Oliver Sartor: Hi, I'm Dr. Oliver Sartor here with UroToday, and we're at the US Prostate Cancer Consensus Conference in Colorado Springs. And we have an interesting topic we're going to be talking about with Dr. Mike Carducci and Tian Zhang, introduce them in a moment. And it's going to be about the introduction of the immunotherapies into the context with prostate cancer for a variety of predictive biomarkers that have been proven to work, at least the FDA says that they're going to work. So first of all, Mike Carducci, Professor at Johns Hopkins, long-term medical oncology lead in all things prostate cancer. Thank you for being here, Mike.

Michael A. Carducci: Good to be here. Thanks, Oliver.

Oliver Sartor: And Tian Zhang, Associate Professor at University of Texas Southwestern. Contributor in many ways to not just prostate, but more, so welcome to you.

Tian Zhang: Thanks, Oliver.

Oliver Sartor: This is going to be structured around certain biomarkers. I want to make sure we have an agreement on what they are. They're going to be mismatch-repair deficiency, TMB high, tumor mutational burden high, MSI High or maybe a selective class of the polymerase epsilon mutations. Is that the right topic?

Tian Zhang: Yes.

Oliver Sartor: Okay, but can you explain MMR, what is MMR? What does that mean?

Michael A. Carducci: Yeah, mismatch-repair, sort of how DNA sort of fixes itself over time. There are certain mutations that keep them from allowing to fix. And if they're not able to fix themselves, the cells go ahead and die. So if you have double hits, drugs that sort of can come in but they die or they make a lot of mistakes. And so the mistakes are really what the immune system can sort of capture. And so these tumor cells mutate a lot and have a lot of variant alleles that you can find in folks who have a mismatch-repair.

Oliver Sartor: What genes are we talking about?

Tian Zhang: There's a bunch of mismatch-repair genes in genetic syndromes who are well-characterized, such as Lynch syndrome. There are genes MLH1, MSH2, MSH6. And when they're in the germline setting, they contribute to these hereditary cancer syndromes. When they're somatic, they can be somatic and occur during the course of a patient's lifetime. They can also contribute to what we call Microsatellite high or MSI High status for a patient.

Oliver Sartor: Tell us what that is, Microsatellite. Is this like a circulatory thing in low orbit? What does that mean?

Tian Zhang: Yeah, Mike said a little bit about these genes when they're mutated, these changes in DNA mutations in each gene, they may not get fixed. And so there are these Microsatellites of changes within the chromosomal DNA. And so we call them High if there are many of these Microsatellites that are mutated. In patients with mutations in MMR genes, they end up having Microsatellite high or MSI High biomarker.

Oliver Sartor: And then we have tumor mutational burden, and Mike had alluded to the fact that they're making mistakes. And so I can envision that if you don't repair very well, the cells live, that these mistakes will accumulate, leading to lots of tumor mutational burden. Now, what is lots? And I'm going to ask you what assay you mean, because circulating tumor DNA and tumor-based DNA is a little bit different. And so, are we talking about 10, 20, 30, 40, 50? What number is high? Because I think clinicians want to know what number you depend on.

Tian Zhang: There is some back and forth about cutoffs, but I think the field recognizes that a high tumor mutational burden is anything more than 10 mutations per megabase of DNA. That's our conventional cutoff. It's one that's been used on the tissue assays. There are some correlates in recent publications of the circulating tumor DNA, using a high cutoff being 16. And so that's sort of the conventional thresholds where we say high tumor mutational burden.

Michael A. Carducci: I think that's important because I'm always confused when I get reports back and it says, "TMB High," and is it 10? And with CTDNA, it's clearly all over the board. You find it much more often because usually these things are less than 5% of the time, pretty rare phenomenon that you see this in prostate cancer. So when you see it, I'm always like, "Huh, is this real?" And so if it is higher, that's important to realize that in CTDNA, you really have to probably look at a higher threshold.

Tian Zhang: I would just add that tumor mutational burden being high can also be driven from other DNA damage pathways. And so there are homologous recombination pathways I'm sure you'll cover with a different segment, BRCA1, BRCA2, for example. And so depending on the pathway to a high tumor mutational burden, I think that's the nuance that we're trying to tease out, which population of patients actually benefit from immunotherapy versus in homologous recombination, we might think about more, the PARP inhibitors.

Oliver Sartor: One more thing on the biomarkers for maybe two more things before we go on. This recent manuscript that looked at combination, so were you TMB high alone with an MSI low and an MMR undetectable, or did you have two out of the three or did you have three out of the three? And that particular manuscript said the more you have, the better it is. So if you have one of these biomarkers, the probability of response is X. If you have two, it's 2X, not exactly but higher. And if you have three, it's even higher. So is that something that resonates with you or should we be looking at this independently or in aggregate?

