A pharmacometrics approach to assess the feasibility of capillary microsampling to replace venous sampling in clinical studies: Tafenoquine case study

Aim Microsampling has the advantage of smaller blood sampling volume and suitability in vulnerable populations compared to venous sampling in clinical pharmacokinetics studies. Current regulatory guidance requires correlative studies to enable microsampling as a technique. A post hoc population pharmacokinetic (POPPK) approach was utilized to investigate blood capillary microsampling as an alternative to venous sampling. Methods Pharmacokinetic data from microsampling and venous sampling techniques during a paediatric study evaluating tafenoquine, a single‐dose antimalarial for P. vivax, were used. Separate POPPK models were developed and validated based on goodness of fit and visual predictive checks, with pharmacokinetic data obtained via each sampling technique. Results Each POPPK model adequately described tafenoquine pharmacokinetics using a two‐compartment model with body weight based on allometric scaling of clearance and volume of distribution. Tafenoquine pharmacokinetic parameter estimates including clearance (3.4 vs 3.7 L/h) were comparable across models with slightly higher interindividual variability (38.3% vs 27%) in capillary microsampling‐based data. A bioavailability/bioequivalence comparison demonstrated that the point estimate (90% CI) of capillary microsample versus venous sample model‐based individual post hoc estimates for area under the concentration‐time curve from time zero to infinity (AUC0‐inf) (100.7%, 98.0‐103.5%) and Cmax (79.7%, 76.9‐82.5%) met the 80‐125% and 70‐143% criteria, respectively. Overall, both POPPK models led to the same dose regimen recommendations across weight bins based on achieving target AUC. Conclusions This analysis demonstrated that a POPPK approach can be employed to assess the performance of alternative pharmacokinetic sampling techniques. This approach provides a robust solution in scenarios where variability in pharmacokinetic data collected via venous sampling and microsampling may not result in a strong linear relationship. The findings also established that microsampling techniques may replace conventional venous sampling methods.