced urothelial carcinoma. There is no clear explanation for those differences. Basically, there is no sure thing in any ball game, and although surprised with the results, I am only slightly surprised, as I have long since recognized that clinical trial design and patient populations often drive clinical trial results just as much as therapeutic drug efficacy and tolerability.
So what do we do about this? We definitely shouldn’t panic. We must recognize that the FDA approvals of nivolumab, durvalumab, and avelumab are also not based on overall survival benefit, but they clearly all offer clinical utility and activity and remain available for our use in the clinic. We also currently have seen a survival benefit with pembrolizumab in this post-platinum setting, and FDA approval is imminent (or perhaps it has already occurred by the time this article is posted). Finally, this should not affect how we think about therapy for those with cisplatin-ineligible disease. Atezolizumab is newly FDA approved in this setting and it otherwise would be considered an unmet need population.4 There are also many combination trials available in the first-line setting, which I highlighted in recent article,5 and we should not feel differently towards those trials. Combination therapy still has enormous potential.
One thing that has happened in my own clinic is mass hysteria among patients either receiving atezolizumab or considering atezolizumab. Everyone reads the internet in this day and age for good or for bad. It is our job as health care providers to help patients interpret such complex and controversial results. What is just as concerning to me is the fact that I now have patients declining to enroll on clinical trials occurring in earlier disease states. Of course they are concerned about a lack of overall survival benefit. However, I make this strong point with my patients. The adjuvant setting where patients are being enrolled are for patients who either did not receive neoadjuvant cisplatin combination chemotherapy and have pT3 disease or greater after cystectomy or patients who had tumors resistant to neoadjuvant cisplatin combination chemotherapy as evidenced by residual pT2 disease or greater after cystectomy. We know these patients, especially the latter, have an extremely high risk of recurrence, and there is no clear standard of care treatment option for them at this time. Hence, from the patient perspective, there is a 50% chance of being randomized to a PD-1 or PD-L1 therapy with a 15-35% chance of antitumor activity, estimated from metastatic disease trials, vs. a 0% chance of antitumor activity with observation if one refuses to enroll on an adjuvant trial because of the recent negative press release with atezolizumab. From the perspective of this physician and researcher, why wouldn’t I enroll patients in a trial with a chance to potentially benefit patients and further science? The ball game is not yet over, as there is still a lot of time on the clock. Don’t give up, and keep enrolling to these trials.
Randomized Phase 3 Adjuvant Trials with PD-1 or PD-L1 Antibodies for Patients with High-risk Urothelial Cancer
- Bellmunt J et al. N Engl J Med 2017; 376:1015-26.
- Powles T et al. Nature 2014; 515:563-7.
- Rosenberg JE et al. Lancet 2016; 387:1909-20.
- Balar A et al. Lancet 2017; 389:67-76.
- Yu EY. Uro-Today. On-line, April 11, 2017
Video Lecture: Immunotherapy in Advanced Urothelial Cancer- Matthew Campbell