Tian Zhang: I think it's a difficult topic. You're referring to a trial called NEPTUNES, where these patients with that higher signatures either with MSI High disease or high tumor mutational burden, or a high infiltrate of the tumor with more immune cells, they qualified for the trial and they were treated with ipilimumab and nivolumab. And so certainly in the MSI High population, those patients do very well with checkpoint inhibitors. Whether we pre-select that main population or whether we go toward this more immune signature, I think that's still to be defined. That particular trial only had about 60 patients, all prostate cancer. But I personally would use more of the tumor agnostic label with pembrolizumab with the folks that we select for MSI High disease.

Oliver Sartor: Mike, you have anything to add to that?

Michael A. Carducci: No, I mean, we'll probably get to it. To me, it's a lot of this stuff and we do these tests, it's like winning the lottery. "Oh, you got something." Patients, 95% aren't not going to have anything when you're looking for particular ones like this, but that 20% that we do find. But you're so excited and if you do have MSI High and TMB given these kind of results, we're saying, "Maybe you are the one who IO therapy, immune therapy would be the right approach for." But if you don't have it or you have mixed generations of it, what's the risk of trying to start giving these agents?

Oliver Sartor: Okay, so 2017, thank you, Mike, for clarifying, the FDA came out with a tumor agnostic indication for the first time, in which you would give pembrolizumab regardless of the tumor if they had these particular signatures. And that was actually a really big deal because we'd never had really a molecular biomarker or a series of biomarkers that would drive a particular therapy. And of course, prostate was among the group of all tumors.

Now, it turns out that we're going to have two sides of an equation talked about here. Mike, as I understand it, you're going to say, "Wait a minute, do we need to jump to that IO right away or let it kind of play out with our normal treatments?" And Tian, as I understand it, you're going to say, "Hey, let's jump. It works, let's go now. Tian, you get first shot at this, articulate. Why do you believe when you find this signature, you should go straight to the IO?

Tian Zhang: Short answer is it works.

Oliver Sartor: I like that answer.

Tian Zhang: The label is for pembrolizumab for MSI High or high tumor mutational burden disease. I think specifically for MSI High, and this is less than 2% of prostate cancers, but when we find it, it's really unique. And those patients who we treat, either in the castration resistant stage or in the hormone sensitive stage, they respond. And so we just recently put together our series at UT Southwestern with prostate cancer folks who have MSI High disease.

Turns out we have had treated a dozen over the last five years or so, and seven of them had complete responses, meaning their PSAs became undetectable, their radiographic features went away. And so I think those are the success stories that we carry forward. And so when we find those very small numbers of patients, we like to use pembrolizumab for those patients because it works.

Oliver Sartor: Now, these are durable, complete remissions? They're ones that get there and stay there? And if so, when do you stop the therapy, two years, three years, 18 years? How long do you need to keep it going?

Tian Zhang: Great question. I think nobody knows. They can be quite durable. In my own practice, I treated two of those dozen patients and one of them was in the castration resistance setting. We did take him off for hypophysitis, so pituitary changes, headaches after the third cycle, but his PSA was undetectable. He was off treatment for about two years before he redeveloped radiographic changes, and so that's when we re-challenged him. And again, about two or three cycles and now he's off. The other patient in my shop, actually, I treated in the very hormone sensitive lymph node positive setting, and he did not want hormone deprivation.

And so, took a look at his tissue profiling and said, "Hey, you've won the lottery." And because he did not want the hormone deprivation, we did pembrolizumab alone. And his PSA also went from 60s to undetectable, the lymph nodes went away. We ended up treating him for about a year because he was tolerating very well and had those discussions of, "This is a data-free zone, we don't know how long to treat. But the longer we treat, the possibility of immune-mediated toxicities might happen." And so took him off after about a year and he remains with undetectable PSAs.

Oliver Sartor: Great story.

Tian Zhang: Yeah.

Oliver Sartor: Great story. Now, Mike, you might have an alternative view, and I wonder if you could articulate? And by the way, I am going to use my memory, which may not be correct by the way, but I think in the label from the FDA, it said that should be used when you've sort of gone through your typical therapies. Now, I don't know if I remember the label right. Is that correct, Mike?

Michael A. Carducci: Yeah, I don't know.

Oliver Sartor: Okay. But in your practice, and you could take it as debate or practice, however you like, you're a little more prone to kind of give the androgen deprivation, let the normal therapies take their course, and then begin to think about the IO in more of a secondary situation? Or let me hear how you might look at this.

