Abstract
Inflammation is strongly associated with cancers. Our research explored the prognostic role of hematological inflammatory indicators in patients with glioblastoma (GBM). Hundred patients were incorporated finally, and we determined the best cutoff values of these blood markers through X-tile first to perform further survival tests. The prognostic role of peripheral blood cell count and corresponding ratios were assessed using the Kaplan-Meier curve and Cox models to identify variables with independent prognostic significance. Then, variables with independent predictive values were incorporated in constructing the nomogram model to realize quantitative prediction for survival. Nomograms were evaluated using Harrell concordance index, receiver operating characteristic curves, and decision curve analysis to assess clinical benefit. Multivariate analysis indicated that a higher platelet-to-basophil ratio (PBR) (>4575) (hazard ratio = 1.819, 95% confidence interval: 1.110-2.980, P = .018) was associated with poorer overall survival in GBM patients. Based on the PBR nomogram, the predictive accuracy was moderate (concordance index = 0.844, area under curve = 0.632). The PBR could serve as a prognostic biomarker for overall survival in patients with GBM, and nomogram models incorporating the PBR could facilitate comprehensive preoperative survival assessment.