The Misclassification of Diffuse Gliomas: Rates and Outcomes

Abstract

Purpose: The integrated histopathologic and molecular diagnoses of the 2016 WHO classification of central nervous system tumors have revolutionized patient care by improving diagnostic accuracy and reproducibility; however, the frequency and consequences of misclassification of histologically diagnosed diffuse gliomas are unknown. Experimental Design: Patients with newly diagnosed ICD-O-3 (International Classification of Diseases) histologically encoded diffuse gliomas from 2010–2015 were identified from the National Cancer Database, the misclassification rates and overall survival (OS) of which were assessed by WHO grade and 1p/19q status. In addition, misclassification rates by isocitrate dehydrogenase (IDH), ATRX, and p53 statuses were examined in an analogous multi-institutional cohort of registry-encoded diffuse gliomas. Results: Of 74,718 patients with diffuse glioma, only 74.4% and 78.8% of molecularly characterized WHO grade II and III oligodendrogliomas were in fact 1p/19q-codeleted. In addition, 28.9% and 36.8% of histologically encoded grade II and III “oligoastrocytomas”, and 6.3% and 8.8% of grade II and III astrocytomas had 1p/19q-codeletion, thus molecularly representing oligodendrogliomas if also IDH mutant. OS significantly depended on accurate WHO grading and 1p/19q status. Conclusions: On the basis of 1p/19q, IDH, ATRX, and p53, the misclassification rates of histologically encoded oligodendrogliomas, astrocytomas, and glioblastomas are approximately 21%–35%, 6%–9%, and 9%, respectively; with significant clinical implications. Our findings suggest that when compared with historical histology-only classified data, in national registry, as well as, institutional databases, there is the potential for false-positive results in contemporary trials of molecularly classified diffuse gliomas, which could contribute to a seemingly positive phase II trial (based on historical comparison) failing at the phase III stage. Critically, findings from diffuse glioma clinical trials and historical cohorts using prior histology-only WHO schemes must be cautiously reinterpreted.