Read the Full Video Transcript

Rashid Sayyid: Hello everyone, and thank you for joining us in this UroToday Journal Club recording. I'm Rashid Sayyid, a Urologic Oncology Fellow at the University of Toronto, and I'm joined today by Zach Klaassen, Associate Professor and Program Director at Wellstar MCG Health, where we'll be discussing the seminal publication, the EV-302 trial that looked at enfortumab vedotin and pembrolizumab in untreated advanced urothelial cancer. It's our opinion that this study ushers in a new era in first-line metastatic urothelial carcinoma treatment.

This study was recently published on March 7th in The New England Journal of Medicine with Dr. Tom Powles as the lead author. Platinum-based combination chemotherapy has been established as the standard of care first-line treatment for patients with locally advanced or metastatic urothelial carcinoma for nearly four decades. But there are major limitations that come with using these combinations, and the primary one is that the median overall survival is only about a year and a half, 15 to 16 months with platinum-based chemotherapy.

Another limiting factor is that patients have to be eligible for platinum, and the Galsky criteria are commonly used in this setting. But even more importantly, we know that cisplatin outperforms carboplatin in this setting, and so cisplatin eligibility also becomes an issue of major concern as well.

The question is, how did we get here? What is the evidence that we have so far to inform previous practice of platinum-based chemo as first-line combination chemotherapy? We started off with MVAC, and MVAC was first reported in 1985 at Memorial Sloan Kettering. It's been shown in a couple of publications to have an overall response rate of about 72% and a complete response rate of 36%. The median overall survival is about 15 months, in line with what we were saying, 15 to 16 months. This pretty much established MVAC as an option for these patients.

Next, the investigators asked the question, is MVAC better than gem-cis, or is the latter better? In 2000, we saw the results of a Phase 3 trial of standard MVAC versus gem-cis in systemic therapy-naive patients with stage IV urothelial carcinoma. In the comparison of these two treatment regimens, there was no difference in overall survival, about 14 to 15 months. There was no significant difference in time to progressive disease, time to treatment failure, or response rate. Pretty much the efficacy is a wash between the two.

But in terms of the safety profile, we saw significantly more grade 3-4 neutropenia, neutropenic fever, and sepsis with MVAC. The consensus was that gem-cis is probably as good as MVAC with a more favorable safety profile, and so we saw in some centers, if not most centers, gem-cis emerging as the first-line treatment.

Okay. Well, can we do better than standard MVAC? In 2001, investigators led by Dr. Sternberg asked the question, is dose-dense MVAC better than standard MVAC? This was again a Phase 3 trial that looked at patients with metastatic or unresectable advanced-stage urothelial carcinoma, and they were randomized to standard MVAC versus high-dose MVAC with G-CSF, with the study design outlined here.

In terms of the efficacy results, we saw an objective response rate that favored dose-dense MVAC of 62% versus 50% with a complete response of 21% versus 9%, so also better as well. We also saw that progression-free survival was superior with dose-dense MVAC with 9.5 versus 8.1 months. Also, the overall survival outcomes favored dose-dense MVAC, albeit not impressively, with a hazard ratio of 0.76, with median overall survival pretty much equivalent. But it was a matter of when the outcomes happened as opposed to if they did. Essentially, dose-dense MVAC outperformed standard MVAC in this setting and became more favorable in those centers.

The next question was, well, can we add something to gem-cis? Can we improve the efficacy by adding agents with different mechanisms of action? Many investigators have proposed adding immune checkpoint inhibitors. For the most part, these have failed. But more recently, in about October, November, presented at ESMO and published concurrently in The New England Journal of Medicine, we saw a trial of nivolumab plus gem-cis in advanced urothelial carcinoma, and this improved the median overall survival by about two to three months. You may note that the median overall survival here is 19 months as opposed to the historic of about 14, 15, 16. That's because we've seen the emergence of novel agents such as maintenance avelumab, and these have prolonged overall survival. As second-line, third-line, fourth-line treatments improve, the median overall survival is going to improve irrespective of first-line chemotherapy.

Taken alone, these results would be very promising, and we would've seen the emergence of nivolumab plus gem-cis as the first-line standard of care, but we also saw better results with enfortumab vedotin plus gem-cis.

Let's talk about enfortumab vedotin and pembrolizumab. So EV is an antibody-drug conjugate that's directed against Nectin-4, whereas pembrolizumab, as many of our listeners will be aware of, is an anti-PD1 immune checkpoint inhibitor. This combination was tested in smaller trials, earlier phase trials. This combination was tested in EV-103 Cohort K, which is a Phase 1b/2 trial of enfortumab vedotin plus pembrolizumab versus enfortumab vedotin in cisplatin-ineligible patients with locally advanced or metastatic urothelial carcinoma. The objective response rate was quite impressive at 65% with a 20% improvement compared to EV alone. The 12 months durable response was also impressive at 65% versus 56%.

