Ashish Kamat: Hello, and welcome to UroToday's Bladder Cancer Center of Excellence. I'm Ashish Kamat, professor of Urologic Oncology in Cancer Research, at MD Anderson Cancer Center. And it's a distinct pleasure to once again welcome, Professor Guru Sonpavde, to join us after GU22, we just finished and wrapped up in San Francisco.

Dr. Sonpavde has done this for the UroToday platform before, where he distills down the best of GU ASCO. And he's going to talk to us today about his top five papers, presentations, groundbreaking things that came out at GU ASCO. And at the end, we'll wrap it up with a few questions, if there are some. Because knowing you, Guru, you're going to have it completely wrapped up for us. So I'm going to hand the stage over to you and take it from here.

Guru Sonpavde: Thank you, Ashish. I want to thank UroToday for the invitation to present the bladder cancer highlights from the GU ASCO symposium. And with apologies for presentations that I probably will miss, because I really wanted to highlight the major five presentations.

These are my disclosures.

So just to give you a little bit of background before getting into the new data at GU ASCO symposium, we, of course, know that in the metastatic disease setting, there have been multiple advances in the past few years. We have switched maintenance avelumab PD-L1 inhibitor, in the first-line metastatic disease space. And of course, we also have a couple of antibody-drug conjugates, enfortumab vedotin, sacituzumab govitecan. And we also have erdafitinib, an FGFR inhibitor, in patients with FGFR2 or three, activating mutations or fusions. So multiple advances in the systemic therapy of metastatic disease.

And also, we have advances in the perioperative space, where we've always had neoadjuvant and adjuvant cisplatin-based combination chemotherapy. But we now also have nivolumab, a PD-1 inhibitor, in the adjuvant disease setting of high-risk muscle-invasive urothelial carcinoma, following surgery. This is both upper tract and bladder.

So just to get into probably, the five of the abstracts and presentations that I thought were the key promising data for new therapies. So one of them is the TROPHY-U-01 trial, which was a trial that combined sacituzumab govitecan with pembrolizumab, in patients who are progressing with metastatic disease, post-platinum-based chemotherapy. And this is one of the cohorts of this Phase II trial. And this is a non-randomized study, with the primary endpoint of objective response.

The rationale for combining an antibody-drug conjugate and an immune checkpoint inhibitor is quite strong. We have strong data for enfortumab vedotin, the other antibody-drug conjugate that's approved, the Nectin-4 targeting antibody-drug conjugate, which combined, pembrolizumab showed a response rate of 73%. And this is in the first line cisplatin-ineligible setting.

Now there is a rationale for irinotecan, a topoisomerase one inhibitor, based on preclinical evidence, showing that it does induce immunogenic cell death, and therefore, might synergize with the PD-1 or L1 inhibitor.

So coming to the data from this study, sacituzumab combined with pembrolizumab appeared active. The response rate was 34%, as you see here, in a cohort of 41 patients. And of course, this is a non-randomized study. 63% of patients have some tumor regression.

So there was regression seen across risk groups. And regardless of visceral metastasis status, there were some durable responses, as you can see on the right-hand side.

The safety profile was consistent with what you might expect. As you might expect, with sacituzumab, you get diarrhea and myelosuppression. And with pembrolizumab, you would get immune-related adverse events. So when you combine these two, you get a little bit of all of these, and so it was quite consistent with what you would expect.

One of the things that jump out at you, is the response rate of 34%. Which comparing across studies, with the EV pembro study, which was the first-line study, it appears much lower. And some of these could be potentially, partly from patient selection. So as you can see here, the median time from the last prior neoadjuvant therapy was 6.8 months. Median time from prior systemic therapy for metastatic disease was quite low, under two months. And the response to prior platinum was low. So patients who had gotten on this had not responded to platinum that well before. So there might be some selection issues.

Now, one of the issues I wanted to bring up, and I had the opportunity to discuss this presentation at the GU ASCO symposium, is that there is a preclinical study looking at irinotecan in a different animal model, in a melanoma model, but it looked like you needed a functional TP53 protein, for a topoisomerase one inhibitor, to induce, to be immunogenic, and induce T-cell mediated killing of these melanoma cells. So maybe there is some of this going on without selection for this factor, of course, in this study.

