PL02.4.A Chromosome 1p19q codeletion frequency, but not survival, varies according to the IDH mutation subtypes: analysis of 1050 IDH-mutated diffuse gliomas

Abstract

Abstract BACKGROUND recurring hotspot mutations in isocitrate dehydrogenase (IDH) 1 enzymes, and to a lesser extent IDH2, are the main oncogenic alteration in most of lower-grade diffuse gliomas and a subset of grade 4 gliomas. Although most commonly represented by the IDH1R132H mutation, non-canonical IDH1/2-nonR132H mutations are present in about 10% of cases. As the neomorphic enzymatic activity can vary depending on the type of the hotspot mutation, a different biological behaviour may be inferred. Nevertheless, the prognostic significance of the different IDH1/2 mutations remains a matter of debate. MATERIAL AND METHODS we queried the OncoNeuroTek tumor bank (Pitié-Salpêtrière Hospital, Paris) to identify all registered cases of IDH-mutated diffuse gliomas. Most relevant clinical and molecular data were collected. RESULTS We identified 1050 IDH mutated diffuse gliomas (481 grade 2 [46%], 459 grade 3 [44%], and 110 grade 4 [10%]), of which 1007 (96%) were IDH1 mutated (934 IDH1R132H [89%] and 73 IDH-nonR132H [7%]) and 43 (4%) were IDH2 mutated (24 IDH2R172K [2%] and 19 IDH2-nonR172K [2%]). The chromosomes 1p/19q codeletion was more frequent in IDH2-mutated tumors (25/42 [60%] versus 350/918 [38%] in IDH1-mutated cases, p=0.005). Nevertheless, only IDH2R172K mutation was associated with the codeletion (18/24 [75%], versus 7/18 [39%] in IDH2-nonR172K-mutated tumors, p=0.02). IDH1-nonR132H tumors showed the lowest rate of 1p/19q codeletion (9/69 [13%], versus 341/849 [40%] in IDH1R132H-mutated cases, p<0.001). At the time of observation, 536 patients were deceased (51%), with a median overall survival (OS) of 115 months (mo., 95% CI 108–125 mo.). Median follow up for censored patients was 77 mo. Codeleted patients had a significantly longer OS compared to non-codeleted ones (179 vs. 96 mo., p<0.001). No significant differences in survival were seen when stratifying according to IDH mutation type (115 mo. for IDH1R132H vs. 136 mo. for IDH1-nonR132H vs. 112 mo. for IDH2R172K vs. 150 mo. for IDH2-nonR172K, p=0.8). No differences were seen when restricting the analysis to codeleted or not-codeleted patients only, respectively. In a multivariate analysis including the main prognostic factors (age, sex, preoperative performance status, tumor grade, surgical resection, midline location, 1p/19q codeletion, and p16 homozygous deletion), no survival difference was associated with any of the IDH mutation subtype. CONCLUSION although significantly different rates of 1p/19q codeletion were seen according to the four main IDH mutation subgroups, these groups does not associate with different survival profiles in our cohort.