Refining Prognostic Stratification in Glioblastoma: A Multivariate Analysis Integrating Corrected Biostatistical Models with Surgical and Therapeutic Factors

INTRODUCTION: Glioblastoma multiforme (GBM) remains the most aggressive primary brain tumor with poor prognosis. Accurate survival analysis is essential for treatment planning and prognostic assessment. This study evaluates the application of corrected Kaplan-Meier and Cox regression methods in analyzing survival outcomes of GBM patients.
METHODS: A retrospective analysis was conducted on 203 GBM patients treated at XXX University Faculty of Medicine Hospital between 2005 and 2011. Patient demographics, clinical characteristics, tumor localization, and treatment protocols were recorded. Survival analysis was performed using classical and corrected Kaplan-Meier methods, and Cox proportional hazards regression was applied to identify independent prognostic factors.
RESULTS: The mean survival time was 20.9 months with a median survival of 17.0 months. Multivariate Cox regression analysis identified age, Karnofsky Performance Status (KPS), tumor localization, and adjuvant treatment protocols as independent prognostic factors. Temporal and parietal lobe localizations were associated with better prognosis. Combined Temozolomide (TMZ) + Altuzan/bevacizumab (ALT) therapy demonstrated a significant survival advantage. Male gender and low postoperative KPS scores increased mortality risk, while high KPS and TMZ+ALT treatment improved survival outcomes.
DISCUSSION AND CONCLUSION: Corrected Kaplan-Meier and Cox regression methods provide reliable tools for survival analysis in GBM patients. These statistical approaches support clinical decision-making and enable more accurate prognostic estimation, particularly when dealing with censored data.

Keywords: Glioblastoma multiforme, survival analysis, Kaplan Meier method, Cox regression, prognostic factors, temozolomide