Non-nutritive sweetener consumption, metabolic risk factors, and inflammatory biomarkers among adults in the Cancer Prevention Study-3 Diet Assessment Sub-Study

Document Type

Journal Article

Publication Date

3-22-2025

Journal

The Journal of nutrition

DOI

10.1016/j.tjnut.2025.03.020

Keywords

artificial sweetener; cancer; cardiometabolic disease; cardiovascular disease; diabetes; diet soda; obesity; sugar

Abstract

BACKGROUND: Non-nutritive sweeteners (NNS) are widely used to replace added sugars, yet their role in metabolic health and chronic disease prevention is debated. OBJECTIVE: Examine associations between NNS consumption, metabolic risk factors, and inflammatory biomarkers. METHODS: This cross-sectional analysis included 624 adults in the American Cancer Society's Cancer Prevention Study-3 Diet Assessment Sub-study (DAS). Consumption of NNS including aspartame, saccharin, sucralose, and acesulfame-potassium were estimated using the mean quantities reported in six 24-hour dietary recalls over one year. Fasting insulin, C-peptide, hemoglobin A1c (HbA1c), leptin, adiponectin, C-reactive protein (CRP), tumor necrosis factor alpha (TNF-α), interleukin-6 (IL-6), and interleukin-10 (IL-10) were measured in fasting blood samples collected twice, six months apart. Multivariable linear regression was used to examine associations between NNS consumption and the average levels of each metabolic or inflammatory biomarkers. Base models were adjusted for age, sex, race, education, smoking, and physical activity; full models were further adjusted for BMI, diet quality (HEI-2020), and energy intake. RESULTS: Over half (55%) of participants reported consuming NNS (mean daily NNS consumption 7, 38, and 221 mg across tertiles). NNS consumption was positively associated with leptin (p-trend = 0.0006) and CRP (p-trend = 0.02) prior to adjustment for BMI, diet quality, and energy intake. No associations between NNS consumption and insulin, C-peptide, HbA1c, adiponectin, TNF-α, or IL-10 were observed. In analyses stratified by BMI, NNS consumption was positively associated with IL-6 among participants with BMI ≥25kg/m, but not BMI <25kg/m. CONCLUSIONS: Findings in the full sample were predominantly null after adjustment for energy intake and BMI; but NNS consumption was positively associated with IL-6 among participants with overweight or obesity. Investigation of mechanisms through which NNS consumption may impact inflammatory pathways is warranted.

Department

Exercise and Nutrition Sciences

Link to Full Text

Share

COinS