https://doi.org/10.1007/s00392-025-02625-4
1Universitätsmedizin der Johannes Gutenberg-Universität Mainz Computergestützte Systemmedizin Mainz, Deutschland; 2Universitätsmedizin der Johannes-Gutenberg Universtität Mainz Institut für Physiologische Chemie Mainz, Deutschland; 3Universitätsmedizin der Johannes Gutenberg-Universität Mainz Präventive Kardiologie und Medizinische Prävention Mainz, Deutschland; 4Universitätsmedizin der Johannes Gutenberg-Universität Mainz Klinische Epidemiologie und Systemmedizin Mainz, Deutschland; 5Johannes-Gutenberg Universität Mainz Institute of Organismic and Molecular Evolution Mainz, Deutschland
Heart failure (HF) is often accompanied by various comorbidities, such as diabetes, hypertension, ischaemic heart disease, hyperlipidaemia, chronic kidney disease (CAD) or atrial fibrillation which further complicate its management and prognosis. Exploring the HF-subtype-specific lipidomic profiles associated with these cardiometabolic comorbidities can shed light on their interplay with HF and potentially uncover new molecular targets. This project aims to investigate the association of lipids with heart failure metabolic comorbidities and their differences between heart failure subtypes using a highly reliable and quantitative lipidomics approach (4D ClinLip).
Samples were obtained from the MyoVasc cohort (NCT04064450), an observational, prospective cohort study on the development and progression of heart failure. Lipidomics was measured at baseline for 1741 participants of which 401 had been diagnosed with HFrEF and 726 with HFpEF. Quantification was carried out using mass spectrometry in a 4D-LC/TIMS-IMS approach. Lipid signatures for HFpEF- and HFrEF-specific signatures per selected comorbidities and biochemical and laboratory parameters were identified using sparse group LASSO regularized regression. Employing lipid-based genome-wide association studies (GWAS), lipid-signatures were transformed into gene-signatures using Baker Institute Lipidomic PheWeb, pathway-analysis of lipid-associated genes was performed using Metascape.
Sparse group LASSO regularized regression selected HFpEF- and HFrEF-specific signatures for 32 cardiometabolic comorbidities and laboratory and biochemical parameters across several lipid classes. While there was a great overlap of lipid-signatures between HFpEF and HFrEF for diabetes, BMI, obesity and HbA1c, surprisingly several lipids were regulated in an opposite manner among HF subtypes. Pathway analysis was performed for both overlapping gene-signatures between the HF subtypes and subtype-specific gene-signatures. Lipids of overlapping pathways between signatures show strong association with metabolic traits, and are currently being investigated. Additionally, both lipid- and gene-signatures of HFpEF and HFrEF specific to Diabetes, HbA1c and C-peptide were compared for a better insight into diabetic and insulin-resistance-specific mechanisms in heart failure. While the resulting pathways are currently still under investigation, comparison of significantly associated lipids between the three clinical traits already show interesting differences as well as commonalities.
The Lipidomics analysis of HF subtypes HFpEF and HFrEF show promising insights into molecular differences. With using 4D ClinLip measurements on the MyoVasc study as well as intersecting the findings with phenome-wide association studies, we hope to uncover new molecular targets. This will ultimately enable more personalized approaches to diagnosis, risk stratification, and treatment selection for patients with HF subtypes HFpEF and HFrEF, specifically focused on diabetes and insulin-resistance.