Background: The real association between extent of resection and outcome in patients with glioblastoma multiforme (GBM) remains unclear.
Objective: The goal of this study was to disclose the effect of gross total resection on survival and establish a scale used for surgical decision making.
Methods: A retrospective review was undertaken of 416 patients who received operation for GBM from 2008 to 2015 in Beijing Tiantan Hospital. To reduce bias in patient selection, propensity score analysis was conducted and 99 pairs of matched GBMs were generated. Survival between different groups was compared using the Kaplan-Meier method, and independent predictors of survival were identified using the Cox proportional hazards model.
Results: Overall, the survival of patients undergoing GTR was significantly longer than those not undergoing GTR (12.0 vs. 9.0 months [p < 0.001] for progression-free survival [PFS], and 20.5 versus 16.0 months [p < 0.001] for overall survival [OS]). In the propensity model, the survival benefit of GTR remained significant, which has been further validated in the multivariate analysis (hazard ratio [HR] 0.613, 95% confidence interval [CI] 0.454-0.827 [p = 0.001] for PFS, and HR 0.475, 95% CI 0.343-0.659 [p < 0.001] for OS). Using a scoring scale based on age, epilepsy, location, tumor size, and Karnofsky performance score, patients were stratified into low-, moderate-, and high-risk cohorts. The survival benefit of GTR could be observed in the low- and moderate-risk cohorts but not the high-risk cohort.
Conclusion: GTR was an independent predictor of increased survival for patients with GBM. The risk scoring scale quantified the clinical significance of operation and helped us to project more personalized surgical strategies for individual patients.