Nintedanib preserves lung growth and prevents pulmonary hypertension in a hyperoxia-induced lung injury model

Abstract

Background: Bronchopulmonary dysplasia (BPD), the chronic lung disease associated with prematurity, is characterized by poor alveolar and vascular growth, interstitial fibrosis, and pulmonary hypertension (PH). Although multifactorial in origin, the pathophysiology of BPD is partly attributed to hyperoxia-induced postnatal injury, resulting in lung fibrosis. Recent work has shown that anti-fibrotic agents, including Nintedanib (NTD), can preserve lung function in adults with idiopathic pulmonary fibrosis. However, NTD is a non-specific tyrosine kinase receptor inhibitor that can potentially have adverse effects on the developing lung, and whether NTD treatment can prevent or worsen risk for BPD and PH is unknown.

Hypothesis: We hypothesize that NTD treatment will preserve lung growth and function and prevent PH in an experimental model of hyperoxia-induced BPD in rats.

Methods: Newborn rats were exposed to either hyperoxia (90%) or room air (RA) conditions and received daily treatment of NTD or saline (control) by intraperitoneal (IP) injections (1 mg/kg) for 14 days, beginning on postnatal day 1. At day 14, lung mechanics were measured prior to harvesting lung and cardiac tissue. Lung mechanics, including total respiratory resistance and compliance, were measured using a flexiVent system. Lung tissue was evaluated for radial alveolar counts (RAC), mean linear intercept (MLI), pulmonary vessel density (PVD), and pulmonary vessel wall thickness (PVWT). Right ventricular hypertrophy (RVH) was quantified with cardiac weights using Fulton's index (ratio of right ventricle to the left ventricle plus septum).

Results: When compared with RA controls, hyperoxia exposure reduced RAC by 64% (p < 0.01) and PVD by 65% (p < 0.01) and increased MLI by 108% (p < 0.01) and RVH by 118% (p < 0.01). Hyperoxia increased total respiratory resistance by 94% and reduced lung compliance by 75% (p < 0.01 for each). NTD administration restored RAC, MLI, RVH, PVWT and total respiratory resistance to control values and improved PVD and total lung compliance in the hyperoxia-exposed rats. NTD treatment of control animals did not have adverse effects on lung structure or function at 1 mg/kg. When administered at higher doses of 50 mg/kg, NTD significantly reduced alveolar growth in RA controls, suggesting dose-related effects on normal lung structure.

Conclusions: We found that NTD treatment preserved lung alveolar and vascular growth, improved lung function, and reduced RVH in experimental BPD in infant rats without apparent adverse effects in control animals. We speculate that although potentially harmful at high doses, NTD may provide a novel therapeutic strategy for prevention of BPD and PH.

Impact: Anti-fibrotic therapies may be a novel therapeutic strategy for the treatment or prevention of BPD. High-dose anti-fibrotics may have adverse effects on developing lungs, while low-dose anti-fibrotics may treat or prevent BPD. There is very little preclinical and clinical data on the use of anti-fibrotics in the developing lung. Dose timing and duration of anti-fibrotic therapies may be critical for the treatment of neonatal lung disease. Currently, strategies for the prevention and treatment of BPD are lacking, especially in the context of lung fibrosis, so this research has major clinical applicability.

Fig. 1. Histology and survival curves for control vs Nintedanib treated rats.

Histological comparisons of preserved, alveolar structure in healthy, control rats on day of life 7 (a) with severe alveolar simplification seen in healthy rats treated with Nintedanib at a dose of 50 mg/kg on day of life 5 (b) are shown here. The survival graph (c) demonstrates the survival curves of healthy, control rats compared with healthy rats treated with Nintedanib at a dose of 10 mg/kg and at a dose of 50 mg/kg. Rats treated with 10 mg/kg of Nintedanib experienced 72% mortality by day of life 5. Rats treated with 50 mg/kg of Nintedanib experienced 100% mortality by day of life 5. Control rats experienced no mortality.

Fig. 2. Nintedanib treatment at 1 mg/kg restored alveolar structure in hyperoxia-exposed rats.

Histological representations of lung structure in saline control rats (a), postnatal hyperoxia-exposed rats (b), control rats treated with 1 mg/kg of Nintedanib (c), and hyperoxia-exposed rats treated with 1 mg/kg of Nintedanib (d) are shown here. Nintedanib treatment was found to prevent alveolar simplification in rats subject to hyperoxia exposure as compared to controls and had no visible adverse effects on alveolarization in healthy, control rats. The micrographs (e) demonstrate that compared with the control rats, postnatal hyperoxia exposure reduced RAC by 64% (p < 0.01) and increased MLI by 108% (p < 0.01). Nintedanib treatment at 1 mg/kg of hyperoxia-exposed rats was found to significantly restore both RAC and MLI back to control values. (RA Control: n = 10, RA+Nintedanib: n = 10, hyperoxia-exposed: n = 20, hyperoxia-exposed+Nintedanib: n = 29).

