DNA methylation-array interlaboratory comparison trial demonstrates highly reproducible paediatric CNS tumour classification across 13 international centres.

in Neuropathology and applied neurobiology by Mihaela Chirica, Philipp Jurmeister, Daniel Teichmann, Arend Koch, Eilís Perez, Simone Schmid, Michèle Simon, Pablo Hernáiz Driever, Carina Bodden, Cornelis M van Tilburg, Emily C Hardin, Cinzia Lavarino, Jürgen Hench, David Scheie, Jane Cryan, Ales Vicha, Francesca R Buttarelli, An Michiels, Christine Haberler, Paulette Barahona, Bastiaan B J Tops, Tom Jacques, Tore Stokland, Olaf Witt, David T W Jones, David Capper

TLDR

  • A prospective laboratory comparison trial demonstrated high technical reproducibility and consistency in DNA methylation profiling for low-grade gliomas across 12 international centres, highlighting the potential for DNA methylation analysis to harmonise brain tumour diagnostics.

Abstract

DNA methylation profiling, recently endorsed by the World Health Organisation (WHO) as a pivotal diagnostic tool for brain tumours, most commonly relies on bead arrays. Despite its widespread use, limited data exist on the technical reproducibility and potential cross-institutional differences. The LOGGIC Core BioClinical Data Bank registry conducted a prospective laboratory comparison trial with 12 international laboratories to enhance diagnostic accuracy for paediatric low-grade gliomas, focusing on technical aspects of DNA methylation data generation and profile interpretation under clinical real-time conditions. Four representative low-grade gliomas of distinct histologies were centrally selected, and DNA extraction was performed. Participating laboratories received a DNA aliquot and performed the DNA methylation-based classification and result interpretation without knowledge of tumour histology. Additionally, participants were required to interpret the copy number profile derived from DNA methylation data and conduct DNA sequencing of the BRAF hotspot p.V600 due to its relevance for low-grade gliomas. Results had to be returned within 30 days. High technical reproducibility was observed, with a median pairwise correlation of 0.99 (range 0.94-0.99) between coordinating laboratory and participants. DNA methylation-based tumour classification and copy number profile interpretation were consistent across all centres, and BRAF mutation status was accurately reported for all cases. Eleven out of 12 centres successfully reported their analysis within the 30-day timeframe. Our study demonstrates remarkable concordance in DNA methylation profiling and profile interpretation across 12 international centres. These findings underscore the potential contribution of DNA methylation analysis to the harmonisation of brain tumour diagnostics.

Overview

  • The study investigated the technical reproducibility and potential cross-institutional differences in DNA methylation profiling for brain tumours, specifically low-grade gliomas, using a prospective laboratory comparison trial.
  • Twelve international laboratories participated in the study, which aimed to assess the consistency of DNA methylation-based classification, copy number profile interpretation, and BRAF mutation status reporting.
  • The study used a centrally selected DNA aliquot, and participants performed DNA extraction, methylation-based classification, and result interpretation without knowledge of tumour histology.

Comparative Analysis & Findings

  • High technical reproducibility was observed, with a median pairwise correlation of 0.99 (range 0.94-0.99) between the coordinating laboratory and participants.
  • DNA methylation-based tumour classification and copy number profile interpretation were consistent across all centres, with unanimous agreement on BRAF mutation status reporting.
  • Eleven out of 12 centres successfully reported their analysis within the 30-day timeframe, demonstrating the feasibility of rapid and reproducible DNA methylation profiling under real-time clinical conditions.

Implications and Future Directions

  • The study's findings underscore the potential contribution of DNA methylation analysis to the harmonisation of brain tumour diagnostics, reducing uncertainty and variability in diagnosis and treatment.
  • Future studies can build on this study's methodology to explore the role of DNA methylation profiling in other cancer types and to develop standardised protocols for clinical implementation.
  • The use of DNA methylation analysis could also facilitate the development of minimally invasive diagnostics, enabling earlier diagnosis and treatment initiation for brain tumour patients.
Read Full Article