Clonal Hematopoiesis of Indeterminate Potential Proteomic Risk Score for Predicting Clinical Outcome in Patients With Heart Failure With Preserved Ejection Fraction.

Abstract

Background: Clonal hematopoiesis of indeterminate potential (CHIP) is common in heart failure with preserved ejection fraction (HFpEF). However, the proteomic signatures linked to CHIP driver mutations and their prognostic implications in HFpEF remain poorly defined. We aimed to identify plasma protein signatures associated with CHIP driver mutations and to develop and externally validate a CHIP-proteomic risk score (CHIP-ProtRS) for predicting clinical outcomes in HFpEF.

Methods: The derivation cohort consisted of 118 TOPCAT (Treatment of Preserved Cardiac Function Heart Failure With an Aldosterone Antagonist) participants with available CHIP sequencing and proteomic profiling using the SomaScan platform. Associations between CHIP mutations and circulating proteins were assessed using gene-specific linear regression adjusted for age, with correction for multiple comparisons. Significant proteins were used to derive gene-specific proteomic classifiers using nested cross-validated least absolute shrinkage and selection operator logistic regression. A composite CHIP-ProtRS was generated using least absolute shrinkage and selection operator Cox regression and validated in a composite of 3 independent HFpEF cohorts (n=654). The primary end point was a composite of death and HF hospitalization.

Results: In TOPCAT, 30.5% of tested participants carried ≥1 CHIP mutation. Significant protein associations were identified for TET2 (7 proteins), DNMT3A (10), and PPM1D (5), all of which were enriched in immune and inflammatory pathways. Gene-specific least absolute shrinkage and selection operator models showed good discrimination for CHIP mutation (area under the curves, 0.752-0.898). The primary end point occurred in 28.0% of the derivation cohort and 34.7% of the validation cohorts. In the Cox-least absolute shrinkage and selection operator combination, only the TET2-derived proteomic signature retained a nonzero coefficient in the CHIP-ProtRS. The CHIP-ProtRS was independently associated with the primary end point after adjustment for age, sex, Meta-Analysis Global Group in Chronic Heart Failure score, NT-proBNP (N-terminal pro-B-type natriuretic peptide), and atrial fibrillation in the derivation (adjusted hazard ratio, 1.42 [95% CI, 1.01-2.00]) and validation cohorts (adjusted hazard ratio, 1.16 [95% CI, 1.01-1.32]).

Conclusions: CHIP is associated with alterations in plasma concentrations of immune/inflammatory proteins. A CHIP-ProtRS is independently associated with adverse outcomes in HFpEF, supporting its potential role as a biomarker of risk and underlying pathobiology.