Cancer Sci. 2026 Jul 6. doi: 10.1111/cas.70463. Online ahead of print.
ABSTRACT
Comprehensive genomic profiling (CGP) tests can identify putative biomarkers for immune checkpoint inhibitors (ICIs), including tumor mutational burden (TMB); however, their clinical utility remains uncertain. In this study, the efficacy and safety of nivolumab treatment based on the CGP results were investigated using a subcohort from the BELIEVE trial (NCCH1901; jRCTs031190104). This trial aims to improve drug accessibility for patients with advanced solid tumors harboring actionable genetic variants through off-label drug administration. Patients for whom off-label nivolumab treatment was proposed based on CGP results were eligible. Nivolumab was administered until disease progression or unacceptable toxicity. The primary endpoint was the best objective response rate (ORR) in patients with measurable disease. Secondary endpoints included progression-free survival (PFS), overall survival (OS), disease control rate (DCR), and safety. Among 60 enrolled patients, 56 patients who received study treatment were included in the analysis. TMB-High was the most frequent biomarker prompting nivolumab treatment (47/56, 83.9%), followed by CD274 (PD-L1) amplification (6/56, 10.7%) and CDK12 inactivating variants (3/56, 5.4%). Among 50 patients with measurable disease, the ORR was 16.0% (95% confidence interval [CI], 7.2-29.1) and the DCR was 40.0% (95% CI, 26.4-54.8). Median PFS and OS were 2.7 months (95% CI, 2.2-4.3) and 7.2 months (95% CI, 4.4-9.0), respectively. Although a subset of patients achieved durable responses, nivolumab treatment guided by CGP results demonstrated limited efficacy. The predictive value of TMB-High may differ across tumor types, and integration with other clinicogenomic factors is warranted to improve its predictive accuracy.
PMID:42410894 | DOI:10.1111/cas.70463