Shilpa Gupta, MD: Let’s switch gears to adjuvant therapy. I’ll pose this question to Andrea. What are the key adjuvant options you offer patients? You’re leading the AMBASSADOR trial, which has completed accrual. Congratulations. Walk us through your standard practice. Are you offering pembrolizumab to everyone? What are you doing for patients with upper-tract disease? Give us an overview.

Andrea B. Apolo, MD: This is an important topic. Adjuvant chemotherapy has been used for years and has shown an improvement in progression-free survival [PFS]. There are important data showing that giving chemotherapy immediately after surgery improves outcomes, although the trial didn’t fully accrue in order to give the overall survival benefit. We see a benefit when patients are treated earlier with systemic therapy. If the patient already received neoadjuvant therapy, then we can talk about immunotherapy. Three trials have asked this question. With CheckMate 274, we saw the updates at ASCO GU [American Society of Clinical Oncology Genitourinary Cancers Symposium] 2023. We saw the prolonged 3-year follow-up of CheckMate 274 showing a persistent improvement in disease-free survival in the patients receiving adjuvant nivolumab vs those receiving placebo. We also have negative trials from the IMvigor010 study with adjuvant atezolizumab vs observation, which didn’t show an improvement in PFS and overall survival.

We completed accrual of the AMBASSADOR study, which is adjuvant pembrolizumab vs observation, and we’re awaiting the data. I discuss adjuvant chemotherapy with patients if they haven’t received chemotherapy. If they’ve already received chemotherapy, then I discuss adjuvant checkpoint inhibitor with nivolumab and discuss the PFS data. If we still don’t have survival data, but the PFS data are positive and show improved outcomes in patients in terms of disease-free survival in the adjuvant setting, that’s something I’ll discuss with patients and offer them.

In terms of the upper-tract disease, that’s a little harder. The reason I say that is we don’t want to base things on subgroup analysis. That’s what we have for the adjuvant checkpoint. In the subgroup analysis of patients with upper-tract disease, it looked as if they didn’t benefit as much as the patients with primary bladder cancer. But this is a subgroup analysis, and I don’t like to make decisions based on that. If a patient has already received chemotherapy, then I also discuss adjuvant nivolumab; if they haven’t, then I discuss adjuvant chemotherapy. We have prospective data through the POUT study that this improves outcomes in our patients.

Shilpa Gupta, MD: In patients who are cisplatin ineligible, the POUT study showed the benefit of gemcitabine-carboplatin, which was less than gemcitabine-cisplatin. Both IMvigor010 and CheckMate 274 didn’t show benefit of immunotherapy. But to your point, the population was capped, and it’s a subgroup analysis. Do you have a preference of gemcitabine-carboplatin vs nivolumab in those patients?

Andrea B. Apolo, MD: That’s a great question. I don’t tend to use gemcitabine-carboplatin. This was a subgroup analysis in the POUT study, where the majority of the benefit was in patients who received a cisplatin-based combination. A subgroup received carboplatin. It looked like there was also a benefit with them, but it was a subgroup analysis and the confidence interval crossed 1. It’s hard to interpret those data, but in practice we haven’t seen the pathologic response rates in the neoadjuvant setting in the trials that have been done. It’s not a practice that I do in the adjuvant setting using carboplatin-based chemotherapy, so at this point I’d offer a checkpoint inhibitor.

Shilpa Gupta, MD: Petros, what are you doing for patients with upper-tract disease in the adjuvant setting?

Petros Grivas, MD, PhD: Great question. I enjoyed all the descriptions Andrea made on this topic. We see these patients in the clinic, and it’s such a hard question. I saw a patient recently, and we had a long discussion regarding the nuances and details. For patients with upper tract-disease, there’s been a huge effort from the corporate groups. The neoadjuvant trial ECOG-ACRIN EA8192 is a phase 3 trial that we’re doing with Jean Hoffman-Censits. She’s a chair of the trial; Vitaly Margulis and I are the co-chairs. Dr Hoffman-Censits is from Johns Hopkins University. This is a neoadjuvant trial with dose-dense MVAC [methotrexate, vinblastine sulfate, doxorubicin hydrochloride, cisplatin] plus or minus durvalumab. It's a phase 3 design. There’s a cohort of 29 patients who are cisplatin ineligible; they get gemcitabine-durvalumab.

I’m going to let the others know that this trial exists. It’s very challenging, and we need dedicated trials in upper-tract disease. That’s a cooperative group effort, and we try to accrue patients. That’s something I think about in the neoadjuvant setting in both cisplatin fit and unfit patients. In patients who come to my clinic after radical nephroureterectomy and have never seen neoadjuvant chemotherapy—a very common scenario—if they are fit for cisplatin after nephroureterectomy, I offer gemcitabine-cisplatin in the adjuvant setting based on the POUT phase 3 trial for disease-free survival and metastases-free survival benefit. The trial was closed because it met the primary end point. It was not powered for overall survival. However, gemcitabine-cisplatin is my go-to for chemotherapy-based patients who have pT2 or T3 or T4 disease and don’t have positive disease in the adjuvant setting.

The big challenge we encounter has to do with cisplatin-ineligible unfit patients. In the absence of a clinical trial—Dr Apolo is leading the amazing effort with the AMBASSADOR trial—we used to accrue patients well; we were 1 of the top accruers. But the PROOF 302 trial closed recently. In the absence of an adjuvant trial, we have a discussion about the uncertainty. The uncertainty goes to carboplatin-gemcitabine. There was a subset of 96 patients in the POUT study. The first block is hard to interpret because it was underpowered with a low number of events. It showed a trend toward disease-free survival benefits with carboplatin-gemcitabine. Because of the small number, it had a confidence interval of close to 1. The interaction tests, as claimed by Dr Alison Birtle, showed that the benefit was there regardless of gemcitabine-cisplatin or gemcitabine-carboplatin. I don’t think we know the answer. There’s huge uncertainty there. I discuss it with the patient, and I try to simplify the uncertainty in layman’s terms.

The same uncertainty applies with nivolumab adjuvantly. For risk clot cannot be overinterpreted. That’s a usual challenge because of small numbers. When we look for the ... of the CheckMate 274 trial, the hazard ratio for the patients with upper-tract disease adjuvantly with nivolumab was more than 1. I can’t overinterpret it. I discuss it with the patient. I have a dialogue about the pros and cons and informed decision-making about carboplatin-gemcitabine, nivolumab, or active surveillance. I help them decide based on their needs, goals, expectations, and personalities in the context of available data and the uncertainty with cisplatin-ineligible patients adjuvantly in upper-tract disease.

Shilpa Gupta, MD: Thank you, Petros. It’s important to enroll in trials, and the ECOG-ACRIN trial will certainly address some of these unmet needs.

Transcript edited for clarity.