Now, what we have to consider is these data in the context of other ongoing Phase III trials in the first-line setting. And of course, EV/pembrolizumab has graduated to the first-line setting, and in a Phase III trial, and this is being compared with gemcitabine-platinum. The other CheckMate 901 trial is the other key trial, which hopefully, we will see results of, sometime this year, comparing nivolumab ipilimumab versus gemcitabine-platinum platinum.

Now there are other combinations ongoing, combining sacituzumab govitecan with other regimens. One is an ipilimumab/nivolumab plus sacituzumab in the first line disease setting, an investigative initiated trial at Moffitt and Dana-Farber. There are also other trials combining sacituzumab with cisplatin alone or cisplatin with avelumab.

There is also an interesting study combining sacituzumab with the other antibody-drug conjugate, enfortumab vedotin. This is two antibody-drug conjugates being combined, which we call the DATA, the double antibody-drug conjugate trial, which is a Phase I trial that we have open here, actually at Dana-Farber.

Yet another combination that might be of interest is PARP inhibitor, combined with sacituzumab govitecan, which has looked interesting across different cancers in a Phase I 1B trial.

My comments, based on these data with this combination, were that, although the response rate looks low, there might be selection issues. It might be a data set that needs to be looked at while controlling for some of these selection factors. But however, we have to consider the real possibility that sacituzumab govitecan may be more myelosuppressive and immunosuppressive, and therefore, may not combine well, or play well, with an immune checkpoint inhibitor, like pembrolizumab. So these are things that we need to consider.

Maybe we need to select patients better for Trop-2 gene expression, instead of just putting everybody on this trial, based on the fact that the protein is over-expressed in almost everybody, but maybe the gene over-expression is also important. Maybe we should consider the genomic context.

Is there an optimal setting? The saci/pembro study was in the secondhand setting. The EV/ pembro study was the firsthand setting. So these are things to consider.

I also think that the cooperative groups, the NCI, need to consider some of these interesting combinations that sometimes industry may not do. Just because some of these drugs are controlled by different companies and therefore, there is sometimes an issue of combining the right drugs with each other.

Now coming to the second trial that I wanted to highlight. This is the neoadjuvant, enfortumab vedotin. This is the other antibody-drug conjugate that's approved in the second and third-line settings. It targets Nectin-4. So this was a straight non-randomized Phase II study, looking at neoadjuvant enfortumab vedotin alone, no immune checkpoint inhibitor, with a primary endpoint of pathologic complete remission.

This was a small study, 22 patients, as you see here. One of the interesting things in this study was a lot of patients were considered cisplatin-ineligible. And this was cisplatin-ineligible patients, of course, because cisplatin eligible patients would get cisplatin with combinations. So, 41% of patients were cisplatin ineligible because of grade two or more hearing loss. That's a little bit unusual, but this is, of course, a small study.

One other thing I wanted to highlight is that clinical T2 N0 zero patients were around two-thirds of patients. And I wanted to highlight that, some of these trials that have come out recently, have had a different mix of clinical T2 N0, and T3, T4 N0. And the reason this is important is that there is a difference in the pathologic complete response rate based on your baseline clinical stage. So when you look across MVAC studies or atezolizumab study I highlight here, there is clearly, a numerical better pathologic complete remission rate in clinical T2 N0 patients, as opposed to clinical T3 or T4 N0 patients. So this is something to consider, and control for in trials.

So coming to this study itself, enfortumab, 22 patients, neoadjuvant, single-agent therapy. 36.4% pathologic complete response rate. So that's quite impressive for a single agent, three cycles of enfortumab vedotin. The toxicity profile was consistent with what's seen. Now, I don't have a slide to show this, but there was some anxiety about three perioperative deaths in this study. So somebody asked Dr. Dan Petrylak, why this was so. Could this be from the neoadjuvant enfortumab? And Dr. Petrylak did not think it could be. This is a small study, so sometimes you do see some numbers that are outliers in small Phase II studies. And as you know, many more toxic agents and regimens have been given in the neoadjuvant setting. But I think that this data does need to be hashed out. We need some more data to explore and look at this preoperative issue a little bit more.

Now, of course, enfortumab vedotin in combination with pembrolizumab has graduated to the neoadjuvant setting in Phase III trials. So these are Phase III trials that are highlighted in these red boxes, which I show you here. Which are combining enfortumab vedotin with various immune checkpoint inhibitors, compared with the standard of care, in the neoadjuvant setting in Phase III trials.

