Risk factors and novel predictive model for metastatic cutaneous squamous cell carcinoma: a population-based analysis

Cutaneous squamous cell carcinoma (cSCC) is one of the most common skin malignancies. Patients with metastatic cSCC (mcSCC) tended to have unfavorable prognosis. However, there is no available models to evaluate the survival outcomes for these patients. This study retrospectively collected mcSCC cases identified from The Surveillance, Epidemiology, and End Results (SEER) database from 2000 to 2018. All patients were randomly divided into the training and validation cohorts. The independent factors which predicted the cancer-specific survival (CSS) were identified by Cox proportional hazard regression analysis, and were used to build nomogram. he discrimination ability of nomograms was evaluated using the concordance index (C-index). The receiver operating characteristic (ROC) curve and the corresponding area under the curve (AUC) at 1, 3, and 5 years were calculated to estimate the predictive accuracy. The performance of this nomogram was evaluated by the calibration curves. The decision curve analyses (DCAs) were conducted to assess the clinical usefulness of the models. Based on the novel nomograms, all patients were classified into low- and high-risk groups. Kaplan-Meier curves were plotted to compare the survival outcomes of two groups. All analyses were conducted by R software (Version 4.1.3). A total of 916 patients were included. Age, marital status, location of the primary sites, number of metastases, T stage, N stage, surgical resection of the primary sites, radiotherapy of the primary sites, and chemotherapy were significantly associated with CSS. A nomogram predicting CSS was built based on these parameters. C-index and ROC curves of nomogram showed promising discriminating and predicting ability for CSS. Survival analysis showed that patients in the low-risk group had significant superior CSS time than those in the high-risk group (median CSS time: 15.0 vs. 5.0 months, p < 0.001). In conclusion, we developed and validated a novel nomogram predicting the CSS in metastatic cSCC patients. This nomogram could be used to assess the prognosis of metastatic cSCC patients, and help clinicians evaluating optimal treatment options for individual patient.