A mechanistic PK/PD model of AZD0171 (anti-LIF) to support Phase II dose selection

Abstract

AZD0171 (INN: Falbikitug) is being developed as a humanized monoclonal antibody (mAb), immunoglobulin G subclass 1 (IgG1), which binds specifically to the immunosuppressive human cytokine leukemia inhibitory factor (LIF) and inhibits downstream signaling by blocking recruitment of glycoprotein 130 (gp130) to the LIF receptor (LIFR) subunit (gp190) and the phosphorylation of signal transducer and activator of transcription 3 (STAT3) and is intended to treat adult participants with advanced solid tumors. LIF is a pleiotropic cytokine (and a member of the IL-6 family of cytokines) involved in many physiological and pathological processes and is highly expressed in a subset of solid tumors, including non-small cell lung cancer (NSCLC), colon, ovarian, prostate, and pancreatic cancer. The aim of this work was to develop a mechanistic PK/PD model to investigate the effect of AZD0171 on tumor LIF levels, predict the level of downstream signaling complex (LIF:LIFR:gp130) inhibition, and examine the dose-response relationship to support dose selection for a Phase II clinical study. Modeling results show that tumor LIF is inhibited in a dose-dependent manner with >90% inhibition for 95% of patients at the Phase II clinical dose of 1500 mg Q2W.

FIGURE 1

Model structure and design. Three‐compartmental model for the blood system, tumor, and tissue along with AZD0171 binding to LIF, LIF binding to LIFR and gp130, and AZD0171 blockade of gp130 interactions with LIF.

FIGURE 2

Model fit to AZD0171 and total LIF clinical data. (a) Sample model fit to plasma concentration of AZD0171 for N = 6 patients and the goodness of fit (observed data vs. model‐predicted values) for all patients, (b) Sample model fit to plasma concentration of total LIF for N = 6 patients and the goodness of fit (observed data vs. model‐predicted values) for all patients.

FIGURE 3

The inhibition of the signaling receptor complex in tumor. The dose‐dependent signaling complex inhibition predicted by the model in 1350 virtual patients in different doses and regiments in Phases I and II clinical trials of AZD0171: 75 mg Q3W, 225 mg Q3W, 750 mg Q3W, 1125 mg Q3W, 1500 mg Q3W, and 1500 mg Q2W.

FIGURE 4

The dynamics of free LIF in plasma and tumor. (a), the dynamics of free LIF in the tumor (black dashed line) and plasma (cyan solid‐line) as well as the tumor concentration of the signaling receptor complex, LIF:LIFR:gp130, (dark green, dashed line) after multiple drug dosing from 240 (h) to 3600 (h) at the clinical dose of 1500 mg Q2W. AZD0171 administrations are indicated by black arrows. (b–f), predicted free tumor LIF dynamics in response to −50% to +50% changes from the estimated values of sensitive parameters on tumor free LIF including clearance of AZD0171 (b), the volume of the tumor (c), LIF synthesis rate in the tumor (d), the steady‐state concentration of gp130 (e) and the internalization rate of unbound LIFR (f). The reference for each parameter in b–f demonstrated by black dashed line and shows either the estimated value by the model or the value of the parameter in the model. Note that the dosing schedule for simulations in b–f is the same as in a; however, for simplicity, the time of administration is only shown in (a) by solid black arrows.