Naltrexone (NTX), an opioid antagonist metabolized by aldo-keto reductase 1C4 (AKR1C4), is prescribed for psychiatric conditions like eating disorders with variable response. Systemic exposure is highly variable in adults, yet no data exist in children.The purpose of this study was to evaluate NTX exposure in adolescents with eating disorders. Adolescents aged 12-21 years with eating disorders underwent postdose blood sampling in the fasted and/or fed state. NTX and primary active metabolite, 6-βnaltrexol, were determined by ultra-high performance liquid chromatography tandem mass spectrometry. Pharmacokinetic parameters were determined by noncompartmental analysis. DNA was genotyped for AKR1C4 missense mutations associated with decreased activity (rs3829125 and rs17134592). Linear mixed effects modeling was performed. In 21 participants, aged 16.9 ± 1.9 years (15-21 years), 81% female participants, maximum concentration (C max ) was 90.4 ± 129 nM/mg/ kg, area under the concentration-time curve from zero to infinity (AUC 0-∞ ) was 166 ± 154 nM h/mg/kg, and varied 63-fold and 21-fold, respectively. Compared with wildtype, those with AKR1C4 allelic variations (n = 7) displayed 3.2-fold higher AUC 0-∞ , four-fold higher C max and delayed time to T max . Linear mixed effects modeling demonstrated a large effect of genotype on AUC 0-∞ (Cohen's d −2.3) and C max (Cohen's d −1.4). Food effect was large for AUC 0-∞ (Cohen's d 2.6), but highly variable and failed to reach significance for C max. The respective model accounted for 82% of the variance in NTX AUC 0-∞ and 46% of the variance in C max . NTX systemic exposure is highly variable in adolescents with eating disorders and modulated, in part, by AKR1C4 genotype and food intake. These findings may, in part, explain the large degree of interindividual variability observed response to NTX.