Development and validation of a nomogram for predicting overall survival of patients with cancer of unknown primary: a real-world data analysis

Background: Cancer of unknown primary (CUP) has a variable prognosis and lacks any standard staging systems. We aim to improve the prediction of survival in patients with CUP by constructing a nomogram based on a real-world, population analysis. Methods: We performed a population analysis of patients diagnosed with CUP between 2010 and 2016 in the Surveillance, Epidemiology, and End Results (SEER) database. Patients with complete study variables were respectively assigned to training and validation cohorts by diagnostic time. A prognostic nomogram was established based on the multivariate Cox proportional hazards model and was evaluated through calculating the Harrell's C-index and plotting calibration curves. Results: In total, 19,543 patients were identified under the selection criteria, and 3,347 cases with complete study variables were included for developing and validating the nomogram. Covariates incorporated in the final nomogram were sex, age, histological type, surgery, radiotherapy, chemotherapy, and the number of metastatic organs. The Harrell's C-index of nomogram was 0.705 (95% CI: 0.692-0.717) for the training cohort and 0.727 (95% CI: 0.703-0.752) for the validation cohort. Conclusions: We developed and validated the first nomogram based on a large population, which showed good prediction ability for predicting overall survival of patients with CUP. The risk stratification based on this nomogram could also help clinicians in treatment planning. This nomogram requires further validation in external cohorts, since important clinical factors such as favorable/unfavorable subset, performance status, lactate dehydrogenase, blood cell counts, or metastatic patterns limited to multiple lymph nodes could not be considered due to the lack of availability of these data.