Suppressing TGF-β activation to reduce lung fibrosis: In search for an efficacious dose regimen for alpha V integrin inhibitors

Authors: Ermakov S, Generaux GT, Singh FA, Chandra A, Siler SQ
Conference: ACoP
Keywords: idiopathic pulmonary fibrosis, lung fibrosis, pulmonary playlist, QSP modeling
Software: IPFsym®
Division: DILIsym Services

Abstract

Objectives: Elevated levels of TGF-β and specifically its activated form is regarded as one of the disease drivers in patients with idiopathic pulmonary fibrosis (IPF). TGF-β can bind to the extracellular matrix and be subsequently
liberated and activated via interaction with aV integrins present on the surface of epithelial and myofibroblast cells in the lungs. Blocking aV integrin molecules with a drug prevents TGF-β liberation and activation, however this may drive concentration of the ECM-bound latent TGF-β higher, thus reducing the drug efficacy. The goal of this study is to model the underlying biology and to identify optimal aV integrin inhibitor regimens that effectively suppress TGF-β activation.

Methods: We developed a computer model describing the dynamics of the TGF-β latent and active forms in the lung tissue that enables the stoichiometric calculation of the inhibitor and substrate binding to aV integrin. The model includes the following features:
• Latent TGF-β is produced by resident cells within the lungs and released into the extracellular fluid (ECF) where a portion can be incorporated into the extracellular matrix (ECM).
• TGF-β is released back to ECF during ongoing ECM turnover.
• Exchange between the tissue and the circulating TGF-β pools is included.
• Activated TGF-β is released upon interaction with aV integrins.
• When the drug is present it competes with TGF-β for aV integrin binding sites and inhibits TGF-β liberation and activation.
• The drug+aV integrin complex can be internalized and degraded inside the cell.
The above mechanisms are modelled by the set of ODEs. Parameters such as baseline concentration of TGF-β, aV integrins, number of TGF-β binding sites in the ECM, drug inhibition constants are taken from or estimated based on available literature data (Balestro 2019, King 2011, Hynes 2009). Drug concentrations are represented, including specified Cmax and Ctrough values, based on the dosing regimen of interest.

Results: An optimal aV integrin inhibitor dosing regimen for IPF patients would persistently decrease the activated TGF-β concentrations. We have simulated effects of aV integrin inhibitor PK profile (Cmax and Ctrough) on
lowering activated TGF-β concentration as well as uncertainty in parameters defining ECM turnover and aV integrins distribution in the tissue. We have identified that 90% suppression of activated TGF-β will require maintaining Ctrough values close to the concentration of available aV integrin molecules in the tissue. Also we found that aV integrin inhibition drives an increase in the latent ECM-bound TGF-β pool. The elevated levels can compete with the drug for aV integrin binding sites. Moreover, therapy cessation will cause a temporary spike in activated TGF-β that could lead to undesired side effects.

Conclusions: We have developed a computer model describing TGF-β dynamics in the lung tissue in response to drugs blocking aV integrins. It was identified that effective suppression of TGF-β activation can be achieved by maintaining trough concentrations close to concentrations of aV integrins in the tissue.

Presented at ACoP 12 November 8-12, 2021

By Sergey Ermakov, Grant Generaux, Fulya Akpinar Singh, Ankit Chandra, Scott Q Siler