Chemotherapy Following Immunotherapy in Bladder Cancer
Panelists: Daniel P. Petrylak, MD, Yale University Cancer Center; Dean F. Bajorin, MD, Memorial Sloan Kettering Cancer Center; Robert Dreicer, MD, MS, Virginia Cancer Center; Arjun V. Balar, MD, Laura and Isaac Perlmutter Cancer Center; Elizabeth R. Plimack, MD, MS, Fox Chase Cancer Center; David I. Quinn, MBBS, PhD, USC Norris Cancer Hospital
Daniel P. Petrylak, MD: From a practical standpoint, and we could throw this question over to the entire panel. What’s your experience with giving chemotherapy after immune therapy? Do you think that these patients are more responsive? About the same level of response? Should these drugs be given together or in sequence?
Robert Dreicer, MD, MS: I would just say that for 40 years, with first-generation immunotherapy followed by chemotherapy, there have been reports in all solid tumors and some hematologic neoplasms of unanticipated levels of activity, never validated in any meaningful way, always anecdotal but lots of anecdotes. I think there’s likely to be something there because now we’re doing prospective studies. For the first time, we’re actually going to be able to see some of that. In a sense, some of them are going to ask the question indirectly because some of those patients will roll over to subsequent therapy. I suspect we will find something there. I suspect the mechanism will be variable. I don’t think it’s going to be one answer.
Elizabeth R. Plimack, MD, MS: I think we have to be very careful though because those who respond beautifully to immunotherapy, we know respond for a long time. And so, if they’re well enough to get the next therapy, that’s a very selected population.
Robert Dreicer, MD, MS: But it might be that patients who actually don’t respond then subsequently get chemotherapy versus immune therapy. That’s the question there.
Elizabeth R. Plimack, MD, MS: That is the more interesting question, and I think parsing that though is challenging. And, while I know we all may have anecdotes to share, I think it’s important to really look at the data and see.
Daniel P. Petrylak, MD: And patient selection, as you’re pointing out, is crucial because if somebody is progressing so rapidly that they can’t get chemotherapy, that’s a very, very different patient.
Elizabeth R. Plimack, MD, MS: Among the progressers, there will be more dropout than among the responders. And so, if you can turn one of those around, that would be impressive. But we may have a smaller pool to look at.
Robert Dreicer, MD, MS: The other question we have to address is we’ve been talking about cisplatin-ineligible in terms of frontline. How do we begin to think about moving this therapy? How do you ask the question when you know you can cure a small subset? How do you do that kind of study? Because I think we’d all agree that if you can cure people using immunotherapy, we would prefer that than giving cisplatin. So, that’s another challenge.
Elizabeth R. Plimack, MD, MS: All 3 randomized trials allow cisplatin-eligible patients.
Robert Dreicer, MD, MS: But whoever puts people that are potentially curable on that will be somewhat biased perhaps, right?
Dean F. Bajorin, MD: I want to put a note of caution here though for the treating oncologists, and that is let’s take a look at your trial with atezolizumab first-line. The response rate was 23%, but their survival is 16 months, right? We go back to the trial you just talked about, which is Maria De Santis’s trial of carboplatin/gemcitabine. Response rate is 42% and their survival is 9 months. And so, this issue of switching therapies for non-responding disease is something we have to be very careful about in the community. Because, I think we are looking at different criteria for success, that it may not be an immediate response. It may be patients who have stable disease for a while. And so, I would caution this switch to a therapy in the non-responding patient as these drugs may become available.
David I. Quinn, MBBS, PhD: I think where we’re going to get help is when the trials come through, where we’ve got a comparator arm of single-agent immunotherapy. We’ll then get to follow the patients and see what happens. I think the issue for the treating community oncologist now is not necessarily the stable patients or indeed the ones that are responding. There are different questions there—when to stop, how long to treat. But for the half of the patients that will blow through your community oncologist sitting there saying, “This is a new therapeutic paradigm for me using whatever checkpoint inhibitor,” I think the message that I give them is that chemotherapy still does appear to be effective. But it depends on what’s going on with your patient. So, that performance status 2 patient or 2.5 patient that you’ve put on who doesn’t respond, we all like the anecdote of the guy who’s hunting near Rob Dreicer’s house. He’s out there doing everything. But for the other patients that you treat, the half of them that blow through, I think the message I give them is they need to work out. We go back to what we’ve done with poor performance patients before. Are they a palliative care patient? Are they a 1-drug patient, where you think you can get it in and get response that might help 20% of them if you’re lucky? Or are they a 2-drug patient, where you should be giving them, say, carboplatin and gemcitabine? I think that paradigm still fits. A lot of the times we sit there looking at the patient asking, “Is this pseudoprogression when the patient’s albumen has dropped from 4 to 3, they’ve become anemic, and clearly their performance status is on the fritz?” When we need to make that kind of decision, I think carboplatin/gemcitabine, the taxanes, still seem to be active in a proportion of patients and they can get a good response. And we do need data on this. But for the treating oncologist who’s sitting out there, you need to go back to not being hypnotized by the checkpoint inhibitor when the patients are progressing and make those decisions on chemotherapy. And you can get a response which, in my experience, is durable to the level of experience of, say, 6 months in a subset of patients.