Christopher Wallis: Hello and thank you for joining us for this UroToday Journal Club. Today, we're discussing the JAVELIN Bladder 100 trial of avelumab maintenance therapy for advanced or metastatic urothelial carcinoma. I'm Chris Wallis, a fellow in urologic oncology at Vanderbilt. And with me today is Zach Klaassen, an assistant professor in the division of urology at the Medical College of Georgia. This is the citation of Dr. Powles' work in the New England Journal of Medicine.

By way of background, most UroToday readers will know that among eligible patients, first-line therapy for those with advanced urothelial carcinoma is a combination of platinum-based therapy, either gemcitabine and cisplatin or MVAC. With this approach, just quite proven benefits, disease progression is still common. The median overall survival is less than 18 months. More recently, immune checkpoint inhibitors, including pembrolizumab and others have shown benefit in second-line therapy among patients with advanced urothelial carcinoma. And in other tumor types, particularly in melanoma, the use of maintenance therapy with an alternative regime apart from the induction course has shown benefit following the induction chemotherapy.

Avelumab is an anti-PD-L1 antibody that has been approved in the treatment of locally advanced or metastatic urothelial carcinoma among patients who have disease progression either during or after their initial induction platinum-based chemotherapy. And so the hypothesis of the JAVELIN Bladder 100 trial was whether avelumab maintenance would improve outcomes in patients who receive first-line platinum-based chemotherapy.

For eligible patients, disease characteristics were as follows, they had to have histologically confirmed either unresectable, locally advanced disease or metastatic urothelial carcinoma, stage IV disease diagnosed at the time of enrollment. They had to have measurable disease per RECIST criteria and no evidence of disease progression while on four to six cycles of initial induction chemotherapy. Further, they had to be within 4 to 10 weeks following their last dose of chemotherapy. Able to be enrolled in a maintenance approach rather than a second treatment course.

Further inclusion criteria included an age of 18 years or greater, good performance stats with an ECOG zero or one. Tumor specimens were available for biologic analyses and adequate hematologic liver and renal function. Patients were excluded if they had received adjuvant or neoadjuvant chemotherapy within 12 months of randomization dates and if they had a contraindication or previous use of immune checkpoint inhibitors.

This is an overview of the study design. As you can see, patients were eligible if they had a complete response, partial response, or stable disease, essentially not progressive disease while on their standard induction chemotherapy, which could be either cisplatin and gemcitabine or carboplatin and gemcitabine, depending on the patient's renal function. Again, there's a treatment interval of four to 10 weeks and then patients were randomized to either receive best supportive care alone or with the addition of avelumab. The primary endpoint of this trial was overall survival and this was analyzed in both the overall intention to treat a population of all randomized patients, as well as a subpopulation of those with a PD-L1 positive disease. And we will describe in a minute how that was defined. Secondary endpoints included progression-free survival, objective response rate, safety, and tolerability, as well as patient-reported outcomes.

As that figure highlighted, the study was designed as a one to one randomization approach, stratified according to the response to first-line chemotherapy, dividing patients who had complete or partial responses from those who had stable disease, as well as by the nature of their metastatic disease, whether it was visceral or non-visceral. Treatment was continued until progression, unacceptable toxicity, or withdrawal of consent. And dose reductions were not allowed, but subsequent doses could be omitted if there were persistent adverse events.

As we said, the primary endpoint was overall survival with a number of other secondary endpoints.

Trial mandated assessments were performed using imaging studies every eight weeks for the first year and then every 12 weeks until the time of progression. Brain imaging was required if there was previous history of brain metastases or if these were suspected. In terms of biomarker assessment, PD-L1 expression was assessed using Ventana assay, and patients were categorized as PD-L1 positive if they had any one of the following three characteristics; at least one-quarter of tumor cells stained positive for PD-L1, at least one-quarter of immune cells stained positive for PD-L1, if more than 1% of the tumor area contained immune cells or if 100% of immune cells stained positive for PD-L1 if less than 1% of the tumor area contained immune cells.

In terms of statistical analysis, the trial was designed as a one-sided testing strategy so they use a one-sided P-value of 0.025. Of this, 0.015 was allocated to overall survival in the overall population, 0.01 in the PD-L1 positive population. They use the two-look, group-sequential design so using alpha-spending was employed to keep from over-testing. The trial was deemed positive if the stratified log-rank test was significant in either the ITT or PD-L1 positive primary study populations. Efficacy analysis was performed in the ITT population while safety analyses were among all exposed patients.

And now we are getting to results so I will hand over to Dr. Klaassen.

Zachary Klaassen: Thanks, Chris. This was a study of 700 patients that were randomized to either avelumab or the control group, which was the best standard of care. You can also see here that just over 1/2 of these patients were defined as PD-L1 positive, including 189 in the avelumab group and 169 in the control group. On the right is the baseline characteristics, it's pretty standard for a metastatic bladder cancer population. The median age was roughly 70 years. You can see that about 1/4 to 1/3 of these patients were actually upper tract urothelial with the rest of them bladder urothelial. You can see that about half of these patients had visceral metastases. The most common first-line chemotherapy regimen was cisplatin in just over 50% and about 40% of these patients actually received gemcitabine plus carboplatin. You can also see here that in terms of the best response to first-line chemotherapy, roughly three-quarters of the patients had a complete response or partial response. And the remaining quarter had stable disease.

