BACKGROUND/PURPOSE: Mycophenolate mofetil (MMF) is a prodrug that is pre-systemically hydrolyzed by esterases to the biologically active mycophenolic acid (MPA) and is mainly inactivated by UDP-glucuronosyltransferases (UGTs) in the liver, intestine and kidney. MMF has been used off-label in SLE because of its favorable side effect spectrum. Optimal MMF dosing in childhood-onset SLE (cSLE) is not well-established. Little is known about the pharmacokinetics (PK) and pharmacogenetics (PG) of MPA in cSLE although in pediatric transplant populations, graft survival is more closely related to individualized dosing based on systemic exposure (PK) to MPA rather than weight-adjusted dosing. Transplant literature has also elucidated the potential influence of genetic variants in UGTs on the large variability of MPA exposure and the occurrence of common MMF-associated adverse events. The purpose of this study is to explore the importance and potential benefits of individualized MMF therapy in cSLE by determining whether MPA PK correlates with cSLE therapeutic response and whether MPA PG influences MMF exposure. METHODS: Full 9-hour pharmacokinetic profiles were obtained from cSLE patients (n = 18; F:M = 17:1; age 10-30 years, 43% Caucasian, 57% African-American, 81% non-Hispanic) who were on stable MMF therapy. The MPA PK parameters dose-normalized area under the curve from 0 to 12 hours (AUC0-12h), peak concentration (Cmax), time to peak concentration (Tmax) and oral clearance (CL/F) were assessed and analyzed by non-compartmental analysis. Genomic DNA was extracted using standard procedures and genotyped for UGT1A8*3 (830G>A), UGT1A9*3 (98T>C), 1A9-2152T>C, 1A9-440C>T, 1A9-331T>C, 1A9-275T>A and UGT2B7-900A>G by TaqMan assay and direct sequencing. Change in SLE disease activity while on MMF was measured using the British Isles Lupus Assessment Group (BILAG) index. RESULTS: Large inter-patient variability in AUC0-12h (mean + S.D.: 38.7 + 19.8 mg hr/L/gMPA, range: 14.9-95.4) and oral clearance (CL/F) of MPA (mean + S.D.: 27.9 + 13.7 L/hr, range: 8.3-59.3) was observed. The AUC0-12h and weight-adjusted MMF dosing were only moderately correlated (r = 0.56, P = 0.01). An AUC0-12h of >30 mg hr/L was associated with decreased BILAG scores while on MMF (P = 0.002). Patients with UGT1A9 -440T (-331C) or UGT2B7 -900G appeared to have a trend toward lower CL/F and higher MPA exposure (AUC) than the wild-type of all evaluated single nucleotide polymorphisms. Patients with UGT1A9 -275A showed relatively high CL/F over a wide range. In an exploratory fashion, wild-type and -275A-carriers were analyzed and showed significantly higher CL/F compared to the rest of the carriers of SNPs 1A9 -440T (-331C) or 2B7 -900G (p < 0.05, Mann-Whitney's U test). CONCLUSION: Our data suggest that weight-adjusted MMF dosing alone does not reliably allow for the prediction of exposure to MPA in cSLE and that MPA exposure measured as AUC0-12h of at least 30 mg hr/L is required for cSLE improvement. UGT1A9 and UGT2B7 polymorphisms accounted in part for inter-individual differences in MPA exposure in patients with cSLE.