Abstract 16835: Plasma Proteomics and Risk of Incident HFpEF and HFrEF: The Atherosclerosis Risk in Communities Study
Victoria Lamberson, Yimin Yang, Peder L Myhre, Pranav Dorbala, Humaira Rasheed, Leo F Buckley, Brian Claggett, Margherita De Marzio, Kimberly Glass, Ron C Hoogeveen, Christie M Ballantyne, Josef Coresh, Bing Yu, Jacob Joseph, Yan Sun, Torbjorn Omland, Amil M Shah- Physiology (medical)
- Cardiology and Cardiovascular Medicine
Introduction: Heart failure with preserved (HFpEF) and reduced (HFrEF) ejection fraction differ in pathophysiology and treatment response. Determining shared and unique protein biomarkers for these phenotypes can aid in understanding their biological mechanisms.
Methods: We studied 4485 participants in the community-based Atherosclerosis Risk in Communities (ARIC) study who were free of HF and had aptamer-based proteomic measurements (SomaLogic, Boulder, CO) at study Visit 5 (2011-2013). For 4955 aptamers corresponding to 4712 unique proteins, we fit separate multivariable Cox proportional hazards regression models to test for associations with HFpEF and HFrEF. Replication of significant associations was conducted using data from the Norwegian Trøndelag Health Study (HUNT, n=3262). Two-sample Mendelian randomization (MR) was used to assess potential causal associations with HFpEF and HFrEF using summary statistics from the Million Veterans Program.
Results: Mean age at Visit 5 was 75±5 years, 58% were women, and 17% reported Black race. At a median follow-up of 7.2 [IQR 5.6-7.8] years, 193 HFpEF and 157 HFrEF events occurred. After Bonferroni correction for multiple testing, 16 aptamers were significantly associated with both HFpEF and HFrEF, 45 were significantly associated with HFpEF only, and 9 aptamers were associated with HFrEF only (Figure). These associations replicated in HUNT (p<0.05) for 6 out of 45 proteins for HFpEF only, and 4 out of 9 proteins for HFrEF only. MR identified a potential causal effect of CD7 on HFrEF. No causal associations were found for HFpEF.
Conclusions: Partially distinct sets of circulating plasma proteins associate with HFpEF and HFrEF. A greater proportion of these associations replicated for HFrEF (4 of 9) than HFpEF (6 of 45) in two distinct population studies, highlighting the heterogeneity of HFpEF. Further study of these proteins may lead to additional insights into the differing biology of these phenotypes.