Read the Full Video Transcript

Sam Chang: Hello, everyone. My name is Sam Chang. I am a Urologic Surgeon in Nashville, Tennessee. I work at Vanderbilt University, and I am very fortunate to have Dr. Gopa Iyer here with me today. He is an Associate Attending at Memorial Sloan Kettering Cancer Center and actually really needs no introduction. The world-famous Department of Oncology and Urology at MSK has been a real leader when it comes to different types of GU oncologies, but I think definitely have made huge inroads when it comes to urothelial carcinoma.

And so there's a recent article in European Urology, looking at perhaps the predictive nature of FGFR3 wild-type versus mutant-type and whether or not it actually impacts or predicts how patients may or may not respond to perioperative chemotherapy. So we have fortunately Dr. Iyer here to kind of go over the highlights of the manuscript, and I'm going to ask him some questions to kind of help folks like myself and like himself when it comes to helping treat these patients. Gopa, thank you again for spending some time with us.

Gopa Iyer: Thank you so much, Sam, and thanks for the opportunity to speak with UroToday today. So I have a few slides I can go over, actually to provide some background and then also go over the actual paper as well. So this slide and this data are very familiar I think to most folks, that we know that pathologic down-staging at the time of cystectomy does correlate with the long-term efficacy of neoadjuvant chemotherapy and also with long-term survival in patients with muscle-invasive bladder cancer. On the left, there is data from the SWOG 8710 study that randomized patients to neoadjuvant cisplatin-based chemotherapy followed by cystectomy, versus upfront cystectomy. And if the survival curves are stratified based upon pathologic response, what we see is that at five years, the overall survival rate is approximately 85% in patients who have a pathologic complete response compared to about 35% to 40% in patients who have the residual muscle-invasive disease at cystectomy.

And then on the right is a retrospective single-institution analysis that looked at patients who received gemcitabine and cisplatin in the neoadjuvant setting. And also showed a similar trend that in patients who have any degree of down-staging to less than muscle-invasive disease, that there was a substantial improvement in both recurrence-free and overall survival with down-staging to neoadjuvant chemotherapy. So these data, which were collected over the last 20 years, really set the stage for folks to search for additional biomarkers of response or resistance to neoadjuvant chemotherapy. So, there have been a couple of efforts now that have looked at mutations within DNA damage response genes. These are genes that encode proteins that help to repair damage to DNA, typically caused by chemotherapy or other mutagens. What these two studies found was that actually, mutations within DDR genes do correlate with response to neoadjuvant chemotherapy.

So in one study, mutations within ERCC2, which is a gene that encodes for a DNA helicase that's involved in the repair of cisplatin-induced damage to DNA was enriched in patients who had pathologic complete responses to neoadjuvant chemotherapy compared to patients who still had the residual muscle-invasive disease. And then another study that was done by folks at Fox Chase showed that mutations in any one of three genes, ATM, RB1, or FANCC were predictive for response to neoadjuvant chemotherapy in both a discovery and a validation cohort. So this was data showing that genetic alterations, certain genetic alterations may be associated with response to chemotherapy.

And one of the genes that is actually altered very frequently in bladder cancer is fibroblast growth factor receptor 3 or FGFR3, which is thought to be involved in cell proliferation. When it is altered or mutated, it's constitutively activated and thought to drive tumor formation in a subset of bladder cancers.

And what's interesting is on this very busy slide, what I'll just sort of point your attention to is the red box there, of FGFR3-altered tumors. And what we see is that the frequency of FGFR3 alterations is very high, over 50%, indeed, about 50% to 80% in patients with low-grade Ta disease. Then again, about 50% in high-grade Ta disease. Then when we go to high-grade T1 and then finally to muscle-invasive disease, it goes down to about 15%. That's what we see in the urothelial TCGA as well, that FGFR3 alterations are found in about 15% of muscle-invasive bladder tumors. What's especially interesting about FGFR3 is that we now have an orally available inhibitor of FGFR3 signaling, erdafitinib that was recently FDA approved in patients with metastatic bladder cancer and FGFR2 or FGFR3 alterations within their tumors.

