Prediction of Survival in the Elderly Patients with Glioblastoma using Cumulative Inflammatory Markers Score.

This retrospective study aimed to explore the prognostic effect of cumulative score based on neutrophil-to-lymphocyte ratio (NLR), platelet-to-lymphocyte ratio (PLR), and fibrinogen in older adults diagnosed with glioblastoma (GBM).  Retrospective study.  Huashan Hospital.  Patients aged over 60 years and diagnosed with GBM between 2010 and 2017.  Results of preoperative routine biochemistry and coagulation blood examinations were reviewed from medical records. Overall survival (OS) was considered a period from first resection surgery until death. Progression-free survival (PFS) was considered a period from initial operation until the date of tumor progression demonstrated in brain magnetic resonance imaging or death from any cause. If no event occurred, the last follow-up appointment was the end of the observation for OS or PFS. The Kaplan-Meier method was used to evaluate survival curves, and prognostic factors were analyzed by the Cox proportional hazards model.  A total of 289 patients were included. Patients with higher levels of fibrinogen, NLR, and PLR had significantly shorter median OS (  = 0.001,  = 0.016, and  = 0.002, respectively) and PFS (  = 0.004,  = 0.022, and  = 0.009, respectively) compared with those with lower levels. Multivariate analyses showed a significant association between higher F-NLR-PLR score and reduced OS (adjusted hazard ratios [aHRs]: 1.356, 95% confidence interval [CI] 1.009-1.822 for scores 1-2 compared with 0; 5.974, 95% CI 2.811-12.698 for score 3 compared with 0). Similarly, a significant association between higher F-NLR-PLR score and reduced PFS was observed (aHR: 1.428, 95% CI 1.066-1.912 for scores 1-2 compared with 0; aHR: 2.860, 95% CI 1.315-6.223 for score 3 compared with 0).  Higher F-NLR-PLR score is associated with reduced OS and PFS in older adults with GBM, which helps identify patients at high risk and guide the individualized treatment in clinical practice.