Portal de Periódicos da CAPES

Radiomic Features Define Risk and Are Linked to DNA Methylation Attributes in Primary CNS Lymphoma

2022; RELX Group (Netherlands); Linguagem: Inglês

10.2139/ssrn.4022235

1556-5068

Karl‐Heinz Nenning, Johanna Gesperger, Julia Furtner, Amelie Nemc, Thomas Roetzer-Pejrimovsky, Seung Won Choi, Christian Mitter, Stefan L. Leber, Johannes Hofmanninger, Johanna Klughammer, Bekir Ergüener, Marlies Bauer, Martina Brada, Tanisa Brandner-Kokalj, Christian F. Freyschlag, Astrid Grams, Johannes Haybaeck, Selma Hoenigschnabl, Markus Hoffermann, Sarah Iglseder, Barbara Kiesel, Melitta Kitzwoegerer, Waltraud Kleindienst, Franz Marhold, Patrizia Moser, Stefan Oberndorfer, Daniel Pinggera, Florian Scheichel, Camillo Sherif, Guenther Stockhammer, Martin Stultschnig, Claudius Thomé, Johannes Trenkler, Tadea Urbanits, Serge Weis, Georg Widhalm, Franz Wuertz, Matthias Preusser, Bernhard Baumann, Ingrid Simonitsch‐Klupp, Do‐Hyun Nam, Christoph Bock, Georg Langs, Adelheid Wöehrer,

Glioma Diagnosis and Treatment

Background: The prognostic roles of clinical and laboratory markers have been exploited to model risk in patients with primary CNS lymphoma (PCNSL), but these approaches do not fully explain the observed variation in outcome. Here we present an extended framework of phenotype-epigenotype correlations that reveal novel prognostic constellations and enable prioritizing epigenetic therapy. Methods: In this retrospective discovery and validation study, we strike a balance between radiomic feature-driven analysis of medical images and supervised bioinformatic integration of DNA methylation profiles. We integrate both data modalities synergistically using machine learning-based prediction and cross-domain alignment models. Ultimately, we validate the most relevant biological associations in tumor tissues and cell lines. Findings: We leverage a cohort of 191 patients across 9 sites in Austria and one site in South Korea, and use T1-weighted contrast-enhanced magnetic resonance imaging to derive a radiomic risk score that consists of 20 features, predominantly including textural features. We determine the risk score as strong and independent predictor of survival (multivariate HR=6·56) confirming its prognostic rating ability in an external validation cohort. Radiomic features align with DNA methylation sites in distinct, biologically meaningful ways, and radiomic risk is predictable from selected DNA methylation sites (AUC=0·78). Ultimately, gene-regulatory differences between radiomically defined risk groups converge on bcl6 binding activity, which is put forth as testable treatment strategy in a subset of patients. Interpretation: The radiomic risk score is a robust and complementary predictor of survival and is reflected at the level of DNA methylation in PCNSL. Assessing risk and selecting epigenetic treatment based on imaging phenotypes represents a huge step forward, and the ability to define radiomic risk groups may provide a concept on which to advance prognostic modeling and precision therapy for this aggressive cancer.Funding Information: This work was supported by the Anniversary fund of the Austrian National Bank under grant # 16725 to AW. Declaration of Interests: MP has received honoraria for lectures, consultation or advisory board participation from the following for-profit companies: Bayer, Bristol-Myers Squibb, Novartis, Gerson Lehrman Group (GLG), CMC Contrast, GlaxoSmithKline, Mundipharma, Roche, BMJ Journals, MedMedia, Astra Zeneca, AbbVie, Lilly, Medahead, Daiichi Sankyo, Sanofi, Merck Sharp & Dome, Tocagen, Adastra, Gan & Lee Pharmaceuticals. The remaining authors KHN, JG, JF, AN, TRP, SWC, CM, SLL, JH, JK, BE, MB, MB, TBK, CFF, AG, JH, SH, MH, SI, BK, MK, WK, FM, PM, SO, DP, FS, CS, GS, MS, CT, JT, TU, SW, GW, FW, BB, ISK, DHN, CB, GL, and AW have no conflicts of interest to declare.Ethics Approval Statement: The study was approved by the institutional review board of the Medical University of Vienna under #1861-2018 and informed consent was obtained from all patients.