Causal effects of early-life traits on lung function trajectory: a Mendelian randomization study

Rationale At least half of chronic obstructive pulmonary disease (COPD) occurs in individuals with subnormal preceding lung function trajectory. Identifying which early-life phenotypes causally influence risk of adverse trajectories could facilitate disease prevention. Objectives Using Mendelian randomization (MR), we investigated whether previous early-life phenotypic associations with lower trajectories are compatible with a causal effect. Methods Outcomes were lower FEV1 trajectory groups from latent profile analysis in participants with ≥2 measurements at ages 8, 15, 24 or 30 in the Avon Longitudinal Study of Parents and Children (% predicted; n=5,165). Nine exposures were prioritized for MR from observational associations with lower trajectories in logistic regression. Two-stage residual inclusion MR, applying polygenic risk scores of variants from the largest published genome-wide association studies, estimated genetically instrumented effects of exposures on odds of lower trajectories. Results Of five trajectories identified, two had lower mean FEV1: Below Average (87.7% predicted) and Persistently Low (72.6% predicted). Observationally, higher birth weight and childhood body mass index (BMI) were associated with lower odds of these trajectories, and pre-school wheeze, childhood aeroallergen sensitization, asthma and airways hyperresponsiveness with higher odds. MR evidence suggested childhood-onset asthma causally increased odds of lower trajectories (OR 9.62; 95% CI 1.49-119.7). MR estimates were directionally concordant with observational analyses but with confidence intervals crossing null for birth weight and aeroallergen sensitization, and were near-null for BMI. Conclusions These findings support a causal role for childhood-onset asthma in subnormal lung function trajectories, suggesting its optimized diagnosis and management as targets for COPD prevention.