So in this study, we sought to evaluate the association between FGFR3 alterations and response to platinum-based chemotherapy, both in the perioperative as well as the metastatic setting. And so we initially looked at a cohort of patients with muscle-invasive bladder cancer who received neoadjuvant cisplatin-based chemotherapy, you can see the concert diagram there on the left. Ultimately we had 72 patients who were evaluable for further analysis. And these were patients who received cisplatin-based neoadjuvant chemotherapy, who all got cystectomy, and for whom we actually had genetic sequencing available from their pretreatment TUR specimens. You can see that in the FGFR3 alterations, there were nine out of 72 tumors, which is about 14% of tumors that were FGFR3-altered and 63 tumors that were FGFR3 wild-type. In terms of clinical characteristics, they were pretty well balanced between the mutant and wild-type cohorts within this group of patients.

The one thing though that I will point your attention to is that there were no complete pathologic responses in the FGFR3-altered tumors compared to the wild-type tumors. And that's what we see in this slide as well. On the top table there, we can see that zero patients had complete responses, three patients actually had down-staging to non-muscle-invasive disease, one patient had carcinoma in situ and two patients had T1 disease. But again, all of the pathologic responses were only seen within the FGFR3 wild-type patients. And then when we look at recurrent-free survival in this patient population following cystectomy, what we can see is that actually, it's statistically significantly worse in the FGFR3-altered patients, but OS is not significantly different between the mutant versus the wild-type patients within this cohort.

Then we also looked at survival as a function of FGFR3 status using the urothelial TCGA patients who had received perioperative chemotherapy, and that's what you see in the top two Kaplan-Meier curves there. So, these were patients from the urothelial TCGA. So all of these patients had muscle-invasive disease, they underwent cystectomy and we have the sequencing data from these patients available to us. And we selected the patients who actually received adjuvant cisplatin-based chemotherapy. What you see interestingly is that the patients who received adjuvant chemotherapy had a much worse, and who had FGFR3 mutations, did much worse in terms of recurrence-free survival compared to those patients who were FGFR3 wild-type. And this was adjusted for the pathologic stage at the time of cystectomy. Overall survival was not statistically different, although those patients also seemed to do worse numerically in the mutant population compared to the wild-type population.

And then when we look at the much larger group of patients who did not receive any adjuvant or perioperative chemotherapy, we actually see the opposite. We see that FGFR3 alterations were associated with improved recurrence-free survival as well as overall survival. Suggesting in general that FGFR3-mutant disease is a more indolent disease compared to wild-type disease, but that it also is more resistant to many of the systemic therapies that we have available.

And then finally in the metastatic cohort of patients, we had 100 patients who were evaluable, who all had genetic sequencing done, and who received platinum-based chemotherapy for their metastatic disease.  Interestingly, the FGFR3-mutant patients were associated with a higher rate of pulmonary metastatic disease, and also in general, lower response rates, numerically at 37% versus 49%. This was not statistically significant, but there definitely was a trend toward lower response rates in those patients. And once again in terms of progression-free survival and overall survival, we didn't see a statistically significant difference between the mutant and the wild-type population. Again, perhaps pointing to the fact that mutant disease is biologically very different than wild-type disease and maybe just more indolent.

So in conclusion, FGFR3 alterations are associated with resistance to neoadjuvant cisplatin-based chemotherapy and also with worse recurrence-free survival following adjuvant chemotherapy. The FGFR3-altered tumors may have a more indolent clinical course compared to wild-type tumors. And then what's very exciting is that FGFR3 inhibitors are currently being examined in the perioperative space for FGFR3-altered bladder cancer. And given that these tumors do tend to be resistant to neoadjuvant chemotherapy, that this space is very exciting actually for FGFR3 inhibition.

So, those are my slides. I will give it back to you, Sam.

Sam Chang: Yeah. Gopa, thank you so much for sharing those insights. I'm going to start with the bigger picture and then kind of hone in perhaps on FGFR3. Your initial slides regarding the different alterations shown by you guys and by Fox chase looking at ERCC2 and the three that Fox chase had pointed out, should all our muscle-invasive bladder cancer patients from their invasive tumors, their TURBT, should they all undergo somatic mutation evaluation?

