Relation between breast cancer and high glycemic index or glycemic load: a meta-analysis of prospective cohort studies

Patrick Mullie, Alice Koechlin, Mathieu Boniol, Philippe Autier, Peter Boyle

Research output: Contribution to journalArticle

33 Citations (Scopus)
67 Downloads (Pure)

Abstract

Breast cancer is the commonest form of cancer in women worldwide. It has been suggested that chronic hyperinsulinemia associated with insulin resistance plays a role in breast cancer etiology. To test the hyperinsulinemia hypothesis, a dietary pattern associated with a high glycemic index and glycemic load, both proxies for chronic hyperinsulinemia, should be associated with an increased risk of breast cancer. A meta-analysis restricted to prospective cohort studies was undertaken using a random effects model with tests for statistical significance, publication bias and heterogeneity. The metric for analysis was the risk of breast cancer in the highest relative to the lowest glycemic index and glycemic load dietary pattern. A dietary pattern with a high glycemic index was associated with a summary relative risk (SRR) of 1.05 (95% CI: 1.00, 1.11), and a high glycemic load with a SRR of 1.06 (95% CI: 1.00, 1.13). Adjustments for body mass index [BMI], physical activity and other lifestyle factors did not influence the SRR, nor did menopausal status and estrogen receptor status of the tumor. In conclusion, the current evidence supports a modest association between a dietary pattern with high glycemic index or glycemic load and the risk of breast cancer.

Original languageEnglish
Pages (from-to)152-159
Number of pages8
JournalCritical Reviews in Food Science and Nutrition
Volume56
Issue number1
Early online date6 Mar 2015
DOIs
Publication statusPublished - 2 Jan 2016

Keywords

  • hyperinsulinemia
  • nutritional epidemiology
  • prevention

Fingerprint Dive into the research topics of 'Relation between breast cancer and high glycemic index or glycemic load: a meta-analysis of prospective cohort studies'. Together they form a unique fingerprint.

Cite this