610 - Neonatal Epigenetics and Trajectories of Attention Problems in Children Born Very Preterm

Publication Number: 4597.610

Icy Liang, The Warren Alpert Medical School of Brown University, Providence, RI, United States; barry lester, The Warren Alpert Medical School of Brown University, east greenwich, RI, United States; Brian S.. Carter, Children's Mercy Hospitals and Clinics, Kansas City, MO, United States; Jennifer Check, Wake Forest School of Medicine of Wake Forest Baptist Medical Center, Winston-Salem, NC, United States; Jennifer Helderman, Wake Forest School of Medicine of Wake Forest Baptist Medical Center, Clemmons, NC, United States; Julie Hofheimer, University of North Carolina at Chapel Hill, Durham, NC, United States; Elisabeth McGowan, Women & Infants Hospital of Rhode Island, Providence, RI, United States; Charles R. Neal, University of Hawaii, John A. Burns School of Medicine, Kailua, HI, United States; Steven L. Pastyrnak, Michigan State University College of Human Medicine, Grand Rapids, MI, United States; Lynne Smith, University of California, Los Angeles David Geffen School of Medicine, Torrance, CA, United States; Thomas M.. O’Shea, University of North Carolina, Chapel Hill, NC, United States; Carmen Marsit, Emory University Rollins School of Public Health, Atlanta, GA, United States; Todd M. Everson, Emory University, Atlanta, GA, United States; Marie Camerota, The Warren Alpert Medical School of Brown University, Providence, RI, United States

Background: Children born very preterm are at increased risk for neurodevelopmental impairments, with a 2 to 4 times greater likelihood of developing attention problems. Epigenetic modifications, particularly DNA methylation (DNAm), are increasingly being explored as predictors of attention problems. Understanding how DNAm relates to trajectories of attention problems prior to school entry could support earlier diagnoses and academic support in children with a history of preterm birth.

Objective: We examined how DNAm at birth is associated with trajectories of attention problems between the ages 2 to 5 years in children born very preterm.

Design/Methods: This analysis included 477 children born < 30 weeks gestational age (GA) from the Neonatal Neurobehavior and Outcomes in Very Preterm Infants study (Table 1). DNAm was measured from buccal swabs obtained at neonatal intensive care unit (NICU) discharge. Child attention problems were evaluated yearly by parents using the Child Behavior Checklist at ages 2 and 3 years and the Behavioral Assessment System for Children at ages 4 and 5 years. We focused on 33 CpG sites we previously found to be associated with attention problems at age 2 in this sample and characterized their relationship with attention problem trajectories estimated from latent growth curve models. All models included cell type, sample plate, site, GA at birth, GA at buccal swab, infant sex, and neonatal medical morbidities as covariates.

Results: Of the 33 CpG sites, 12 were significantly associated with both the trajectory intercept and slope (Figure 1). Seven of these sites were positively associated with the intercept and negatively associated with the slope, indicating higher DNAm was associated with higher attention problems at age 2 but a decrease in attention problems from age 2 to 5. The other 5 CpG sites were negatively associated with the intercept and positively associated with the slope. Several of the significant CpG sites were located in genes previously shown to be related to neurodevelopmental traits and mental health (e.g., COG4, DENND1C, BCL7A). Other CpG sites were located in genes related to telomere length, educational attainment, substance use, and physical health (Table 2).

Conclusion(s): Neonatal DNAm is associated with changes in attention problems in children born very preterm. Epigenetic measurements at birth could potentially identify children born very preterm at highest risk for attention-deficit/hyperactivity disorder and other neurodevelopmental disorders, enabling earlier intervention and prevention efforts.

Demographics and Clinical Characteristics of the Study Sample
GA = gestational age, GED = general equivalency diploma, HS = high school, NICU = neonatal intensive care unit, SES = socioeconomic status. *Serious brain injury included parenchymal echo density, periventricular leukomalacia, or ventricular dilation diagnosed via cranial ultrasound.

Coefficients for each CpG Site in Association with Attention Trajectory Parameters
Parameter estimates from 33 univariate, conditional latent growth curve (LGC) models are depicted in a dot and whisker plot ordered by increasing slope values. Significant estimates are bolded and do not cross the dotted vertical line. Estimates are represented by a dot, and standard error bars depict 95% confidence intervals.

Parameter Estimates from Univariate Latent Growth Curve Models
Univariate models investigated 33 CpG sites as predictors in association with trajectory parameters. Gene annotations were derived from the Illumina manifest file. CpGs without gene annotations are intergenic. Bolded values are significant at p < 0.05 for both intercept and slope estimates.