In the same year, we saw in JCO, again another Phase 1b/2 trial in cisplatin-ineligible patients with locally advanced or metastatic urothelial carcinoma. The objective response rate here was even better at 73% with a CR of 16%. The median duration of response was over two years, 26 months, and also the median overall survival we see here is 26 months. Typically, cisplatin-ineligible patients are patients with poor outcomes, typically unhealthier; they have other comorbidities. Even in this unhealthy population, we were able to outperform the historic 15 to 16 months median overall survival. Clearly, a very strong signal that EV plus pembrolizumab is a very promising combination in this setting.

The objective of the EV-302 trial was to evaluate the efficacy and safety of the combination EV plus pembrolizumab in a cohort of patients with previously untreated, locally advanced, or metastatic urothelial carcinoma, importantly irrespective of cisplatin eligibility.

As I mentioned, this included patients with locally advanced or metastatic urothelial carcinoma of the upper or lower tracts, and they had a wide eligibility criteria. This could include patients with pure urothelial carcinoma, mixed type, or pure variant disease. Importantly, patients were unselected for Nectin-4 and PD-L1 expression. This is important because it facilitates the widespread adoption of such a combination in clinical practice where patients' tumors don't have to be sent for testing for these two markers. It's much easier and facilitates the utilization of these agents in clinical practice across various settings. Prior systemic therapy in the neoadjuvant and/or adjuvant setting was permitted as long as it wasn't a PD-1 or PD-L1 agent used.

Now, this is the trial design and the treatments that were used. The most important thing is patients were all eligible for platinum, enfortumab vedotin, pembrolizumab, and then they were randomized in a one-to-one fashion to either EV plus pembro. I want to note that there was no maximum treatment cycle for EV, and so this was at the clinician's discretion. Even after the maximum 35 cycles of pembro were done, the physicians could continue EV in their patients.

On the other hand, you had chemotherapy that could be cisplatin or carboplatin depending on the cisplatin eligibility. Important to note, they're all platinum-eligible, but they could be either cisplatin-eligible or ineligible. They obviously received gemcitabine for a maximum of six cycles.

Now, there were dual primary endpoints of progression-free survival that were assessed by a blinded independent central review using the RECIST criteria, and the follow-up involved cross-sectional imaging every nine weeks for the first year and a half and then every 12 weeks. Overall survival was also another primary endpoint in this trial. Secondary endpoints were overall response, complete or partial, the duration of response, the time to pain progression, subsequent anti-cancer therapy.

It's important to note that initially, maintenance therapy with avelumab was considered a second-line agent, so subsequent therapy, but the investigators introduced an amendment that did not consider avelumab a second-line or subsequent therapy. It's very important to note. They looked at safety outcomes as is typical. They also looked at quality of life and patient-reported outcomes. But this is going to be reported in a later publication down the line.

In terms of statistical analysis, as is typical, they looked at the intent-to-treat populations, so this is any patient that would have been randomized to the treatment. We look at time-to-event outcomes, so progression or survival. These are summarized using Kaplan-Meier methods, and then the log-rank test is used to compare treatment groups. They also went above and beyond and utilized Cox proportional hazards regression modeling and generated hazard ratios and 95% confidence intervals.

The point of this is to control for any potential imbalances or confounders between the groups that may sway the outcomes in one direction versus the other. It's an added layer of robustness to these statistical analyses. In terms of the alpha level or the type 1 error or the false positive error, they assigned it at 0.005 for progression-free survival and 0.045 or 4.5% for overall survival. This is a common misconception among readers when they look at different outcomes and see, "Oh, well this outcome has a p-value less than 0.05," but it all depends on how much alpha was assigned to the certain outcome. Although a comparison between two arms may reach a significance of less than 0.05, if the p-value or the type 1 error alpha that was assigned was, let's say, three percent, two percent, then in effect this did not reach significance. It's important to note here for overall survival, the critical p-value is 0.045.

The target sample size for this study was 860, and this would allow 90% power to detect a difference in either overall survival or progression-free survival with a hazard ratio of about 0.7 to 0.73.

At this point, I'll turn it over to Zach to go over the results and discussion, and he'll help frame the results of this study in the greater context of the treatment of urothelial carcinoma.

Zach Klaassen: Rashid, thanks so much for the outstanding historical background and context of the EV-302 trial. This is the consort diagram for this trial. As you can see, there were 886 patients that were randomized, including 442 to EV plus pembro and 444 to chemotherapy. At the bottom of this figure, we can see that there were no patients in the chemotherapy arm that remained on treatment. However, there were 144 patients that remained on treatment in the EV plus pembrolizumab arm.

