The following is a summary of “Cell-free DNA in maternal blood and artificial intelligence: accurate prenatal detection of fetal congenital heart defects,” published in the January 2023 issue of Obstetrics and Gynecology by Singh, et al.

One crucial process for regulating gene expression in the development of the heart is DNA cytosine nucleotide methylation, also known as epigenomics and epigenetics. Fetal congenital cardiac abnormalities may be found using a combination of artificial intelligence and whole-genome epigenomic analysis of circulating cell-free DNA in maternal blood. For a study, researchers sought to develop minimally invasive methods for detecting embryonic congenital heart abnormalities, including genome-wide DNA cytosine methylation and artificial intelligence studies of circulating cell-free DNA.

In the prospective investigation, circulating cell-free DNA was subjected to whole-genome cytosine nucleotide methylation profiling utilizing the Illumina Infinium MethylationEPIC BeadChip array. For the purpose of identifying congenital hearts, various artificial intelligence methods were examined. To better understand the etiology of isolated congenital heart abnormalities, the Ingenuity Pathway Analysis tool was utilized to pinpoint gene pathways that were epigenetically changed and crucial in congenital heart defect pathogenesis.

There were 26 matched controls and 12 instances of isolated, nonsyndromic congenital cardiac abnormalities. In instances of congenital cardiac defects compared to controls, 5,918 cytosine nucleotides containing 4,976 genes exhibited substantially different methylation, which is defined as a P value of<.05 and a change in whole-genome cytosine nucleotide methylation of ≥5%. Excellent congenital heart defect predicting accuracy was attained using artificial intelligence analysis of the methylation data (areas under the receiver operating characteristic curve, ≥0.92). For instance, an artificial intelligence model that included five whole-genome cytosine nucleotide indicators had a 98% sensitivity and 94% specificity, with an area under the receiver operating characteristic curve of 0.97 (95% CI, 0.87-1.0). The following crucial cardiac developmental processes—”cardiovascular system development and function,” “cardiac hypertrophy,” “congenital heart abnormality,” and “cardiovascular disease”—were all associated with epigenetic modifications in genes and gene networks. This gave the results biological plausibility.

Using artificial intelligence and DNA methylation analysis of circulating cell-free DNA for the prediction of prenatal congenital heart defect, the study reported on the viability of minimally invasive detection of fetal congenital heart defect. The results also supported the notion that epigenetic modifications played a significant role in the development of congenital cardiac defects.

Reference: ajog.org/article/S0002-9378(22)00618-4/fulltext