High-Grade Urothelial Carcinoma With Deep Muscle Invasion

Matt Galsky, MD, Tisch Cancer Institute at Mount Sinai, New York, NY, discusses the course of treatment he would take for a patient with high-grade urothelial carcinoma with deep muscle invasion.

Transcript:

Hi, I'm Matt Galsky from the Icahn School of Medicine at Mount Sinai. I'm a medical oncologist, professor of medicine, and I focus on the treatment of patients with bladder cancer.

This is a case of a man aged 68 years with a 30-year smoking history who presents with gross hematuria. He undergoes a CAT scan of the chest, abdomen, and pelvis, which reveals a bladder mass, and he's referred to a urologist. The urologist pursues a workup including a cystoscopy and a TURBT. The TURBT reveals muscle-invasive urothelial cancer. His creatinine is 1.1. He has no history of hearing loss, no history of neuropathy, and he's recommended to undergo treatment with neoadjuvant cisplatin-based chemotherapy.

He receives 4 cycles of gemcitabine plus cisplatin and tolerates treatment reasonably well, develops grade 1 sensory peripheral neuropathy on treatment. And he subsequently undergoes a radical cystectomy. Surgical pathology reveals residual T3N0 urothelial cancer. PD-L1 expression is performed with the 22C3 antibody clone and the CPS, the combined positive score is 20 and he undergoes circulating tumor DNA testing using the Signatera assay to assess for molecular residual disease, and ctDNA testing is undetectable.

What should we do for this individual? Should we recommend treatment with adjuvant gemcitabine plus carboplatin? Should we recommend adjuvant radiation therapy, adjuvant atezolizumab or adjuvant nivolumab?

Adjuvant chemotherapy, this is a patient who received cisplatin-based chemotherapy and had residual cancer in the surgical specimen. Pathological T3 disease, we know that is associated with a reasonably high risk of metastatic recurrence. And no randomized clinical trials have assessed non-cross resistant cytotoxic chemotherapy in patients who've received neoadjuvant chemotherapy. There have been a few retrospective studies, but no prospective studies. Therefore, giving this patient adjuvant gemcitabine-carboplatin after residual disease with neoadjuvant gemcitabine plus cisplatin wouldn't be the right approach.

Adjuvant radiation is something that has been evaluated in a series of trials in the past, and there is an NCCN guideline suggesting that adjuvant radiation can play a role in some patients. We know that individuals with bladder cancer can be at risk for local recurrence in the pelvis in addition to distant recurrence, but this isn't a treatment strategy that's routinely employed and for this patient, there's higher level evidence that would suggest other treatments might be beneficial.

How about adjuvant atezolizumab, the PD-L1 inhibitor? We know that based on the IMvigor010 study that randomized patients who had received neoadjuvant chemotherapy and had residual T3 or higher disease like this patient or who didn't receive neoadjuvant chemotherapy and where cisplatin-ineligible patients were randomized to adjuvant atezolizumab versus observation. And we know that study did not show a benefit in terms of disease-free survival or overall survival, so adjuvant atezolizumab would not be the right approach for this patient.

Then, there's adjuvant nivolumab. We know based on the CheckMate 274 study, enrolling a similar patient population to what I just described for the IMvigor010 study by randomizing patients to adjuvant nivolumab versus placebo. We know that that study demonstrated a significant improvement in disease-free survival in the all-comer population and in the subset of patients with tumors with high levels of PD-L1-expression leading to FDA approval of adjuvant nivolumab. Adjuvant nivolumab would be appropriate for this patient.

Does the PD-L1 testing sway us? Does the ctDNA testing sway us? Well, the PD-L1 testing that was done in the setting of the CheckMate 274 study was using the 28-8 antibody clone and assessing scoring of PD-L1 on tumor cells. Not completely the same as the 22C3 antibody clone and CPS testing that's commonly reported in pathology labs. That said, there is a publication showing that when using CPS scoring in the context of the CheckMate 274 cohort, there was also an enrichment in the effect of benefit with adjuvant nivolumab versus placebo in patients with high PD-L1 expression based on CPS scoring.

What about ctDNA? We know from the IMvigor010 study, a retrospective analysis demonstrated that patients without detectable ctDNA did not seem to benefit from adjuvant atezolizumab, whereas those with detectable ctDNA did benefit. The challenge though is that's a retrospective exploratory analysis and we know from CheckMate 274 that there was a benefit demonstrated in all randomized patients. Until we have prospective data demonstrating the clinical utility of ctDNA testing premature to use that to identify patients who should or should not receive adjuvant immune checkpoint blockade.

This is a patient for whom adjuvant immune checkpoint blockade, which is now a standard strategy, should certainly be considered.