Lifestyle behaviors and immune cell characteristics in type 2 diabetes risk: a Mendelian randomization study

To investigate the causal associations of lifestyle behaviors and immune cell characteristics with type 2 diabetes mellitus (T2DM) risk, and to explore potential immune-mediated pathways.

We conducted bidirectional two-sample Mendelian randomization (MR) and mediation analysis using large-scale GWAS summary statistics. We assessed 48 lifestyle behaviors, 731 immunophenotypes, and T2DM. Genetic instruments were selected using tiered thresholds: for lifestyles/T2DM, primarily p < 5 × 10⁻⁸, relaxed to p < 5 × 10⁻⁶ when fewer than 5 genome-wide significant SNPs were available; for immune traits, a suggestive threshold of p < 1 × 10⁻⁵ was applied due to limited sample size. Primary MR analyses were conducted using inverse variance weighted (IVW) fixed-effects and random-effects models, with method selection based on Cochran's Q heterogeneity test. Horizontal pleiotropy was evaluated via MR-Egger intercept test and MR-PRESSO global test. Method-specific Benjamini-Hochberg FDR correction was applied separately for each model. Associations surviving p-FDR < 0.05 were considered confirmed; nominal p < 0.05 without FDR correction were deemed exploratory.

Three lifestyle behaviors showed confirmed causal effects on T2DM (p-FDR < 0.05): dried fruit intake (protective, OR = 0.52, p-FDR = 1.44 × 10⁻³) and strenuous sports (protective, OR = 0.047, p-FDR = 5.16 × 10⁻³) decreased risk, while smoking initiation increased risk (OR = 1.18, p-FDR = 5.63 × 10⁻⁴). Screening 731 immunophenotypes revealed that only one trait survived stringent p-FDR < 0.05 (CD25 on IgD+ CD24 + B cell, OR = 1.017, p-FDR = 7.84 × 10⁻⁵), with seven reaching suggestive significance (0.05 ≤ p-FDR < 0.20) and 19 remaining at nominal significance (p < 0.05, p-FDR ≥ 0.20). Analysis of lifestyle-immune links identified one confirmed association (smoking initiation increased CD25 on CD39 + CD4+ T cells, p-FDR = 0.028), one suggestive association (dried fruit intake decreased lymphocyte count, OR = 0.33, p-FDR = 0.083), and two exploratory associations (p < 0.05, p-FDR > 0.20). No complete mediation pathway (lifestyle→immune→T2DM) satisfied p-FDR < 0.05 in both steps. Exploratory mediation analyses (based on nominal associations) suggested potential indirect effects via CD25 + CD39+CD4 + T cells (mediated proportion 2.1%), absolute lymphocyte count (-2.1%), and CD24 + IgD + B cells (1.9%).

This study provides high‑confidence genetic evidence for causal effects of lifestyle on T2DM. Immune-mediated pathways remain exploratory due to limited statistical power and require confirmation in larger immune cell GWAS cohorts.

The online version contains supplementary material available at 10.1007/s40200-026-01922-7.