Materials and methods

Study design and patient selection

This retrospective, observational cohort study was conducted using clinical records from the Neuro-Oncology Unit at the Hospital de Clínicas “Dr. Manuel Quintela”. The study was designed and reported in accordance with the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) guidelines. We screened all clinical records of adult patients with histologically confirmed supratentorial hemispheric GB who were initially evaluated between January 2009 and December 2017. Administrative and clinical follow-up for survival status extended through December 31, 2018.

Inclusion criteria were adult patients (≥18 years) with histologically confirmed supratentorial hemispheric GB who received standard multimodal frontline therapy. Frontline protocol compliance was defined as a minimum radiotherapy dose of 40 Gy delivered concurrently with daily TMZ, followed by at least three cycles of adjuvant TMZ maintenance. Patients who received radiotherapy or chemotherapy alone, who received no tumor-directed treatment, or who completed their protocol at another institution were excluded.

Treatment protocol and operational definitions

Extent of surgical resection was categorized as gross total resection (GTR, R0) when postoperative neuroimaging confirmed macroscopically complete tumor removal; subtotal resection (STR, R1-R2) included cases with residual tumor and diagnostic stereotactic biopsies. Standard protocol compliance was defined as concurrent chemoradiotherapy - fractionated radiotherapy (60 Gy over six to seven weeks) plus concurrent daily oral TMZ (75 mg/m²) - followed by adjuvant maintenance, defined as oral TMZ (150-200 mg/m²) for five consecutive days every 28 days for a planned six cycles. OS was defined as the interval from primary surgical intervention to death from any cause, or to the last verified clinical follow-up for censored patients.

Statistical analysis

Continuous variables are expressed as means ± standard deviation (SD), and categorical variables as frequencies and percentages. OS probability was estimated using the Kaplan-Meier method, and differences between groups were assessed with the log-rank test. Univariate and multivariate survival analyses were performed using the Cox proportional hazards model. Given the limited sample size, the multivariate model was restricted to four pre-specified clinical variables to avoid overfitting, following the events-per-variable rule. Statistical significance was set at a two-tailed p < 0.05. All analyses were performed using STATA version 12.0 (StataCorp LLC, College Station, TX).

Ethical considerations

The study protocol was reviewed and approved by the Institutional Ethics Committee of the Hospital de Clínicas “Dr. Manuel Quintela” (approval number CEI-HC-2019-04). Informed consent was waived given the retrospective, anonymized nature of the study. The research was conducted in accordance with the Declaration of Helsinki. All authors confirm that this study involved human participants and their clinical data.