Search the PBPK Model Repository

  Absorption Model

• First-Order

  Volume of Distribution

• Full PBPK (Method 2)
• Note: A Kp scalar (0.3) was used in the model

  Route of Elimination

• Biliary = 66%; Renal= 44%

  Perpetrator DDI

• None

  Validation

• Seven clinical studies describing single and multiple dose exposure of doxycycline were used to verify the PBPK model. The model predicted AUC values in 86% of studies within 2-fold (100% if one simulated/observed ratio is rounded down from 1.52 to 1.5), of which 57% were within 1.5-fold. 

  Limitations

• Model is not verified at doses below 100 mg or about 200 mg (dose-linearity of doxycycline is uncertain)
• Model assumes hyclate, monohydrate and hydrochloride formulations are bioequivalent
• Model is not developed for the prediction of IV doxycycline
• Model was developed and verified primarily in healthy volunteer studies (except Newton et al. 2005); appropriateness of extrapolation to acute malaria patients is unknown

  Updates in V19

• Updated in vitro­ data
  • fu: 0.142 -> 0.23
  • B:P: 1.5 -> 0.78
• Converted from minimal PBPK model to full PBPK model
• Elimination changed from user input IV clearance to retrograde clearance with biliary clearance and additional hepatic clearance

Absorption Model

• First-Order

Volume of Distribution 

• Full PBPK (Method 2)

Routes of Elimination

• 89% MAO (entered using ‘user-UGT’ as a surrogate in the Simulator), 11% CYP2D6

Perpetrator DDI

• CYP1A2 Inhibitor (in vitro)

Validation

•  6 studies with single (15 to 45 mg) and multiple (15 mg QD) dosing. 100% of Cmax and AUC values within 1.5-fold.
• No clinical DDI studies to verify contribution of metabolic routes

Limitations

•  The active metabolites of primaquine have not characterized due to their instability. Therefore, a PBPK model for active metabolites cannot be developed in their own right.
• Qualitative data suggests a role of P-gp, however, Jmax and Km values have not been measured.
• There is evidence of enantiomer specific metabolism for primaquine which has not been considered in the current model.

Updates in Version 19

• Updated in vitro protein and blood binding data and subsequent back calculation of CLint (retrograde approach)
  •  f_u: 0.19 -> 0.26
  • B:P: 1 -> 0.82

• Converted from minimal PBPK model to full PBPK model

  Absorption Model

• First-Order

  Volume of Distribution

• Full PBPK (Method 2)

Note: Kp scalar used

  Route of Elimination

• UGT1A9 (50%); UGT2B7(50%)

  Perpetrator DDI

• CYP1A2 Inhibitor

  Validation

• One clinical study describing single dose exposure of DHA was used to verify the PBPK model.  100% of studies were within 2-fold, of which 100% were within 1.5-fold. 

  Limitations

• The absorption model does not consider the formation of ‘DHA from artesunate’ mechanistically. Instead, an optimized ka and fa were applied to the DHA model to describe the observed plasma concentration-time curve of DHA. The remainder of the DHA model was identical to the DHA model which is described above.
• The model does not account for the differences in plasma fraction unbound observed in patients compared to healthy volunteers.
• Verification needed for perpetrator DDI assessment as literature data is unavailable at this time.

  Updates in V19

• Updated in vitro data
  • Propranolol Papp: 30 cm/s x 106
• Converted model to full PBPK with Vss predicted through Method 2
• Updated retrograde clearance