Detection of diagnostic and prognostic methylation-based signatures in liquid biopsy specimens from patients with meningiomas
Grayson Herrgott, James M. Snyder, Ruicong She, Tathiane M. Malta, Thaís S. Sabedot, Ian Lee, Jacob Pawloski, Guilherme Gozzoli Podolsky Gondim, Karam Asmaro, Jiaqi Zhang, Cara E. Cannella, Kevin Nelson, Bartow Thomas, Ana C. deCarvalho, Laura Hasselbach, Kelly Tundo, Rehnuma Newaz, Andrea Transou, Natalia S. Morosini, Victor Francisco, Laila Poisson, Dhananjay Chitale, Abir Mukherjee, Maritza Mosella, Adam Robin, Tobias Walbert, Mark L. Rosenblum, Tom Mikkelsen, Steven N. Kalkanis, Daniela Pretti da Cunha Tirapelli, Daniel J. Weisenberger, Carlos Gilberto Carlotti, Jack Rock, Ana Valeria Castro, Houtan Noushmehr
Abstract
Recurrence of meningiomas is unpredictable by current invasive methods based on surgically removed specimens. Identification of patients likely to recur using noninvasive approaches could inform treatment strategy, whether intervention or monitoring. In this study, we analyze the DNA methylation levels in blood (serum and plasma) and tissue samples from 155 meningioma patients, compared to other central nervous system tumor and non-tumor entities. We discover DNA methylation markers unique to meningiomas and use artificial intelligence to create accurate and universal models for identifying and predicting meningioma recurrence, using either blood or tissue samples. Here we show that liquid biopsy is a potential noninvasive and reliable tool for diagnosing and predicting outcomes in meningioma patients. This approach can improve personalized management strategies for these patients.