Article Text

Download PDFPDF
Original research
Associations between metabolites and pancreatic cancer risk in a large prospective epidemiological study
  1. Rachael Stolzenberg-Solomon1,
  2. Andriy Derkach2,
  3. Steven Moore1,
  4. Stephanie J Weinstein1,
  5. Demetrius Albanes1,
  6. Joshua Sampson2
  1. 1 Metabolic Nutrition Branch, Division of Cancer Epidemiology and Genetics, National Cancer Institute, Rockville, Maryland, USA
  2. 2 Biostatistics Branch, Division of Cancer Epidemiology and Genetics, National Cancer Institute, Rockville, Maryland, USA
  1. Correspondence to Dr Rachael Stolzenberg-Solomon, Metabolic Nutrition Branch, National Cancer Institute Division of Cancer Epidemiology and Genetics, Rockville, MD 20850, USA; rs221z{at}


Objective To assess whether prediagnostic metabolites were associated with incident pancreatic ductal adenocarcinoma (PDAC) in a prospective cohort study.

Design We conducted an untargeted analysis of 554 known metabolites measured in prediagnostic serum (up to 24 years) to determine their association with incident PDAC in a nested case-control study of male smokers (372 matched case-control sets) and an independent nested case-control study that included women and non-smokers (107 matched sets). Metabolites were measured using Orbitrap Elite or Q-Exactive high-resolution/accurate mass spectrometers. Controls were matched to cases by age, sex, race, date of blood draw, and follow-up time. We used conditional logistic regression adjusted for age to calculate ORs and 95% CIs for a 1 SD increase in log-metabolite level separately in each cohort and combined the two ORs using a fixed-effects meta-analysis.

Results Thirty-one metabolites were significantly associated with PDAC at a false discovery rate <0.05 with 12 metabolites below the Bonferroni-corrected threshold (p<9.04×10–5). Similar associations were observed in both cohorts. The dipeptides glycylvaline, aspartylphenylalanine, pyroglutamylglycine, phenylalanylphenylalanine, phenylalanylleucine and tryptophylglutamate and amino acids aspartate and glutamate were positively while the dipeptides tyrosylglutamine and α-glutamyltyrosine, fibrinogen cleavage peptide DSGEGDFXAEGGGVR and glutathione-related amino acid cysteine-glutathione disulfide were inversely associated with PDAC after Bonferroni correction. Five top metabolites demonstrated significant time-varying associations (p<0.023) with the strongest associations observed 10–15 years after participants’ blood collection and attenuated thereafter.

Conclusion Our results suggest that prediagnostic metabolites related to subclinical disease, γ-glutamyl cycle metabolism and adiposity/insulin resistance are associated with PDAC.

  • cancer epidemiology
  • epidemiology
  • pancreatic cancer

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.


  • Contributors Conception and design of the study: RS-S, JS. Collection of data: RS-S (ATBC, PLCO metabolite data), DA (ATBC only). Analysis and interpretation of data: RS-S, AD, JS, SM. Drafting of the manuscript: RS-S, JS, AD. Critical revision of the manuscript for important intellectual content: RS-S, AD, JS, SJW, DA, SM. All authors read, revised and approved the final draft.

  • Funding This work was supported by the Intramural Research Program of the National Cancer Institute, National Institutes of Health, Department of Health and Human Services and by the US Army Medical Research Acquisition Activity through the Peer Review Cancer Research Program Discovery Award under Award No. W81XWH-12-1-0369.

  • Disclaimer The opinions, interpretations, conclusions and recommendations are those of the author and are not necessarily endorsed by the US Army. The Department of Defense sponsors had no role in the design of the study; the collection, analysis and interpretation of the data; the writing of the manuscript and the decision to submit the manuscript for publication.

  • Competing interests None declared.

  • Patient consent for publication Not required.

  • Ethics approval The ATBC was approved by the NCI Special Studies IRB protocol number OH95CN012. The PLCO was approved by the NCI Special Studies IRB protocol number OH97CN041.

  • Provenance and peer review Not commissioned; externally peer reviewed.

  • Data availability statement Data are available on reasonable request. Ethical restrictions on human subjects’ data prevents our posting the data used for this analysis. Biomedical research scientists from recognised research institutions can contact us directly to request data as bona fide researchers by emailing corresponding author.