This table here shows the subsequent anticancer therapy. You can see in the middle here, the overall population of avelumab. 42.3% had subsequent anticancer therapy, whereas the best standard care alone had 61.7%. And you can see here that in the PD-L1 positive population, nearly two-thirds of patients in the best standard of care alone had a subsequent anticancer therapy. When looking at any PD-1 or PD-L1 inhibitor, only 6.3% of the avelumab patients had that treatment whereas nearly half of the patients in the best standard of care did.

Looking at the primary outcome, which is overall survival in the intention to treat population. This was over a median follow up of 19 months for both the control and the intervention arm. You can see here that there is an early splitting of the curves between avelumab and the control group with a median overall survival for avelumab of 21.4 months, with a 95% confidence interval of 18.9 to 26.1 months versus the control arm, which was 14.3 months at a 95% confidence interval of 12.9 to 17.9 months, corresponding to a stratified hazard ratio of death of 0.69 with a significant confidence interval of 0.56 to 0.86.

They also did an overall survival analysis in the PD-L1 population. Once again, we see a very drastic splitting of the curves early between avelumab and the control group. In this population, the immediate overall survival for avelumab was actually not reached and was 17.1 months in the control group with an even more significant hazard ratio of 0.56 with a 95% confidence interval of 0.40 to 0.79.

This is a forest plot of a subgroup analysis looking at overall survival in the intention to treat population. You can see the asterisks on the left here are the significant subgroup analyses. You can see that patients older than 65 years of age, male patients, those with excellent performance status, ECOG zero, white patients, those that were treated in Europe, and interestingly here we see first-line chemotherapy, right in the middle of the slide, both gemcitabine plus cisplatin and gemcitabine plus carboplatin were significant for overall survival. This is important because carboplatin is often thought of as an inferior treatment as opposed to GemCis. Not surprisingly, those with complete response or partial disease also had a significant benefit as with those with non-visceral metastases and those with PD-L1 positive status, as you mentioned already. Those subgroups with asterisks all favored avelumab versus standard care alone.

Looking at the progression-free survival analysis in the intention to treat population, you can see here that the median PFS for avelumab was 3.7 months versus the control group of 2.0 months with a significant hazard ratio of 0.62 and 95% confidence interval of 0.52 to 0.75. Similarly, a very similar-looking curve with the PD-L1 positive population for progression-free survival. We can see the avelumab PFS median was 5.7 months compared to the control group of 2.1 months and a hazard ratio once again, significant 0.56, 95% confidence interval 0.43 to 0.73.

This is another forest plot looking at the intention to treat population for progression-free survival. Lots of asterisks on the left, as you can see, even more than the overall survival. Basically, the majority of these patients had a benefit of avelumab maintenance therapy.

We'll take some time looking at this table. This is the response in the overall population and the PD-L1 positive population. You can see here the confirmed objective response percent-wise for the avelumab group was 9.7% compared to only 1.4% for the control group, which is even more drastic in the PD-L1 positive population, 13.8% versus 1.2%. Looking at the confirmed best overall response, complete response in the avelumab overall population was 6% and up to 9.5% in the PD-L1 positive population. Partial response, 3.7%, and 4.2% in the PD-L1 positive population and stable disease were quite similar between both the intervention and control group in the overall population and PD-L1 positive population. As you can see here, disease control, 41.1% for the overall population compared to 27.4 and even slightly more impressive in the PD-L1 positive population with 43.9% in the avelumab group and 27.8% in the control group.

Looking at the adverse event profile, not surprisingly when you're comparing an intervention versus essentially a control best standard of care, there were more adverse events in the avelumab group. 98% of patients in any grade versus 77% in the control group. Looking at grades greater than three, adverse events 47.4 in the avelumab group and 25.2 in the control group. Most commonly for the avelumab group, you can see here was fatigue, pruritus, urinary tract infection, diarrhea, and arthralgias, but otherwise relatively well tolerated in general.

The next table will look at serious treatment adverse events and you can see here that there was 27.9% in the avelumab group and 20% in the best standard of care alone group. Most commonly, a serious treatment adverse event was a urinary tract infection at 4.7% of the population, followed by acute kidney injury and hematuria.

Several important discussion points from this trial. The JAVELIN trial showed that patients with advanced urothelial carcinoma who had non-progressed disease with platinum chemotherapy had significantly longer overall survival with maintenance avelumab versus best supportive care alone. And this was evident both in the overall and PD-L1 positive population. This is despite the more frequent use of subsequent therapy in the control group. Patients treated with avelumab also had longer progression-free survival and this is quite impressive as I showed you in the OSPFS and forest plots, the PD-L1 expression in two more immune cells appears to be quite predictive. We saw impressive results with the PD-L1 population. This is basically the immune checkpoint inhibitor therapy moving from the second-line into the maintenance, but we've also seen some initial results in the first-line setting from the IMvigor 130 trial where the final PFS analysis published earlier this year showed a significant advantage for the combination of atezolizumab plus chemo versus chemotherapy alone with a hazard ratio of 0.82 and a significant confidence interval.

We are continuing to see more trials coming out. We've had several agents approved in the second-line setting and we are starting to see it move back into the earlier disease phase in terms of the maintenance for this trial, as well as a combination in the first-line setting.

In conclusion, first-line maintenance avelumab in patients with advanced urothelial carcinoma with the non-progressive disease with platinum-based therapy, chemotherapy had a significantly longer overall survival than best supportive care alone. Certainly, this is a new standard of care. This is very exciting for our bladder cancer patients and we look forward to being able to treat them with this maintenance therapy.

Thank you very much for your attention. We hope you enjoyed this UroToday Journal Club.