METABOLOMIC PREDICTORS OF INCIDENT GALLSTONE DISEASE

Background & Aims
Gallstone disease is highly prevalent and costly in the USA: over 40 million people are affected with estimated annual health costs of $4 billion. Established risk factors for gallstone disease include overweight/obesity and metabolic syndrome, both of which have been associated with alterations in plasma metabolites. However, data are limited on the prediagnostic plasma metabolome in relation to gallstone disease. This study aims to identify circulating metabolites that are associated with incident gallstone disease in two prospective cohorts and validate findings in an external independent cohort.
Methods
We included 9960 female participants enrolled in the prospective Nurses’ Health Study (NHS) and the Nurses’ Health Study II (NHS II) who were free of gallstone disease at time of baseline metabolomic profiling (1990 for NHS and 1999 for NHS II) and followed through 2021. Metabolomic profiling was performed using liquid chromatography-tandem mass spectrometry platforms at the Broad Institute. Multivariable logistic regression and enrichment analysis were used to identify metabolites and metabolite groups that are associated with incident gallstone disease, adjusting for BMI, rapid weight loss, AHEI diet quality score, smoking, alcohol, activity, parity, menopausal hormone use & thiazide use. The p-value was set at 0.05 and adjusted for multiple testing by controlling the false discovery rate (FDR). Findings were externally validated in the Women’s Health Initiative (WHI) cohort which included 1866 female participants who were free of gallstone disease at time of baseline Broad Institute metabolomic profiling (from 1993-1998) with follow-up through 2005.
Results
Among participants with metabolomic profiling of prediagnostic blood specimens, we identified 1585 cases of incident gallstone disease in NHS/NHS II and 170 cases in WHI. Mean age was 54.8 years in NHS/NHSII and 66.9 years in WHI. After multivariate adjustment for lifestyle risk factors, 27 metabolites (14 positive, 13 negative) were significantly associated with incident gallstone disease in the discovery NHS/NHS II cohorts and subsequently validated in the WHI cohort after adjustment for multiple testing (Table 1). Enrichment analysis in NHS/NHS II revealed that triacylglycerols & diacylglycerols were positively associated with incident gallstone disease while plasmalogens, cholesterol esters & carnitines were inversely associated. These findings were also replicated in WHI (Figure 1).
Conclusions
This study identifies several metabolites (particularly triglycerides & plasmalogen glycerophospholipids) that are associated with incident gallstone disease in women with validation in an external cohort. These findings could further the understanding of the etiopathogenesis of gallstone disease as well as serve as potential biomarkers for risk prediction and preventive strategies.