A Precision Medicine Approach to Natriuretic Peptide-Guided Risk Prediction and Treatment: A Retrospective Analysis in the STOP-HF Cohort

Abstract

Introduction: Individual studies show that NP-guided treatment is an effective intervention for the prevention of HF. We hypothesised that this approach may be optimised utilising a precision medicine approach. Methods: A systematic literature review and meta-analysis were performed to establish the effectiveness of NP-guided treatment versus standard care. The original research work uses data from the STOP-HF cohort, a prospective, longitudinal study population comprising patients over-40, with risk factors for developing HF. We evaluated the incremental benefit of molar NT-proBNP/BNP ratio in predicting HF/LVD. We developed a genetic adjustment (GA) score of genetic variants associated with NT-proBNP levels and cardioprotection (rs198389, rs6676300, and rs13107325). We tested its association with outcomes and used it to generate genetically adjusted NT-proBNP which was evaluated for the prediction of HF/LVD. Finally, we defined a cohort of patients with Stage B HF and relatively low BNP (< 20 pg/mL) and evaluated the clinical, genetic, and echocardiographic characteristics of these patients. Results: The systematic review and meta-analysis indicate that NP-guided treatment is an effective intervention for preventing cardiovascular and all-cause hospitalisation, and ventricular dysfunction, although the effect on mortality is unclear. Molar NT-proBNP/BNP ratio was inversely associated with risk of HF and/or LVD but did not improve risk prediction metrics compared to individual NP measurements. The GA score was significantly associated with risk of prevalent hypertension and left atrial volume index (LAVI), but did not improve risk prediction for HF/LVD beyond standard NT-proBNP. Patients with Stage B HF and low BNP were more likely to have a lower GA score and had earlier features of Stage B HF than those with BNP > 20 pg/mL yet were still at risk due to increased LAVI. Conclusions: Personalised approaches to NP-guided risk prediction using molar NT- proBNP/BNP ratio and genetically adjusted NT-proBNP did not improve the prediction of HF/LVD beyond current NP-based screening. GA score evaluation in HF prevention programmes may help identify patients with relatively low BNP levels and Stage B HF who would be missed according to current thresholds for referral for echocardiography.