Automated Digital Quantification of Pulmonary Fibrosis in Human Histopathology Specimens

ABSTRACT

RATIONALE Pulmonary fibrosis is a progressive disease characterized by abnormal accumulation of extracellular matrix protein, mesenchymal cells, and immune cells in the alveolar interstitium. Methods quantifying fibrosis severity in lung histopathology samples are semi-quantitative, subjective, and analyze only portions of sections. We sought to determine whether automated computerized imaging analysis shown to continuously measure fibrosis in mice could be applied in human samples to quantify pulmonary fibrosis severity and immunostained cells in whole tissue sections. METHODS: Digital images of entire tissue sections from lung explants or biopsies from patients with Hermansky-Pudlak syndrome pulmonary fibrosis (HPSPF) or idiopathic pulmonary fibrosis (IPF) stained with picrosirius red and alcian blue or immunostained with anti-CD68 antibody were analyzed using a dedicated software program to quantify fibrosis, collagen, and macrophage content. Automated fibrosis quantification was compared to pulmonary function measurements or Ashcroft score. RESULTS: Automated quantification of pulmonary tissue density and fibrosis scores of HPSPF lung explants (n=3) was significantly higher than that of IPF lung explants (n=3) or biopsies (n=4). These values were also significantly higher in IPF lung explants than in IPF biopsies. Mean Ashcroft score in HPSPF or IPF explants was significantly higher than that of IPF biopsies. In contrast to automated quantification, Ashcroft scores of HPSPF and IPF explants did not differ significantly from each other. Automated tissue density and fibrosis scores correlated with lung function values and Ashcroft scores. Automated quantification of collagen content was similar in the three groups. Immune cell aggregates were identified in hematoxylin and eosin stained tissue;automated quantification of CD68 immunolabeled cells was significantly higher in HPSPF explants than in IPF biopsies. CONCLUSIONS: Severity of pulmonary fibrosis in stained human tissue can be automatically quantified in entire lung sections using a dedicated software program. This reader-independent method provides a more accurate assessment of pulmonary fibrosis than Ashcroft scores. Automated quantification of fibrosis, collagen content, and immunostained cells in fibrotic lung tissue sections can be performed simultaneously. Robust automated digital image analysis of human lung samples enhances the available tools to quantify and study pulmonary fibrosis and has potential application in preclinical studies and clinical investigations focusing on fibrotic lung disease and other diffuse pulmonary diseases, such as sarcoidosis, hypersensitivity pneumonitis, pneumoconiosis, or COVID-19.