P09.35 Gross total resection of glioblastoma multiforme: influencing factors & survival outcomes - a single centre experience

Abstract

BACKGROUND: Gross total resection (GTR) of glioblastoma multiforme (GBM) can be variably defined. One definition is tumour residuum < 0.175 cm3 that has notably been used by Stummer and colleagues. We studied factors influencing the extent of resection according to this definition of GTR and attendant overall survival (OS) and functional performance. METHODS: Consecutive patients who underwent debulking surgery for histologically proven GBM between September 2013 and February 2016 by subspecialty surgeons at a single institution were included. Data were collected on demographics, tumour location, tumour size pre- and post- debulking using volumetric analysis of magnetic resonance imaging, histology, oncological therapy and functional performance. Survival data and correlations were calculated using Kaplan-Meier and linear regression analyses. RESULTS: 100 patients were analysed with a mean age of 59.9 years (range, 17 - 85 years), and a male:female ratio of 1:1. Mean tumour size pre- and post - operatively were 37.2 cm3 and 1.78 cm3, respectively, representing a mean 95% extent of resection. Median OS across the cohort was 14.9 months. Pre-operative WHO performance status was 0-1 in 60% of cases and 2 in 37% of cases, and there was no significant change in these proportions post-operatively (p > 0.05). GTR with tumour residuum < 0.175 cm3 was achieved in 66% of small tumours (< 37.2 cm3) whilst in 34% of large tumours (> 37.2 cm3) (p < 0.05). Furthermore, volume of tumour residuum was significantly correlated with pre-operative tumour volume on linear regression analysis (p = 0.01). GTR was achieved in 53% versus 29% of tumours involving a single versus multiple lobes respectively (p = 0.08). There was no difference in the number of GTRs achieved amongst tumours of dominant versus non-dominant lobes. There was no significant difference in median OS between patients who had undergone GTR versus subtotal resection as per Stummer's definition (15.1 vs 14.9 months, respectively, Log-Rank, p = 0.40). We note that in this cohort nearly two-thirds of patients achieved > 99% extent of resection. Comparison of 24 patients with complete resection versus 52 patients with near total resection (thin rim of enhancement) versus 24 patients with subtotal resection did yield significant differences in survival (15.9 vs 13.2 vs 11.1 months respectively, Log-Rank p = 0.03). CONCLUSIONS: Factors influencing GTR include pre-operative size of the tumour and involvement of more than one lobe. GTR as per Stummer's definition does not appear to confer a survival advantage in our cohort but the vast majority of patients had very small tumour residuum. Categorising patients with respect to complete, near total and subtotal resection did yield small but significant differences in survival. Further prospective studies are needed to elucidate the most meaningful definition of gross total resection.