Petros Grivas (PG) and Helen Moon (HM)

PG: Hello everybody, I’m Dr Petros Grivas, I’m a Medical Oncologist in Seattle. I’m a Professor and serving as the Clinical Director of the Genitourinary (GU) Cancers program at the University of Washington and Fred Hutch Cancer Center. I’m super thrilled and excited to talk today with an esteemed dear friend and colleague, Dr Helen Moon. Dr Moon, you want to introduce yourself?

HM: Of course. Thank you everyone for tuning in. My name is Helen Moon. I’m a Medical Oncologist practicing out of Southern California with Kaiser Permanente, and I am currently the Director for GU Oncology Research at Kaiser Permanente.

PG: Fantastic. The title of the podcast today will be “A Podcast on Real-World Evidence with Avelumab First-Line Maintenance and Treatment Sequencing in Locally Advanced or Metastatic Urothelial Carcinoma.” Helen, we’re going to discuss a lot of things, right?

HM: Right. I think what we would like to touch on in this podcast is data from real-world studies that’s been coming online of avelumab first-line maintenance treatment in patients with advanced urothelial cancer. And the data that’s relevant to the effectiveness, safety, treatment sequencing, and something that’s dear to my heart—healthcare resource utilization and its impact on patients. One key caveat to be highlighted is that treatment options vary in different countries, so some of the conversations about healthcare utilization are really [for] a US-based system. So, for our listeners, the points discussed may be affected by your local considerations. So, Dr Grivas, you are the world expert in this—can you provide some background information on advanced urothelial carcinoma?

PG: Thank you so much, Dr Moon, great to work with you and many others in the field. Urothelial carcinoma refers to cancers that arise in the cells that line the urinary tract system [1, 2]. Approximately 90% of urothelial carcinoma tumors arise in the bladder [2], but we always remember that [urothelial carcinoma can also arise in] other parts, for example, the upper urinary tract, like kidney, pelvis, and ureter and, of course, urethra, which is less common [2]. The most common type is bladder cancer, as we discussed. Bladder cancer is very common in the United States; it is the sixth most common cancer site, with more than 83,000 new cases and around 17,000 deaths from bladder cancer estimated in 2024 [3]. Moreover, more than 4500 cancers arise each year in the ureter and other urinary organs, resulting in more than 1000 deaths [4]. So, a very common and important cancer.

When we talk about metastatic urothelial carcinoma, this is usually a lethal phenotype of disease with a few exceptions. Key risk factors include older age, male sex, smoking, and chronic inflammatory conditions of the bladder resulting from exposure to occupational hazards, environmental chemicals, chronic infections, chronic inflammation because of catheterization—for example, neurogenic bladder with chronic trauma, injury, and catheterization in the bladder—and genetic predisposition [1, 2, 5, 6]. Genetic factors may play a role in more than we thought, so we are doing more and more germline and genetic testing that can help patients and their families, both in terms of the patient benefit but also in terms of cascade testing and cancer prevention in the broader family, for genetic predisposition. Bladder cancer can cause substantial morbidity and is among the most expensive cancers to treat based on the lifetime per-patient costs, due in part to its high recurrence rate and ongoing need for invasive monitoring, surveillance, diagnostic testing, and treatment [7, 8].

Advanced urothelial carcinoma refers to patients who have developed what we call stage IV disease usually, stage III disease in some cases, which includes unresectable locally advanced or metastatic urothelial carcinoma and, as I mentioned before, has a relatively poor prognosis [1,2,3]. For example, historically, what we were quoting for years, the 5-year overall survival rate for metastatic urothelial carcinoma in the US has been less than 10% [3]. We’re trying to change this with the research going on, and we’ll discuss today some of those research results. Survival rates are likely to be very similar for metastatic urothelial carcinoma arising from nonbladder sites like the upper tract or urethra. I would say we treat metastatic urothelial cancer usually the same regardless of where it started, the organ of origin, bladder, upper tract, or urethra. Factors associated with shorter survival in older studies of platinum-based chemotherapy include poor performance status, medical comorbidities, and organ dysfunction, visceral metastases—for example, if you have metastasis to the lung or the liver—and of course, bone metastasis is also a poor prognostic factor [2, 9, 10].

