Immunotherapy in Second-Line Advanced Bladder Cancer
Insights From: Daniel J. George, MD, Duke University Medical Center
Daniel J. George, MD: 2017 into 2018 is going to be a transition period in bladder cancer, particularly for patients who are in the second-line setting. Still today, many of our patients are getting a platinum-based chemotherapy in either the neoadjuvant setting, adjuvant setting, or frontline setting. And so, many patients are developing platinum-refractory disease and are good candidates for I-O therapy. Today, we actually have 5 I-O therapies that are FDA approved for management of these platinum-refractory patients, and these include pembrolizumab, atezolizumab, nivolumab, avelumab, and durvalumab. So, how we use these agents and how we select these agents is a bit controversial. Several of these agents are based on accelerated approval from single-arm phase II data. They are still very good data. They are still very reproducible data, and it’s very exciting to see how these agents may add to this armamentarium.
But on paper, it’s difficult by the results to really tease out if there’s a patient population that might do better on a PD-L1 inhibitor—like atezolizumab, avelumab, or durvalumab—or if there’s a group of patients who really do better on a PD-1 inhibitor, like nivolumab or pembrolizumab. So, that’s one differentiating factor. Another differentiating factor is schedule. Many of these are on an every-2-week schedule, others are on an every-3-week schedule. And most patients will tell you, as much as they like their physician and their practice, they’d much prefer to come in every 3 weeks than every 2 weeks, particularly if they’re responding, they’re doing well, and we’re looking at a very durable response.
For me, and being in a referral center like Duke, an every-3-week schedule is a big advantage. But there are situations where seeing patients on a more regular basis, treating them more frequently, might have its advantages in the community in particular. So, I can imagine where every-2-week schedules might have some advantages for some folks. Lastly, there are the pre-medications, the administrations, and the time it takes to infuse these therapies, which can vary as well. So, I think there are multiple factors.
Interestingly, probably the thing that’s changing the most is the frequency of dosing, and we’ll probably see more data in 2018 showing some alternative regimens, every-4-week, even every-6-week dosing, potentially, with some of these agents. So that may change some of our attractiveness to one agent or another. But ultimately, what we really want to see is efficacy. We want to see biomarkers to enrich the response, we want to see efficacy improve beyond the 20% to 30% response rate, and we’re going to want to see combinations that really enhance not just the response but the complete response and the duration of response, the really long-term benefits that our patients most desire.
KEYNOTE-045 was a remarkable study in many ways. First, it was the first study in decades to show a survival benefit in metastatic bladder cancer patients in an unselected population. These were patients who were platinum refractory and who obviously have a heterogeneous tumor biology, wide spectrum of metastatic sites of disease and prognosis. And the control arm here, we saw an overall survival of roughly 7 months, so really a pretty poor-risk population overall. And yet in the control arm, in the pembrolizumab arm, we saw an overall response rate of 27% and a survival benefit of roughly 3 months with a hazard ratio of 0.73.
Now, a 3-month survival benefit, improving from 7 months to 10 months, may not sound like much, and many patients may find that somewhat discouraging, but in an unselected population, that’s remarkable. Because what that’s telling you is that you’re really changing the survival curve in a dramatic way for a subset of patients. There are going to be some patients who don’t benefit from either therapy and have a very poor prognosis, and you can see that in the Kaplan-Meier curves where these curves both drop down for the first 30% or 40% of patients and then they start to separate. And what’s remarkable is even though there’s a subset of patients who were spawned and probably benefit to this second-line chemotherapy, there’s even a greater benefit associated with I-O therapy.
And we see that driving the overall separation and hazard ratio here. It’s the bottom two-thirds of this population that really benefit the most and that, to me, is exciting. Of course, we’d love to help all the patients, but we recognize there are some in this population who are probably too sick and we are too late to really help. But what’s remarkable is the durability of this response in probably that bottom 30% to 40% of patients. They’re not only responding to therapy, that curve is extending out. So, we’ve pushed out the median, but we’ve pushed out the bottom half of that curve even further. And that’s a population of patients who, historically, maybe we’ve made some anecdotal inroads on, but we’ve never seen a population benefit like this and it’s really remarkable. And it gives us a tremendous base to build off of because the pembrolizumab was well tolerated.
Here’s a population of patients we would have expected to be perhaps sicker and more vulnerable, at risk for complications than maybe some of the frontline studies that have been done with this agent. And yet, we saw a really very low rate of grade 3 toxicities, we saw a 10% rate of dropout for toxicity. The vast majority of patients were able to tolerate this therapy and tolerate it well for a long period of time to see that really extended survival benefit. So, to me, that’s a really remarkable feat in a population as sick, as diverse, and as genetically complicated as platinum-refractory bladder cancer is.