A microenvironment-determined risk continuum refines subtyping in meningioma and reveals determinants of machine learning-based tumor classification
Sybren L. N. Maas, Yiheng Tang, Eric Y. Stutheit-Zhao, Ramin Rahmanzade, Christina Blume, Thomas Hielscher, Ferdinand Zettl, Salvatore Benfatto, D Calafato, Martin Sill, J Kada Benotmane, Yahaya A Yabo, Felix Behling, Abigail K. Suwala, Helin Kardo, Michael Ritter, M. Peyre, Roman Sankowski, Konstantin Okonechnikov, Philipp Sievers, Areeba Patel, D. E. Reuss, Mirco Friedrich, Damian Stichel, Daniel Schrimpf, T P P van den Bosch, Katja Beck, Hans‐Georg Wirsching, Gerhard Jungwirth, C. Oliver Hanemann, K Lamszus, Nima Etminan, A. Unterberg, C. Mawrin, Marc Remke, Olivier Ayrault, Peter Lichter, Guido Reifenberger, Michael Platten, Tim Kacprowski, Markus List, Josch K. Pauling, Jan Baumbach, Till Milde, Rachel Grossmann, Zvi Ram, Miriam Ratliff, Jan‐Philipp Mallm, Marian C. Neidert, Eelke M. Bos, M Prinz, Michael Weller, Till Acker, Felix J. Hartmann, Matthias Preusser, G. Tabatabai, Christel Herold-Mende, Sandro M. Krieg, David T. W. Jones, Stefan M. Pfister, Wolfgang Wick, Michel Kalamarides, Andreas von Deimling, Dieter Henrik Heiland, Volker Hovestadt, Moritz Gerstung, Matthias Schlesner, Felix Sahm
Abstract
Classification of tumors in neuro-oncology today relies on molecular patterns (mostly DNA methylation) and their machine learning-supported interpretation. Understanding the process of algorithmic interpretation is essential for safe application in clinical routine. This is paradigmatically true for the most common primary intracranial tumor in adults, meningioma. Here, by applying multiomic profiling and multiple lines of orthogonal computational evaluation in multiple independent datasets, we found that not only tumor cell characteristics but also incremental changes in the tumor microenvironment (TME) have impact on epigenetic meningioma classification and clinical outcome. Besides revealing the decisive role of non-neoplastic cells in the CNS methylation classifier, this challenges the model of distinct meningioma subgroups toward a TME-determined risk continuum. This refines current controversies in molecular meningioma subtyping. In addition, we apply these learnings to devise and validate a simple diagnostic approach for increased clinical prediction accuracy based on immunohistochemistry, which is also applicable in resource-limited settings.