麻豆传媒

SAN FRANCISCO -- Almost half of patients with untreated cisplatin-ineligible advanced urothelial carcinoma responded to the combination of pembrolizumab (Keytruda) and cabozantinib (Cabometyx), but that result fell short of prespecified statistical requirements for a positive study.

Results of the phase II single-arm PemCab trial showed 16 of 35 evaluable patients responded to the combination (45.7%), including five complete responses. However, the trial design required at least one more objective response to exclude the prespecified 95% confidence intervals, said Rohit K. Jain, MD, of the Moffitt Cancer Center in Tampa, Florida, at the Genitourinary Cancers Symposium.

He noted that in the phase III trial, treatment with the tyrosine kinase inhibitor (TKI) lenvantinib (Lenvima) plus pembrolizumab did not improve outcomes against pembrolizumab alone in this patient population.

"This could suggest that select TKIs with immune-modulating activity targeting the TAM kinases might have better activity in this disease. Indeed, the phase III trial is evaluating the combination of cabozantinib and maintenance avelumab (Bavencio)," said Jain.

"Further evaluation of cabozantinib and pembrolizumab may be warranted, informed by principles of precision medicine," he added. "Potentially a more tolerable and potent VEGF TKI similar to cabozantinib -- zanzalintinib -- may be an important direction of investigation to improve the therapeutic index."

The PemCab combination's activity was "interesting but not groundbreaking," said invited discussant Parminder Singh, MD, of the Mayo Clinic in Phoenix.

"The combination demonstrated an overall response rate of 45% and 80% of the patients did have tumor shrinkage," said Singh. "However, it did not meet the specified 17 patient responses in the design of the trial. We have to keep this in context, as close to 60% of the patients had fatigue, diarrhea, palmar/plantar erythrodysesthesia [PPE], which we all know can affect the quality of life significantly. That raises the question of how high the bar should be for clinical benefit to overcome the impact on quality of life."

Singh went on to summarize a list of negative clinical trials involving VEGF inhibitors for urothelial carcinoma, including ramucirumab (Cyramza) and bevacizumab (Avastin) paired with chemotherapy. The advent of multikinase inhibitors renewed interest in evaluating TKIs alone and in combination with immuno-oncology agents. However, the aforementioned lenvatinib-pembrolizumab study was negative, now followed by the PemCab trial.

"Can we say that this is the end of VEGF pathway in combination in urothelial carcinoma, or do we need to explore them in a different clinical setting?" Singh continued. "In the evolving paradigm of first-line therapy, their role is questionable."

TKIs in combination with immunotherapy might have a role as maintenance therapy, where the combination of cabozantinib and nivolumab (Opdivo) has been shown to "reinvigorate the immune response" in immuno-oncology-refractory patients. The agents also might be worth exploring in the third-line setting as salvage therapy, possibly informed by predictive biomarkers, Singh concluded.

Patients with cisplatin-ineligible bladder cancer have limited options, as even approved therapies have limitations on their use, Jain noted. Single-agent VEGF receptor TKIs, including cabozantinib, have demonstrated activity in previously treated urothelial carcinoma. Preclinical and clinical studies showed that cabozantinib has "encouraging safety and efficacy" in combination with PD-1 inhibitors. Collectively, the evidence provided a rationale for studying cabozantinib and pembrolizumab in the first-line setting.

involved cisplatin-ineligible patients with a PD-L1 combined positive score ≥10, those who were platinum ineligible regardless of PD-L1 status, and patients who had refused cisplatin-based chemotherapy. They received pembrolizumab and cabozantinib concurrently, and the primary endpoint was overall response rate (ORR).

Statistical considerations were derived from past studies of chemotherapy in metastatic urothelial carcinoma, which produced ORRs of 36-38%, and pembrolizumab, which had an ORR of 26%. With a study population of 35 patients, the lower bound of the 95% confidence intervals would extend no more than 26% from the observed proportion. With 17 or more responses, the confidence interval would exclude 32%.

The study population had a median age of 72.5, and men accounted for three-fourths of the patients. Primary tumor location was in the upper tract in 52.8% of patients and the bladder in 47.2%. Three-fourths of the patients had visceral metastases, and the most common reason for cisplatin ineligibility was related to creatinine clearance (69.4%). Seven patients refused cisplatin-based therapy.

The combination led to an ORR of 45.7%. The 95% confidence intervals for ORR were 30.5%-61.8%, which did not exclude 32% as per the protocol requirements. In addition to the 16 responses, nine patients had stable disease, resulting in a clinical benefit rate of 71.4%.

The median time to response was 2.7 months, and with a median follow-up of 20.2 months, the median duration of response was 14.7 months. Median progression-free survival was 7.6 months, and median overall survival was 17.1 months.

The most common all-grade adverse events were fatigue (63.9%), diarrhea (58.3%), PPE (41.7%), pruritus (38.9%), anorexia (33.3%), and nausea (33.3%). Three patients (8.3%) had thromboembolic events. The most common grade ≥3 AEs were proteinuria (5.6%), anorexia (5.6%), and PPE (5.6%), followed by fatigue, diarrhea, nausea, and oral mucositis (2.8% each).

Comment