Survival Outcomes and Prognostic Factors of Central Nervous System Malignancies in Adult Patients: A Single-Center Retrospective Cohort Study From a Tertiary Hospital in Bangladesh.

Background and objective Central nervous system (CNS) malignancies carry a high mortality burden, and longitudinal survival data from low- and middle-income countries remain sparse. We aimed to characterise overall survival (OS) and identify key prognostic factors in adult patients with primary CNS malignancies managed at a tertiary referral hospital in Bangladesh. Methods This retrospective cohort study included consecutive adult patients aged 18 years or older with histopathologically confirmed primary CNS malignancies treated between June 2018 and December 2023. Patients with severe, life-limiting comorbidities, as defined in the Methods section, and those lost to follow-up within 30 days were excluded. OS was estimated using the Kaplan-Meier method, group differences were assessed by the log-rank test, and a multivariable Cox proportional hazards model was used to identify independent prognostic factors. Reporting followed the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) guidelines. Results A total of 151 adult patients were analysed (mean age: 47.3 ± 13.8 years; 100 (66.2%) men, 51 (33.8%) women; male-to-female ratio: 1.96:1). Median follow-up among censored patients was 29.3 months (interquartile range: 10.4-43.7), and 47 (31.1%) patients remained alive at last follow-up. Median OS for the entire cohort was 19.6 months (95% confidence interval (CI): 14.3-27.6). Histopathological diagnosis stratified survival (log-rank p=0.003): median OS was 12.4 months (95% CI: 9.2-19.6) for glioblastoma (n=79), 21.9 months (95% CI: 14.7-39.7) for grade II-III astrocytoma (n=51), 113.8 months (95% CI: 1.1-113.8) for ependymoma (n=7), and 153.3 months (95% CI: 24.4-153.3) for oligodendroglioma (n=14); the latter two estimates are based on small event numbers and should be interpreted with caution. World Health Organization (WHO) tumour grade stratified survival (log-rank p=0.003): median OS was 55.0 months (95% CI: 27.8-124.8) for grade II (n=41), 24.4 months (95% CI: 19.2-66.7) for grade III (n=22), and 12.2 months (95% CI: 9.2-17.9) for grade IV (n=88). Age was the strongest determinant (log-rank p<0.001): median OS was 66.7 months (95% CI: 38.5-124.8) for patients aged 18-40 years (n=48), 18.1 months (95% CI: 12.1-24.4) for those aged 41-60 years (n=73), and 7.7 months (95% CI: 5.4-9.8) for those aged 60 years or older (n=30). The extent of surgical resection was not associated with survival (log-rank p=0.620). On multivariable Cox regression, each 10-year increment in age conferred an adjusted hazard ratio of 1.84 (95% CI: 1.52-2.23, p<0.001), the only independent predictor of mortality. Conclusion Survival from adult CNS malignancies in this Bangladeshi cohort remained lower than benchmarks reported from high-income settings, particularly for glioblastoma. Age was the dominant independent prognostic factor. The findings underscore the urgent need for routine molecular diagnostics, expanded radiotherapy and adjuvant chemotherapy capacity, and a national cancer registry to guide neuro-oncological care in Bangladesh.