OBJECTIVE

To develop a nomogram prediction model for ovarian cancer prognosis using tumor proliferation and immune-related biomarkers.

METHODS

Between January 2018 and December 2023, clinical data were collected from 140 patients diagnosed with epithelial ovarian cancer (EOC). These patients were randomly allocated into a training cohort consisting of 84 patients and a validation cohort comprising 56 patients, adhering to a 6:4 ratio. Immunohistochemical staining assessed Ki67, epidermal growth factor receptor (EGFR), and programmed death-ligand 1 (PD-L1) expression. Lasso-Cox regression identified variables for the nomogram model. Model performance was evaluated using time-dependent receiver operating characteristic (ROC) curves, concordance index, calibration curves, and decision curve analysis. Kaplan-Meier survival analysis assessed the prognostic value of the model's risk score.

RESULTS

Lasso-Cox regression identified seven variables for constructing the nomogram prediction model: maximum tumor diameter, KI67 positive rate, pathological grade, N stage, M stage, KI67 ≥ 20%, and PD-L1 > 0. The area under the ROC curve (AUC) for 1-, 4-, and 6-year overall survival (OS) in the training set were 0.908, 0.940, and 0.965, respectively, with a concordance index of 0.85 [95% confidence interval (CI): 0.80-0.90]. In the validation set, the AUCs for 1-, 4-, and 6-year OS were 0.835, 0.802, and 0.832, respectively, with a concordance index of 0.72 (95% CI: 0.65-0.79). Calibration curves demonstrated good agreement between predicted and actual outcomes (as indicated by non-significant Hosmer-Lemeshow tests, all P>0.05). The model outperformed TNM staging in clinical benefit. High-risk scores correlated with poorer overall and progression-free survival (P<0.01). These findings suggest the nomogram can effectively stratify patients and predict prognosis.

CONCLUSION

The successful development and validation of a nomogram prediction model based on tumor proliferation and immune-related biomarkers offers an efficient and straightforward clinical tool. This tool holds promise for enabling personalized treatment strategies for patients with ovarian cancer.