Poster #85 - Xiaotong Yang

Cell type proportions rather than DNA methylation in the cord blood show statistically significant associations with severe preeclampsia

Xiaotong Yang1✝, Wenting Liu1✝, Zhixin Mao1, Yuheng Du1, Cameron Lassiter2, Fadhl M. AlAkwaa1, Paula A Benny2, Lana X Garmire1* 1. Department of Computational Medicine and Bioinformatics, University of Michigan, Ann Arbor, MI 2. University of Hawaii Cancer Center, Epidemiology, Honolulu, HI 3. Department of Neurology, University of Michigan, Ann Arbor, MI ✝ These authors contributed equally to the work * corresponding author

Background Preeclampsia is a severe pregnancy complication that threatens maternal and neonatal health. Previous epigenome-wide association studies (EWAS) in cord blood produced inconsistent results, likely due to inadequate adjustment for confounders. Methods We analyzed DNA methylation in cord blood from newborns in a multi-ethnic Hawaiian cohort (24 cases, 38 controls), rigorously adjusting for maternal age, BMI, parity, and estimated cell proportions. Two public datasets were re-analyzed with the same adjustments and combined with ours in a pooled analysis (58 cases, 71 controls). To disentangle preeclampsia effects from gestational age, we also included idiopathic preterm samples (n=11). Results After controlling for cell proportions and clinical covariates, previously reported CpG methylation changes associated with severe preeclampsia largely disappeared in our dataset, in the re-analyzed public datasets, and in the combined meta-analysis. This null finding held even after including idiopathic preterm controls. Instead, severe preeclampsia primarily influenced immune cell composition, particularly CD8T and natural killer (NK) cells. Furthermore, gestational age progression itself was associated with significant shifts in granulocyte, nRBC, CD8T, and B cell proportions. Conclusions Our findings demonstrate that reported differential methylation patterns in cord blood are artifacts of unadjusted confounders, especially cell type heterogeneity and gestational age. Severe preeclampsia is not linked to robust DNA methylation changes but instead to altered immune cell proportions. These results underscore the need for rigorous confounder adjustment in EWAS.