HOUT-06. SYSTEMIC METABOLIC BIOMARKERS OBTAINED FROM COMPUTED TOMOGRAPHY (CT)-BASED BODY COMPOSITION AND SERUM GLUCOSE PREDICT SEX-SPECIFIC SURVIVAL IN PATIENTS WITH GLIOBLASTOMA

Abstract

Abstract In many cancers including glioblastoma, males have higher incidence and mortality. Although the mechanisms underlying this phenomenon have yet to be elucidated, there are inherent sex differences in metabolism. Males not only have higher glucose metabolism relative to females, but males also have higher amounts of abdominal visceral fat, a well-known metabolic biomarker for poor cardiovascular and cancer outcomes that could contribute to enhanced tumorigenesis. Previously, we identified that in renal cell carcinoma, females with male-pattern visceral fat did significantly poorly compared to all other patients (PMID 29558292) and that in lower grade gliomas, tumor glycolysis specifically stratified males (PMID: 28768910). We hypothesized that in glioblastoma, visceral fat would selectively stratify females and serum glucose would selectively stratify males. We retrospectively analyzed abdominal/pelvic computed tomographic (CT) datasets for the normalized relative visceral fat area (rVFA) and random glucose measurements at diagnosis from 159 males and 105 females with glioblastoma. Females with rVFA > 30.5% had significantly reduced median overall survival (OS) (11.6 months, n=63) vs females with rVFA < 30.5% (20.9 months, n=42, p=0.006), that was close to the median rVFA of 32.7%. For males, the optimized rVFA threshold was higher than the median (43.6%); males with rVFA > 58.9% had significantly reduced median OS (10.7 months, n=14) compared to males with rVFA < 58.9% (20.5 months, n=145, p=0.02). However, after adjusting rVFA for age and tumor size, it was an independent predictor in females, but not in males. Conversely, there was a male-specific survival effect of glucose: males with prediabetic random serum glucose greater than 179.5 mg/dL had shorter OS (9.9 months vs 20.9 months; p=0.007), where no glucose level could stratify females. These data support the use of metabolism to predict sex-specific cancer outcomes and suggest that interventions that target metabolism can improve sex-specific outcomes in glioblastoma.