This is baseline characteristics for this trial. This is broken down into two slides just based on the size of this table, and you can see that the median age for these patients was 69 years of age. The majority of these patients, over three-quarters, were male. Roughly two-thirds were white, and this was a worldwide study, so roughly 20% were Asian. When we move down to the middle of this table, ECOG performance status, the majority of these patients, over 95%, were ECOG zero or one. Again, creatinine clearance, roughly 50-50 in terms of excellent creatinine clearance versus somewhat poor creatinine clearance. When we move down to the disease status at randomization, the majority of these patients, roughly 95%, were metastatic at the time of randomization.

The second part of this slide highlights that interestingly, almost one-third of patients in the EV plus pembro arm were upper tract urothelial compared to roughly 23% in the chemotherapy arm. Over 85% of patients were urothelial carcinoma. The majority of the remainder were urothelial carcinoma with mixed subtypes. When we look at the site of metastases, roughly one-quarter had lymph node-only disease. When we talk about visceral metastases, the most common area of visceral metastases was the lung, at roughly one-third of the patients. I've highlighted the cisplatin eligibility status in this table. As Rashid mentioned, this was an important component of the trial design, is that regardless of cisplatin eligibility, patients were enrolled in this trial, and it was roughly 50-50, slightly more cisplatin-eligible than ineligible. There are some key subgroup analyses that we'll talk about in subsequent slides that distinctly look at the cisplatin eligibility patients.

This is the primary outcome of progression-free survival. You can see enfortumab vedotin in sort of this gold color and chemotherapy in gray. The median PFS for enfortumab vedotin plus pembrolizumab was 12.5 months versus 6.3 months in the chemotherapy arm, with a statistically significant benefit for EV plus pembro of a hazard ratio of 0.45 and a 95% confidence interval of 0.38 to 0.54.

This is the progression-free survival subgroup analyses, and you can see in this table, looking at all the baseline demographics, that basically everything is to the left of the hazard ratio of 1.0, which shows that these all favored enfortumab vedotin plus pembrolizumab regardless of age, race, sex, ECOG status, etc. Impressive subgroup analyses for progression-free survival favoring EV plus pembro.

As I mentioned, this is the subgroup analyses looking at PFS for cisplatin-eligible versus ineligible, and we can see very similar to the intention-to-treat population, both of these subgroups of patients benefited from the combination therapy. Median of 14.6 months and a statistically significant hazard ratio of 0.48 for cisplatin-eligible compared to a median PFS of 10.6 months for EV plus pembro in the cisplatin-ineligible. Again, a statistically significant hazard ratio of 0.43.

Moving to overall survival, again, we can see a consistent and early splitting of the Kaplan-Meier curve for EV plus pembro versus chemotherapy. The median overall survival of EV plus pembro was 31.5 months compared to 16.1 months for chemotherapy, and a 53% reduction in the risk of mortality with a hazard ratio of 0.47, 95% confidence interval of 0.38 to 0.58. Impressively, here we can see the landmark analyses that at 12 months, the survival rate was 78.2% for the combination compared to 61.4% for chemotherapy, and even at 18 months, a 69.5% survival rate for EV plus pembro versus 44.7 months. This really is a significant clinically and statistical benefit for enfortumab vedotin plus pembrolizumab with regards to overall survival.

Similar to the progression-free survival subgroup analyses, the majority of these patients showed benefit for enfortumab vedotin plus pembrolizumab across the subgroups. I've highlighted several important subgroups of interest. Patients with liver metastases, typically having a poor prognosis, but we see a benefit for EV plus pembro here, with a hazard ratio of 0.47. Regardless of PD-L1 expression, there's a benefit for EV plus pembro. Again, for cisplatin eligibility status, we see a benefit for both, which we'll highlight in the subsequent slide looking at the overall survival benefit for both of these subgroups. A median overall survival of 31.5 months for cisplatin-eligible patients, hazard ratio 0.53, and a median overall survival that is not yet reached for cisplatin-ineligible, hazard ratio of 0.43. Both of these are statistically significant.

This is overall response and duration of response, and I've highlighted several important notes that we'll talk about. In terms of the complete response rate, 29.1%, complete response rate for EV plus pembro compared to only 12.5% for chemotherapy. When we talk about confirmed overall response, over two-thirds of patients in the EV plus pembro arm, 67.7%, versus 44.4% in the chemotherapy arm, and a median duration of response that is not yet reached in the combination group versus seven months in the chemotherapy group.

