EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Prognostic pan-cancer and single-cancer models: A large-scale analysis using a real-world clinico-genomic database

Sarah F McGough iD, Svetlana Lyalina, Devin Incerti, Yunru Huang, Stefka Tyanova, Kieran Mace iD, Chris Harbron, Ryan Copping, Balasubramanian Narasimhan, Robert Tibshirani

PLOS One · 13.02.2026 · 21 (2) · e0341355

DOI: 10.1371/journal.pone.0341355 ↗

Kurz gefasst

Fachartikel in der Zeitschrift PLOS One, online erschienen am 13.02.2026. Verfasst von Sarah F McGough und 9 weiteren Autor:innen. Beteiligte Einrichtung: Computational Sciences, Genentech Research and Early Development, Genentech, Inc., South San Francisco, California, United States of America. Hauptthemen laut MEDLINE: Neoplasms. Die Arbeit wurde bisher 2-mal zitiert (Crossref) und nennt 84 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: NIGMS NIH HHS, Genentech, Inc..

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Prognostic models in oncology have a profound impact on personalized cancer care and patient profiling, but tend to be heterogeneously developed and implemented in narrow patient cohorts. Here, we develop and benchmark multiple machine learning models to predict survival in pan-cancer and 16 single-cancer settings using a de-identified clinico-genomic database of 28,079 US patients with cancer. We identify key predictors of cancer prognosis, including 15 shared across seven or more cancer types, revealing strong consistency in cancer prognostic factors. We demonstrate that pan-cancer models generally outperform or match single-cancer models in predicting survival and risk stratifying patients, especially in smaller cancer cohorts, suggesting a unique transfer learning advantage of pan-cancer models. This work demonstrates the potential of pan-cancer approaches in enhancing the accuracy and applicability of prognostic models in oncology, paving the way for more personalized and effective cancer care strategies.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

NeoplasmsHumansPrognosisPredictive Learning ModelsGenomicsMachine LearningDatabases, Genetic

Autor:innen und Institutionen

  • Computational Sciences, Genentech Research and Early Development, Genentech, Inc., South San Francisco, California, United States of America.
  • Safety Science, Product Development, Genentech, Inc., South San Francisco, California, United States of America.
  • Data Sciences, Product Development, Genentech, Inc., South San Francisco, California, United States of America.
  • Data Sciences, Product Development, F. Hoffmann-La Roche, Ltd., Welwyn Garden City, United Kingdom.
  • Department of Statistics, Stanford University, Stanford, California, United States of America.
  • Department of Biomedical Data Sciences, Stanford University, Stanford, California, United States of America.

ORCID-Profile: Sarah F McGough, Kieran Mace

Förderung

  • NIGMS NIH HHS
  • Genentech, Inc.

Zitieren

APA

McGough, S. F., Lyalina, S., Incerti, D., Huang, Y., Tyanova, S., Mace, K., Harbron, C., Copping, R., Narasimhan, B., & Tibshirani, R. (2026). Prognostic pan-cancer and single-cancer models: A large-scale analysis using a real-world clinico-genomic database. PLOS One, 21(2), e0341355. https://doi.org/10.1371/journal.pone.0341355

Vancouver

McGough SF, Lyalina S, Incerti D, Huang Y, Tyanova S, Mace K, et al. Prognostic pan-cancer and single-cancer models: A large-scale analysis using a real-world clinico-genomic database. PLOS One. 2026;21(2):e0341355. doi:10.1371/journal.pone.0341355

BibTeX
@article{McGough2026,
  author = {McGough, Sarah F and Lyalina, Svetlana and Incerti, Devin and Huang, Yunru and Tyanova, Stefka and Mace, Kieran and Harbron, Chris and Copping, Ryan and Narasimhan, Balasubramanian and Tibshirani, Robert},
  title = {Prognostic pan-cancer and single-cancer models: A large-scale analysis using a real-world clinico-genomic database},
  journal = {PLOS One},
  year = {2026},
  volume = {21},
  number = {2},
  pages = {e0341355},
  doi = {10.1371/journal.pone.0341355}
}
Kostenlos starten →