IMG-14. DIAGNOSTIC UTILITY OF THE T2-FLAIR MISMATCH SIGN FOR DNETS AND ITS ASSOCIATION WITH FGFR1 ALTERATIONS

Abstract

Abstract BACKGROUND The T2-FLAIR mismatch sign, hyperintense T2-weighted signals with corresponding hypointensity on FLAIR sequence, has emerged as a potential diagnostic tool for identifying specific brain tumor subtypes. We tested the diagnostic utility of the sign in dysembryoplastic neuroepithelial tumors (DNETs) with and without FGFR1 alterations. METHODS We evaluated the T2-FLAIR mismatch sign in a cohort of 43 patients with definitive or probable DNET. Low-grade glioma patients with definite non-DNET diagnoses were chosen as control (n=43). Receiver operative curve analysis was performed, and the corresponding odds ratios were calculated. RESULTS Of the 43 DNET cases, 35 were definite, and 8 were probable DNETs. FGFR1 alterations were confirmed in 9 definitive DNETs. The area under the curve (AUC) for diagnosing DNET with the T2-FLAIR mismatch sign was 0.884 (DeLong CI: 0.2, 0.95) (Accuracy: 0.88, 95% CI: 0.80, 0.94; Sensitivity: 0.84; Specificity: 0.95); for definite DNET it was 0.891 (DeLong CI: 0.82, 0.96) (Accuracy: 0.90, 95% CI: 0.81, 0.95; Sensitivity: 0.87; Specificity: 0.94); for probable DNET it was 0.852 (DeLong CI: 0.69, 1;) (Accuracy 0.92, 95% CI: 0.81, 0.98; Sensitivity: 0.84; Specificity: 0.75), and for DNET with FGFR1 alterations was 0.977 (DeLong CI: 0.95, 1) (Accuracy 0.96, 95% CI: 0.87, 1; Sensitivity: 1; Specificity: 0.81). The odds of having DNET (definite +probable) if the T2-FLAIR mismatch sign was positive was 81.96 (CI: 16.25, 852.48, P<0.001), and DNET with FGFR1 is 315.4 (CI: 312.3, 318.5, P<0.001). CONCLUSION The T2-FLAIR mismatch sign is a reliable radiological biomarker for DNET and strongly predicts FGFR1 alterations in these tumors.