Methylation biomarkers can distinguish pleural mesothelioma from healthy pleura and other pleural pathologies

Molecular Oncology 2025 November 14 [Link]

Janah Vandenhoeck, Nele De Meulenaere, Thomas Vanpoucke, Joe Ibrahim, Dieter Peeters, Suresh Krishan Yogeswaran, Wen Wen, Paul Van Schil, Jeroen M H Hendriks, Jo Raskin, Jan van Meerbeeck, Guy Van Camp, Ken Op de Beeck

Abstract

Pleural mesothelioma (PM) is a rare and aggressive cancer that often requires multiple diagnostic procedures before a definitive diagnosis can be made. To improve diagnostic accuracy, we developed a DNA methylation-based biomarker assay capable of distinguishing PM from healthy pleura and other pleural pathologies. Using Infinium EPIC array data, we identified 744 hypermethylated CpG sites in PM as candidate biomarkers. These were validated in silico using external datasets, yielding a high mean AUC of 0.935. Clinical validation was performed using IMPRESS, a novel bisulfite-free methylation detection technique that enables simultaneous analysis of thousands of CpG sites. A two-step classifier approach was applied: the first model differentiated tumoral from nontumoral pleura with 89.2% sensitivity and 93.5% specificity, while the second model distinguished PM from pleural metastases with 85.2% sensitivity and 100% specificity. These results demonstrate that our methylation-based biomarker panel offers a highly accurate and minimally invasive tool for differentiating PM from other pleural conditions, potentially streamlining the diagnostic process and improving clinical decision-making.