Abstract
BACKGROUND: Given the limited treatment options recommended for cancer of unknown primary (CUP), especially the role of immune checkpoint inhibitors (ICIs), our study aimed to evaluate the efficacy of ICIs and identify associated genomic biomarkers in these patients. METHODS: This retrospective, multicenter, real-world analysis included individuals with oncologist-confirmed CUP cases treated with ICIs across four hospitals in China. Clinical outcomes, safety and biomarkers were analyzed. RESULTS: Between January 2016 and November 2023, 124 patients were enrolled. Of these, 117 patients underwent combination therapy, predominantly ICIs with taxane-platinum based chemotherapy (54.84%), while 7 received monotherapy. After a median follow-up of 18.6 months, the median progression free survival (PFS) and overall survival (OS) were 23.20 and 38.86 months, respectively. According to ESMO guideline, patients were stratified into favorable (n = 41) and unfavorable subset (n = 83). The favorable subset demonstrated significantly longer PFS and OS than the unfavorable subset (median PFS NR vs. 9.7 months, P < 0.001; median OS NR vs. 23.73 months, P < 0.001). There are 31 patients in our cohort whose PD-L1 detection results are available, while 25 patients was eligible for TMB assessment. Better clinical efficacy of ICIs was apparent for tumors with a higher PD-L1 expression (CPS ≥ 20) and a greater tumor mutation burden (> 12 mutations/Mb). Multivariate analyses revealed that higher ECOG performance status, the presence of visceral metastasis, and KRAS mutation were independently associated with inferior PFS and OS. Immune-related adverse events occurred in 49 (39.52%) patients, with one developing grade 3 pneumonia. CONCLUSIONS: The application of ICIs showed encouraging efficacy with an acceptable safety profile, suggesting its potential as an additional therapeutic option for CUP patients. Identifying predictive markers for ICIs response remains essential to enhance therapeutic strategies in CUP management.