1521-009X/43/6/803–811$25.00 http://dx.doi.org/10.1124/dmd.114.061713 DRUG METABOLISM AND DISPOSITION Drug Metab Dispos 43:803–811, June 2015 Copyright ª 2015 by The American Society for Pharmacology and Experimental Therapeutics Mechanistic Investigation of the Preclinical Pharmacokinetics and Interspecies Scaling of PF-05231023, a Fibroblast Growth Factor 21–Antibody Protein Conjugate s Craig Giragossian, Chandra Vage, Jun Li, Kathleen Pelletier, Nicole Piché-Nicholas, Manoj Rajadhyaksha, Jennifer Liras, Alison Logan, Roberto A. Calle, and Yan Weng Pfizer Inc., Groton, Connecticut (C.G., C.V., Ju.L., K.P., M.R., A.L.); and Pfizer Inc., Cambridge, Massachusetts (N.-P.N., Je.L., R.A.C., Y.W.) Received October 21, 2014; accepted March 16, 2015 ABSTRACT PF-05231023, a long-acting fibroblast growth factor 21 (FGF21) analog, was generated by covalently conjugating two engineered [des-His1, Ala129Cys]FGF21 molecules to a nontargeting human IgG 1k scaffold. The pharmacokinetics (PK) of PF-05231023 after i.v. and s.c. administration was evaluated in rats and monkeys using two enzyme-linked immunosorbent assays with high specificity for bi- ologically relevant intact N termini (NT) and C termini (CT) of FGF21. Intact CT of FGF21 displayed approximately 5-fold faster systemic plasma clearance (CL), an approximately 2-fold lower steady-state volume of distribution, and at least 5-fold lower bioavailability compared with NT. In vitro serum stability studies in monkeys and humans suggested that the principal CL mechanism for PF-05231023 was degradation by serum proteases. Direct scaling of in vitro serum degradation rates for intact CT of FGF21 underestimated in vivo CL 5-fold, 1.4-fold, and 2-fold in rats, monkeys, and humans, respectively. The reduced steady-state volume of distribution and the bioavailability for intact CT relative to NT in rats and monkeys were compatible with proteolytic degradation occurring outside the plasma compartment via an unidentified mechanism. Human CL and PK profiles for intact NT and CT of FGF21 were well predicted using monkey single-species allometric and Dedrick scaling. Physiologically based pharmacoki- netic models incorporating serum stability data and an extravascular extraction term based on differential bioavailability of intact NT and CT of FGF21 in monkeys improved accuracy of human PK predictions relative to Dedrick scaling. Mechanistic physiologically based phar- macokinetic models of this nature may be highly valuable for predicting human PK of fusion proteins, synthetically conjugated proteins, and other complex biologics. Introduction Fibroblast growth factor 21 (FGF21) is a 19-kDa endocrine hormone that modulates lipid and glucose homeostasis (Belouski et al., 2010; Gimeno and Moller, 2014). The N terminus of FGF21 displays affinity for several fibroblast growth factor receptors (FGFRs), including FGFR1c, FGFR3c, and FGFR4, whereas the C terminus binds to the membrane-associated cofactor b-klotho. Truncation of N termini (NT) by six or more residues, or C termini (CT) by two or more residues, decreased in vitro potency more than 10-fold, suggesting that potent functional activity was derived from both termini of FGF21 (Micanovic et al., 2009; Yie et al., 2009). In diabetic animal models and patients with type 2 diabetes, recombinant fibroblast growth factor 21 (rFGF21) and its analogs demonstrated dose-dependent reductions in low-density lipoprotein cholesterol, apolipoproteins, fasting triglycerides, fasting insulin, and body weight, as well as dose-dependent elevations of high-density lipoprotein cholesterol and adiponectin (Adams et al., 2013; Gaich et al., 2013; Kharitonenkov et al., 2013; Smith et al., 2013). Although dose-dependent reductions in blood glucose have been observed in diabetic rodent and nonhuman primate models, thus far only a trend toward lower fasting blood glucose has been observed in human clinical studies (Gaich et al., 2013). Although the precise mechanism by which FGF21 exerts its pharmacological actions has not been deduced, both NT and CT are believed to be important for potent regulation of these processes. The highly desirable pharmacology profile of FGF21 has generated significant interest as a potential therapy for type 2 diabetes and/or dyslipidemia; however, the full therapeutic potential of FGF21 may be limited by its short in vivo persistence. Systemic plasma clearance (CL) of rFGF21 in mice and monkeys was comparable to the glomerular filtration rate in each respective species, resulting in half- lives on the order of 0.5 to 2 hours (Kharitonenkov et al., 2007). Furthermore, CL of NT and CT immunoreactivity was indistinguish- able in mice, suggesting that passive renal filtration was the primary CL mechanism for rFGF21 (Hager et al., 2013). Consistent with this hypothesis, endogenous levels of FGF21 were elevated more than 15- fold in patients with impaired renal function compared with controls (Stein et al., 2009). At the time of data generation, all authors were employees and stockholders of Pfizer Inc. Pfizer Inc. provided support in the form of salaries for the authors but did not have any additional role in the study design, data collection and analysis, decision to publish, or preparation of the manuscript. dx.doi.org/10.1124/dmd.114.061713. s This article has supplemental material available at dmd.aspetjournals.org. ABBREVIATIONS: ADA, anti-drug antibody; CL, systemic plasma clearance; CT, C termini; FGF21, fibroblast growth factor 21; FGFR, fibroblast growth factor receptor; HRP, horseradish peroxidase; mAb, monoclonal antibody; NT, N termini; PBPK, physiologically based pharmacokinetics; PF-05231023, [des-His1, Ala129Cys]FGF21-human IgG1 conjugate; PK, pharmacokinetics; rFGF21, recombinant fibroblast growth factor 21; t 1/2 , elimination half-life; V ss , steady-state volume of distribution. 803 http://dmd.aspetjournals.org/content/suppl/2015/03/24/dmd.114.061713.DC1 Supplemental material to this article can be found at: at ASPET Journals on April 2, 2017 dmd.aspetjournals.org Downloaded from