M Chetty, R Rose, L Li, K Machavaram, M Jamei, I Gardner.
Simcyp Limited, Blades Enterprise Centre, Sheffield, UK.
Objectives: To determine whether physiologically based pharmacokinetic (PBPK) models, which incorporate covariates and interindividual variability in systems parameters, are effective in predicting whether fixed dosing or weight based dosing is more appropriate for specific monoclonal antibodies (mAbs).
Methods: The Simcyp Population Based Simulator was used to simulate concentration-time profiles for omalizumab (150mg and 300mg) and efalizumab (1mg/kg and 10mg/kg) using the study designs of published clinical studies1,2. The suitability of the models was verified by comparison of the simulated pharmacokinetic (PK) profiles with those observed clinically. Using these models and 500 virtual healthy volunteers, PK profiles were simulated for each mAb using both dosing options and four single doses (ie. omalizumab 2mg/kg and 4mg/kg and efalizumab 75mg and 750mg, in addition to the above doses). The means of the area under the plasma concentration versus time curve (AUC0-t) and maximum plasma concentration (Cmax) for each weight group were compared for variability with the two dosing options after stratifying population by weight (40 – 50 kg; 51-60 kg; 61-70kg; 71-80kg; 81-90 kg; 91-100 kg; 101-110kg; ≥111kg). Variability was evaluated by the fold difference (>2 fold was considered to be significant) between the lowest and highest value for each of the above PK parameters, using the mean of the predicted values for the different weight groups.
Results: Observed PK profiles for omalizumab and efalizumab were successfully recovered with the Simcyp minimal PBPK model for therapeutic proteins and the mechanistic FcRn model, respectively. Variability in Cmax and AUC of efalizumab was significantly higher when the fixed doses were used. No clear trends in variability between the weight groups using the two dosing approaches was observed for omalizumab. The fold differences in AUC (>2 fold) and Cmax (>2 fold) suggested that weight based dosing is more appropriate for efalizumab. Decisions on dosing for omalizumab may require further investigation since there is no clear advantage of one approach over the other. These predictions correspond with dosing recommendations for these mAbs, where weight based dosing had been used for efalizumab while omalizumab dosing is based on an algorithm with mg/kg and IgE level.
Conclusions: This preliminary study suggests that simulations using PBPK modelling can be useful in predicting suitability of dosing options for mAbs.
References:
[1] Riviere et al., J Bioequiv Availab, 2011, 3(6):144-50[2] Mortensen et al., J Clin Pharmacol, 2005, 45(3): 286-298
Reference: PAGE 23 (2014) Abstr 3147 [www.page-meeting.org/?abstract=3147]
Poster: Methodology - Covariate/Variability Models