Gopa Iyer: So, that's a great question. I think my short answer to that would be yes, but not because we should be utilizing that data to decide whether patients should receive neoadjuvant chemotherapy. Because those are being tested in prospective clinical trials, but only in a clinical trial setting.

Sam Chang: Right.

Gopa Iyer: So, I would still utilize some sort of clinical data, such as what is their clinical stage, radiographic stage, et cetera to help me decide. And of course, if they're cisplatin-eligible, to decide on neoadjuvant chemotherapy prior to cystectomy.

Sam Chang: Right.

Gopa Iyer: The reason I would favor doing genetic testing is because of the FDA approval of erdafitinib for the metastatic space. And so I feel like the earlier that we have that data, the more helpful it might be for a patient if they do end up developing metastatic disease and they have an FGFR3 alteration. A lot of those patients don't have a lot of time to wait for genetic sequencing to come back, in order to get them started on erdafitinib. So if we already have it, we can use that to justify starting that patient on treatment there.

Sam Chang: I agree, I think that's fantastically important. And so when you at this point with the data that you've generated, single-institution, retrospective, all those types of things. At this point, does it serve for you a little bit as a tiebreaker? For instance, someone comes in, you're planning on neoadjuvant chemotherapy, but maybe he has a little bit of neuropathy, a little bit of hearing loss, a little bit, renal function may be borderline. You don't want to give platinum, you don't have a clinical trial open. Which I know is not true, because you guys have clinical trials in every space. But say you're in the situation where you're not sure that you are going to give neoadjuvant chemotherapy and you have this data showing that this patient is actually in the smaller subset, that has a mutation in FGFR3. Would that be enough to trigger you to say, "You know what? Let's go ahead and not do neoadjuvant.  He is right on the borderline, let's proceed with cystectomy. Let's see what we find and go from there."? Is that, am I jumping the gun a little too quickly there? Or what do you think?

Gopa Iyer: So, I would actually maybe do something a little different.

Sam Chang: Okay.

Gopa Iyer: So personally, if a person had an FGFR3 alteration, I would not let that make a decision for me about neoadjuvant chemotherapy on those sort of bubble patients. I think if they had an ERCC2 mutation, I might be more likely to push a little bit for giving neoadjuvant chemotherapy because I do think the data is much stronger for that. Because it's based on more than just one institution, even though it is retrospective, they are much larger numbers compared to FGFR3 and resistance. The other thing that really impacts my decision, to be honest, especially in those sort of borderline patients is that we do now have adjuvant nivolumab as an option for patients who are at high risk for recurrence after cystectomy. And so instead of trying to make somebody cisplatin-eligible, let's say by putting in a PCN or doing other things that are going to impact their quality of life, it might be worth it to consider going straight to cystectomy. And then based on the pathologic stage, consider adjuvant nivolumab, if they truly are high-risk.

Sam Chang: Got it. Understood. And so with ... let's extrapolate. Because you can already tell I'm already ... I like to extrapolate. Let's switch gears and look at or consider upper tract disease, which your group has considered quite extensively. And there, I may be simplifying it, but it's almost the opposite. The majority tend to have FGFR3 alterations in the upper tract as opposed to within the bladder. So that being the case, we think ... and if we extrapolate, all right? Perhaps then a little bit more indolent course, perhaps then maybe not as good a response. But at this point too early to tell, to extrapolate this study for the upper tract? Okay. Got it.

Gopa Iyer: I would say it's too early to tell. And I would also like to say that I mean the sort of up-staging concern at the time of nephro-U is a big issue. So, I'm a big proponent of surgery. If a patient can receive neoadjuvant chemotherapy, I will give it. But I think otherwise, upfront surgery. And then there is a clinical trial, the PROOF 302 study which is actually testing another FGFR inhibitor, infigratinib. It is a placebo-controlled study, but it's an option for patients who have FGFR3-altered tumors in the adjuvant setting.

Sam Chang: And is that an oral medication, or is that-

Gopa Iyer: It is.

Sam Chang: That's an oral as well?

Gopa Iyer: Yeah.

Sam Chang: Got it.

Gopa Iyer: It's an oral medication that I believe is for about a year-

Sam Chang: Got it.

Gopa Iyer: Of treatment after surgery. Yeah.

