System Maintenance Notice

Please note that maintenance work is scheduled between 4 AM and 5 AM - CET. During this time, you may experience reduced performance. Thanks for your patience.

Publications

Journal Article Ahead of print

PubMed: 41637936 DOI: 10.1016/j.ebiom.2026.106147

Autoencoders reveal polyunsaturated fatty acids (PUFA)-Related metabolic signature linked to cancer risk.

EBioMedicine 124:106147

Authors: Marie Breeur, Joshua Atkins, Laia Peruchet-Noray, Lisa Bonheme, Nicolas Alcala, Laure Dossus, Mazda Jenab, Mattias Johansson, Sabina Rinaldi, Ruth C Travis, Christian Bork, Christina C Dahm, Anne Tjønneland, Anja Olsen, Sabine Naudin, Seehyun Park, Therese Truong, Verena Katzke, Charlotte Le Cornet, Matthias B Schulze, Marcela Prada, Carlotta Sacerdote, Benedetta Bendinelli, Claudia Agnoli, Fabrizio Pasanisi, José María Gálvez-Navas, Marcela Guevara, Alicia K Heath, James Yarmolinsky, Marc J Gunter, Pietro Ferrari, Karl Smith-Byrne, Vivian Viallon

Affiliations

  • Nutrition and Metabolism Branch, International Agency for Research on Cancer, Lyon, France; Cancer Epidemiology Unit, Oxford Population Health, University of Oxford, Oxford, United Kingdom.
  • Cancer Epidemiology Unit, Oxford Population Health, University of Oxford, Oxford, United Kingdom.
  • Nutrition and Metabolism Branch, International Agency for Research on Cancer, Lyon, France.
  • Genetic Epidemiology Branch, International Agency for Research on Cancer, Lyon, France.
  • Department of Cardiology, Aalborg University Hospital, Denmark; Department of Clinical Medicine, Aalborg University, Denmark.
  • Department of Public Health, Aarhus University, Denmark.
  • Danish Cancer Institute, Denmark; Department of Public Health, University of Copenhagen, Denmark.
  • Danish Cancer Institute, Denmark; Department of Public Health, University of Århus, Denmark.
  • Paris-Saclay University, UVSQ, Inserm, Gustave Roussy, CESP, Villejuif, France.
  • Division of Cancer Epidemiology, German Cancer Research Center (DKFZ), Heidelberg, Germany.
  • German Institute of Human Nutrition, Department of Molecular Epidemiology, Nuthetal, Germany.
  • Unit of Cancer Epidemiology Città della Salute e della Scienza University-Hospital, Turin, Italy.
  • Clinical Epidemiology Unit, Institute for cancer research, prevention and clinical network (ISPRO) Florence, Italy.
  • Epidemiology and Prevention Unit, Fondazione IRCCS Istituto Nazionale dei Tumori, Milan, Italy.
  • Dipartimento di Medicina Clinica e Chirurgia, Federico II University, Naples, Italy.
  • Biosanitary Research Institute ibs.GRANADA, Granada, Spain; Centro de Investigación Biomédica en Red en Epidemiología y Salud Pública (CIBERESP), Madrid, Spain; Andalusian School of Public Health, Granada, Spain.
  • Centro de Investigación Biomédica en Red en Epidemiología y Salud Pública (CIBERESP), Madrid, Spain; Instituto de Salud Pública y Laboral de Navarra, Pamplona, Spain; Navarra Institute for Health Research (IdiSNA), Pamplona, Spain.
  • Cancer Epidemiology and Prevention Research Unit, School of Public Health, Imperial College London, London, United Kingdom.
  • Nutrition and Metabolism Branch, International Agency for Research on Cancer, Lyon, France; Department of Epidemiology and Biostatistics, School of Public Health, Faculty of Medicine, Imperial College London, London, UK.
  • Nutrition and Metabolism Branch, International Agency for Research on Cancer, Lyon, France.

Abstract

BACKGROUND

Metabolomics is a valuable tool for characterising biological mechanisms involved in cancer development, but produces complex datasets with intricate interdependencies. While linear dimension reduction techniques such as principal component analysis (PCA), have proven useful to summarise informative hidden patterns, biological evidence suggests metabolic relationships extend beyond linearity. Non-linear dimension reduction techniques, such as autoencoders (AEs), may identify more meaningful components.

METHODS

We applied AEs and PCA to metabolomic data available for 5828 matched case-control pairs from 8 cancer-specific case-control studies nested within the European Prospective Investigation into Cancer and Nutrition (EPIC) cohort, and compared their performance. We evaluated the association between components identified by AEs and PCA with cancer risk, and explored the biological interpretation of components through their association with genetic factors and selected biomarkers.

FINDINGS

PCA and AEs showed similar reconstruction performance. PCA's first component (PCA.1) captured phosphatidylcholines (PCs) as the primary source of variability and was associated with cancer risk. Conversely, AEs decomposed PC metabolism into two components, one of which exhibited a stronger association with cancer risk than PCA.1. Unlike PCA.1, this component was strongly associated with genetic variants mapping to the TMEM258 and FADS genes, key in polyunsaturated fatty acids (PUFA) biosynthesis and regulation. Consistently, the AE component demonstrated stronger associations with circulating omega-3 and omega-6 PUFA levels than PCA.1.

INTERPRETATION

Linear methods remain adequate for general dimension reduction. However, AEs better captured specific pathways, identifying a component reflecting perturbations in PUFA metabolism associated with cancer risk.

FUNDING

World Cancer Research Fund (IIG_FULL_2022_013).

MeSH

Biomedical Research, Neoplasms, Metabolomics, Fatty Acids