DNA methylation-based classification of hematolymphoid neoplasms

作者：Annapurna Saksena, Christin Siewert, Rust Turakulov, Omkar Singh, Zied Abdullaev, Ting Zhou, Jeremiah X. Karrs, João Victor Alves de Castro, Deanna Riley, Shannon Skarshaug, Valerie Zgonc, Lana Sultanaeva, Ruifeng Teng, Liqiang Xi, Robert J. Kreitman, Andreas von Deimling, Martí Duran‐Ferrer, Jose Ignacio Martin-Subero, Elı́as Campo, Benjamin Englert, Julia Richter, Wolfram Klapper, Ioannis Anagnostopoulos, A Rosenwald, Marion Wobser, F Fend, Leticia Quintanilla‐Martínez, Gerardo Ferrer, Manel Esteller, Miguel Á. Piris, Michael Hummel, Stefania Pittaluga, Mark Raffeld, Elaine S. Jaffe, David Capper, Kenneth Aldape · 发表于：Blood Advances · 年份：2026 · DOI：10.1182/bloodadvances.2024015275 · 被引用次数：1 · 研究领域：Epigenetics and DNA Methylation、Digital Imaging for Blood Diseases、AI in cancer detection

ABSTRACT: Accurate pathologic diagnosis of hematolymphoid neoplasms (HLN) is often challenging because of their complexity and heterogeneity. Genome-wide DNA methylation profiling has emerged as a valuable tool for tumor classification and diagnosis across malignancies, such as central nervous system neoplasms. In this study, we explored the role of DNA methylation-based profiling in HLN. We generated the largest crossplatform HLN methylome cohort to date (1156 samples) and identified 44 reproducible methylation classes (MCs) that aligned closely with entities defined by the 5th edition of the World Health Organization (WHO) Classification/the International Consensus Classification (ICC), including subgroups with clinical and biological relevance. Copy number alterations were also inferred for all MCs. Additionally, a machine learning-based DNA methylation classifier was developed and validated on an independent test set, demonstrating a modest 58% high-confidence score rate, nevertheless, with a robust 97% concordance with the original diagnosis in these high-confidence score cases. Additionally, in discrepant high-confidence score cases, although few, the methylation classifier demonstrated its potential utility as an adjunct, in which, on additional review, the diagnosis was revised in favor of the methylation prediction in most of the cases (5/8 discrepant cases). Tumor purity was a significant contributor for a substantial proportion of low-confidence score samples (scor...