Abstract Body: Background: Evidence suggests that dairy-derived saturated and trans fatty acids may contribute to observed inverse associations of dairy intake with risk of metabolically associated fatty liver disease (MAFLD) and type 2 diabetes mellitus (T2DM). Longitudinal studies examining the association between circulating dairy fatty acids and phenotypes underlying these disorders are limited. This study aimed to investigate the 9-year longitudinal associations of 15:0 (pentadecanoic acid), t16:1n9 (trans palmitoleic acid), t18:1n11 (trans vaccenic acid) and c9,t11 CLA (conjugated linoleic acid) in four serum fractions with insulin sensitivity, beta cell function and alanine aminotransferase (ALT in U/L, a biomarker of liver function). Methods: In the PROMISE cohort (n=460), insulin sensitivity was assessed using Matsuda’s insulin sensitivity index (ISI) and the homeostasis model assessment of insulin sensitivity (HOMA2_S). Beta cell function was determined using the insulinogenic index divided by HOMA insulin resistance (IGI/IR) and the Insulin Secretion-Sensitivity Index-2 (ISSI-2). Serum ALT was measured at each visit. Baseline serum fatty acids (mol%) in the cholesteryl ester (CE), non-esterified fatty acid (NEFA), phospholipid (PL) and triglyceride (TG) pools were analyzed using gas chromatography flame-ionization detection. Longitudinal associations were analyzed using covariate-adjusted generalized estimating equations with P values adjusted using the Benjamini-Hochberg False Discovery Rate (FDR). Results: In the TG pool, 18:1n11t was significantly associated with logISSI-2 (β= 3.09, CI: [0.04, 6.24], p<0.05). Significant inverse associations were observed between 15:0 (CE) and logALT (β= -3.30, CI: [-5.33, -1.24], p<0.01). In the PL pool, 15:0 was positively associated with logISSI-2 (β=5.37, CI: [1.71, 9.16], p<0.01) and logIGI/IR (β=10.49, CI: [4.35, 16.99], p<0.01). After FDR adjustment, 15:0 (CE) continued to be inversely associated with logALT (p<0.05). Similarly, 15:0 (PL) was significantly associated with logISSI-2 and logIGI/IR (p<0.05). Fatty acids in the NEFA and TG pools were not significantly associated with outcomes. Conclusions: Longitudinal associations of dairy-derived serum fatty acids with liver health and beta cell function were detected across serum pools. Our findings expand the literature on the potential role of dairy-derived saturated and trans conjugate fatty acids in cardiometabolic phenotypes underlying MAFLD and T2DM.