Proteomics-based computational modelling to personalise HF management
31 Aug 2024
The Young Investigator Award Sessions continue in the Research Gateway and one presentation in particular captures the essence of this year’s Congress Spotlight, aiming to understand disease heterogeneity to enable tailored management of advanced heart failure (HF) in the future.
Doctor Niklas Beyhoff (University of Oxford - Oxford, UK) presents a study that used a computational model of patient-specific myocardial metabolism1 to assess individual bioenergetic phenotypes and explore their clinical implications. Using proteomic analysis of left ventricular (LV) biopsies, individual energetic states were reconstructed for 48 patients with advanced HF (mean LV ejection fraction [LVEF] 20±10%; non-ischaemic aetiology: 72.9%; females: 52.1%) undergoing heart transplantation (n=25) or LV assist device (LVAD) implantation (n=23) and 17 sex-matched non-failing controls in the study cohort. The bioenergetic impact of alterations in substrate availability and myocardial workload were simulated, and the model’s ability to predict myocardial reverse remodelling after LVAD implantation was assessed.
Computational modelling indicated impaired ATP production capacity in patients with HF (p<0.01), although there was considerable interindividual variation, ranging from normal values to substantial reductions. While overall oxygen consumption was lower in failing hearts, the amount of ATP generated per mole of oxygen did not differ. A shift from fatty acid to glucose utilisation was observed in most patients, depending on substrate availability and myocardial workload. These key findings were validated in an external proteomic dataset comprising 44 advanced HF samples and 28 non-failing samples.
Notably, when the clinical implications of the substrate shift were investigated in the study cohort, the ratio of fatty acid to glucose oxidation predicted absolute LVEF recovery ≥ 10% after LVAD implantation (C-index 0.93, p=0.003) and correlated significantly with improvement in postoperative LVEF (r=0.67, p<0.05). ATP production capacity was not associated with reverse remodelling after LVAD implantation.
Taken together, the results gained from proteomics-based computational modelling identified the ratio of fatty acid to glucose utilisation as a predictor of myocardial recovery after LVAD implantation. This finding suggests the potential for substrate manipulation as a therapeutic approach in a subset of patients with deranged fuel use. Furthermore, these results highlight the value of computational assessment of myocardial metabolism to improve understanding of HF heterogeneity, enhance individual risk stratification and enable more personalised clinical decision-making in the future.
References
- Berndt N, et al. Circulation. 2021;144:1926–1939.