So coming to neoadjuvant EV, is the data interesting and promising? Yes. I think the pathologic complete remission of 36% is interesting. I think antibody conjugate development in earlier disease settings is rational. And yeah, as you may know, it's also being looked at in the non-muscle-invasive bladder cancer setting with intravesical therapy. So that is an interesting study. Now pathologic complete remission is impressive with enfortumab, but you must recall that this really doesn't mean anything unless the pathologic complete remission translates to improve long-term survival and cures. Right? So we don't have those data. We need longer follow up from this trial.

We also should consider, always reporting the pathologic complete remissions, based on baseline clinical stage, which was not done in this presentation. So I think the clinical T2 N0, clinical T3, to T4 N0, have vastly different pathologic complete remission rate. You can stack your trial full of clinical T2 N0 patients, and get a high path CR rate. So I think that this is important to pay attention to.

I'm of course, and we are all very eager to look at the data for the Phase III trial, combining neoadjuvant EV with immune checkpoint inhibition. And I think precision medicine should be implemented in this space. These drugs, EV, sacituzumab, and antibody-drug conjugates that have been developed across all comers, based on the fact that the protein targets, Nectin-4 and Trop-2, are present in the vast majority of patients. However, when you're looking at the neoadjuvant setting, where you really are looking for a great depth of response and cures, maybe you want to select patients even better. Maybe you should look at Trop-2, or Nectin-4 gene, over-expression to select these patients, because there is data in the translation studies showing that gene over-expression might improve the depth of response.

Now getting to the third study I wanted to highlight. This is a study, also in the metastatic disease setting. This is a study that compared pembrolizumab with, or without, lenvatinib, which is an oral TKI, that targets VEGFRs, as well as FGFR1. This was a Phase III trial, conducted in the first-line space of platinum ineligible patients and cisplatin-ineligible PDL-1 high patients. Regrettably, this is a design, I show you here. They were planning to accrue 350 patients per arm. Now, regrettably, this study was closed early, based on the data safety monitoring committee recommendations. And we will come to why that was so, and there was not an improvement in outcomes, obviously.

This study was well balanced. It finally accrued around 245 patients per arm. There were a ton of patients who were not eligible for any platinum. So they were really a platinum ineligible population, or a chemo ineligible population, more than 80% of patients, as opposed to cisplatin-ineligible, PDL-1 high. So that may have compromised the three, and led to this progression-free survival, which was not better, with the combination of lenvatinib/pembrolizumab compared with pembrolizumab plus placebo, despite promising Phase II data earlier.

And similarly, overall survival also was not different, around a median of one-year median overall survival in both arms. And you look at response rate, also around a 30% response rate in each of these two arms.

When you look at the toxicity profile, nothing jumped out. I think that the toxicity profile for the combination of lenvatinib/pembrolizumab is consistent with what would be expected with these two agents, with some GI toxicity, some hypothyroidism, which has been seen with both of these agents.

When you look at the immune-related adverse events, again, consistent with what you would expect.

So to conclude, this trial was obviously, disappointing. The combination of VEGF and PD-1 inhibition has looked promising in Phase II studies in this disease. And of course, this combination is approved in many other cancers, especially RCC, which, we GU oncologists treat commonly. So this is a disappointing study. And maybe the study was compromised by the 80% of patients being platinum ineligible. So chemo is ineligible. So maybe we need to pay attention to our selection of patients.

Now getting to the next study of interest, this is a drug that's called trastuzumab deruxtecan. This is an antibody-drug congregate that targets HER2. This is approved in other cancers, notably, HER2-positive breast cancer. And this was looked at in combination with nivolumab in HER2-expressing urothelial carcinoma.

So this is a study of multiple cohorts. And the cohort of interest in this study is cohorts three and four, which included patients with metastatic urothelial carcinoma who were progressing after chemotherapy. So after platinum chemotherapy. And they had IHC 3+ or 2+ disease. There was also a cohort four, that was HER2 low. This drug may have activity, even in HER2 low patients, based on the fact that the payload toxin is released in the tumor macro environment, and so you might not need HER2 high-expressing cells for activity of this drug. But this cohort was closed early, with only four patients accrued, because of difficulty in accrual, actually. So there were 30 patients in the HER2 high, 2+ or 3+ IHC, that we are going to focus on.

