Clinical and VASARI Features to Predict CDKN2A/B Homozygous Deletion in IDH-Mutant Astrocytomas: A Multicenter Study

BACKGROUND AND PURPOSE: Cyclin-dependent kinase inhibitor (CDKN)2A/B homozygous deletion holds an important value in the prognostic analysis of isocitrate dehydrogenase (IDH)-mutant astrocytomas. This study aimed to develop and validate a prediction model based on clinical and Visually AcceSAble Rembrandt Images (VASARI) MRI features for identifying CDKN2A/B homozygous deletion status in IDH-mutant astrocytomas. MATERIALS AND METHODS: Preoperative MR images of 122 patients with astrocytoma (101 without CDKN2A/B homozygous deletion, and 21 with CDKN2A/B homozygous deletion) were retrospectively collected as a training set. Eighteen patients with astrocytomas from another center (14 without CDKN2A/B homozygous deletion, and 4 with CDKN2A/B homozygous deletion) were enrolled as an external test set. VASARI features in MR images of patients with astrocytoma were evaluated by radiologists. Prediction models for identifying CDKN2A/B homozygous deletion status based on clinical and VASARI features were constructed by using logistic regression. RESULTS: Multivariate regression analysis showed that definition of enhancing margin (OR: 13.19, P = .003) was an independent predictor of CDKN2A/B homozygous deletion in IDH-mutant astrocytomas. The area under the curve of prediction model in the training set and external test set was 0.84 (95% CI, 0.73–0.96) and 0.96 (95% CI, 0.86–1.00), respectively. The optimal cutoff value of the nomogram calculated by using the Youden index was 124.20 points. Under the cutoff value, the prediction model exhibited 85% accuracy, 67% sensitivity, and 88% specificity in the training set, and 83% accuracy, 75% sensitivity, and 86% specificity in external test set. CONCLUSIONS: The nomogram model based on clinical and VASARI MRI features was useful for prediction of CDKN2A/B homozygous deletion status in IDH-mutant astrocytomas.