| [Objective] To develop and validate a survival prediction model for high-grade gliomas using Lasso-Cox regression. [Methods] The clinical data of 309 patients with high-grade glioma who underwent postoperative intensity-modulated radiotherapy combined with chemotherapy and other comprehensive treatments at Jiangsu Cancer Hospital from January 2016 to December 2023 were retrospectively collected. They were randomly divided into a training set (n = 216) and a validation set (n = 93) at a ratio of 7:3. In the training set, the least absolute shrinkage and selection operator (LASSO) was employed to screen the influencing factors of overall survival (OS), which were subsequently incorporated into the multivariate Cox proportional hazards regression analysis to construct a nomogram model. The discrimination, calibration, and clinical utility of the prediction model were evaluated using the concordance index (C-index), calibration curve, and clinical decision curve (DCA), respectively. The Kaplan-Meier survival curve was utilized to assess the risk stratification ability of the nomogram model for patients.[Results] Histologic grade, MGMT promoter status, intervals between surgery and radiotherapy, monocyte level before radiotherapy, and platelet to lymphocyte ratio before radiotherapy are independent influencing factors for the overall survival of patients with high-grade glioma. The C-index of the nomogram model in the training set and validation set are 0.705 (95%CI: 0.658-0.752) and 0.661 (95%CI: 0.583-0.739), respectively. The calibration curve is well fitted, and the clinical decision curve shows that the model has high clinical utility. The nomogram model has significant discrimination power for patients with different risk stratification (P<0.001).[Conclusions] The nomogram model can effectively predict overall survival in high-grade glioma patients following postoperative intensity-modulated radiotherapy combined with chemotherapy. |
