Preclinical concept studies showing advantage of an inhaled anti-CTGF/CCN2 protein for pulmonary fibrosis treatment

Abstract

Inhaled therapeutics have high potential for the treatment of chronic respiratory diseases of high unmet medical need, such as idiopathic pulmonary fibrosis (IPF). Preclinical and early clinical evidence show that cellular communication network factor 2 (CCN2), previously called connective tissue growth factor (CTGF), is a promising target for the treatment of IPF. In recent phase 3 clinical trials, however, systemic CCN2 inhibition failed to demonstrate a clinically meaningful benefit. Here, we present the preclinical profile of the inhaled anti-CCN2 Anticalin® protein PRS-220. Our study demonstrates that efficient pulmonary delivery directly translates into superior efficacy in relevant models of pulmonary fibrosis when compared to systemic CCN2 inhibition. Moreover, we present a holistic approach for the preclinical characterization of inhaled PRS-220 from state-of-the art in vitro and in vivo models to novel human ex vivo and in silico models, highlighting the advantage of inhaled drug delivery for treatment of respiratory disease.

Fig. 1. Structural evolution and binding properties of PRS-220.

a Structural models comparing NGAL (Anticalin scaffold, PDB ID: 1DFV), an intermediate, and PRS-220 highlighting the evolution of residue variations towards high affinity binding to CCN2 and favorable biophysical properties (higher Tm). Yellow sticks: disulfide bond. Red and orange spheres: residues that were mutated in Lcn2 to generate the intermediate. Blue spheres: Cα atoms of residues that remained fixed during subsequent engineering steps towards PRS-220. Graphs were generated with Pymol. b Binding of PRS-220 to recombinant CCN2, CCN2 N-terminal fragment (IGFBP & vWC) and C-terminal fragment (TSR & CT domain) by ELISA (n = 3 experiments, mean ± SD). c Competition of PRS-220 and the Anticalin scaffold (NGAL ctrl) with pamrevlumab for binding to CCN2 (IC₅₀ = 0.35 ( ± 0.05) nM) and competition of pamrevlumab with itself for binding to CCN2 (IC₅₀ = 1.14 ( ± 0.03) nM) analyzed by ECLA (n = 3 experiments, mean ± SD). d Experimental setup of the competition ECLA. Created in BioRender. Pavlidou, M. (2025) https://BioRender.com/y92y274e Quantification of PRS-220 binding to CCN2 expressed by TGF-β1-activated normal human lung fibroblasts (n = 3 experiments, mean+SD). Dotted line represents background signal in presence or absence of 5 ng/mL TGF-β1 with anti-scaffold detection antibody only. f Schematic view of air-liquid interface (ALI) cell culture model for investigation of mucus interaction. Created in BioRender. Pavlidou, M. (2025) https://BioRender.com/j57y046g Confocal imaging (63x) of PRS-220 mucus penetration in ALI cultures of cystic fibrosis derived human bronchial epithelial cells (HBECs) at baseline, 0.5 min (PRS-220) or 5 min (pos. ctrl) after adding Alexa Fluor 647-labeled PRS-220 or mucus binding microspheres (positive ctrl) to the apical side. Representative images of n = 3 independent experiments with mucus = green, air surface liquid (ASL) =magenta, PRS-220/pos. ctrl  = blue. h PRS-220 mucus interaction by Quartz Crystal Microbalance with Dissipation (QCMD) monitoring of the frequency shift (=changes in mucus mass) and dissipation shift (=changes in mucus hydration). Mucus solution was perfused over a chip, followed by perfusion of 10 µM PRS-220, Poly-L-lysine (PLL) as positive control for mucus interaction and PBS. Source data are provided as a Source Data file.

Fig. 2. Pulmonary delivery of PRS-220 achieves superior lung tissue penetration when compared to systemic delivery of an antibody.

