Can the impact of childhood adiposity on disease risk be reversed? A Mendelian randomization study

Objective To evaluate whether early life adiposity has an independent effect on later life disease risk or whether its influence is mediated by adulthood body mass index (BMI). Design Two-sample univariable and multivariable Mendelian randomization. Setting The UK Biobank (UKB) prospective cohort study and four large-scale genome-wide association study (GWAS) consortia. Participants 453,169 participants enrolled in the UKB and a combined total of over 700,000 individuals from different GWAS consortia. Exposures Measured BMI during adulthood (mean age: 56.5) and self-reported adiposity at age 10. Main outcome measures Coronary artery disease (CAD), type 2 diabetes (T2D), breast cancer and prostate cancer. Results Individuals with genetically predicted higher BMI in early life had increased odds of CAD (OR:1.49, 95% CI:1.33-1.68) and T2D (OR:2.32, 95% CI:1.76-3.05) based on univariable MR (UVMR) analyses. However, there was little evidence of a direct effect (i.e. not via adult BMI) based on multivariable MR (MVMR) estimates (CAD OR:1.02, 95% CI:0.86-1.22, T2D OR:1.16, 95% CI:0.74-1.82). In the MVMR analysis of breast cancer risk, there was strong evidence of a protective direct effect for early BMI (OR:0.59, 95% CI:0.50-0.71), although adult BMI did not appear to have a direct effect on this outcome (OR:1.08, 95% CI:0.93-1.27). Adding age of menarche as an additional exposure provided weak evidence of a total causal effect (UVMR OR:0.98, 95% CI:0.91-1.06) but strong evidence of a direct causal effect, independent of early and adult BMI (MVMR OR:0.90, 95% CI:0.85-0.95). Weak evidence of a causal effect was observed in the MVMR analysis of prostate cancer (early life BMI OR:1.06, 95% CI:0.81-1.40, adult BMI OR:0.87, 95% CI:0.70-1.08). Conclusions Our findings suggest that increased CAD and T2D risk attributed to early life adiposity can be mitigated if individuals reduce their weight in later life. However, having a low BMI during childhood may increase risk of breast cancer regardless of changes to weight in later life, with timing of puberty also putatively playing an important role.