Transcript:Daniel P. Petrylak, MD: We’d like to move on to talking about other checkpoint inhibitors. Betsy, would you like to discuss the data with durvalumab in muscle-invasive bladder cancer and metastatic bladder cancer?

Elizabeth R. Plimack, MD, MS: There was actually an update presented at this meeting. There was a poster update of the durvalumab data. And, as with all of these, the trials enrolled really quickly. And so, as there’s more mature follow-up, we’re seeing many updates at multiple different settings. The initial data that were presented at ASCO last year involved about 60 patients and showed a really nice response rate in the PD-L1-positive group. This is a subset of over 100 patients, most of which were treated second-line, a couple had no prior therapy. What’s tricky about the durvalumab data set is that they amended the enrollment criteria—so first it was all-comers, then it was PD-L1-positive, then it was back to all-comers. So, I think we have to remember this is a heterogeneous data set, and quite possibly enriched for a better group of responders just given the way that they enrolled. It is given on a 2-week schedule. But one of the most interesting findings on the poster yesterday was that this trial was designed to give treatment for 1 year and then cease. And so, they just now have patients who’ve moved beyond that 1 year. When they look at the responders, those who made it to 1 year and stopped, have all continued to respond, now, albeit, with very short follow-up.

So, again, one of the interesting things about all of these data sets is since durability is one of the key benefits, that’s something that can really only be seen over time. And so, we’re really only going to know the true benefit 3 to 5 years from now. But I think the way the study was designed gives us an opportunity for an early look at cessation of therapy. And just a point that Rob touched upon previously is, how little can we give and can we make this really the only treatment patients need for a long time and shorten the duration for both patient benefit and convenience, cost, etc? So, I think those are the things to look for. Again, it’s another checkpoint inhibitor. It probably works similarly. The response rate is, again, right in that 20% range. Overall survival right in the 14-month range. We’re seeing very much consistent data across the checkpoint inhibitors and very little to differentiate them in terms of efficacy in my view.

Daniel P. Petrylak, MD: I think the important point about stopping after 1 year from this trial is it is really going to help us. But the problem is that we’re never going to see a randomized study, which is the best way to go about this of ceasing treatment or giving treatment intermittently because it’s going to simply take too long. And we really don’t have any intermediate endpoints to look at. Remember, progression-free survival is fairly short with most of these studies, about 2 to 3 months. So, it’s forcing us to rethink our clinical trial designs for this disease. In fact, with the explosion of all these agents, it concerns me that a lot of these trials may start slowing down in their accrual because we have only a limited number of patients that we can put on the study. We have to be really, really smart about our trial designs and what we’re trying to accomplish because I believe about 12,000 patients die from metastatic disease per year, at least within that particular ballpark. It’s not as many as patients with metastatic prostate, lung, or breast cancer, but we’ve got to be very, very prudent about what we’re doing, what we’re trying to accomplish, and how we’re going to try to design…

Elizabeth R. Plimack, MD, MS: And prioritize the important questions in terms of trials.

Daniel P. Petrylak, MD: Exactly, exactly. I think there are, again, some important data taken from this particular study. Dr. Quinn, what about CTLA4? This was the first checkpoint approved in melanoma. Ipilimumab, tremelimumab, what are the roles for these agents in the treatment of metastatic bladder cancer?

David I. Quinn, MBBS, PhD: The bottom line is we don’t know, but we’re trying to work it out. We had a very nice study led by the Hoosier Oncology Group that looked at gemcitabine/cisplatin with the CTLA4 inhibitor, ipilimumab. It didn’t appear like that was going to move us forward. It wasn’t a good addition, and we’re not going to move forward into a phase III study of that one. Where we’re seeing both ipilimumab and tremelimumab is in the first line combined with a PD-1 or a PD-L1 inhibitor. So, the new trials are going to look at those combinations first-line, and the trials that we’re seeing such as the DANUBE study, the IMvigor-310, and the KEYNOTE-361 are going to be interesting because they’ll give us different comparisons.

So, in the DANUBE study, at least we’ve got the combination of tremelimumab and durvalumab, a PD-L1 inhibitor that will give us an idea compared to the PD-L1 inhibitor alone about what the activity is. We also have ipilimumab versus nivolumab compared to standard-of-care chemotherapy, either carboplatin- or cisplatin-based therapy starting. This is still in development. I will just caution the treating oncologists to say that there may be some evidence that CTLA4 is part of a resistance spectrum or activating spectrum for PD-1 or PD-L1. But these patients do not have melanoma and they are a very different population. If you’re adding ipilimumab, it needs to be done in a trial. If you’re adding tremelimumab, it needs to be done in a trial because there is additional toxicity whenever you add anything. We know that from the melanoma data. We’ve seen a number of patients transferred to our inpatient unit with severe toxicity because they’ve been given melanoma doses of ipilimumab combined with a randomly selected PD-1 or PD-L1 inhibitor. Their average duration when they survive in hospital is 5 weeks, and they do not do well after that. So, this is an area of active development. We’re doing the trials, we need to accrue to them, and until we have definitive trial data, my advice is not to do it ad hoc at home.

Daniel P. Petrylak, MD: That raises a very, very important point about toxicity and the management of toxicities with immune therapy. I think, in general, when you add a CTLA4 to a PD-1 or PD-L1, you’ll about double the toxicities, at least what we’ve seen from the lung data and the melanoma data. And remember, our patients with bladder cancer generally tend to be a little bit less medically fit, they tend to be older, and they tend to have multiple medical problems as well. I think that this also reflects the importance of developing a marker or series of markers to tell us who’s going to respond to monotherapy. Because, you don’t want to ratchet up your treatment in a person who’s frail. And certainly, it may not be the right thing to do in that situation. So, I think management of toxicities is important.

Transcript Edited for Clarity