Let me ask, Dr Moon—in your opinion, what are the recommended first-line therapy options for advanced urothelial carcinoma today?

HM: Well, that was a really masterful review of the background, and I will comment on that piece you said that, historically, we always considered [urothelial carcinoma] a highly lethal cancer with very poor survival because, as you well know, for decades treatment has been essentially the same, right—platinum-based therapy. However, in recent years, several treatments [have been] approved for patients with advanced urothelial carcinoma, really enabling individualization of strategies that’s tailored to each patient’s needs. In National Comprehensive Cancer Network (NCCN) and European Society for Medical Oncology (ESMO) guidelines, and the version we’re quoting from was updated in 2024, the preferred first-line option is combination therapy with an ADC, an antibody-drug conjugate, enfortumab [vedotin] (EV) and an immune checkpoint inhibitor, pembrolizumab [2, 11]. Alternatively, there’s a recommendation for first-line therapy that’s built off the efficacy of the decade-long experience with platinum-based chemotherapy [2, 11, 12]. This includes adding avelumab, which is an immune checkpoint inhibitor, which is given as a switch-maintenance treatment for cisplatin-eligible or -ineligible patients who are progression free after platinum-based chemotherapy [2, 11]. And this particular approach of switch-maintenance avelumab has been approved in this particular indication since June 2020 in the United States [13]. Alternatively, cisplatin-eligible patients may receive the [immune checkpoint] inhibitor nivolumab in combination with cisplatin and gemcitabine, and that is also followed by nivolumab monotherapy in a maintenance-like approach; this option was approved in the United States in March of 2024 [2, 11, 14]. Each of these treatment options was approved based on really high-quality phase 3 trials that showed a significant overall survival benefit compared to conventional platinum-based chemotherapy alone. Particularly, EV-302 [which] showed that [benefit] for enfortumab [vedotin] plus pembrolizumab [15], and JAVELIN Bladder 100 that showed significant benefit from avelumab first-line maintenance [16, 17]. The nivolumab plus gemcitabine and cisplatin [regimen] was studied and published in CheckMate 901 [18].

For avelumab first-line maintenance, several real-world studies have also been reported from different countries that really form the body of literature that can confirm and extend these clinical trial findings in a more heterogeneous population [19,20,21,22,23,24,25,26,27,28]. Previous studies in large US-based populations demonstrated real-world outcomes with platinum-based chemotherapy alone, with median overall survival of 19–20 months seen with [cisplatin-based] chemotherapy and approximately 10–11 months with carboplatin-based chemotherapy in cisplatin-ineligible patients [29, 30]. So, for the rest of this podcast, we’ll really focus our discussion on findings of the real-world studies of avelumab maintenance. But first, I want to start with an overview of JAVELIN Bladder 100, so a reminder for our audience that is the phase 3 study that got this particular approach FDA approved. Then we will follow with a detailed discussion of the real-world studies.

So, I think no one in the world is better than you to ask, because I believe you’re the senior author on this trial, right? What are the key findings of the JAVELIN Bladder 100 trial?

PG: Thank you, Dr Moon. This goes back in time, almost 5 years ago, in 2020, when the pandemic had started; we had the American Society of Clinical Oncology (ASCO) plenary session with Professor Powles and other team members, and we presented for the first time the practice-changing data from the JAVELIN Bladder 100 trial. This was, as you mentioned, a randomized phase 3 trial that enrolled approximately 700 patients with unresectable, locally advanced or metastatic urothelial carcinoma who were progression free after four, five, or six cycles of first-line induction cisplatin-gemcitabine or carboplatin-gemcitabine [16]. After a treatment-free interval between 4 and 10 weeks, between the end of chemotherapy and randomization, patients were randomized 1:1 to receive avelumab (anti-programmed death ligand 1 [PD-L1]) plus best supportive care or best supportive care alone [16], which was what we were doing at the time, after the completion of chemotherapy; we were doing active surveillance for those patients before the results of JAVELIN Bladder 100.

