Abstract
PURPOSE: Brain tumour classification is a rapidly evolving field, with diagnostic evaluation integrating the latest in molecular testing techniques. As data in brain tumour registries and repositories are collected in real time, neuro-oncology researchers face clear challenges when analysing tumour cohorts diagnosed according to differing standards over time. This study aims to evaluate the impact of an evolving tumour classification system on both our institutional registry and widely used multi-institutional repositories in glioma translational research. METHODS: Clinicopathological data, including molecular profiles, were obtained from the Sydney Brain Tumour Bank registry (1993-2025). We sourced available clinicopathological and molecular classification data from the Rembrandt and Gravendeel datasets, the Chinese Glioma Genome Atlas (CGGA) and The Cancer Genome Atlas (TCGA). All cases were reclassified according to the WHO Classification of Tumours of the Central Nervous System (5th edition). RESULTS: Between 37% and 100% of cases diagnosed prior to the 2016 WHO Classification (revised fourth edition) require additional molecular testing for accurate diagnosis and grading. In contrast, the majority of cases in datasets established after the 2016 Classification met the WHO 2021 Classification criteria (61% to 97%). Two cohorts that consistently failed to meet 2021 requirements over time were high-grade gliomas in patients under 55 years and histological grade 2/3 IDH-mutant gliomas. CONCLUSION: An evolving tumour classification system necessitates regular review and reclassification of brain tumour datasets to ensure that brain cancer research is accurate and equitable. The reclassified datasets are provided for use by neuro-oncology researchers worldwide.