Placental somatic mutation in human stillbirth and live birth: A pilot case-control study of paired placental, fetal, and maternal whole genomes.

A high frequency of single nucleotide somatic mutations in the placenta has been recently described, but its relationship to placental dysfunction is unknown. We performed a pilot case-control study using paired fetal, maternal, and placental samples collected from healthy live birth controls (n = 10), live births with fetal growth restriction (FGR) due to placental insufficiency (n = 7), and stillbirths with FGR and placental insufficiency (n = 11). We quantified single nucleotide and structural somatic variants using bulk whole genome sequencing (30-60X coverage) in four biopsies from each placenta. We also assessed their association with clinical and histological evidence of placental dysfunction. Seventeen pregnancies had sufficiently high-quality placental, fetal, and maternal DNA for analysis. Each placenta had a median of 473 variants (range 111–870), with 95 % arising in just one biopsy within each placenta. In controls, live births with FGR, and stillbirths, the median variant counts per placenta were 514 (IQR 381–779), 582 (450–735), and 338 (245–441), respectively. After adjusting for depth of sequencing coverage and gestational age at birth, the somatic mutation burden was similar between groups (FGR live births vs. controls, adjusted diff. 59, 95 % CI -218 to +336; stillbirths vs controls, adjusted diff. −34, −351 to +419), and with no association with placental dysfunction (p = 0.7). We confirmed the high prevalence of somatic mutation in the human placenta and conclude that the placenta is highly clonal. We were not able to identify any relationship between somatic mutation burden and clinical or histologic placental insufficiency. • Somatic mutation is highly prevalent in the human placenta. • 95 % of comatic variants were present in just one of four biopsies in each placenta. • Somatic mutation estimates were driven by sequencing depth and gestational age. • Two variant calling algorithms yield very different somatic variation estimates. • We found no association between somatic variant burden and clinical outcome. [ABSTRACT FROM AUTHOR]