Predictive Factors for Changes in Serum Irisin Concentrations after Different Exercise Interventions

This study compared serum irisin concentration changes following 16 weeks of different exercise protocols (hill, running, cycling, and combined) versus control in 289 previously sedentary adults. We identified phenotypic and genetic predictors to establish an irisin response prediction model for precise exercise prescription. Phenotypic measures (serum irisin, maximal oxygen consumption, and lipids) and fibronectin type-III domain containing protein-5 genen single nucleotide polymorphism (rs16835198) genotyping were performed pre/post-intervention. Serum irisin significantly increased post-exercise overall (p<0.05), but with substantial interindividual variability (Δ range:-4.98 to 20.77 ng/mL; 28.03% responders, Cohen's d≥0.2). Under the specific (and unstandardized) exercise volumes utilized in this study, hill and combined training elicited the most significant increases (p<0.01). Cohen's d-based irisin change effect size significantly correlated with baseline maximal oxygen consumption (r=0.164 and p<0.01) and irisin levels (r=0.481 and p<0.05). It was also linked to FNDC5 rs16835198 (p=0.041), with the TT genotype exhibiting greater responses than GT/GG genotypes. A model incorporating baseline serum irisin concentrations (β=0.522), maximal oxygen consumption (β=0.107), exercise protocols (β=0.207), and rs16835198 genotype (β=0.116) explained 27.1% of the variance in the effect size of Δirisin. In short, exercise-induced irisin elevation exhibited marked individual variability. Baseline irisin, maximal oxygen consumption, FNDC5 rs16835198 genotype, and exercise protocol are significant predictors, accounting for 27.1% of the individual differences in training response, informing precision exercise prescription. Keywords: Irisin; Individual differences; Effect of exercise intervention; prediction model.