This study was designed to develop a risk model for disease recurrence among cervical cancer patients who underwent neoadjuvant chemotherapy and radical surgery. Data for 853 patients were obtained from a retrospective study and used to train the model, and then data for 447 patients from a prospective cohort study were employed to validate the model. The Cox regression model was used for calculating the coefficients of the risk factors. According to risk scores, patients were classified into high-, intermediate-, and low-risk groups. There were 49 (49/144, 34%) recurrences observed in the high-risk group (with a risk score >= 2.65), compared with 3 (3/142, 2%) recurrences in the low-risk group (with a risk score < 0.90). Disease-free survival (DFS) was significantly different (log-rank p < 0.001) among the three risk groups; the risk model also revealed a significant increase in the accuracy of predicting 5-year DFS with the area under the ROC curve (AUC = 0.754 for risk model vs 0.679 for FIGO stage system); the risk model was also validated with data from the prospective study (log-rank p < 0.001, AUC = 0.766). Both high-risk and intermediate-risk patients can be more effectively identified by this risk model.