This is the demographics of this patient population. Nothing jumped out at me. Most of the patients, 30 of the 34, I'm going to talk about, is the IHC 2+/3+, cohort three. Looking at the efficacy. As you can see, this combination is active, the nivolumab plus trastuzumab deruxtecan, with a response rate of 36.7% in 30 patients.

Also, you will note, that in those who are IHC 3+, so the most high-expressing tumors, these patients had the highest response rate. Five of eight of these patients responded.

Also of note, although there were only four patients in cohort four, which is the low IHC 1+ patients, two out of these four patients had a response. Again, that's promising, because that suggests that maybe, even the IHC 1+ ought to be developed for this drug, based on the fact that it seems active even in the low IHC patients, just as it was hypothesized, based on the mechanism of activity of this drug, the release of the payload in the environment, which leads to killing even low IHC target expressing cells.

When you look at the regression of tumor, significant number of patients had some regression of tumor.

When you look at the safety profile, really, nothing jumped out. The safety profile was consistent with what's seen with nivolumab and trastuzumab deruxtecan. You can see that 30% of patients did discontinue the drug, or one of these two drugs, because of toxicities. The key toxicities are interstitial lung disease and LV dysfunction. And that's because trastuzumab deruxtecan can cause interstitial pneumonitis and cardiac dysfunction. So as you can see here, they saw an interstitial lung disease incidence of 23.5%. But most of these were grades one and two, so that's reassuring. They also saw LV dysfunction grade three in one patient, so that's 3.3%. So low, in terms of cardiac dysfunction.

Now, one of the interesting findings in the biomarker studies in this trial was that those with the CPS of one or higher PDL-1 expression, and those with high blood-based TMB, tumor mutation burden, seem to benefit more, so that's interesting. So maybe we have a biomarker for an even better selection of these patients.

So overall, this is a promising combination, trastuzumab deruxtecan plus nivolumab, appeared active in post platinum in patients with progressive disease, and especially in PDL-1 high CPS, one or higher, and in blood-based TMB high patients. So this combination might warrant further evaluation in these patients. As you know, we've had a remarkably disappointing history with HER2 targeting in this disease in larger trials. So maybe this drug actually does give us some hope for therapy in this area.

So next, I'm going to jump to a couple of PARP inhibitors, really quickly. This is the first of the two trials that looked at a PARP. This is the BAYOU trial. This is a Phase II randomized trial, which looked at durvalumab alone, or in combination with olaparib, a PARP inhibitor, for the first-line therapy of platinum ineligible patients with metastatic urothelial carcinoma.

So the background, of course, is that this is a disease in which there are DNA damage repair alterations seen, the frequency of around a quarter of patients who have these. And you would expect that these kinds of alterations would make these tumors more sensitive to PARP inhibition.

This is a randomized Phase II study, which looked at the first line platinum ineligible patients. And it could be also cisplatin-ineligible patients or carboplatin ineligible. So not just platinum ineligible, actually. And when you look at these patients, the PFS, the primary endpoint in this study was PFS.

There were around 76 patients in each arm. Nothing jumped out at me with this.

And as you can see here, the HRR, homologous recombination alteration status, found in approximately 20% of patients per group, this was the panel of HRR changes that were examined, based on the foundation of one assay.

So just jumping quickly, to the overall primary endpoint. This is progression-free survival. So the overall progression-free survival in all comers in this first-line study did not meet the primary endpoint. The median PFS was around four months for durva/placebo or durva/olaparib. However, lots of interest is that, in this exploratory endpoint, in patients with the homologous recombination repair defects, there was a significant improvement in progression-free survival, going from 1.8 months for durva/placebo, to 5.6 months with durva/olaparib. So this was quite significant. Again, this is an exploratory endpoint. Again, you see the subanalysis showing a benefit in the HRR altered patients.

When you look at overall survival, and the intention to treat the all-comer population, no difference. However, in the HRR population, a difference in the survival. Again, these are not well powered, but you can see one of the other interesting findings was that, in the control arm of HRR mutated patients, they seem to be the lowest survival. So it appeared, that patients who had somatic HRR alterations in the tumor had the worst prognosis. So this is an unmet need population, perhaps, that could benefit by adding olaparib.

