Abstract
INTRODUCTION: Bronchodilator responsiveness (BDR) is associated with progression to COPD. Genetic risk for COPD, summarized by polygenic risk scores (PRS), predicts low lung function and COPD. However, it remains unclear whether genetic predisposition to COPD is related to BDR and whether PRS and BDR together influence lung function decline in individuals at risk for the disease.
METHODS: We analyzed data from COPDGene participants with a smoking history and normal spirometry at study enrollment. We cross-sectionally examined the association of a PRS with 2005-BDR-FEV1 % (change relative to pre-bronchodilator) and 2021-BDR-FEV1 % (change relative to predicted). We also examined the association of PRS, 2005-BDR-FEV1 %, and 2021-BDR-FEV1 % with progression to COPD and longitudinal FEV1 decline between enrollment and follow-up adjusted for demographics, smoking history, and FEV1 at enrollment.
RESULTS: PRS did not correlate with 2005-BDR-FEV1 % in 1446 African Americans (AA) but PRS correlates with BDR in both unadjusted (rho = 0.01, P < 0.001) and adjusted analysis in 3378 non-Hispanic Whites (NHW). NHW participants with BDR had higher PRS than those without. Models including 2005-BDR-FEV1 % demonstrated better accuracy than those including PRS (Area under the curve: 0.762 vs 0.743 in NHW; 0.693 vs 0.653 in AA). BDR models also outperformed PRS models for longitudinal FEV1 decline. Mediation analysis showed that about one third of the PRS effect on FEV1 decline in NHW was explained through BDR.
CONCLUSIONS: BDR is more strongly associated with progression to COPD and FEV1 decline than PRS, and part of the PRS effect is mediated through BDR.