The following is a summary of “Validation of an Integrated Genetic‐Epigenetic Test for the Assessment of Coronary Heart Disease,” published in the November 2023 issue of Cardiology by Philibert et al.

Coronary heart disease (CHD) remains a leading global cause of mortality, yet existing diagnostic tools are often invasive, costly, and fail to provide comprehensive post-diagnosis therapeutic guidance. Leveraging insights from the Framingham Heart Study, previous in silico studies hinted at the potential of integrated genetic–epigenetic tools as a novel method for CHD assessment. In this investigation, utilizing data from two additional cohorts comprising 449 cases and 2067 controls, researchers refined a machine learning model to identify symptomatic CHD better. 

The study group translated and validated the earlier findings into a clinically deployable method by implementing an artificial intelligence-guided approach integrating inputs from 6 methylation-sensitive digital polymerase chain reaction assays and 10 genotyping assays. Their method yielded promising results, demonstrating an overall average area under the curve of 82%, with sensitivity and specificity of 79% and 76%, respectively, across three test cohorts. Analysis of specific cytosine-phospho-guanine loci underscored their correlation with critical risk pathways associated with atherosclerosis, implying potential targeted therapeutic strategies. 

This scalable integrated genetic–epigenetic approach holds promise for CHD diagnosis, outperforming existing methods and offering personalized insights into CHD therapy. Moreover, owing to the adaptable nature of DNA methylation and the simplicity of methylation-sensitive digital polymerase chain reaction techniques, these findings suggest avenues for precision epigenetic strategies in monitoring CHD treatment response.

Source: ahajournals.org/doi/10.1161/JAHA.123.030934