Clonal hematopoiesis of indeterminate potential risk score for predicting clinical outcome in patients with heart failure with preserved ejection fraction
M Kim, V Subramanian, A Pandey, J Chirinos, H Tavolinejad, M Sarmiento Bustamante, B PourmussaAbstract
Background
Clonal hematopoiesis of indeterminate potential (CHIP) is prevalent among patients with heart failure with preserved ejection fraction (HFpEF) and has been linked to systemic inflammation. However, the plasma proteomic profiles associated with specific CHIP driver mutations and their prognostic relevance in HFpEF are not well characterized.
Purpose
To identify circulating protein signatures associated with CHIP driver mutations and to develop and externally validate a CHIP-related proteomic risk score (CHIP-ProtRS) for predicting adverse clinical outcomes in HFpEF.
Methods
The derivation cohort included 118 participants from TOPCAT. Associations between CHIP mutations and plasma protein levels were examined using gene-specific linear regression with adjustment for multiple testing. Proteins significantly associated with individual CHIP driver genes were used to construct gene-specific proteomic classifiers via nested cross-validated LASSO logistic regression. A composite CHIP-ProtRS was subsequently derived using LASSO Cox regression and externally validated in three independent HFpEF cohorts (n=654). The primary endpoint was a composite of death and heart failure hospitalization.
Results
Among TOPCAT participants, 30.5% carried at least one CHIP mutation. Distinct proteomic signatures were identified for TET2 (7 proteins), DNMT3A (9 proteins), and PPM1D (6 proteins), predominantly related to immune and inflammatory pathways. Gene-specific LASSO models demonstrated good discrimination for CHIP mutation status (AUC range 0.743–0.830). The CHIP-ProtRS was independently associated with the primary endpoint in the derivation cohort (adjusted HR 1.42, 95% CI 1.04–1.95) and in the external validation cohorts (HR 1.15, 95% CI 1.01–1.31) (Table 1). Risk stratification by CHIP-ProtRS tertiles was consistent across derivation and validation cohorts (log-rank p=0.034 and p=0.001, respectively) (Figure 1).
Conclusion
CHIP is associated with distinct alterations in circulating immune and inflammatory proteins in HFpEF. A proteomic risk score derived from these signatures is independently associated with adverse clinical outcomes, supporting the potential role of CHIP-ProtRS as a biomarker of risk and underlying disease biology in HFpEF.Table 1.Multivariable Cox analysis Figure 2.Cumulative incidence curves