After at least 2 years of follow-up in all patients, the median overall survival from randomization was approximately 2 years—it was 23.8 months—with avelumab plus best supportive care versus 15 months with best supportive care alone, and the median progression-free survival (PFS) also favored the avelumab arm, 5.5 versus 2.1 months' median PFS, respectively [17]. In a post hoc, exploratory, descriptive analysis, median overall survival measured from the start of first-line chemotherapy was 29.7 months with avelumab plus best supportive care and 20.5 months with best supportive care alone [31, 32]. Again, this was an exploratory post hoc analysis and has to be interpreted with caution because the trial enrolled a selected patient population without progression after first-line chemotherapy, similar to real-world studies of avelumab first-line maintenance that we will discuss in a second, which means that time on chemotherapy prior to randomization can introduce an immortal time bias because patients have to be alive to get into the avelumab maintenance phase. Immortal time refers to time included in the observation period when the outcome of interest, for example, death, when you talk about overall survival, could not have occurred [33], which here relates again to overall survival after chemotherapy was received but before maintenance treatment was started. Because, as I mentioned, the trial selected patients on enrollment who were progression free after first-line chemotherapy. It is also important to note that very few patients received subsequent treatment with antibody-drug conjugates, such enfortumab vedotin, or other targeted therapies, like erdafitinib [34], due to the time period where the JAVELIN Bladder 100 trial was enrolled. Just to remind the audience, the trial was presented for the first time in June 2020, so it was kind of the earlier days, before many of the new, recently approved agents were available, when the trial was conducted. In the JAVELIN Bladder 100 trial, efficacy benefits were observed irrespective of characteristics of platinum-based chemotherapy, including which platinum agent, cisplatin or carboplatin, was used; number of treatment cycles; achievement of complete or partial response versus stable disease, which is a stratification factor, by the way; or the time interval between the end of chemotherapy and the start of the switch [avelumab] maintenance therapy [16, 17, 32, 35, 36]. Efficacy benefits were also observed irrespective of patient characteristics such as older age or high body mass index (BMI) [17, 35, 37, 38]. Longer survival was observed in patients with nonvisceral metastases or lymph node-only disease. Those subsets also appear to have benefit with avelumab switch maintenance [39].

In safety analyses, the long-term safety profile of avelumab first-line maintenance was confirmed [17, 40]. After at least 2 years of follow-up in all patients, treatment-related adverse events of any grade had occurred in 78% of patients and were grade 3 or higher in 19% of patients [17, 40]. Safety findings were consistent across all patient subgroups analyzed, and there was no new safety signal observed [36,37,38]. Analysis of patient-reported outcomes showed that health-related quality of life was relatively stable during avelumab treatment [41, 42]. In addition, a post hoc analysis from the JAVELIN Bladder 100 trial showed that a quality-adjusted time without cancer symptoms or toxicity, what we call Q-TWiST, which is a measure of “quality survival” that incorporates efficacy, toxicity, and the patient-reported outcomes, was 22% longer with avelumab plus best supportive care versus best supportive care alone [43].

Overall, these trial findings provided the first data supporting the use of first-line platinum-based chemotherapy followed by avelumab first-line maintenance as a switch-maintenance approach in patients without progression on first-line induction chemotherapy for patients with advanced urothelial carcinoma.

So, Dr Moon, based on this context of the JAVELIN Bladder 100 trial, can you tell us why it’s important, in addition to the prospective clinical trials, to also perform real-world studies in general? And what are the limitations, since you have done a lot of work in that setting?

