Objective Neurocognitive Performance Before Adjuvant Treatment and Survival in Adult-Type Diffuse Glioma: A Systematic Review and Meta-Analysis
Year: 2026
Type: preprint
Study designs: Systematic Review, Meta-Analysis
Abstract: Background. Self-reported cognitive functioning predicts survival in glioma, but self-report is confounded by mood and insight. Whether objective neurocognitive test performance carries independent prognostic information has not been synthesised. Methods. We systematically searched MEDLINE (via PubMed) to 2 October 2026 for cohorts, registries and secondary trial analyses of adults with adult-type diffuse glioma that related standardised cognitive test performance, measured before adjuvant therapy, to overall survival (OS) or progression-free survival (PFS). Risk of bias was assessed with QUIPS. Hazard ratios (HRs) for impaired versus unimpaired performance were pooled with random-effects models (REML, Hartung–Knapp–Sidik–Jonkman). Certainty was rated with GRADE for prognostic factors. Results. Of 400 records, 57 reports were assessed in full and 25 reports from 20 independent cohorts (approximately 3,900 patients) were included. In the primary analysis (9 cohorts, 1,468 analysed patients), objective cognitive impairment was associated with shorter OS: HR 1.91 (95% CI 1.53–2.40; I² = 3%; 95% prediction interval 1.46–2.50). Results were similar for adjusted estimates only (HR 2.09, 1.75–2.49; k = 6) and in high-grade glioma (HR 2.19, 1.66–2.90; k = 5), and weaker and imprecise in lower-grade glioma (HR 1.60, 0.67–3.84; k = 3). In the few cohorts reporting domain-specific estimates, impaired executive function (HR 2.11, 1.12–3.97; k = 4) and verbal fluency (HR 2.37, 1.29–4.37; k = 2) predicted shorter OS; memory was similar in magnitude but imprecise (HR 2.30, 0.83–6.35; k = 3). Five of seven further cohorts reporting continuous or time-limited metrics showed worse outcomes with poorer performance. Certainty of evidence was low. Conclusions. Objective neurocognitive impairment before adjuvant treatment is associated with roughly twice the hazard of death in adult diffuse glioma, an association that persisted in most adjusted analyses and was clearest in high-grade disease. Adjustment sets were incomplete and risk of bias was often high, so independence from established prognostic factors needs confirmation in individual-participant analyses. Brief standardised testing should be evaluated in prognostic models and trial stratification.
Source: Preprints.org
Language: English
Subfield: Oncology
Field: Medicine
Domain: Health Sciences
Open Access status: green
OpenAlex ID: W7236606975
