3.2 Population Pharmacokinetic Base Model
All iberdomide and M12 concentrations were converted to molar concentrations for modeling purpose. The starting structural model of iberdomide was a two-compartment model with first order absorption and elimination based on the observed concentration-time profile (Figure 1). Incorporating a lag time improved the model fitting by significantly decreasing OFV (-576). A first order conversion rate (Kpm) was used to account for the transformation of iberdomide to M12. M12 PK was described by a one-compartment model with first order elimination. Of note, to rule out the occurrence of parameter identifiability issues existing in typical parent-metabolite popPK models which has been discussed in previous publication 12, the volume of distribution of M12 was assumed to be similar as that of iberdomide (V1/F). The model structure was shown in Figure 2. During the modelling process, iberdomide and M12 PK data were fitted simultaneously into the model.
In addition, the inter-individual variability (IIV) associated with the PK parameters in the model were estimated as log-normally distributed with a non-zero covariance, as described in the method section. Based on modeling fitting, proportional error model was selected for both iberdomide and M12 PK. Equation 2 described the error model, where y1 and y2 denoted observed concentration of iberdomide and M12, respectively; Cp and Cm represented the model predicted concentration of iberdomide and M12; b1 and b2 were proportional parameters; e was the base of the natural logarithm.
Equation 2: \(y1=Cp+b1*Cp*e\)
\(y2=Cm+b2*Cm*e\)
In conclusion, a two-compartment model with first order absorption rate constant incorporating a lag time for iberdomide and a first order conversion rate linking the one-compartment metabolite model was selected as the base model.