HM: Yes, as we’ve discussed before, I’m really quite passionate about real-world studies, and to define that, what we mean is observational, retrospective studies, not interventional studies. Because I find that it really helps complement findings from interventional or randomized clinical control trials and assess the impact of these new treatments in relevant patient populations [44, 45]. These kinds of studies, real-world studies, can confirm the clinical effectiveness and provide an expanded safety assessment in heterogeneous populations, which can include patients who would not have been eligible for clinical trial enrollment [44]. Frequently, we talk about how they, in the United States, a very small percentage—something like 1–3% of our oncological patients—are actually ever enrolled on a clinical trial [recently updated estimate, 7% [46]], which means that the clinical trial patients are a very unique and possibly healthier population; so, [for] drugs as they’re approved—does that benefit extend to the real world? And specifically, [does that benefit extend to] those [patients] with poor performance status or with other medical comorbidities [44]? Real-world studies can also generate data on treatment patterns and sequencing. Once something gets approved, how is it used? Is it first line? Is it second line? And how are clinicians in the United States thinking about these options? It can also provide insights into whether treatment use is consistent with the guidelines that we provide, and provide data on any cost and healthcare resource utilization such as hospitalizations and emergency room department [visits], which not only impacts the bigger system but also really impacts the patient experience [44, 45]. So, a patient who is alive but, alive, hospitalized, and going to the emergency [room] every single week would have a very different quality [of life] than someone who’s at home and able to maintain their activities of daily living.

Having said all of this—very positive things about real-world studies—there are obviously inherent limitations [44, 45, 47]. Number one, real-world studies, by definition, have no randomization and have no blinding. It limits the potential for whether or not what you’re observing as an outcome, positive or negative, is due to the actual treatment you have in mind that you’re looking at or other confounding factors. Is it suburban distance to the emergency room, other health and medical issues, financial toxicity, and so on? And specifically, real-world populations really have greater potential for sampling and selection bias and confounding bias, which is when these outcomes are affected by factors other than the treatment that you’re looking at. And data quality can also sometimes be lower because of missing or unknown patient data due to poor compliance with follow-up or just, in general, incomplete data collection, in addition to, sometimes, recall bias for retrospective studies. We all know adverse events are often under-reported in real-world studies compared to clinical trials, so potentially we miss those type of signals. Other limitations in some real-world studies may include a very small population size, sometimes the follow-up duration is fairly limited, and there’s a lack of patient-reported outcome[s], [and] the analysis of data is not really collected for research purposes, to kind of satisfy that level of clarity, which may result in misclassifications of treatment line [44, 45, 47]. So, there’s a lot of problems potentially with [real-world studies]. However, even noting all of that, real-world studies can expand, complement, and validate the findings of clinical trials, and treatment decisions should be based on the totality of all the data that we can gather. The more we know, the more we know, right? The summary of the benefits and limitations of real-world studies are also provided in the infographic that is attached to this podcast, which graphically shows you what we are talking about.

So, Dr Grivas, can you tell us what has been reported in the real-world studies of avelumab first-line maintenance in a United States-based population?

PG: Thanks, Dr Moon. Thank you for doing such a comprehensive job outlining the characteristics and the importance and limitations of real-world studies. And on that note, several real-world studies have examined avelumab first-line maintenance following first-line platinum-based chemotherapy in US populations [20,21,22]. Additional details for all the studies to be discussed are provided in the infographic that you mentioned, and we encourage the audience to see that infographic; it’s very helpful.

PATRIOT-II was a retrospective, multicenter chart review study that included 160 patients who received avelumab first-line maintenance at 37 geographically dispersed sites, including community oncology practices and academic centers [20]. I had the honor to lead that study, along with many colleagues—and Dr Moon, you were a big part of this study and thank you for your contributions for the really important work you did. Eight percent of patients had Eastern Cooperative Oncology Group performance status (ECOG PS) 2 or higher, and first-line chemotherapy included cisplatin in about 60% of patients and carboplatin in about 40% of patients. So, slightly more than half of patients had cisplatin as part of chemotherapy [in the] induction phase. Median overall survival was approximately 2 years—24.4 months—and median PFS was 5.4 months, both measured from the start of avelumab treatment [20]. So, I would argue in line with what we saw with the JAVELIN Bladder 100 trial [17]. Median overall survival from the start of first-line chemotherapy—again, exploratory, observational endpoint here in the selected population who had response or stable disease with first-line chemotherapy before avelumab started—was 30.5 months [20], again, in line with what we saw in the exploratory analysis of the JAVELIN Bladder 100 trial [31]. Treatment-related adverse events of any grade were reported in 39% of patients, including immune-related adverse events in 22% of patients [