Abstract
BACKGROUND: Recent studies conducted by the RANO resect group established the importance of resection beyond the contrast-enhancing (CE) area (supramarginal resection) as a critical element in the management of glioblastoma (GBM). Our study aims to employ this concept in our cohort and emphasize the importance of the EOR of FLAIR and non-CE tumor for survival outcome. MATERIAL AND METHODS: We analyzed 445 patients (187 female, 258 male) with a median age of 64.3 years (range: 26.9 - 86.6 years) and median presurgical Karnofsky index (KPI) of 80, who underwent surgical resection of newly diagnosed glioblastoma. The MGMT status was unmethylated in 48.1%. Fluorescence guidance (FG) was used in 57.7%. Supramarginal resection was determined by volumetry of the pre-and postsurgical FLAIR- or non-contrast enhancing (non-CE) tumor volume as previously defined. According to the classification of the RANO-Resect group, we divided patients in following groups: “biopsy” (without tumor reduction), “submaximal CE resection”, “ maximal CE resection” and “supramarginal CE resection”. Outcome analysis included progression-free (PFS) and overall survival (OS). RESULTS: 95 (31.6%) patients underwent complete resection of the contrast-enhancing tumor (CE); of these, 74 (62.1%) received a supramarginal resection. Preoperative CE tumor volume correlated with preoperative FLAIR volume (p=0.0001); a higher FLAIR volume was associated with a lower KPI (p=0.021). The extent of supramarginal resection was significantly higher in the FG group (p=0.046). Patients undergoing supramarginal resections showed better OS (p=0.0001) and PFS (p=0.017). Postoperative FLAIR volume significantly correlated negatively with OS (p=0.004). Multivariate Cox regression analysis revealed a progressive decrease in the hazard of death (p=0.00001, HR: 0.74) and progression (p=0.012, HR: 0.82) for each RANO resect category from incomplete to supramarginal resection. Moreover, supramarginal resection (p=0.024) was found to be an independent positive prognostic factor for OS, next to preoperative KPI (p=0.0001), age (p=0.0001), and MGMT status (p=0.0001). in a Cox proportional hazards regression model residual FLAIR volume and residual nCE volume resulted as statistically significant predictors of OS. CONCLUSION: Supramarginal resection was beneficial for both, PFS and OS in patients with newly diagnosed GBM. The hazard of death and progression decreased steadily with each RANO resect category. Residual FLAIR and nCE volume and RANO resect categories are significant predictors of OS and PFS. Fluorescence guidance significantly increased the extent of supramarginal resection.