Survival analysis and nomogram for pulmonary sarcomatoid carcinoma: an SEER analysis and external validation

Uncommon and particularly deadly, pulmonary sarcomatoid carcinoma (PSC) is an aggressive type of lung cancer. This research aimed to create a risk categorisation and nomogram to forecast the overall survival (OS) of patients with PSC. To develop the model, 899 patients with PSC were taken from the Surveillance, Epidemiology, and End Results database from the USA. We also used an exterior verification sample of 34 individuals with PSC from Fujian Provincial Hospital in China. The Cox regression hazards model and stepwise regression analysis were done to screen factors in developing a nomogram. The nomogram's ability to discriminate was measured employing the area under a time-dependent receiver operating characteristic curve (AUC), the concordance index (C-index) and the calibration curve. Decision curve analysis (DCA) and integrated discrimination improvement (IDI) were used to evaluate the nomogram to the tumour-node-metastasis categorisation developed by the American Joint Committee on Cancer (AJCC-TNM), eighth edition, and an additional sample confirmed the nomogram's accuracy. We further developed a risk assessment system based on nomogram scores. Six independent variables, age, sex, primary tumour site, pathological group, tumour-node-metastasis (TNM) clinical stage and therapeutic technique, were chosen to form the nomogram's basis. The nomogram indicated good discriminative ability with the C-index (0.763 in the training cohort and 0.746 in the external validation cohort) and time-dependent AUC. Calibration plots demonstrated high congruence between the prediction model and real-world evidence in both the validation and training cohorts. Nomogram outperformed the AJCC-TNM eighth edition classification in both DCA and IDI. Patients were classified into subgroups according to their risk ratings, and significant differences in OS were observed between them (p<0.001). We conducted a survival analysis and nomogram for PSC. This developed nomogram holds potential to serve as an efficient tool for clinicians in prognostic modelling.