Comparison between conventional and "clinical" assessment of experimental lung fibrosis

Abstract

Background: Idiopathic pulmonary fibrosis (IPF) is a treatment resistant disease with poor prognosis. Numerous compounds have been demonstrated to efficiently prevent pulmonary fibrosis (PF) in animal models but only a few were successful when given to animals with established fibrosis. Major concerns of current PF models are spontaneous resolution and high variability of fibrosis, and the lack of assessment methods that can allow to monitor the effect of drugs in individual animals over time. We used a model of experimental PF in rats and compare parameters obtained in living animals with conventional assessment tools that require removal of the lungs.

Methods: PF was induced in rats by adenoviral gene transfer of transforming growth factor-beta. Morphological and functional changes were assessed for up to 56 days by micro-CT, lung compliance (measured via a mechanical ventilator) and VO2max and compared to histomorphometry and hydroxyproline content.

Results: Standard histological and collagen assessment confirmed the persistent fibrotic phenotype as described before. The histomorphological scores correlated both to radiological (r2 = 0.29, p < 0.01) and functional changes (r2 = 0.51, p < 0.0001). VO2max did not correlate with fibrosis.

Conclusion: The progression of pulmonary fibrosis can be reliably assessed and followed in living animals over time using invasive, non-terminal compliance measurements and micro-CT. This approach directly translates to the management of patients with IPF and allows to monitor therapeutic effects in drug intervention studies.

Figure 1

Representative lung histology. (A) naïve lungs, (B) day 14, (C) day 21, (D) day 35, (E) day 56 and (F) day 225 after intratracheal AdTGF-β1 administration (40×). (G) Fibrotic index (Ashcroft). (G) Hydroxyproline content per mg dry lung. Group comparison was performed using one way ANOVA with Dunnett's multiple comparison test. All values are given as mean, SE, n = 4–6 in each group, * p < 0.05, ** p < 0.01, *** p < 0.001 vs. naïve.

Figure 2

Computed tomography. (A). Representative example of one AdTGF-b1 exposed rat, scanned at day 0, 21, 35 and 56. Axial slice of CT-scan (upper panel) and 3D reconstructed lungs with fibrotic areas color-coded in green (medium panel). Corresponding histograms of total lung voxels from same animal (lower panel; insert represent percentage of fibrotic tissue. (B). Average percentage of voxels different from naïve lungs (All values are given as mean, SE, n = 4 to 6 at each time-point).

Figure 3

Correlation Histology – Computed tomography. (A). Low magnification histology 225 days after intratracheal AdTGF-β1 administration (10×) corresponding to CT image of axial slice (B) and to area indicated with red arrow in 3D reconstructed lungs from same animal (C). See insert in Fig 1F for higher magnification of same lung.

Figure 4

Lung function. (A) Average pressure/Volume curves from naïve and AdTGF-β1 exposed rats. The curves from fibrotic animals demonstrate marked shifts downward and to the right, indicating stiffer lungs. Note: The upward deviation of the curve at day 14 at pressures above 60 cm H2O reflect the upper limits of the pressure transducer. (B) The parameter K was reduced at all time-points. (C) Lung stiffness was derived from the PV loop in Figure 4A and characterized as the volume of air needed to reach a pressure of 20 cm H₂O. All values are given as mean, SE, n = 4–6 per group.

Figure 5

Correlations. Correlations between (A) histological fibrotic score and the parameter K (n = 57), (B) fibrotic score and absolute fibrotic volume (n = 20), (C) parameter K and absolute fibrotic volume (n = 18) and (D) fibrotic score and VO₂max (n = 41).