Metabolites (Jun 2023)

Dietary Inflammatory and Insulinemic Potentials, Plasma Metabolome and Risk of Colorectal Cancer

  • Dong Hoon Lee,
  • Qi Jin,
  • Ni Shi,
  • Fenglei Wang,
  • Alaina M. Bever,
  • Jun Li,
  • Liming Liang,
  • Frank B. Hu,
  • Mingyang Song,
  • Oana A. Zeleznik,
  • Xuehong Zhang,
  • Amit Joshi,
  • Kana Wu,
  • Justin Y. Jeon,
  • Jeffrey A. Meyerhardt,
  • Andrew T. Chan,
  • A. Heather Eliassen,
  • Clary B. Clish,
  • Steven K. Clinton,
  • Edward L. Giovannucci,
  • Fred K. Tabung

DOI
https://doi.org/10.3390/metabo13060744
Journal volume & issue
Vol. 13, no. 6
p. 744

Abstract

Read online

The inflammatory and insulinemic potentials of diets have been associated with colorectal cancer risk. However, it is unknown whether the plasma metabolite profiles related to inflammatory diets, or to insulinemic diets, underlie this association. The aim of this study was to evaluate the association between metabolomic profile scores related to the food-based empirical dietary inflammatory patterns (EDIP), the empirical dietary index for hyperinsulinemia (EDIH), and plasma inflammation (CRP, IL-6, TNFα-R2, adiponectin) and insulin (C-peptide) biomarkers, and colorectal cancer risk. Elastic net regression was used to derive three metabolomic profile scores for each dietary pattern among 6840 participants from the Nurses’ Health Study and Health Professionals Follow-up Study, and associations with CRC risk were examined using multivariable-adjusted logistic regression, in a case-control study of 524 matched pairs nested in both cohorts. Among 186 known metabolites, 27 were significantly associated with both the EDIP and inflammatory biomarkers, and 21 were significantly associated with both the EDIH and C-peptide. In men, odds ratios (ORs) of colorectal cancer, per 1 standard deviation (SD) increment in metabolomic score, were 1.91 (1.31–2.78) for the common EDIP and inflammatory-biomarker metabolome, 1.12 (0.78–1.60) for EDIP-only metabolome, and 1.65 (1.16–2.36) for the inflammatory-biomarkers-only metabolome. However, no association was found for EDIH-only, C-peptide-only, and the common metabolomic signatures in men. Moreover, the metabolomic signatures were not associated with colorectal cancer risk among women. Metabolomic profiles reflecting pro-inflammatory diets and inflammation biomarkers were associated with colorectal cancer risk in men, while no association was found in women. Larger studies are needed to confirm our findings.

Keywords