Elizabeth Plimack, MD, recaps recent advances in bladder cancer and discusses the need for diverse approaches to further improve patient outcomes.
Elizabeth Plimack, MD
Due to the success of pivotal trials such as KEYNOTE-045, 5 immunotherapy agents have been approved for the treatment of patients with urothelial carcinoma so far in 2017—but this does not mean that treatment options are diverse, says Elizabeth Plimack, MD.
Nivolumab (Opdivo), atezolizumab (Tecentriq), avelumab (Bavencio), durvalumab (Imfinzi), and pembrolizumab (Keytruda) have all achieved FDA approval status, but Plimack says none have differentiated themselves.
“Multiple times on expert panels we get asked the difference between these drugs and we just can’t say yet,” Plimack explains, “There may be a difference between the data, or the way the studies were designed—but we do not see a major difference between the drugs themselves.”
In an interview with OncLive, Plimack, chief of the division of genitourinary medical oncology and director of genitourinary research at Fox Chase Cancer Center in Philadelphia, recapped recent advances in bladder cancer and discussed the need for diverse approaches to further improve patient outcomes.Plimack: There were 4 abstracts presented [at the 2017 ASCO Annual Meeting] all of which were very interesting. The first one presented was actually an update on The Cancer Genome Atlas (TCGA). That is where centers across the country, and really the world, donated tissue samples for analysis and they did a deep-dive into varying biomarkers and reported on that. The data is really too dense to sort of encapsulate in a brief blurb, but the paper will be forthcoming. It is extending the numbers into the 400-range, before it was in the 100-range, so we are really going to learn a lot from these data which are publicly available to investigators, and is an excellent resource that should be highlighted.
The second abstract that was discussed was the KEYNOTE-045 study. This was originally published in the New England Journal of Medicine in February. It is a randomized phase III study of pembrolizumab versus chemotherapy in metastatic urothelial carcinoma. This is really the only data that we have seen from a randomized trial in urothelial carcinoma, and fortunately this study was positive and then we saw pembrolizumab get approved in the second-line setting based on this study.
We were very interested in seeing an update based on a fresher data cut, and not surprisingly, the overall survival benefit is maintained with further follow-up. We do see at the [18-month] mark, a more stable estimate of overall survival with at least 10% to 15% benefit for patients treated with pembrolizumab versus chemotherapy. That is certainly encouraging. What we also saw that I thought was interesting was that all patients on the chemo arm have come off of treatment, compared to the 33 patients in the pembrolizumab arm who stayed on treatment. So, this sort of speaks to the tail of the curve that we are seeing—even though we sometimes see responses with chemotherapy, and, indeed the progression-free survival was better with chemotherapy, we know from clinical experience and now from this study, that those responses aren’t durable. Occasionally, and what we hope for, is that responses with pembrolizumab are durable.
We also saw an update from the KEYNOTE-052 study, which investigated pembrolizumab in the frontline setting. We have seen these data previously at ESMO and ASCO GU, but this update, interestingly, looked specifically at biomarkers. The PD-L1 biomarker had been looked at in this study, and in fact, for the pembrolizumab bladder program, defined as part of the KEYNOTE-052 study with a cutoff of 10%, and they represented those data [at ASCO].
In addition, they looked at a gene signature that predicted for response. And it looks like that might select out responders a little bit better, whereas the negative predictive value of that test does look a little bit stronger than the PD-L1. What we’d love to see is those data placed on the KEYNOTE-045 data where we have randomized data to see how that biomarker holds up in a randomized trial. We are really urgently looking for biomarkers that are better than PD-L1 in bladder cancer that can help us select treatment—especially as more novel therapies come around—so that patients have options. Right now, it is less important because we have so few options for patients, but I am pretty confident in the next year or so that will change. It’s interesting. In the last year, we have seen 5 new drugs approved for bladder cancer, which is phenomenal, especially compared to what had been decades prior to that. The problem is, they really all do the same thing. So, while we have a wealth of options, in a sense they are all the same option.
What we are looking for are novel therapies, and there were 2 really interesting abstracts presented at [ASCO] looking at novel approaches, 1 combined with checkpoint inhibitors and 1 that shows there might be success after checkpoint failure.
The first is combining epacadostat with pembrolizumab. There was a lot of data presented on epacadostat at ASCO 2017. It is an IDO inhibitor so it preps the immune microenvironment for better immune response with checkpoint inhibitor. This particular trial was interesting in that it showed a response rate that was higher than we typically see with checkpoint inhibitors—around the 30% rate.
Notably, when we combine [these agents], we see an increase in immune-mediated toxicity. Although the numbers are small with this phase Ib basket study, we did not see a surge in immune-related toxicities—no colitis, no pneumonitis. We did see rash, but again these weren't clinically impactful and most patients were treated with either topical steroids or a brief course of oral steroids and allowed to continue treatment. I think this regimen warrants further attention, both because it is active in multiple tumor types, but also because that efficacy does not seem to be offset by toxicities. So, hopefully, that will hold out in larger studies and I know those are planned for this combination. Another agent that is novel and different—that is not an immune-acting agent—is the ASG-22CE compound; it is an antibody-drug conjugate targeting nectin-4. These data were presented by Dr Dan Petrylak, and we were very encouraged that this study did allow for checkpoint failures, and, in fact, showed an even higher response in checkpoint failures than the rest of the group, even though the numbers were small. So, we are looking for expansion work in this area and with this agent.