Frontline Treatment for Metastatic Urothelial Carcinoma

Petros Grivas, MD, PhD: Let’s focus on metastatic disease. I’ll start with first-line therapy. What are the first-line options in patients who are eligible for first-line treatment? Obviously in metastatic disease, there’s no curative intent. But do you use chemotherapy? Do you use immunotherapy? I will make a quick parenthesis to relate this to our prior discussion. The role of PD-L1 testing isn’t clear, especially with localized disease. We don’t use PD-L1 at all for localized disease. Do you use it for metastatic disease? Does it help you with the selection of first-line therapy in the metastatic setting?

Neeraj Agarwal, MD: That’s a great question, and I’d also like your opinion on this. We’re driven by ongoing trials, especially the molecularly targeted trials. We have several open at Huntsman Cancer Institute as well. It’s a standard of care for us to obtain comprehensive genomic profiling of the bladder tumors when we see patients for the first time. When we see them in the localized setting, we obtain the genomic profile, which includes PD-L1 testing results. When we see the patients for the first time in metastatic setting, we obtain the genomic profile.

Although PD-L1 testing is a reflex test in our institution by our pathologists, they do it because of the FDA mandate that came a couple of years ago that high levels are mandatory before we can order single-agent PD-1/PD-L1 inhibitors. It’s done on a reflex basis. More common is the increasing use of comprehensive genomic profiling for all our patients with all solid tumors, which has made this decision much easier. We get the results whether we’re looking for it or not. That’s the answer to your first question.

In regard to first-line therapy, I don’t think it helps [with selection], despite all the recent advances in the second- and third-line settings. The first-line therapy space has remained pretty stagnant over the last 5 or 6 years. Other than the results of the IMvigor210 and KEYNOTE-052 trials bringing atezolizumab and pembrolizumab for our patients who aren’t eligible for platinum-based chemotherapy, the first-line space is the same as it was 10 years ago. Decision-making is pretty easy in this space. Patients who don’t match any of the following criteria—ECOG performance status of 2 or higher, creatinine clearance of less than 60 mL/min, grade 2 neuropathy or hearing loss or higher, and class 3 heart failure—are eligible for cisplatin. I would say that 50% of my patients don’t have any of these cisplatin-unfit criteria.

Of the other 50% of patients, 40% still remain eligible for carboplatin-based chemotherapy. I really want to see who’s ineligible for any platinum. Obviously, we don’t have defined criteria that are validated and accepted commonly. The remaining 10% will not be eligible for any platinum, either cisplatin or carboplatin. What’s your experience?

Petros Grivas, MD, PhD: Neeraj, you covered it so well. I completely agree with you. The criteria to assess cisplatin fitness is the Galsky criteria published in the Journal of Clinical Oncology in 2011, based on consensus agreement of different medical oncologists who treat patients with urothelial cancer. You covered them well. Age is not 1 of those criteria. It’s mainly patients with higher ECOG performance status in whom it’s very difficult to give cisplatin.

With creatinine clearance, it’s interesting to see the variability across different providers. Some people feel they are comfortable to give cisplatin with a creatinine of 15 mL/min or higher, and many of us may do that. But the classical Galsky criteria talk about 60 mL/min or higher to be fit for cisplatin. That’s a little variable. Hearing loss grade 2 or higher don’t match the criteria, although some patients may still want to get cisplatin with grade 2 hearing loss. Obviously, neuropathy grade 2 or higher is a no-go for me for cisplatin, along with severe heart failure, as you mentioned.

Based on these criteria, I agree that about half the patients I’ve seen in practice are fit for cisplatin. An additional 40% or so in my practice are carboplatin-only. The so-called platinum-unfit, meaning unfit for cisplatin and carboplatin, is about 10%. In community practices, this percentage may be higher. It may be up to 20%, because some of those patients may not end up coming to see us in academic centers. Overall, 10% to 20% of patients are not fit for any platinum agents, cisplatin or carboplatin. For these 10% to 20% of patients, in the United States, we have the option of giving a single-agent checkpoint inhibitor, either atezolizumab or pembrolizumab, regardless of PD-L1 expression. That’s a difference compared with Europe, where PD-L1 expression is needed and you have the dilemma or discussion of how to utilize checkpoint inhibitors in the first-line setting. In Europe, PD-L1 testing is required even for platinum-unfit patients.

The other point I will make is because we try to improve on outcomes, we have many clinical trials looking at combinations of checkpoint inhibitors plus something else. However, the vast majority of patients in my practice—the 90% who are fit for cisplatin or carboplatin—end up getting chemotherapy in the frontline setting and not a checkpoint inhibitor. This is based on the JAVELIN trial for avelumab that we published in September 2020 that saw the switch maintenance avelumab immunotherapy significantly prolong overall survival and progression-free survival after response to stable disease with platinum-based chemotherapy. This study isn’t a direct comparison with a checkpoint inhibitor by itself. But it has shown us the best data we’ve seen in the first-line setting.

To make it simple for the audience, if someone in my practice is fit for cisplatin chemotherapy, I try to give cisplatin-based chemotherapy, like gemcitabine-cisplatin followed by avelumab maintenance. If they’re not fit for cisplatin but are fit for carboplatin, I use gemcitabine-carboplatin followed by avelumab maintenance for response of stable disease. For those who aren’t fit for any platinum, I use a checkpoint inhibitor, atezolizumab or pembrolizumab. Obviously, clinical trials come into play. Do you have the same approach?

Neeraj Agarwal, MD: Absolutely. The JAVELIN 100 trial data changed my practice from this perspective. We used to perform PD-L1 testing specifically to look for who may be eligible for single-agent therapy with atezolizumab or pembrolizumab if they were not cisplatin eligible. With the JAVELIN trial data coming up, even though all carboplatin-based chemotherapy regimens have lower overall survival associated with them, compared with cisplatin-based chemotherapy regimens. I know that I have a checkpoint inhibitor available to be used for both of these categories of patients.

Taking a step back, I still have no reason to believe we should avoid chemotherapy in bladder cancer, with this being such a chemotherapy-sensitive disease. I try my best to utilize chemotherapy in patients with bladder cancer as much as I can. If they are eligible, use chemotherapy. Then we have avelumab available based on JAVELIN 100 data. That’s my approach.

Transcript edited for clarity.