This study describes the disposition of cyclosporine in 24 children undergoing HSCT after IV administration and oral administration of 2 different formulations: Neoral® and Sandimmune®. Whole-blood samples for cyclosporine measurement were obtained after the first IV cyclosporine dose on day −1 and after the morning dose on each of the first 2 days of oral cyclosporine administration. Neoral® was given on the first day of oral cyclosporine administration, and a single dose of Sandimmune® was given in the morning of the second day. Oral cyclosporine doses were identical and were given on an empty stomach. Cyclosporine concentration-time data were analyzed by noncompartmental methods. A total of 24 children completed the study. After IV administration of 1.5± 0.07 mg/kg, the mean cyclosporine maximum concentration (Cmax), volume of distribution at steady state, elimination half-life, clearance, and area under the concentration-versus-time curve (AUC) were 1103.4± 787.71 μg/L, 2.6± 1.52 L/kg, 4.2± 2.49 hours, 0.56± 0.255 L/hr/kg, and 2852± 1198 μg· hr· L−1, respectively. Children took their first oral cyclosporine dose on average 29.4± 6.88 days after their first IV dose. The mean Cmax observed after Neoral® administration was significantly higher than after Sandimmune® administration (595 ±349.7 vs 486± 363.0 μg/L; P= .043), as was AUC 0–12 hr (3432± 1563 vs 3144± 1780 μg· hr· L−1; P= .022). There was no difference in time to reach Cmax (tmax) between the formulations. The coefficients of variation in Cmax, tmax, and AUC were all lower with Neoral® than with Sandimmune®, but remained substantial. The cyclosporine concentration at hour 4 after oral Neoral® and Sandimmune® administration exhibited the strongest correlation with AUC0–12 hr. However, unlike Sandimmune®, the trough cyclosporine concentration after Neoral® administration showed the weakest relationship with AUC (Spearman’s ρ coefficient= 0.568; P= .027). In children undergoing HSCT, cyclosporine absorption is lower than that usually seen in children who have undergone solid organ transplantations. The relationship between cyclosporine concentrations during the dosing interval and AUC differs depending on the oral formulation administered. Thus, Neoral® dose adjustment based on trough concentration may not be appropriate. The link between pharmacokinetic parameters and clinical outcomes such as GvHD must be studied further.