Sam Chang: Got it.

Gopa Iyer: Which I could wax poetic about that for a while, but I won't. Because I think that's the big issue with just FGFR inhibitors in general, is that they are toxic. And it is hard right now with the ones that we have available to tolerate them, so it will be interesting to see how many of those patients can make it a whole year on a-

Sam Chang: Especially in an adjuvant setting.

Gopa Iyer: Exactly.

Sam Chang: Got it.

Gopa Iyer: Exactly. Exactly.

Sam Chang: Got it. Yeah.

Gopa Iyer: Yeah.

Sam Chang: And this will be, I'm going to end with this question then ask you for your key kind of take-home messages. But is there any push at this point at looking at these medications in a neoadjuvant setting? Understanding their toxicity, also understanding their effectiveness, are there studies ongoing now or are there institutions looking at starting them? Tell me about that kind of situation?

Gopa Iyer: Yeah. No, that's a great question. There are studies that are looking to, primarily looking at the cis-ineligible space, where there isn't a neoadjuvant option for those patients right now. We have several clinical trials of immunotherapy that have been reported out, but there's no standard of care still in the cis-ineligible space. So I think that's the first space that a lot of companies are looking into, and Janssen is the one that's really looking into it. Since they are the ones that have erdafitinib, and that's the only FDA-approved drug.

Sam Chang: Their drug.

Gopa Iyer: Probably right now they'll want to focus on monotherapy, just because it is difficult to tolerate, and only give it for a short period of time.

Sam Chang: Period of time.

Gopa Iyer: Exactly, so that we don't in any way impact the safety of getting to surgery in a timely fashion.

Sam Chang: Got it. Got it.

Gopa Iyer: Then there's also one interesting study that we are hoping to open at Memorial, which is actually a window of opportunity trial of erdafitinib, for a month only in patients who have non-muscle invasive bladder cancer and who have FGFR3 alterations. And Eugene Pietzak will be the PI of that study here. Because it's only one month, we're hopeful that there won't be much in the way of toxicities at all in those patients. But it's a way to try to either avoid a timely cystectomy or prevent further local treatments, by doing a month of treatment with erdafitinib.

Sam Chang: And then looking for a signal at least? I mean that's it, right?

Gopa Iyer: Yeah. It's a marker study. So they're really just looking-

Sam Chang: Correct.

Gopa Iyer: Yeah, they're going to do a TUR prior and post. To see how much residual disease there is, yeah.

Sam Chang: Yeah, I think that's fantastic. So Gopa, kind of key points here before we finish up? What do you want to kind of emphasize regarding either FGFR3 alterations or mutation analysis as a whole for urothelial carcinoma, upper tract, and bladder?

Gopa Iyer: Yeah, no, fantastic question. I think that we definitely need to be doing genetic analysis on all patients with either muscle-invasive or metastatic disease. And hopefully the earlier, the better. So now that we have an FDA-approved targeted therapy for FGFR3, that information will be available for patients if they do develop metastatic disease. The erdafitinib, I mean even though it's toxic, it is an active drug. And the median time to respond was about 1.4 months in the metastatic space. Suggesting that really patients, even those patients that have the rapidly progressive disease can actually benefit quite a bit from that treatment. I think that the next step moving forward, a lot of companies are looking into coming up with FGFR3-specific inhibitors. As opposed to erdafitinib, which inhibits all the FGFR's.

Sam Chang: For all of the FGFRs, got it.

Gopa Iyer: Exactly.

Sam Chang: Got it. Got it.

Gopa Iyer: So hopefully reduce toxicity, and that may also lay the groundwork for more combination trials with those drugs as a backbone in the perioperative space as well.

Sam Chang: Well, Gopa, I mean just a few minutes here, I think everyone will understand kind of your trajectory. And how much you've been able to achieve in a short period of time at your relatively youthful stage and age for ... no, we really appreciate the time that you've spent with us. Obviously, we look forward to a lot of important findings coming out from your group at Memorial and thank you again for spending some time with us and enlightening us.

Gopa Iyer: Oh, thank you so much for the opportunity. It's always great to see you and speak with you, Sam, and hopefully, we can do that in person soon.

Sam Chang: Sounds great.