Abstract
BACKGROUND: Patients diagnosed with cancer of unknown primary (CUP) have limited treatment options, resulting in a dismal prognosis. The efficacy of immune checkpoint inhibitors (ICIs) in these patients is still unclear. METHODS: We collected the clinical data of these patients retrospectively and multicentrically. In total, 190 patients diagnosed with CUP in 6 hospitals were collected, of which 58 patients received immunotherapy. RESULTS: The median overall survival (OS) was 17.3 months. The OS of the ICIs-treated group was significantly longer than that of the non-ICIs group in unfavorable subgroup, 29.27 months and 10.43 months, respectively (HR = 0.435, 95% CI 0.267–0.709; p = 0.0006). In the first-line systemic treatment, the OS and progression-free survival (PFS) in the ICIs group were significantly longer than that of the non-ICIs group. The OS was 45.53 months versus 12.03 months (HR = 0.431, 95% CI 0.259–0.717; p = 0.004), and the PFS was 11.33 months versus 5.43 months (HR = 0.558, 95% CI 0.351–0.889; p = 0.021). The objective response rate (ORR) and disease control rate (DCR) of ICIs group were higher too. Finally, a predictive nomogram model was developed to accurately forecast the response of patients with CUP to ICIs. CONCLUSIONS: In this cohort of unfavorable subgroup, immunotherapy appears to extend the survival of CUP patients who received chemotherapy or systemic therapy. One model could predict the time to recurrence following immunotherapy in the cohort. TRIAL REGISTRATION: Retrospective registration of local ethics committees. SUPPLEMENTARY INFORMATION: The online version contains supplementary material available at 10.1186/s12885-025-14778-6.