Body composition phenotypes, and specific adiposity and lean mass indicators, and their lifestyle determinants among adult childhood cancer survivors - The iBoneFIT adult study.

BACKGROUND & AIMS: Childhood cancer survivors (CCSs) face elevated long-term cardiometabolic risk due to altered body composition. This study aimed to identify distinct body composition phenotypes among adult CCSs and to examine their associations with lifestyle behaviors.

METHODS

In this cross-sectional study, 92 adult CCSs (54% male; median age 35 years) underwent dual-energy X-ray absorptiometry to measure body composition. Physical activity (PA) and sedentary time were measured using ActiGraph accelerometers. Factor and hierarchical cluster analyses identified body composition phenotypes, from BMI, waist-to-hip ratio (WHR), visceral fat area (VFA), body fat percentage, fat mass index (FMI), lean mass index (LMI), total body lean mass, total body fat mass, and subcutaneous adipose tissue. Multinomial and linear regressions adjusted for sex and age, assessed associations between phenotypes, lifestyle factors, and individual body composition indicators.

RESULTS

Four body composition phenotypes were identified: muscle-dominant (21%), high adiposity (16%), thin-low muscle (27%), and low adiposity-preserved lean (36%). Male CCSs were more prevalent in the muscle-dominant phenotype, whereas female participants predominated in the thin-low muscle phenotype. Older age was associated with the muscle-dominant phenotype (RRR = 1.08, p = 0.032) and with higher LMI (β = 0.07, p = 0.013), WHR (β = 0.003, p = 0.001), and VFA (β = 1.61, p = 0.003). Lifestyle behaviors were not associated with phenotypic classification. However, moderate-to-vigorous PA was positively associated with LMI (β = 1.05, p = 0.008) and total lean mass (β = 4.15, p = 0.008), while light-intensity (β = -0.48, p = 0.048) and total PA (β = -0.46, p = 0.032) were inversely associated with body fat percentage.

CONCLUSIONS

Distinct body composition phenotypes were identified among adult CCSs. While lifestyle behaviours did not differentiate phenotypes, PA was favourably associated with key body composition measures.