Also of interest is that, in the HRR altered group, the response rate with durva/olaparib was 35.3%. Really, that's an impressive response rate for this combination in this group, while there was not a big difference in the overall intention to treat population. And this, numerically, looks better, but this is driven really by the HRR group doing much better by combining olaparib with durvalumab.

The safety profile was consistent with what is expected from olaparib. This is overall, a tolerable drug, with manageable toxicities.

So to end, the secondary analyses suggest that there is a potential PFS benefit with durvalumab combined with olaparib, in patients who have somatic HRR operations. So maybe, PARP inhibitors can benefit, at least this subgroup of patients, combined with immune checkpoint innovation.

I will point out, and I did not have time to get into this, is that there is a history of negative studies with PARP inhibition in bladder with rucaparib not being active in the post platinum setting, in both the wild type and the altered patients, as single-agent therapy.

Now getting to the last study I wanted to highlight. This is the ATLANTIS study, which is a double-blind study that looked at maintenance, which maintenance PARP inhibition, following platinum chemotherapy in stable or responding disease.

This is a study, similar to kind of the STAMPEDE, which is a Phase III study, but this is more, this is a Phase II study here, the ATLANTIS study, where following stable or responding disease after platinum-based chemotherapy, patients are directed towards these biomarker positive groups. So this is the DNA damage repair positive group, where patients are sent to the rucaparib trial, in which there is, within this cohort, there is a randomization to placebo or rucaparib. No immune checkpoint inability here. This is a single-agent rucaparib with the primary endpoint of progression-free survival. And this is the switch-maintenance setting.

They were targeting hazard ratio of 0.5. And most of the patients had had prior CR or PR, with platinum-based chemotherapy, before they went on to maintenance rucaparib or placebo.

So when you look at the primary endpoint, there was a significant trend in improvement of PFS, the hazard ratio is 0.07. Did not quite meet the less than 0.05, but this is a significant trend for improvement with rucaparib. The median PFS is 35.3 weeks, compared to 15.1 weeks with placebo. So that's an impressive median PFS. If you remember the median PFS with switch-maintenance avelumab, that was only 3.7 months in the JAVELIN Bladder 100 trial. So that's around 16 weeks. So you can see that this is about twice the median PFS seen with the avelumab in the switch-maintenance JAVELIN Bladder 100 trial.

Overall survival in this study was not statistically significant. Again, this is not well powered.

The safety profile is consistent with what you would expect with rucaparib. Nothing remarkable, other than the routine side effects that we do expect with this drug, some fatigues, and nausea. So very manageable toxicity profile.

So to conclude, in this study, maintenance rucaparib following stable or responding disease and platinum-based chemotherapy, extended PFS in patients with metastatic disease, that exhibited a DNA damage repair alteration. And this could be somatic or germline, mostly was somatic. And maintenance rucaparib was tolerable. So this, I think, also gives hope to PARP inhibition in a biomarker-selected group of metastatic disease patients.

With all that, thank you. And I will await questions as they come. This is my contact information if you want to shoot me questions. Thank you.

Ashish Kamat: So Guru, once again, you've covered the top abstracts and top presentations at GU ASCO, and on behalf of our audience, I want to thank you. I'm sitting here listening to you, and I was following GU ASCO virtually, because I couldn't be there on purpose in person, due to the attendance ban from our institution. But the way you covered it really helps distill it. I want to thank you again, on behalf of our audience.

I don't really have any specific questions about the presentations that you give, because as I said, you covered it so well. But just for the benefit of our listeners, some that might not be as familiar with some of these data, I want to ask you a few general questions. And one is, how should the reader, or the viewer, be evaluating these results of combination therapies, or therapies that are coming after gem/cis backbone, when studies, including the recent VESPER study, suggest that an MVAC backbone might be, in some ways, it should be, the standard of care, but it isn't. Would you have any pearls of wisdom for people looking at that issue?

Guru Sonpavde: Yeah. Thanks, Ashish. There's always been a debate about MVAC dose-dense and MVAC versus gemcitabine/cisplatin. So I look at it in two directions. So one is, the metastatic disease setting, where gem/cis, versus MVAC, appears similar in the metastatic disease, first-line, for efficacy. And the toxicity profile appeared much better for gem/cis. So this is mostly a palliative setting, the metastatic disease setting. So there, I tend to go with gem/cis, just because it's much more tolerable.

