Objective:To develop a 5-year mortality risk prediction model for patients with small cell neuroendocrine carcinoma of the cervix(SCNEC).Methods:Based on the Surveillance,Epidemiology,and End Results(SEER)database and R software version 4.3.3,variables were screened via Lasso regression,followed by multivariable logistic regression and stepwise regression to develop a 5-year mortality risk prediction model for SCNEC patients.The Akaike Information Criterion(AIC),C-index,receiver operating characteristic(ROC)curve,Hosmer-Lemeshow test,and calibration curve were employed to evaluate the model.Results:Age,M stage,surgical status,and lymph node metastasis were ultimately selected as variables to construct the 5-year mortality risk prediction model for SCNEC patients.The model demonstrated superior predictive performance compared to FIGO staging(P＜0.01).The Hosmer-Lemeshow test yielded a P-value＞0.05.The C-index values for the training and validation sets were 0.808 and 0.755,respectively,with the areas under the ROC curves of 0.826 and 0.744.The calibration curves of the model fluctuated near the diagonal line,indicating good agreement between predicted and observed outcomes.The decision curve analysis demonstrated significant clinical net benefit.Results showed that higher mortality risk was associated with advanced age,M1 status,lymph node metastasis,and lack of surgical opportunity.Conclusions:The model exhibits good discriminatory power and accuracy,providing significant benefits to patients.Enhanced management should be implemented for patients with advanced age,distant metastasis,lymph node metastasis,or ineligibility for surgery.Lymph node metastasis is an independent risk factor for 5-year mortality in patients with SCNEC.

Objective:To develop a 5-year mortality risk prediction model for patients with small cell neuroendocrine carcinoma of the cervix(SCNEC).Methods:Based on the Surveillance,Epidemiology,and End Results(SEER)database and R software version 4.3.3,variables were screened via Lasso regression,followed by multivariable logistic regression and stepwise regression to develop a 5-year mortality risk prediction model for SCNEC patients.The Akaike Information Criterion(AIC),C-index,receiver operating characteristic(ROC)curve,Hosmer-Lemeshow test,and calibration curve were employed to evaluate the model.Results:Age,M stage,surgical status,and lymph node metastasis were ultimately selected as variables to construct the 5-year mortality risk prediction model for SCNEC patients.The model demonstrated superior predictive performance compared to FIGO staging(P＜0.01).The Hosmer-Lemeshow test yielded a P-value＞0.05.The C-index values for the training and validation sets were 0.808 and 0.755,respectively,with the areas under the ROC curves of 0.826 and 0.744.The calibration curves of the model fluctuated near the diagonal line,indicating good agreement between predicted and observed outcomes.The decision curve analysis demonstrated significant clinical net benefit.Results showed that higher mortality risk was associated with advanced age,M1 status,lymph node metastasis,and lack of surgical opportunity.Conclusions:The model exhibits good discriminatory power and accuracy,providing significant benefits to patients.Enhanced management should be implemented for patients with advanced age,distant metastasis,lymph node metastasis,or ineligibility for surgery.Lymph node metastasis is an independent risk factor for 5-year mortality in patients with SCNEC.