Michael A. Carducci: Yeah, I think for me, patients present in a whole lot of different ways. Because of the FDA approval for pembrolizumab and Microsatellite High or MMR deficient disease, and the role for PARP inhibitors for BRCA1 or BRCA2, clearly we're testing folks, saying we should be testing folks who have recurrent or metastatic disease, high grade Gleason, folks with Lynch family, things like that so that you know. There are people who are progressing under our watch and we know. But somebody, when did they progress? Was it diagnosis or when they have metastatic disease? What's the extent of their disease? And now all of a sudden, "Hey, I've got something," because you won the lottery. "Do I do it now?" You're well, you don't have any symptoms. We can't see it on scan.

With PET scans, you're getting this group of patients that are PSMA positive biochemical recurrence, and then you got PSMA negative biochemical recurrence. They're probably not the ones that are going to be MSI High. Or the folks that are progressing at a clip, maybe it makes sense to sort of try these. But to me, prostate cancer is prostate cancer. You're going to get benefits, so it is those folks who really want to avoid hormonal therapy. And we have this whole field whiplash about, "Hormonal therapy is great, it works. It extends survival. We got to use it earlier and earlier such as the EMBARK data."

And reality is these are men who are well and asymptomatic and we can't see the disease and, "How much therapy should we be giving them?" It's the same thing with IO therapy. You talked about hypophysitis or pituitary, there are side effects. The duration of therapy, these are given every three weeks, six weeks, depending on which drug you're using and schedule. These are making folks come in a lot. The response rates are working. Most of the data's in castrate-resistant disease where people do get responses. Many of them are durable, some are sort of relatively short-lived. So do you try figure out where they are? Does the patient have a window that sort of says, "You're asymptomatic, let's try?"

I worry about the patient who comes in with widely metastatic disease, symptomatic, you're going to probably start hormonal therapy, then you're going to get your sequencing. Then do you add them in, do you stop? So I think we just don't know. And sort of maybe when things are quiet or progressing slowly, that's the time. I think we learned from Sipuleucel-T, folks with low burden, low volume may be the ones that sort of have it. So maybe I switch over and say, "Yes, it's the time to give it." So I think the data just continuing to evolve, there are possibilities. The numbers of patients are small, so I think it's reasonable and the labels help us. If it's the right thing for that patient, I think we should be able to try it.

Oliver Sartor: One of the things that I was thinking about a bit earlier, we kind of jumped to the pembrolizumab label, but are there other forms of IO that might be considered, or is it kind of pembro in your practice or not pembrolizumab in your practice?

Tian Zhang: It's usually pembro in my practice because of the label. Not aware that there's another label for MSI High disease with a different checkpoint inhibitor. There are some data sets now with ipilimumab and nivolumab, for example, in prostate cancer. But we have not gone through the courses of either off-label use, compassionate use type of applications for those patients. I don't know if you've-

Michael A. Carducci: Yeah, it's all been clinical studies for us. We've had nivolumab for folks with MMR disease or in early disease, just to see what the response rate is. But it took us a really, really long time to find enough patients to get the drug, to see how nivolumab did, so it's hard. It's almost case report-ish because it takes so long. And most of the data that you look at for these sort of things is pooled. You sort of take your 12 and City of Hope's 15 and Hopkin's nine, and then you add them all together and you sort of say, "Hey, this stuff probably is really active," but it is still sort of piecemeal and mostly castrate-resistant disease.

Oliver Sartor: I'm going to be wrapping up shortly. And so I wondered, Tian, if you could kind of articulate your position succinctly and then Mike, the same? And this way we can hear both sides of the equation.

Tian Zhang: Sure. In prostate cancer, there's a select few who have MSI High or high tumor mutational burden disease. And when we capture those on their molecular profiling, there is an on-label use of pembrolizumab. And when we give it, it works quite well for patients with prostate cancer.

Michael A. Carducci: I think we need to be looking for those patients, even though they're pretty rare. The issue about timing is when do we start? Do you start as early as biochemical relapse or add it in with therapies or instead of therapies for metastatic hormone-sensitive disease? I think those are the sweet spots. It's sort of where we might be able to avoid hormonal therapy, but that's research still. We just don't know if that makes sense. The labels say yes, you can do it. And it may work for some patients, but others, maybe get your standard of care started and then figure out what's the right time to use the pembrolizumab in this case.

Oliver Sartor: Good, you both make good points. Thank you, Tian, thank you, Mike. Enjoyed the conversation.

Michael A. Carducci: Thank you, Oliver.

Tian Zhang: Thanks.