Model-based interspecies scaling and impact of anti-drug antibodies on the pharmacokinetics of the therapeutic protein CB 4332
Rami Ayoun Alsoud (1), Natacha Le Moan (2), Lars Holten-Andersen (2), Hans Lennernäs (1), Tom Knudsen (2), Ulrika SH Simonsson (1)
(1) Department of Pharmaceutical Biosciences, Uppsala University, Uppsala, Sweden; (2) Catalyst Bioscience, South San Francisco, California, USA
Introduction: CB 4332 is a recombinant complement factor I (CFI) protein developed to restore CFI levels in individuals with complement system deficiency. Due to the small target population in CFI deficiency, initial dose selection for first-in-human (FIH) studies is important for successful clinical development. The extrapolation of protein therapeutics pharmacokinetics (PK) from animals to humans can have large inherent variability relative to traditional allometric scaling typically used for small molecules. In particular, this is due to issues related to species-specific immunogenicity to the human xenoprotein [1]. The formation of anti-drug antibodies (ADA) can alter the proteins PK [2], e.g. by increasing clearance. To accurately estimate and scale its clearance to humans, it is therefore important to account for the potential impact from ADA when analyzing/modeling PK. Traditional scaling approaches, e.g. scaling the PK with three or more species and scaling to a human-equivalent dose (HED), do not take ADA formation into account. Nonlinear mixed-effects modelling makes use of pooled multi-species data maximizing the use of information, to characterize the PK and interspecies scaling simultaneously while accounting for the potential effects of ADA.
Objective: Compare different traditional and model-based allometric scaling approaches with respect to ADA formation and species differences to optimally inform FIH dose selection..
Methods: Pharmacokinetic studies were conducted with CB 4332 in mouse, rat and nonhuman primates after both intravenous (IV) and subcutaneous (SC) administration. Data from all species were fitted simultaneously using allometric scaling where the exponents were estimated. Lower-than-expected CB 4332 plasma concentrations were observed in the terminal elimination phase in all rats and some nonhuman primates after IV and SC administration indicating ADA formation. Apparent ADA-dependent clearance could not be detected in mice where sparse sampling was applied. Positive ADA titers were detected in plasma from nonhuman primates confirming the ADA signal from the PK analysis and hence the immunogenicity of the human CB 4332 xenoprotein in this species. To characterize the ADA-dependent clearance, an additional clearance parameter was estimated along with the time of its onset [3]. As only a subset of the nonhuman primates showed signs of ADA formation, a mixture model was used with two populations representing the presence and absence of ADA-dependent clearance in this species. To investigate the impact of the different allometric exponents on human exposure, simulations of human exposure were completed using the final model exponents as well as using exponents reported in the literature for therapeutic proteins.
Results: The final three-species PK model was a two-compartment PK model with the allometric exponents estimated at 0.78 and 1.13 for clearance and volume of distribution, respectively. Both estimated exponents were well within the range of exponents reported in the literature of therapeutic proteins [4-6]. An exponential model for the increase of rat clearance with increasing CB 4332 concentrations was supported by the data and improved the fit of the model to the rat data. The ADA-dependent clearance in rats and nonhuman primates were estimated to 4 ml/h/kg and 11 ml/h/kg, respectively, vs. the non-ADA-dependent clearance of 0.8 ml/h/kg. The onset of the ADA-dependent clearance was detectable around day 5 following first dose in rats and day 10 after initial dosing in nonhuman primates. The proportion of nonhuman primates with ADA was predicted to be approximately 20%. Simulations using the final model with estimated exponents in addition to exponents reported in the literature revealed that predicted human CB 4332 exposure can significantly vary with the different exponents. This may impact FIH dose selection.
Conclusions: Model-based interspecies scaling of CB 4332 PK allowed the prediction of human PK using all information from mice, rats, and nonhuman primates after IV and SC administration. To optimally describe CB 4332 PK, the model accounted for the impact of ADA formation and other species-related differences in clearance. Model-based interspecies scaling may be superior to traditional allometric scaling, which uses the PK parameters derived through means of noncompartmental analyses and body size without accounting for the potential impact of immunogenicity or other species-related differences.
References:
[1] Food and Drug Administration. Guidance for Industry: Considerations for the Design of Early-Phase Clinical Trials of Cellular and Gene Therapy Products. 2015.
[2] Chirmule N, Jawa V, Meibohm B. Immunogenicity to therapeutic proteins: impact on PK/PD and Efficacy. AAPS Journal. 2012;14(2):296–302.
[3] Doshi S, Wang H, Chow, V. Establishing PK Equivalence Between Adalimumab and ABP 501 in the Presence of Antidrug Antibodies Using Population PK Modeling. Clinical Therapeutics. 2022;44(1):111-122.
[4] Deng R, Iyer S, Theil FP, Mortensen DL, Fielder PJ, Prabhu S. Projecting human pharmacokinetics of therapeutic antibodies from nonclinical data: What have we learned? MAbs. 2011;3(1):61–6.
[5]Huh Y, Smith DE, Rose Feng M. Interspecies scaling and prediction of human clearance: Comparison of small- and macro-molecule drugs. Xenobiotica. 2011;41(11):972–87.
[6] Wang W, Prueksaritanont T. Prediction of human clearance of therapeutic proteins: simple allometric scaling method revisited. Biopharm Drug Dispos. 2010;31:253–263.