Differential DNA Methylation of the Brain-Derived Neurotrophic Factor Gene is Observed after Pediatric Traumatic Brain Injury Compared to Orthopedic Injury

Pediatric traumatic brain injury (TBI) triggers biological changes that may differ from those observed in non-brain injuries. BDNF DNA methylation (DNAm) may serve as a novel, dynamic biomarker of the brains response and help identify TBI-specific epigenetic patterns relevant to later recovery. Therefore, the purpose of this study was to examine whether BDNF DNAm differed between children with TBI and those with orthopedic injury (OI, comparison group) acutely and over time. Data were derived from the Epigenetic Effects on TBI Recovery (EETR) study, a prospective, longitudinal cohort study conducted at UPMC Childrens Hospital of Pittsburgh. Children aged 3 to 18 years hospitalized at a minimum of overnight for complicated mild to severe TBI or OI without head trauma were enrolled. Exclusion criteria included prior hospitalization for TBI, pre-existing neurological or psychiatric conditions, or sensory or motor impairments precluding study participation. Blood samples were collected during hospitalization (mean=31.6 hours post-injury) and at 6 (mean=216.9 days) and 12 months (mean=405.9 days) post-injury. The primary outcome variable was DNAm assessed via pyrosequencing at five quality-controlled CpG sites in the BDNF gene (chromosome 11, hg38 positions 27722033, 27722036, 27722047, 27701612, and 27701614). The primary exposure was injury type (TBI vs. OI), with severity (measured via Glasgow Coma Scale [GCS]) examined as a secondary exposure within the TBI group. Primary covariates included age, sex, and race; secondary covariates included pubertal status, age-adjusted BMI, non-head injury severity, socioeconomic status, and psychosocial adversity. The final analysis sample included n=189 participants with TBI and n=105 participants with OI. Participants were 66.3% male, 83.2% White, and had a mean age of 10.6 ({+/-}4.4) years at the time of enrollment. Acutely, children with TBI showed significantly lower DNAm at three of five sites (3.17%- 5.83% lower; p=0.0044 to 6.48E-06) while controlling for age, sex, and race. One site remained significantly lower at 12 months (8.56% lower; p=0.0045); no significant differences were observed at 6 months. Observed differences remained robust across sensitivity models adjusting for secondary covariates. GCS-measured TBI severity was not associated with DNAm at any time point. These findings suggest that BDNF DNAm differs between children with TBI and those with OI, particularly in the acute period. BDNF DNAm differences may reflect early biological responses that are specific to TBI.