NIH grant: Molecular Profiling and Risk Assessment of Preeclampsia Using Cell-Free DNA (1R01HL183263-01)

retained text

NIH grant 1R01HL183263-01: Molecular Profiling and Risk Assessment of Preeclampsia Using Cell-Free DNA
Institute: {'code': 'HL', 'abbreviation': 'NHLBI', 'name': 'National Heart Lung and Blood Institute', 'admin_org_id': '12500068', 'admin_funding_url': ''}. PI: Gavin  Ha. Award: $1697415.
Start: 2026-07-01T00:00:00.
PROJECT SUMMARY
Preeclampsia (PE) is a major worldwide contributor to perinatal morbidity. Prediction of PE before disease
onset is challenging, and current methods either lack the necessary sensitivity, or implementation has been
limited due to the number and complexity of measures. Additionally, current approaches that rely solely on
assessing underlying risk factors struggle to identify individuals who have no clinically evident risk factors (e.g.,
chronic hypertension, pregestational diabetes, obesity, etc.) but go on to develop PE. As the placenta plays a
well-defined role in the developmental origins of PE, our group leveraged cell-free DNA (cfDNA) circulating in
maternal plasma to identify differential placental and other tissue signatures indicative of PE risk. These
signatures rely on the epigenetic features of cfDNA, as informed by nucleosome positioning, which dictates
cfDNA fragmentation patterns. Nucleosome positioning is cell- and tissue-specific allowing for a real-time and
non-invasive assessment of multiple tissue sites with one blood sample. Using this framework, we developed
PEARL (Preeclampsia Early Assessment of Risk from Liquid biopsy) which predicts PE from the same data
generated from early pregnancy (≤16 weeks) prenatal cfDNA screening (PDNAS) for aneuploidy. In our
preliminary work, we achieved favorable performance (AUC ≥0.84) for the prediction of PE resulting in preterm
delivery using cfDNA sequence data generated for PDNAS, negating the need for additional samples or
laboratory workflows. We also identified that those who go on to develop PE have increased endothelial tissue
signatures detectable early in pregnancy (≤16 weeks). This proposal is a direct extension of this work. In Aim
1, we will first improve the performance of PEARL using a three-pronged approach designed to use deeper
sequenced (30x) cfDNA from early in pregnancy and at the time of PE, coupled with studies of placental
chromatin accessibility, to define robust and durable epigenetic signatures associated with PE. We will
translate these signatures back to low-coverage (0.5x) sequence data that is routinely generated through
PDNAS by training and validating on a new and distinct set of thousands of samples. In Aim 2, given our
observation of increased endothelial tissue signatures early in pregnancy in those that develop PE, we will test
the hypothesis that this represents a common feature in those who develop PE, both with and without
underlying traditional risk factors. We will employ novel computational tools on multi-omics data sources
(transcriptome, methylome, genome) from both placenta tissue and paired plasma cfDNA, followed by
detection of these differential epigenetic and expression signatures in early pregnancy low coverage (0.5x)
cfDNA sequence data. Results of this proposal will refine a PE prediction tool rooted in the underlying
pathophysiology of the disease that uses non-invasive means of sampling, providing an accessible screening
modality that can eventually be used to mitigate PE risk and its sequelae in all PE subtypes.

document fetched 2026-07-12 08:14