Molecular Signatures for Survival Prediction in Glioma: A Prospective Data Analysis from an LMIC Center

Introduction: Glioma characterization and follow-up is underreported from low-and-middle-income country (LMIC) centers within the literature. With emphasis on molecular markers for survival prediction, there is a need for robust data exploring molecular epidemiology in these countries. In Pakistan particularly, there is a significant gap in glioma outcomes reporting and survival analysis.

Methods: 200 cases (194 patients) were enrolled from 2019 onwards; histopathological and molecular analysis was performed on formalin-fixed paraffin-embedded (FFPE) blocks for IDH, P53, ATRX, and Ki-67 immunohistochemical (IHC) markers. Survival analysis was calculated using the Kaplan-Meier method. Prior diagnoses according to the 2016 WHO criteria were revised after updated molecular markers and review by a board-certified neurohistopathologist.

Results: 53 cases were reclassified according to the updated 2021 WHO criteria (26.5%): 45 within the same grade, 6 cases were reclassified to a higher grade glioma and 4 to a lower grade glioma diagnosis. 12 glioblastoma cases (18.4%) were reassigned to a new classification while 9 cases were re-classified as GBM. IDH wild type (IDH-WT) gliomas had a significantly worse overall survival (log-rank = 0.0001), with median OS 14.2 months, 1-year survival rate of 95% for IDH-M and 72% for IDH-WT. Significant survival differences were seen for P53 mutations in grade 4 gliomas (p=0.04), ATRX in IDH-mutant (log-rank = 0.026, 1-year survival rate: 80% (Loss), 100% (Retained)), and 1p19q co-deletion for Grade 3 oligodendroglioma (log-rank = 0.049). MGMT methylation status was available for 35 patients – a significant survival difference was seen on overall comparison (p=0.004), however, this could not be demonstrated for IDH-WT glioma only.

Conclusion : Our re-classification of glioma diagnosis based on molecular markers shows a subset of patients that were misdiagnosed – particularly GBM cases which were a different pathology on re-examination, and similarly 9 cases which were in fact GBM. This emphasizes the need for molecular data particularly in LMICs where such data remains underreported. Our data also shows better OS in P53-mutant glioma particularly in grade 4 glioma and a significant role by ATRX status in survival.