a PK of PRS-220 in BALF, lung tissue, and plasma at 1 h, 4 h, 8 h and 24 h after delivery to the lungs of healthy mice by oropharyngeal aspiration (100 µg/mouse, n = 4 animals per timepoint, mean ± SD). b PK of pamrevlumab in BALF, lung tissue, and plasma 1 h, 8 h, 24 h and 96 h after intravenous delivery to healthy mice (100 µg/mouse, n = 4 animals per timepoint, mean ± SD). c Comparison of lung tissue and plasma peak concentrations (C_max) of PRS-220 and pamrevlumab determined by non-compartmental analysis of PK data shown in (a, b). d Alexa Fluor 647-labeled PRS-220 (PRS-220-AF647) and pamrevlumab (pamrevlumab-AF647) imaged in lung tissue sections 2 h after administration to healthy mice. PRS-220-AF647 was delivered to the lung by oropharyngeal aspiration and pamrevlumab-AF647 was administered intravenously (both at 100 µg/mouse). Figure shows representative images of n = 4 animals per treatment group for PRS-220 and pamrevlumab (both in red) and cell nuclei (DAPI, blue) at 4x and 80x magnification. e Alexa Fluor 647-labeled PRS-220 (100 µg/mouse) delivered to fibrotic lungs of mice at day 21 after bleomycin challenge by oropharyngeal aspiration imaged by light sheet microscopy. Figure shows representative whole left lung imaging of PRS-220 2 h after dosing (scale bars 500 µm). Experiment was performed with n = 5 animals (vehicle) and n = 3 animals (all other treatment groups). f Magnified 2D sections generated from 3D scanned lung (scale bars 500 µm & 150 µm) shown in (e). g Compound accumulation of PRS-220 delivered to the lung and pamrevlumab administered intravenously (both 100 µg/mouse) in fibrotic lung tissue determined by light sheet imaging & quantitative analysis. Data is shown as mean ± SD including single data points of individual mouse lungs (n = 5 (vehicle)); n = 3 (PRS-220 2 & 24 h, pamrevlumab 8 & 24 h); n = 2 (PRS-220 8 h, pamrevlumab 2 h). Source data are provided as a Source Data file.

Fig. 3. Inhaled delivery of PRS-220 achieves high, dose-dependent lung exposure with low systemic bioavailability in vivo.

a Scheme outlining the design of a study to investigate the PK profile of daily (QD) inhaled PRS-220 in lung tissue and serum of healthy rats. Rats were treated daily with PRS-220 at 3 different dose levels by nose-only inhalation for 4 days. Created in BioRender. Pavlidou, M. (2025) https://BioRender.com/e13e683b Serum and lung exposure of free, non-target bound PRS-220 was determined over time after daily treatment with 3 different dose levels by nose-only inhalation for 4 days. Graph shows PRS-220 levels above the detection limit as mean ± SD of n = 6 animals (3 male and 3 female) per timepoint in serum and lung tissue. Figure legend indicates the mean achieved delivered dose levels during the study. c Scheme outlining the design of a study to investigate the PK profile of a single intravenous (i.v.) administration of PRS-220 in lung tissue and serum of healthy rats. Created in BioRender. Pavlidou, M. (2025) https://BioRender.com/z18p244d PRS-220 levels upon single intravenous administration at a dose level of 2 mg/kg were assessed in lung tissue and serum. Graph shows PRS-220 levels above the detection limit as mean ± SD of n = 6 animals (3 male and 3 female) per timepoint in serum and lung tissue. Source data are provided as a Source Data file.

Fig. 4. Deep penetration of fibrotic lung tissue by PRS-220 upon inhalation translates into superior attenuation of fibrosis in vivo.

a Fibrotic remodeling of the lung assessed 41 days after silica challenge in comparison to saline challenge of mouse lungs by Hematoxylin and Eosin staining. b PK profile of PRS-220 in lung tissue and plasma delivered via nose-only inhalation at 3 dose levels (0.2, 1 and 5 mg/kg targeted delivered dose) to healthy or silica challenged mice. Graph shows data of n = 4 animals per group and timepoint as mean ± SD. c Immunofluorescence imaging of PRS-220 (magenta) 2 h after inhaled delivery to fibrotic, silica-challenged mouse lungs (day 41, targeted delivered dose 5 mg/kg). Cell nuclei are stained with DAPI (blue). d Co-immunofluorescence staining of PRS-220 (magenta), club cell marker CC10 (green) and e smooth muscle cell & myofibroblast marker aSMA (green) in lung tissues from c. f Localization of PRS-220 (magenta) in relation to cells expressing its target assessed by co-immunofluorescence imaging of PRS-220 and CCN2 (green) in lung tissues from c. g Scheme of the in vivo efficacy study in mice. Created in BioRender. Pavlidou, M. (2025) https://BioRender.com/x52k357h Total collagen deposition in the lung by measuring lung hydroxyproline. Individual data points of n = 3 (saline) and n = 6 (all other) animals per treatment group including mean ± SD are shown (One-way ANOVA with Šídák’s multiple comparisons test; *p ≤ 0.05, **p ≤ 0.01, ns (not statically significant) p > 0.05). i Lung tissue mRNA gene expression of Timp1 and j Ccl2 with Rn18s serving as reference gene. Individual data points of n = 3 (saline) and n = 6 (all other) animals per treatment group including mean ± SD are shown (One-way ANOVA with Šídák’s multiple comparisons test; *p ≤ 0.05, **p ≤ 0.01, ***p ≤ 0.001, ns (not statically significant) p > 0.05). For all imaging analyses, representative images of in total n = 4 mice per treatment group are shown. Source data are provided as a Source Data file.

