Sill Martin, Schrimpf Daniel, Patel Areeba, Sturm Dominik, Jäger Natalie, Sievers Philipp, Schweizer Leonille, Banan Rouzbeh, Reuss David, Suwala Abigail, Korshunov Andrey, Stichel Damian, Wefers Annika K, Hau Ann-Christin, Boldt Henning, Harter Patrick N, Abdullaev Zied, Benhamida Jamal, Teichmann Daniel, Koch Arend, Hench Jürgen, Frank Stephan, Hasselblatt Martin, Mansouri Sheila, Díaz de Ståhl Theresita, Serrano Jonathan, Ecker Jonas, Selt Florian, Taylor Michael, Ramaswamy Vijay, Cavalli Florence, Berghoff Anna S, Bison Brigitte, Blattner-Johnson Mirjam, Buchhalter Ivo, Buslei Rolf, Calaminus Gabriele, Dikow Nicola, Dohmen Hildegard, Euskirchen Philipp, Fleischhack Gudrun, Gajjar Amar, Gerber Nicolas U, Gessi Marco, Gielen Gerrit H, Gnekow Astrid, Gottardo Nicholas G, Haberler Christine, Hamelmann Stefan, Hans Volkmar, Hansford Jordan R, Hartmann Christian, Heppner Frank L, Driever Pablo Hernaiz, von Hoff Katja, Thomale Ulrich W, Tippelt Stephan, Frühwald Michael C, Kramm Christof M, Schüller Ulrich, Schittenhelm Jens, Schuhmann Martin U, Stein Marco, Ketteler Petra, Ladanyi Marc, Jabado Nada, Jones Barbara C, Jones Chris, Karajannis Matthias A, Ketter Ralf, Kohlhof Patricia, Kordes Uwe, Reinhardt Annekathrin, Kölsche Christian, Lamszus Katrin, Lichter Peter, Maas Sybren L N, Mawrin Christian, Milde Till, Mittelbronn Michel, Monoranu Camelia-Maria, Mueller Wolf, Mynarek Martin, Northcott Paul A, Pajtler Kristian W, Paulus Werner, Perry Arie, Blümcke Ingmar, Plate Karl H, Platten Michael, Preusser Matthias, Pietsch Torsten, Prinz Marco, Reifenberger Guido, Kristensen Bjarne W, Kool Marcel, Hovestadt Volker, Ellison David W, Jacques Thomas S, Varlet Pascale, Etminan Nima, Acker Till, Weller Michael, White Christine L, Witt Olaf, Herold-Mende Christel, Debus Jürgen, Krieg Sandro, Wick Wolfgang, Snuderl Matija, Aldape Ken, Brandner Sebastian, Hawkins Cynthia, Horbinski Craig, Thomas Christian, Wesseling Pieter, von Deimling Andreas, Capper David, Pfister Stefan M, Jones David Tw, Sahm Felix
medRxiv. 2025 May 29:2025.05.28.25328344. doi: 10.1101/2025.05.28.25328344.
DNA methylation-based classification is integral to contemporary neuro-oncological diagnostics, as highlighted by the current World Health Organization (WHO) classification of central nervous system (CNS) tumors. We introduce the Heidelberg CNS Tumor Methylation Classifier version 12.8 (v12.8), trained using 7,495 methylation profiles, thereby expanding recognized tumor types from 91 classes in the previously published v11 to 184 subclasses in v12.8. This expansion was primarily driven by novel tumor types discovered in our large website-derived repository and through global collaborations, further elucidating the heterogeneity of CNS tumors. Utilizing a random forest-based methodology, the classifier was rigorously validated through five-fold nested cross-validation, achieving a 95% subclass-level accuracy and a Brier score of 0.028, indicative of well-calibrated probability estimates. The hierarchical output structure facilitates comprehensive interpretation, allowing clinicians to assess subclass and aggregate class-level probabilities for informed decision-making. Comparative analyses demonstrate that v12.8 surpasses previous versions as well as traditional WHO-based diagnostics across diverse tumor cohorts. These advancements underscore the enhanced precision and practical utility of the updated Heidelberg CNS Tumor Methylation Classifier, reinforcing the pivotal role of DNA methylation profiling in personalized neuro-oncological care.
正如世界卫生组织（WHO）当前对中枢神经系统（CNS）肿瘤的分类所强调的那样，基于DNA甲基化的分类是当代神经肿瘤学诊断的重要组成部分。我们推出了海德堡中枢神经系统肿瘤甲基化分类器版本12.8（v12.8），该分类器使用7495个甲基化谱进行训练，从而将已识别的肿瘤类型从先前发布的v11版本中的91类扩展到v12.8版本中的184个子类。这种扩展主要是由我们从大型网站衍生库中发现的新型肿瘤类型以及通过全球合作推动的，进一步阐明了中枢神经系统肿瘤的异质性。利用基于随机森林的方法，该分类器通过五重嵌套交叉验证进行了严格验证，子类水平的准确率达到95%，布里尔得分0.028，表明概率估计校准良好。分层输出结构便于全面解读，使临床医生能够评估子类和总体类水平的概率，以便做出明智的决策。比较分析表明，在不同肿瘤队列中，v12.8优于以前的版本以及传统的基于WHO的诊断方法。这些进展强调了更新后的海德堡中枢神经系统肿瘤甲基化分类器的更高精度和实际效用，强化了DNA甲基化谱在个性化神经肿瘤护理中的关键作用。
