Spatially resolved transcriptomics and graph-based deep learning improve accuracy of routine CNS tumor diagnostics

Journal article

M. Ritter, C. Blume, Yiheng Tang, A. Patel, B. Patel, N. Berghaus, Jasim Kada Benotmane, J. Kueckelhaus, Y. Yabo, Junyi Zhang, E. Grabis, Giulia Villa, David Niklas Zimmer, Amir Khriesh, P. Sievers, Z. Seferbekova, F. Hinz, V. Ravi, M. Seiz-Rosenhagen, M. Ratliff, C. Herold-Mende, Oliver Schnell, J. Beck, Wolfgang Wick, Andreas von Deimling, M. Gerstung, D. Heiland, F. Sahm
Nature Cancer, vol. 6(2), 2025, pp. 292-306

DOI: 10.1038/s43018-024-00904-z

Semantic Scholar DOI PubMedCentral PubMed

Cite

APA Click to copy
Ritter, M., Blume, C., Tang, Y., Patel, A., Patel, B., Berghaus, N., … Sahm, F. (2025). Spatially resolved transcriptomics and graph-based deep learning improve accuracy of routine CNS tumor diagnostics. Nature Cancer, 6(2), 292–306. https://doi.org/ 10.1038/s43018-024-00904-z

Chicago/Turabian Click to copy
Ritter, M., C. Blume, Yiheng Tang, A. Patel, B. Patel, N. Berghaus, Jasim Kada Benotmane, et al. “Spatially Resolved Transcriptomics and Graph-Based Deep Learning Improve Accuracy of Routine CNS Tumor Diagnostics.” Nature Cancer 6, no. 2 (2025): 292–306.

MLA Click to copy
Ritter, M., et al. “Spatially Resolved Transcriptomics and Graph-Based Deep Learning Improve Accuracy of Routine CNS Tumor Diagnostics.” Nature Cancer, vol. 6, no. 2, 2025, pp. 292–306, doi: 10.1038/s43018-024-00904-z.

BibTeX Click to copy

@article{m2025a,
  title = {Spatially resolved transcriptomics and graph-based deep learning improve accuracy of routine CNS tumor diagnostics},
  year = {2025},
  issue = {2},
  journal = {Nature Cancer},
  pages = {292-306},
  volume = {6},
  doi = { 10.1038/s43018-024-00904-z},
  author = {Ritter, M. and Blume, C. and Tang, Yiheng and Patel, A. and Patel, B. and Berghaus, N. and Benotmane, Jasim Kada and Kueckelhaus, J. and Yabo, Y. and Zhang, Junyi and Grabis, E. and Villa, Giulia and Zimmer, David Niklas and Khriesh, Amir and Sievers, P. and Seferbekova, Z. and Hinz, F. and Ravi, V. and Seiz-Rosenhagen, M. and Ratliff, M. and Herold-Mende, C. and Schnell, Oliver and Beck, J. and Wick, Wolfgang and von Deimling, Andreas and Gerstung, M. and Heiland, D. and Sahm, F.}
}

Abstract

The diagnostic landscape of brain tumors integrates comprehensive molecular markers alongside traditional histopathological evaluation. DNA methylation and next-generation sequencing (NGS) have become a cornerstone in central nervous system (CNS) tumor classification. A limiting requirement for NGS and methylation profiling is sufficient DNA quality and quantity, which restrict its feasibility. Here we demonstrate NePSTA (neuropathology spatial transcriptomic analysis) for comprehensive morphological and molecular neuropathological diagnostics from single 5-µm tissue sections. NePSTA uses spatial transcriptomics with graph neural networks for automated histological and molecular evaluations. Trained and evaluated across 130 participants with CNS malignancies and healthy donors across four medical centers, NePSTA predicts tissue histology and methylation-based subclasses with high accuracy. We demonstrate the ability to reconstruct immunohistochemistry and genotype profiling on tissue with minimal requirements, inadequate for conventional molecular diagnostics, demonstrating the potential to enhance tumor subtype identification with implications for fast and precise diagnostic workup.