PATH-17. DISTINCT MOLECULAR SUBCLASSES OF H3F3A-WILDTYPE, EGFR-ALTERED PEDIATRIC-TYPE DIFFUSE MIDLINE GLIOMAS WHO CNS GRADE 4

Abstract

Abstract OBJECTIVES Histone gene H3K27-altered diffuse midline glioma (DMG) are malignant tumors that occur in all age groups. In this study, we report comprehensive genomic profiling from H3F3A-wildtype, EGFR-altered DMG, as a distinct and newly recognized subset of pediatric-type DMG. METHODS Tumors were profiled in the comprehensive genomic profiling (CGP) program at Foundation Medicine between 2013-2020. Information from pathology reports, histopathology review, and patient clinical data was assessed. RESULTS We collected demographic and genomic data from 39 pediatric patients with H3F3A-wildtype, EGFR-altered HGG (17 females, 22 males; median age: 8.5 years, range 1-18 years). Female patients were younger at first diagnosis compared to male (median age 7 years vs. 10 years). All cases were microsatellite stable (MSS). The EGFR alterations consisted of 30 mutations and 9 amplifications. The mutations were distributed across the entire gene with no clear hotspot location. Our genomic data converged to identify three distinct molecular patterns. The first and most common group contained TP53 mutations (n = 17, 43.5%), showed no association with patient sex (8 females, 9 males), contained ATRX mutations (n = 3) and CDKN2A deletions (n = 5); these tumors did not harbor pathogenic mutations in TERTp, PIK3CA/PIK3R1, BCOR/BCORL1, STAG2, or SETD2. The second group featured TERTp-mutant tumors (n = 10), were more common in males (70%), and often demonstrated additional mutations in PIK3CA/PIK3R1 (n = 4), BCOR/BCORL1 (n = 4), CDKN2A/B (n = 4), and SETD2 (n = 2). The third group (n = 12) lacked TERTp and TP53 mutations and had a heterogeneous spectrum of non-recurrent mutations, including one CDKN2A/B deletion. CONCLUSIONS We have identified three distinct molecular subclasses that defined specific genomic tumor subgroups in pediatric-type EGFR altered DMG. Overall, these data increase our understanding of the pathobiology of this DMG subset and can guide the design of clinical trials.