Introduction
The prevalence of heart failure with preserved ejection fraction (HFpEF) is increasing.
This disease has posed diagnostic and therapeutic challenges; mortality and morbidity
remain unacceptably high. Unbiased molecular studies are necessary to identify better
diagnostic biomarkers.
Hypothesis
Large-scale discovery proteomics with a multiplex immunoassay platform will identify
novel molecular mechanisms that differentiate HFpEF from patients without heart failure
(HF).
Methods
We performed proteomic analyses (Olink) of 552 proteins on 88 HFpEF cases and 88 no
HF controls from individuals referred for cardiac catheterization at Duke University.
HFpEF was defined as history of HF, ejection fraction >45% and diastolic dysfunction
class ≥1; no-HF controls had no diastolic dysfunction. Univariate logistic regression
was used to identify proteins associated with HFpEF; associated proteins at a false
discovery rate <0.05 were then tested in a multivariable model adjusted for age, gender,
body mass index (BMI), systolic blood pressure (SBP), diabetes (DM) and creatinine
and included in a LASSO regression to create a sparse protein model. Incremental improvement
in model prediction from the clinical model + NT-proBNP was assessed using change
in the C-statistic. Gene Set Enrichment Analysis (GSEA) was used to identify pathways
associated with HFpEF.
Results
As expected, HFpEF cases were older and had higher BMI, SBP, creatinine and prevalence
of DM (p<0.05) compared to no-HF controls. Univariate analysis identified 228 proteins
associated with HFpEF; 77 remained nominally significant in the multivariable model
(p<0.05). Key proteins included those involved in angiogenesis, fibrosis and inflammation.
Most were more abundant in HFpEF cases than in no-HF controls, including Lysosomal
integral membrane protein 2 (p=0.01), Neutrophil gelatinase-associated lipocalin (p=0.01),
Urokinase plasminogen activator surface receptor (p=0.001), Serine/threonine-protein
kinase receptor R3 (p=0.02) and Interleukin-1 receptor antagonist protein (p=0.002).
LASSO yielded a 20-protein model; when added to the clinical model + NT-proBNP this
LASSO proteomic model improved the C-statistic from 0.82 (Figure black line) to 0.92
(red line p=0.0001). GSEA revealed fatty acid metabolism and inflammatory pathways
to be enriched.
Conclusions
Biomarkers of angiogenesis, fibrosis, fatty acid metabolism and inflammation are associated
with HFpEF and improve discriminative capabilities on top of clinical factors and
NT-proBNP. These findings highlight the importance of these pathways in HFpEF and
identify potential novel circulating diagnostic biomarkers.
To read this article in full you will need to make a payment
Purchase one-time access:
Academic & Personal: 24 hour online accessCorporate R&D Professionals: 24 hour online accessOne-time access price info
- For academic or personal research use, select 'Academic and Personal'
- For corporate R&D use, select 'Corporate R&D Professionals'
Subscribe:
Subscribe to Journal of Cardiac FailureAlready a print subscriber? Claim online access
Already an online subscriber? Sign in
Register: Create an account
Institutional Access: Sign in to ScienceDirect
Article info
Identification
Copyright
© 2020 Published by Elsevier Inc.