With regards to subgroup analyses for overall response, again on the right side this time, favoring EV plus pembro across all of these subgroups. We see a benefit favoring the combination therapy with regards to overall response and an absolute difference compared to chemotherapy, roughly between 13 and 33% improvement for EV plus pembro across all of these subgroups.

One of the patient-reported outcomes they looked at was time to pain progression, which was 14.2 months in the EV plus pembro arm compared to 10 months in the chemotherapy arm. There is no difference between these two groups, which I think shows that this is well tolerated. Chemotherapy for metastatic disease is often difficult for patients, and it's not necessarily easy for patients with EV plus pembro, but showing no difference in time to pain progression for this important patient-reported outcome, with a hazard ratio of 0.92.

In terms of subsequent therapy, we see that there was 32.6% of patients that remained on treatment in the EV plus pembro arm compared to zero in the chemotherapy arm. The patients in the EV plus pembro arm, 31.7%, received subsequent therapy which was most commonly cisplatin systemic therapy at 24.9% of these patients. We see that in the chemotherapy arm, the most common subsequent therapy was PD-1 or PD-L1 inhibitor-containing therapy at 58.6%. Of note, too, in terms of the maintenance therapy, avelumab, for chemotherapy patients, 30.4% received avelumab.

In terms of treatment-related adverse events, I've highlighted at the top here in terms of grade 3+ adverse events, it was 55.9% in the EV plus pembro arm and was 69.5% in the chemotherapy arm, so more grade 3+ adverse events in the chemo arm. Then the rest of these treatment-related adverse events I've grouped together. There's a group here including peripheral sensory neuropathy, pruritus, alopecia, maculopapular rash that were higher in the EV plus pembro arm compared to the following here: nausea, anemia, hyperglycemia, neutropenia, thrombocytopenia that were higher in the chemotherapy arm. We do see some trade-off in terms of types of treatment-related adverse events, depending on whether the patient received EV plus pembro or chemotherapy.

In terms of treatment-related adverse events leading to discontinuation, this was in 35% of patients receiving EV plus pembro, most commonly secondary to peripheral sensory neuropathy at 10.7%, compared to 18.5% discontinuing for chemotherapy, so slightly higher discontinuation rates secondary to treatment-related adverse events for the EV plus pembro arm.

I think it's important to take a step back and look at the historical context as Rashid already beautifully delineated in the introduction, but really look at this in a single snapshot of where we've been and where we are. In terms of the mid-80s, we had median overall survivals of roughly 14.8 months for MVAC, which was quite similar to the gem-cis and dose-dense MVAC median overall survivals of the mid-2000s. Then fast forward to previously with nivo plus gem-cis, which was recently published as Rashid mentioned. The median overall survival was slightly better than what we've seen at 21.7 months, but it just shows how impressive the EV plus pembro 31.5 months is. This is a drastic improvement over chemotherapy, nearly double in this trial, which also fits into the historical context of just an impressive improvement in overall survival with this new combination.

Again, let's step back and look at all of these impressive wins for the EV plus pembro versus chemotherapy arm. We see there are really eight important points, sort of a take-home message slide for how good this combination therapy is. This combo led to a 55% lower risk of disease progression or death, a 53% lower risk of death, a significantly higher percentage of overall response, with the majority of these responses ongoing for 12 to 18 months, a significantly higher complete response rate, 29.1% versus 12%, efficacy across all pre-specified subgroups including cisplatin-eligible or ineligible patients... As we know, roughly 50% of patients are cisplatin-ineligible... No new safety signals for the combination and lower grade 3+ adverse events for EV plus pembro compared to chemotherapy.

Some additional discussion points that I think are important: The treatment with EV plus pembro, the PFS and OS were nearly double those observed with chemotherapy, and the median OS in the EV plus pembro group was at least as good as seen in the previous Phase 2 trials, so there is some consistency there. This may be attributable to the inclusion of cisplatin-eligible patients in EV-302, which typically do have better survival outcomes.

Finally, the OS magnitude of benefit is stable in EV-302 with nearly 75% of the survival events already occurring in this final analysis. At the time of the data cutoff, 31.7% of EV plus pembro patients had received subsequent therapies, the majority of which was platinum-based chemotherapy as second-line therapy. As we've mentioned, 32.6% continued to receive trial treatment.

In conclusion, the EV-302 trial showed a significant survival benefit of enfortumab vedotin plus pembrolizumab as compared with chemotherapy in patients with previously untreated, locally advanced, or metastatic urothelial carcinoma. For the first time in decades, we now have a new combination therapy in the first line metastatic urothelial carcinoma regardless of cisplatin eligibility. This was approved by the FDA on December 15, 2023.

Thank you very much for your attention. We hope you enjoyed this UroToday Journal Club discussion of the EV-302 trial, recently published in The New England Journal of Medicine.