Integrating Molecular Profiling into Glioma Diagnosis: Implications of the WHO-CNS5-2021 Classification
Andrés Felipe Patiño Aldana1, Omar Echeverría2, Ana María Chinchilla2, Nora Contreras2, Camilo Peralta2, Nicolás Caballero2, Hanna Valentina Tovar1, José Manuel Palacio1, Verónica Uribe2, Matteo Mineo2, Fernando Velandia3, Daniel Silgado4, William Mauricio Riveros Castillo5, Alejandro Oyono Ondo Méndez1, Dora Janeth Fonseca Mendoza2
1Clinical Research Group. SiBIO Research Incubator, 2Centro de Investigación en Genética y Genómica (CIGGUR), Universidad del Rosario, 3Consultant Neuropathologist, 4Departamento de Biología Molecular, Genética Molecular de Colombia, 5Centro de Investigación en Neurocirugía (CIEN), Hospital Universitario Mayor Mederi & Hospital Universitario la Samaritana
Objective:
We aimed to conduct a comprehensive molecular characterization of gliomas following the WHO-CNS5-2021 classification and to compare it with histopathological evaluations.
Background:
Gliomas are the most frequent primary malignant brain tumors. The current classification includes histopathologic features and molecular biomarkers with defined roles in pathogenesis, therapeutics, and prognosis. Adopting these translational methods in clinical practice represents an educational, clinical, and resource challenge that may hinder its implementation. Understanding how this paradigm shift might change previous histopathological diagnoses could improve adherence to molecular diagnosis practices.
Design/Methods:
We conducted a descriptive study. We analyzed glioma DNA using Sanger sequencing, multiplex ligation-dependent probe amplification (MLPA), and methylation-specific MLPA (MS-MLPA). We assessed mutations in IDH1, IDH2, TERT, EGFR amplification, homozygous CDKN2A deletion, 1p/19q codeletion and the methylation status of the MGMT promoter.
Results:
We included 22 Colombian patients, 41% of whom were male. Glioblastoma was the most frequent histopathological diagnosis (68.2%, n=15), followed by astrocytoma (9%, n=2) and oligodendroglioma (4.5%, n =1). After molecular characterization, most patients were identified as IDH-wildtype (59% n=13) and were classified as glioblastoma IDH-Wildtype WHO grade 4. We identified non-canonical IDH1 mutations (p.R132S) in 14% (n=3) samples. MGMT promoter was hyper-methylated in 40% of the glioblastomas IDH-wildtype. The discordance rate between histopathological and molecular diagnosis was 21%, primarily due to tumors initially described as glioblastomas being reclassified as astrocytomas IDH-mutant.
Conclusions:
Our findings highlight the critical importance of integrating molecular profiling into diagnostic workflow for gliomas. One out of five gliomas were reclassified with a more benign phenotype using the WHO-CNS5-2021 molecular classification. This discrepancy underscores the limitations of traditional methods and emphasizes the need for routine molecular assessment to ensure accurate diagnosis. A broader adoption of molecular diagnosis might enhance diagnosis precision and therapeutic strategies. Further molecular profiling is needed to refine classification systems and identify new biomarkers.
Disclaimer: Abstracts were not reviewed by Neurology® and do not reflect the views of Neurology® editors or staff.