Efficacy and biomarker analysis for nivolumab in cancer of unknown primary: results from an investigator-initiated expanded access program

Abstract Background Cancer of unknown primary (CUP) remains a challenging malignancy with a poor prognosis. Although immune checkpoint inhibitors have shown activity in CUP, available data remains limited. We describe clinical outcomes and a comprehensive biomarker analysis from an expanded access trial of nivolumab in CUP. Methods Patients with CUP were enrolled regardless of treatment history and received intravenous nivolumab (240 mg every 2 weeks or 480 mg every 4 weeks) until disease progression, unacceptable toxicity, or withdrawal. Enrollment concluded after regulatory approval of nivolumab for CUP in Japan. The primary objective was safety; secondary endpoints addressed efficacy, and exploratory biomarker analyses were performed. Results A total of 53 patients (31 previously treated, 22 untreated) received nivolumab. Adverse events (AEs) occurred in 77.4% of patients, including 24.5% grade 3–4 and 15.1% serious AEs; no treatment-related deaths occurred. The objective response rate was 17.0% (95% confidence interval [CI] 8.1–29.8%) overall and 20.5% (95% CI 9.8–35.3%) in the response-evaluable subset. Median progression-free survival was 4.5 months (95% CI 2.9–7.9) and median overall survival was 17.5 months (95% CI 11.9–25.1). Higher PD-L1 expression was associated with better efficacy, whereas no meaningful differences were observed across estimated tissue-of-origin subgroups. Conclusions This study provides additional prospective evidence consistent with the preceding NivoCUP trial, supporting the reproducibility of the efficacy and safety of nivolumab for patients with CUP. These findings support PD-1-targeted therapy as a feasible treatment option for this rare malignancy. Trial registration Japan Registry of Clinical Trials (jRCT) identifier, jRCT2051200146.