Now, the neoadjuvant is a little bit of a different situation, where you're trying to cure patients, and you might actually want to give a more toxic regimen, if it's more effective, and if it can cure more patients. Now, the VESPER trial. Now we have, of course, have data in a Phase III trial with MVAC, which has improved survival in the SWOG 8710 landmark trial, from almost two decades ago.

Dose-dense MVAC, of course, has been used a lot, since, in the metastatic disease setting, it did appear a little bit better than MVAC, at least in the tail of the curve. And so, the VESPER trial is the key trial that is brought up. So the VESPER trial was a trial that looked at perioperative for muscle-invasive bladder cancer, perioperative gem/cis, versus dose-dense MVAC. However, it was mostly a neoadjuvant trial of around 85% of patients. The other 15% were adjuvant patients. So there was a kind of heterogeneity of phase context of the perioperative therapy. The other confounding issue in this study is, that they use six cycles of dose-dense MVAC, and compared it with four cycles of gem/cis. Now, that right there is a big no-no to me. Because the standard of care is four cycles, three to four cycles, of dose-dense MVAC. And so, you're kind of biasing and making an unlevel playing field here, with six cycles of dose-dense MVAC.

So what the study found is, in the primary comparison, the study was negative, because a disease-free survival was not different in all comers. However, when they looked at the neoadjuvant subgroup, which was unplanned, it was kind of a secondary endpoint, not a primary endpoint, not even the co-primary endpoint, it was positive. So there was an improvement for six cycles of dose-dense MVAC, versus four cycles of GAGC, when given neoadjuvant, dose-dense MVAC was more toxic. We know that it does have more toxicities, in terms of stomatitis, mucositis, hair loss, and infection, cytopenias.

So I look at it as I mostly still do gem/cis. However, given the higher response rate, and downstaging rate with dose-dense MVAC, if I have a patient who is node-positive N1 disease, I might want to give dose-dense MVAC, just because of the higher downstaging rate, you might render some of these patients operable if you reduce the burden of disease. So I kind of use dose-dense MVAC sparingly, perhaps in younger, for the patients, and in patients who are N1 disease.

Ashish Kamat: Yeah, I know. Great points. And again, my question was more, just again, to highlight the points that you made, which is the more use of cycles. But also, just to keep in mind, that the backbone that goes in with immunotherapy, or combination therapy, or coming in after, doesn't necessarily reflect what happened before, because we're using targeted agents. Right?

And along those lines, and again, just in short for our viewers, what are your thoughts on this sort of downward shift in the definition of cisplatinum ineligible? It used to be that you had to have strict criteria, creatinine clearance, and nowadays, even patients who refuse cisplatinum, are considered cisplatinum ineligible. So some words of caution, I guess, from your side to our viewers, based on that aspect, and the readouts of these trials?

Guru Sonpavde: I agree that there has been a slippery slope toward broadening the definition of platinum ineligibility. I think that one of the things to remember is, we still think that cisplatinum-based combination is what's been shown to improve the cure rates of patients with muscle-invasive bladder cancer. One situation where it does come up is in the adjuvant setting, where nivolumab is now approved. So nivolumab is approved for high-risk muscle-invasive disease. And that trial was post neoadjuvant/cis, or patients had to be cisplatin-ineligible, or they could be refusing cisplatin. So the thing is, if you see a cisplatin eligible patient, who's not received neoadjuvant cisplatin, I'm still going with cisplatin-based combination adjuvant therapy, just because the long term tracker and survival improvement have been shown with perioperative cisplatin. Whereas with adjuvant nivolumab, all we have is improved BFS disease-free survival, and we don't have anything, we haven't seen anything yet on survival. Maybe there is survival. I do expect, improved BFS with an immune checkpoint inhibitor to be more likely to translate to improved survival, just because the benefits tend to be more durable, but we haven't seen it yet.

Ashish Kamat: Guru, you know we could go and keep chatting forever. But in the interest of time, I want to thank you for taking the time and spending it with us, and this is great, as always. Once again, thank you.

Guru Sonpavde: Thank you, Ashish.