PAR2 genetic variation stratifies biomarker dynamics and rehospitalisation risk after acute heart failure
E Ghanbari, S Sritharan, A Doerner, Z Chen, A Diagel, V Bargou, M S Anker, J E Weber, N Kraenkel, S D Anker, U Landmesser, F Edelmann, U RauchAbstract
Background
Recovery after acute heart failure (AHF) is highly heterogeneous, and the biological determinants of impaired biomarker resolution and adverse post-discharge trajectories remain incompletely understood. Genetic regulation of inflammatory and metabolic pathways may contribute to these differences.
Methods
We studied 135 patients hospitalized with AHF from the prospective BeLOVE cohort with paired biomarker and proteomic profiling at admission and 90-day follow-up. Longitudinal trajectories of NT-proBNP, troponin, and ~3,000 plasma proteins were analysed using linear mixed-effects models stratified by the F2RL1 cis-eQTL rs1529505. Clinical outcomes were assessed in up to 20,649 heart failure patients from UK Biobank using Cox and negative binomial models
Results
rs1529505 showed strong, dose-dependent associations with reduced F2RL1 expression across tissues, including whole blood (NES −0.40, p = 4.4×10⁻⁹¹), liver (NES −0.39, p = 9.3×10⁻¹⁷), and heart left ventricle (p = 4.1×10⁻⁶). In BeLOVE, rs1529505 significantly modified post-acute biomarker trajectories. NT-proBNP showed a genotype×time interaction (F(1,111.3) = 5.17, p = 0.025), with a marked decline in non-carriers (p = 1.0×10⁻⁵) but a blunted reduction in carriers, who had higher levels at follow-up (p = 0.0399). Troponin showed a similar pattern (genotype×time F(1,18) = 5.12, p = 0.0363): levels declined in non-carriers (β = −162.9, p = 0.0211) but not in carriers, resulting in higher follow-up levels (interaction β = +376.5, p = 0.0363). Proteomic profiling identified a genotype-dependent increase in circulating CES1 (interaction β = 0.57, p = 0.019), embedded in a coherent liver-derived secretory module. In UK Biobank, each copy of the PAR2-increasing rs1529505 C allele was associated with higher risk of cardiac hospitalization after HF diagnosis (HR 1.20, 95% CI 1.05–1.37; p = 0.0075), corresponding to ~44% higher risk in CC versus TT homozygotes. In negative binomial models, each C allele was associated with a 2.1% higher rate of any cardiac hospitalizations (IRR 1.02, 95% CI 1.00–1.04) and a 2.8% higher rate of HF re-hospitalizations (IRR 1.03, 95% CI 1.00–1.06).
Interpretation
Genetic upregulation of PAR2 signaling is associated with impaired biomarker resolution, a distinct hepatic metabolic proteomic signature, and higher risk and burden of post-discharge hospitalizations after AHF. These findings mark a genotype-dependent disease progression after acute HF warranting further clinical validation.