Fig. 3. Micrographs demonstrate that compared to RA controls, rats from the hyperoxia group had increased total respiratory system resistance by 94% (p < 0.01) and reduced total respiratory system compliance by 75% (p < 0.01).

Nintedanib treatment at 1 mg/kg of hyperoxia-exposed rats was found to significantly restore total respiratory resistance back to control values but did not significantly increase compliance. (RA control: n = 50, RA+Nintedanib: n = 20, hyperoxia-exposed: n = 19, hyperoxia-exposed+Nintedanib: n = 24).

Fig. 4. Nintedanib treatment at 1 mg/kg improved pulmonary vessel density in hyperoxia-exposed rats.

Histological representations of pulmonary vasculature in in saline control rats (a), postnatal hyperoxia-exposed rats (b), control rats treated with 1 mg/kg of Nintedanib (c), and hyperoxia-exposed rats treated with 1 mg/kg of Nintedanib (d) are shown here. Nintedanib treatment visually preserved pulmonary vasculature in control rats and improved pulmonary vessel density in hyperoxia-exposed rats. The micrographs (e) demonstrate that compared to control rats, hyperoxia exposure significantly reduced PVD by 65% (p < 0.01). Nintedanib treatment at 1 mg/kg of hyperoxia-exposed rats was found to restore PVD back to control values. (RA control: n = 5, RA+Nintedanib: n = 3, hyperoxia-exposed: n = 6, hyperoxia-exposed+Nintedanib: n = 5).

Fig. 5. Nintedanib treatment at 1 mg/kg reduced pulmonary vessel wall thickness and RVH in hyperoxia-exposed rats.

Histological representations of pulmonary vessel wall thickness in saline control rats (a), postnatal hyperoxia-exposed rats (b), control rats treated with 1 mg/kg of Nintedanib (c), and hyperoxia-exposed rats treated with 1 mg/kg of Nintedanib (d) are shown here. Vessels chosen for measurements are marked by red arrows. Nintedanib treatment visually preserved pulmonary vessel wall thickness in control rats and reduced the thickness in hyperoxia-exposed rats. The micrographs (e) demonstrate that hyperoxia significantly increased RVH by 118% (p < 0.01) and PVWT by 170% (p < 0.02) as compared to controls. Nintedanib treatment at 1 mg/kg of hyperoxia-exposed rats was found to significantly decrease pulmonary vessel wall thickness and RVH back to control values. (For RVH, RA control: n = 20, RA+Nintedanib: n = 10, hyperoxia-exposed: n = 20, hyperoxia-exposed+Nintedanib: n = 29). (For PVWT, RA control: n = 5, RA+Nintedanib: n = 4, hyperoxia-exposed: n = 4, hyperoxia-exposed+Nintedanib: n = 4).

Fig. 6. Protein expression detected by western blot analysis demonstrated that compared to RA controls, hyperoxia exposure increased phosphorylated Src by 122% (p < 0.02).

Protein expression of day 5 lungs (a) showed no significant difference between phosphorylated Src kinase expression over total Src kinase expression in the hyperoxia-exposed group compared with the hyperoxia-exposed group that received Nintedanib treatment. (RA control: n = 4, RA+Nintedanib: n = 4, hyperoxia-exposed: n = 4, hyperoxia-exposed+Nintedanib: n = 4).

Fig. 7. Inflammatory and apoptotic markers were reduced with Nintedanib treatment at 1 mg/kg.

PCR analysis of gene expression in day 5 lungs showed  that inflammatory cytokines IL1a and TNFa were significantly elevated in the hyperoxia group compared with the RA control group by 452% (p < 0.05) and 668% (p < 0.01), respectively (a, b). The gene expression of apoptotic marker Caspase 3 (c) was also significantly elevated by 516% (p < 0.01) in the hyperoxia group compared with the RA control group (Fig. 7c). Nintedanib treatment of the hyperoxia group restored IL1a, TNFa, and Caspase 3 levels back to control values. (RA control: n = 4, RA+Nintedanib: n = 4, hyperoxia-exposed: n = 4, hyperoxia-exposed+Nintedanib: n = 4).