EPID-06. ASSOCIATIONS BETWEEN GERMLINE GENETIC VARIANTS AND OVERALL SURVIVAL IN PATIENTS WITH GLIOMA

Abstract

Abstract BACKGROUND In addition to somatic genetic variation in tumors, germline genetic variation can better define cancer susceptibility risk, guide therapy, and predict survival. Most prior research into associations between germline genetic variants and survival outcomes in patients with glioma has been limited by small sample sizes and to high-grade glioma only. This study is the first to use data from Brigham and Women’s Hospital (BWH) and to include both low-grade and high-grade glioma cases to explore associations between germline genetic variants and overall survival. METHODS This study included 211 patients enrolled at BWH from April 2010 to January 2020. Ninety-two candidate germline genetic variants were identified via literature review. Fifty-five variants with minor allele frequency >0.05 were included preliminarily, for which bivariate Cox proportional hazards regression models were employed. Twenty-two variants with P< 0.5 were included in the final multivariable model, adjusted for patient sex, age, race, and glioma grade. Stratified sub-analyses were also conducted by sex. Associations were considered statistically significant at P< 0.05. RESULTS Among all patients, homozygous C/C genotype at rs17655 (ERCC5 gene) was significantly associated with worse overall survival (hazard ratio=2.61, P=0.0095). Among females only, worse survival was associated with homozygous T/T genotype at rs1381057 (POLQ, hazard ratio=2.58, P=0.046). Among males only, heterozygous status for the A genotype at rs487848 (POLQ, hazard ratio=0.31, P=0.018) and the A genotype at rs1468923 variant (PIK3CA, hazard ratio=0.51, P=0.016) were associated with improved survival. CONCLUSION We identified statistically significant associations between overall survival and four variants in genes involved in DNA repair and the PIK3 pathway, three of which were sex-specific. Our results contribute to improving patient stratification in glioma and may lead to increased targeting of treatment strategies.