P11.65.B TEMPORAL MUSCLE THICKNESS AS A PROGNOSTIC MARKER IN A REAL-LIFE COHORT OF NEWLY DIAGNOSEDMGMT PROMOTER METHYLATED GLIOBLASTOMA PATIENTS: A MULTICENTRIC IMAGING ANALYSIS

Abstract BACKGROUND In cranial magnetic resonance imaging (MRI) assessed temporal muscle thickness (TMT) correlates with lumbar skeletal muscle mass, which serves as a surrogate marker of sarcopenia. Prior research has identified TMT as a prognostic marker in glioblastoma. However, incorporation into clinical practice is challenging due to the heterogeneity of previous studies and the absence of established cutoff values. To address this, we conducted a multicenter analysis to validate recently proposed sex-specific cutoff values in a homogeneous cohort of newly diagnosed MGMT promoter-methylated glioblastoma patients uniformly treated under real-life conditions with combined CCNU/temozolomide. We also included a balanced control cohort for comparison. MATERIAL AND METHODS TMT was measured at baseline using initial preoperative and postoperative MRI and during the course of the disease using the first MRI after completion of radiotherapy in the first therapy-line. Using TMT and previously reported cutoff values from the literature, patients were classified into "at risk of sarcopenia" or "normal muscle status" cohorts. To correlate TMT with survival, we conducted Kaplan-Meier and multivariable Cox regression analyses. RESULTS A total of 126 patients were analyzed, with 66 patients treated with combined CCNU/temozolomide and 60 patients treated with temozolomide single-drug treatment. In both cohorts, patients with normal muscle mass at baseline had significantly longer survival. Specifically, in the CCNU/temozolomide cohort median progression-free survival (mPFS) was 16.7 months in the normal muscle mass group versus 6.8 months in the at-risk group (p=0.0008); median overall survival (mOS) was 44.2 months in the normal muscle mass group versus 16.7 months in the at-risk group (p=0.0046). In the temozolomide single-drug cohort, mPFS was 12.4 months in the normal muscle mass group versus 7.7 months in the at-risk group, and mOS was 29.5 months in the normal muscle mass group versus 17.4 months in the at-risk group (p=0.0006). Using multivariable Cox regression analysis, we found normal muscle mass (HR: 0.39, 95% CI: 0.23-0.66; p=0.0003) and initial age at diagnosis of less than 50 years (HR: 0.42, 95% CI: 0.22-0.79; p=0.0067) were statistically significant prognostic markers. Longitudinally, patients with lacking TMT decline during the course of the disease had the best survival outcomes. CONCLUSION This analysis suggests TMT to serve as a prognostic marker in glioblastoma patients treated with CCNU/temozolomide or single-drug treatment with temozolomide. However, for TMT assessment to be implemented in a generalizable manner, such as for patient selection for therapeutic measures, further validation in prospective controlled trials is necessary.