Fig. 5. Lung tissue distribution of nebulized PRS-220 in ex vivo ventilated and perfused human lungs.

a Scheme of the ex vivo lung perfusion (EVLP) set-up with the nebulizer set-up in line with the ventilation tubing. Created in BioRender. Pavlidou, M. (2025) https://BioRender.com/f59h729b Picture of the actual EVLP experiment. c Overview immunofluorescence imaging of EVLP lung biopsies 4 h after nebulized delivery of Alexa Fluor 647-labeld PRS-220 (magenta). Cell nuclei (stained with DAPI (blue) and scale bar 1 mm). d Co-staining of lung tissue sections for CCN2 (red), smooth muscle actin (SMA, green) and DAPI (blue) at baseline (0 h, serving as negative control) and 2 and 4 h after nebulized delivery of Alexa Fluor 647-labeld PRS-220 (magenta). Biopsies from two different regions of the lung were analyzed at the terminal 4 h timepoint (Biopsy A and B). Figure shows images at 2 different magnifications (scale bars 500 µm for upper images and 100 µm for images below). e Co-immunofluorescence imaging of PRS-220 (magenta), CCN2 (red), platelet-derived growth factor α (PDGFRα, green) and DAPI (blue). f Co-immunofluorescence imaging of PRS-220 (magenta), CCN2 (red), keratin-5 (KRT5, green) and DAPI (blue). Figure shows representative images from independent experiments performed with n = 3 human donor lungs.

Fig. 6. In silico inhalation study of PRS-220 in 3 human subjects shows a broad deposition pattern of PRS-220 containing aerosol in healthy and IPF lungs.

a CT image analysis and segmentation of fibrotic tissue (red) and b airways, lungs, and lobes (left) as first step of the subject-specific model generation process. Generation of the full 16-generation tree of conducting airways (right) and viscoelastic alveolar clusters (Supplementary Fig. 4) based on segmented volumetric information and a physiology-based, space-filling growth algorithm. c Final pattern of deposited PRS-220 aerosol particles after simulation of airflow, particle transport and deposition for one complete breathing cycle. d Mass fraction of deposited aerosol per category: Mouth/throat (green), conducting airways (orange), alveolar tissue (blue), and exhaled (brown). e Normalized concentration of deposited aerosol mass per volume for healthy vs. fibrotic regions of the lungs (mild case = orange; severe case = red). Values smaller/>100% indicate under/over-proportional deposition in the respective sub-volume.

Fig. 7. CCN2 inhibition by PRS-220 attenuates fibrotic remodeling in a human precision cut lung slice (PCLS) model of IPF.

a Histopathological analysis of CCN2 expression in IPF lung tissue. Representative images of lung tissue sections from n = 16 different IPF patients are shown. b Experimental procedure of PCLS model. Created in BioRender. Pavlidou, M. (2025) https://BioRender.com/b00o965c Experimental design of an ex vivo proof of concept study to investigate the anti-fibrotic effect of CCN2 inhibition by PRS-220 in PCLS from IPF patients. PCLS were treated daily with PRS-220 (100 nM and 500 nM), pamrevlumab (100 nM) and nintedanib (2 µM). Treatment with PRS-220 vehicle served as a control. PCLS were harvested at day 5 for analysis of Collagen1A1 protein levels. Created in BioRender. Pavlidou, M. (2025) https://BioRender.com/y24t356d Collagen1A1 (COL1A1) protein analysis by Western blot with GAPDH serving as a loading control. Figure shows representative immunoblots of experiments performed with lung tissue from n = 3 IPF patients with n = 4-5 technical replicates per IPF donor. Arrows indicate COL1A1 protein bands subjected to densitometric analysis. e Densitometric analysis of COL1A1 signal in Western blots. Single data points show integrated density values (IDV) of COL1A1 signal normalized to loading control GAPDH of all technical replicates from PCLS of in total n = 3 IPF lung tissue donors (n = 4 technical replicates per treatment group for donor 1 & 2 and n = 5 per treatment group for donor 3; Kruskal-Wallis test with Dunn’s multiple comparison; ***p < 0.001, ns (not statistically significant) p > 0.05). Source data are provided as a Source Data file.