Exploring the effect of antenatal depression treatment on children's epigenetic profiles: Findings from a pilot randomized controlled trial

Background: Children prenatally exposed to maternal depression more often show behavioral and emotional problems compared to unexposed children, possibly through epigenetic alterations. Current evidence is largely based on animal and observational human studies. Therefore, evidence from experimental human studies is needed. In this follow-up of a small randomized controlled trial (RCT), DNA-methylation was compared between children of women who had received cognitive behavioral therapy (CBT) for antenatal depression and children of women who had received treatment as usual (TAU). Originally, 54 women were allocated to CBT or TAU. A beneficial treatment effect was found on women's mood symptoms. Findings: We describe DNA methylation findings in buccal swab DNA of the 3-7-year-old children (CBT(N) = 12, TAU(N) = 11), at a genome-wide level at 770,668 CpG sites and at 729 CpG sites spanning 16 a priori selected candidate genes, including the glucocorticoid receptor (NR3C1). We additionally explored associations with women's baseline depression and anxiety symptoms and offspring DNA methylation, regardless of treatment. Children from the CBT group had overall lower DNA methylation compared to children from the TAU group (mean δβ = - 0.028, 95% CI - 0.035 to - 0.022). Although 68% of the promoter-associated NR3C1 probes were less methylated in the CBT group, with cg26464411 as top most differentially methylated CpG site (p = 0.038), mean DNA methylation of all NR3C1 promoter-associated probes did not differ significantly between the CBT and TAU groups (mean δβ = 0.002, 95%CI - 0.010 to 0.011). None of the effects survived correction for multiple testing. There were no differences in mean DNA methylation between the children born to women with more severe depression or anxiety compared to children born to women with mild symptoms of depression or anxiety at baseline (mean δβ (depression) = 0.0008, 95% CI - 0.007 to 0.008; mean δβ (anxiety) = 0.0002, 95% CI - 0.004 to 0.005). Conclusion: We found preliminary evidence of a possible effect of CBT during pregnancy on widespread methylation in children's genomes and a trend toward lower methylation of a CpG site previously shown by others to be linked to depression and child maltreatment. However, none of the effects survived correction for multiple testing and larger studies are warranted. Trial registration: Trial registration of the original RCT: ACTRN12607000397415. Registered on 2 August 2007