Real-world evidence on the management and outcomes of glioblastoma in elderly patients: a monocentric retrospective analysis

Abstract Background Glioblastoma (GBM) carries a poor prognosis, and older patients and those with impaired performance status are often underrepresented in randomized trials. This study evaluated real-world treatment patterns, survival outcomes, and factors associated with outcome in older patients with GBM. Methods We retrospectively analyzed 63 patients aged ≥ 65 years with a diagnosis of GBM managed at a single-center radiation oncology department between 2014 and 2023. Overall survival (OS) and progression-free survival (PFS) were calculated from diagnosis. Multivariable Cox models included age, sex, Karnofsky Performance Status (KPS), radiologically confirmed gross total resection (GTR), and temozolomide (TMZ) administration as clinically prespecified covariates. Complete-case analyses were performed. Proportional hazards assumptions were assessed using Schoenfeld residuals, with additional time-varying and exploratory sensitivity analyses addressing diagnostic classification, treatment selection, concomitant TMZ exposure, and treatment timing. Results Median OS was 7.0 months (95% CI 5.3–14.4) and median PFS was 5.3 months (95% CI 3.9–7.8). In the multivariable OS model ( n = 56; 50 deaths), increasing age was associated with higher mortality (HR 1.09, 95% CI 1.02–1.17; p = 0.011), whereas female sex was associated with a lower hazard of death (HR 0.45, 95% CI 0.23–0.91; p = 0.025). Omission of TMZ was associated with higher mortality (HR 2.26, 95% CI 1.16–4.41; p = 0.017), while GTR showed a directionally favorable association that did not reach conventional statistical significance (HR 0.52, 95% CI 0.27–1.01; p = 0.054). In the PFS model ( n = 56; 53 events), increasing age (HR 1.07, 95% CI 1.01–1.14; p = 0.027) and omission of TMZ (HR 2.34, 95% CI 1.20–4.55; p = 0.013) were associated with higher hazards, whereas female sex (HR 0.45, 95% CI 0.23–0.89; p = 0.022) and higher KPS (HR 0.98, 95% CI 0.96–1.00; p = 0.039) were associated with lower hazards. Resection status violated the proportional hazards assumption for PFS; time-varying analysis indicated an early favorable association of GTR that attenuated over time. Sensitivity analyses based on concomitant TMZ exposure and survival measured from radiotherapy initiation yielded similar associations. Conclusions Survival remained limited in this real-world cohort of older patients with GBM. Age, sex, performance status, resection status, and TMZ administration showed varying associations with outcome, with TMZ showing the most consistent association across analyses. Given the small retrospective cohort and substantial treatment selection, these findings should be considered hypothesis-generating rather than causal.