Association between HOMA-IR and coexisting MASLD and type 2 diabetes: Insights from NHANES and network pharmacology

Insulin resistance is a common metabolic feature of metabolic dysfunction-associated steatotic liver disease (MASLD) and type 2 diabetes (T2D), but its cross-sectional association with the coexistence of these conditions remains incompletely characterized.

To evaluate the associations between the homeostasis model assessment of insulin resistance (HOMA-IR) and the concurrent presence of MASLD and T2D, and to explore potential demographic heterogeneity and generate hypotheses regarding candidate biological pathways.

Population-based cross-sectional study using data from the National Health and Nutrition Examination Survey 2017-March 2020 pre-pandemic cycle.

Adults with fasting glucose, fasting insulin, vibration-controlled transient elastography, and covariate data were included. Participants were classified as having neither condition, MASLD alone, T2D alone, or both. Multivariable logistic regression evaluated HOMA-IR continuously and by quartiles. Interaction and stratified analyses were conducted by sex and race/ethnicity. Secondary analyses examined concurrent MASLD and prediabetes among participants without T2D. Exploratory mediation analyses used fasting serum insulin as the exposure to assess whether glycemic and lipid indicators statistically accounted for part of the association. Network pharmacology was used as a hypothesis-generating approach to predict candidate targets and enriched pathways.

Among 1,527 participants, each 1-unit increase in HOMA-IR was associated with 59.9% higher odds of concurrent MASLD and T2D versus neither condition (odds ratio [OR] 1.599, 95% confidence interval [CI] 1.491-1.715), 20.2% higher odds versus MASLD alone, and 14.5% higher odds versus T2D alone. When T2D alone was the reference, the association differed by race/ethnicity (P for interaction <0.01). Each 1-unit increase in HOMA-IR was also associated with 22.5% higher odds of concurrent MASLD and prediabetes versus prediabetes alone. Sex-related heterogeneity was observed in secondary comparisons involving concurrent MASLD and prediabetes. The odds of comorbidity were progressively higher across increasing HOMA-IR quartiles (P for trend <0.001). Fasting plasma glucose and hemoglobin A1c statistically accounted for 48.1% and 12.1%, respectively, of the total association between fasting serum insulin and concurrent MASLD and T2D. Hypothesis-generating network pharmacology predicted PRKACA, MAPK1/3, PIK3CA, and SRC as candidate hub targets and suggested candidate pathways related to PI3K-Akt, FOXO, and MAPK signaling.

Higher HOMA-IR was associated with higher odds of concurrent MASLD and T2D. The magnitude of the associations varied by race/ethnicity and the selected reference group, while sex-related heterogeneity was observed in secondary analyses of concurrent MASLD and prediabetes.