M. Stroh (1), B.L. Millard (2), H. Lu (1), V. Huels (1), B. Zheng (1), L. Desnoyers (1), J. Richardson (1), J.F. Apgar (2), M. Will (1), W. Michael Kavanaugh (1), and R. Humphrey (1)
(1) CytomX Therapeutics, South San Francisco, CA, USA, (2) Applied BioMath, Concord, MA, USA.
Objectives:
Monoclonal antibodies (mAb) targeting the programmed cell death (PD-1) pathway have shown anticancer activity in different tumor types, though with associated toxicities especially in combination with other immuno-oncology agents [1]. Probody™ therapeutics (Pb-Tx) are mAb prodrugs designed to reduce on-target toxicities [2]. A mask inhibits antigen binding of the Pb-Tx in the periphery, and this mask can be removed by tumor-associated proteases to produce an active mAb. CX‑072 is an anti‑programmed cell death ligand 1 (PD‑L1) Pb‑Tx under clinical evaluation in cancer patients [3].
The study aim is to develop a quantitative systems pharmacology (QSP) model for CX‑072 clinical translation and to compare predicted versus observed preliminary PK/PD from ongoing phase 1/2 study PROCLAIM‑CX‑072 (NCT03013491).
Methods:
A QSP model was developed, calibrated against monkey PK data, and used to project human PK/PD. Human and monkey PK samples were collected at prespecified times following CX‑072 administration; plasma analytes included intact CX‑072, activated CX‑072 and the sum of intact and activated CX‑072 (total CX‑072). Receptor occupancy (RO) was estimated in cancer patients based upon levels of activated/unmasked CX-072 in postdose tumor biopsy lysates as measured by capillary electrophoresis immunoassay [4].
Models were implemented using KroneckerBio v. 0.5 (https://github.com/kroneckerbio) and expressed as a system of ordinary differential equations with the following form:
dx/dt=k+Ax+B (x⊗x)
where k is a vector of 0th order rate constants, A is an n by n matrix of 1st order rate constants, and B is an n by n matrix of second order rate constants. Parameter estimation and simulations were performed using MATLAB v. 2015b (Mathworks, Natick MA).
Results:
The QSP Pb‑Tx model was developed based on proposed mechanisms of Pb‑Tx distribution, elimination, activation and binding. The QSP Pb-Tx model captures events both at the Pb-Tx- and compartmental- levels. Unique to the Pb-Tx, reversible breathing events and irreversible cleavage reactions are both captured in the model to yield multiple species that may interact with PD-L1 with differential affinities. Common to other mAb pharmacology models [5], the multiple states of the Pb-Tx distribute to the plasma, peripheral, and tumor compartments.
The Pb-Tx QSP model was calibrated to intact and total CX-072 circulating levels following administration of CX-072 20, 60, and 200 mg/kg to cynomolgus monkeys. Preliminary human single‑dose (SD) PK data following CX‑072 0.03‑30.0 mg/kg appeared consistent with QSP model predictions: CX‑072 circulated predominantly as intact CX‑072 (96% intact at 30 mg/kg) with no strong trending of intact CX‑072 elimination kinetics with dose. In contrast to simulations for the Pb-Tx (intact CX-072), the parental mAb (CX-075) QSP model simulations suggest evidence of target mediated drug disposition (TMDD) as has been noted for other PD-L1 mAbs such as atezolizumab [6]. Preliminary tumor RO estimates were observed to be consistent with QSP predictions following CX‑072 3 ‑ 30 mg/kg.
Conclusions:
The CX‑072 QSP model captures the determinants of Pb‑Tx PK/PD and corroborates preliminary clinical PK/PD findings from PROCLAIM‑CX‑072: as designed, CX‑072 circulates predominantly as intact CX‑072 and likely is not strongly influenced by TMDD following CX‑072 0.03 ‑ 30 mg/kg SD.
References:
[1] Iwai Y, Hamanishi J, Chamoto K, Honjo T. Cancer immunotherapies targeting the PD-1 signaling pathway. J Biomed Sci. 2017;24(1):26.
[2] Desnoyers LR, Vasiljeva O, Richardson JH, Yang A, Menendez EE, Liang TW, et al. Tumor-specific activation of an EGFR-targeting probody enhances therapeutic index. Sci Transl Med. 2013;5(207):207ra144.
[3] Autio KA, Arkenau H-T, O’Neil BH, Bendell JC, El-Khoueiry A, Strauss J, et al. Preliminary results of the first-in-human, dose-finding PROCLAIM-CX-072 trial of the PD-L1 Probody therapeutic CX-072 as monotherapy in patients (pts) with advanced solid tumors. Journal of Clinical Oncology. 2018;36(15_suppl):3071.
[4] Lyman, S. et al., “Preliminary Evidence of Intratumoral Activation and Immunomodulatory Effect of CX-072, a Probody Therapeutic Antibody Prodrug Targeting PD-L1, in a Phase 1/2 Trial.” Presented at SITC 2018. November 2018. Washington, D.C.
[5] Baxter LT, Zhu H, Mackensen DG, Jain RK. Physiologically based pharmacokinetic model for specific and nonspecific monoclonal antibodies and fragments in normal tissues and human tumor xenografts in nude mice. Cancer Res. 1994;54(6):1517-28.
[6] Stroh M, Winter H, Marchand M, Claret L, Eppler S, Ruppel J, et al. Clinical Pharmacokinetics and Pharmacodynamics of Atezolizumab in Metastatic Urothelial Carcinoma. Clin Pharmacol Ther. 2017;102(2):305-12.
PROBODY is a trademark of CytomX Therapeutics, Inc.
Reference: PAGE 28 (2019) Abstr 8918 [www.page-meeting.org/?abstract=8918]
Poster: Drug/Disease Modelling - Oncology