Identification of the Metabolic Signature of Atrial Fibrillation for Personalized Prevention

Overview

Atrial fibrillation (AF) is a major public health problem. The efficacy of the existing techniques is limited in the more aggressive forms. It is therefore necessary to develop approaches, in particular the identification of relevant biomarkers, to prevent the onset, recurrence or progression of AF in at-risk patients. The objective of this study is to describe the longitudinal metabolic and biomolecular signature of AF in patients eligible for cardiac ablation.

Atrial fibrillation (AF) is a major public health problem. Its prevalence exceeds 2%. The main aim of drug treatment is to prevent the onset of stroke and heart failure, but side effects often require discontinuation, and contraindications limit their use. Rhythm control strategies based on catheter ablation have led to significant progress in incident AF, improving quality of life. Nevertheless, the efficacy of these techniques is limited in the more aggressive forms. Significant recurrence rates are reported one year after ablation, and access to them is often reserved for symptomatic patients due to their invasive and costly nature. It is therefore necessary to develop approaches to prevent the onset, recurrence or progression of AF in at-risk patients. While the pathophysiology of AF involves metabolic remodelling that can be observed in animal and human models, no clinically relevant metabolites have been identified as biomarkers of the risk of AF onset or progression, with a view to preventive and personalized management. In response to this unmet need, this project aims to develop a method for assessing the risk of AF recurrence, combining the identification of a metabolic signature of the arrhythmia and the patient, with a machine learning approach to aggregate conventional risk factors and metabolic biomarkers. A longitudinal clinical study will be conducted on patients scheduled for AF ablation, to monitor changes in their metabolic signature over 12 months, in parallel with arrhythmia progression. Using machine learning, the study team will establish and validate a classifier retrospectively stratifying patients with or without recurrent AF, and compare this method with canonical risk stratification. This will enable to consider personalized management of patients at risk of recurrence, with the aim of reducing human and economic costs.

Conditions

Interventions

Eligibility

Locations (1)

Outcomes

Central Contacts