Overview of PPK model
The MPA concentrations were well described by a two-compartment model parameterized as Ka (0.45 h-1), CL (9.86 L/h), Vc (19.69 L), Vp (408.32 L), and Q (23.01 L/h). Previous reports in children supported this result where Ka ranged from 0.39 to 5.16 h-1, CL ranged from 6.42 to 25.3 L/h, Vc ranged from 4.75 to 64.7 L, Q ranged from 3.74 to 25.6 L/h and Vp ranged from 16.8 to 411.0 L (15). The variability of Ka was 164.2%, where previous reports showed Ka were highly variable in pediatric patients ranging from 20.5% ~ 308.4%. However, only age was identified as a significant covariate on absorption in current pediatric dataset. Our findings indicated that dosage forms significantly affected Ka, and its inclusion in final model resulted in a 57.2% reduction of inter-individual variation. The fastest absorption was dispersible formulation, followed by capsule, and then EC-MPS. Extremely variable and unpredictable pharmacokinetic curves were found in kidney transplant recipients who were given EC-MPS, whereas those who were treated with MMF had regular pharmacokinetic profiles (16). Therapeutic drug monitoring should be considered when switching between EC-MPS and MMF. When it comes to MPA metabolism, the CL was lower than that reported in AAV patients (17.28±9.24 L/h), because most of patients in our study had severe kidney damage (7). The CL in this study was similar to that reported in patients with idiopathic nephrotic syndrome (CL=9.7 L/h) but lower than that conducted in systemic lupus erythematosus (CL=25.3 L/h),kidney and liver transplant (CL=12.7~22.0 L/h), demonstrating the importance of underlying disease on drug metabolism (15). In addition, we did not find a significant effect of body weight on CL, perhaps because the use of a per kg for dose. Last but not least, a combination of glucocorticoids is inevitable in ANCA treatment. Our study revealed that glucocorticoids had a dose-dependent effect on absorption, but were not included in final model (Supplemental Figure 1). A significant increase on MPA CL has been reported in renal transplant and lupus nephritis patients with concomitant use of corticosteroids compared to those with a corticosteroid free (17-19).