Fibroblast Activation Protein-Specific PET/CT Imaging in Fibrotic Interstitial Lung Diseases and Lung Cancer: A Translational Exploratory Study

Abstract

Interstitial lung diseases (ILDs) comprise over 200 parenchymal lung disorders. Among them, fibrosing ILDs, especially idiopathic pulmonary fibrosis, are associated with a poor prognosis, whereas some other ILDs, such as sarcoidosis, have a much better prognosis. A high proportion manifests as fibrotic ILD (fILD). Lung cancer (LC) is a frequent complication of fILD. Activated fibroblasts are crucial for fibrotic processes in fILD. The aim of this exploratory study was to evaluate the imaging properties of static and dynamic fibroblast activation protein (FAP) inhibitor (FAPI) PET/CT in various types of fILD and to confirm FAP expression in fILD lesions by FAP immunohistochemistry of human fILD biopsy samples and of lung sections of genetically engineered (Nedd4-2^-/- ) mice with an idiopathic pulmonary fibrosislike lung disease. Methods: PET scans of 15 patients with fILD and suspected LC were acquired 10, 60, and 180 min after the administration of 150-250 MBq of a ⁶⁸Ga-labeled FAPI tracer (FAPI-46). In 3 patients, dynamic scans over 40 min were performed instead of imaging after 10 min. The SUV_max and SUV_mean of fibrotic lesions and LC were measured and CT-density-corrected. Target-to-background ratios (TBRs) were calculated. PET imaging was correlated with CT-based fibrosis scores. Time-activity curves derived from dynamic imaging were analyzed. FAP immunohistochemistry of 4 human fILD biopsy samples and of fibrotic lungs of Nedd4-2^-/- mice was performed. Results: fILD lesions as well as LC showed markedly elevated ⁶⁸Ga-FAPI uptake (density-corrected SUV_max and SUV_mean 60 min after injection: 11.12 ± 6.71 and 4.29 ± 1.61, respectively, for fILD lesions and 16.69 ± 9.35 and 6.44 ± 3.29, respectively, for LC) and high TBR (TBR of density-corrected SUV_max and SUV_mean 60 min after injection: 2.30 ± 1.47 and 1.67 ± 0.79, respectively, for fILD and 3.90 ± 2.36 and 2.37 ± 1.14, respectively, for LC). SUV_max and SUV_mean decreased over time, with a stable TBR for fILD and a trend toward an increasing TBR in LC. Dynamic imaging showed differing time-activity curves for fILD and LC. ⁶⁸Ga-FAPI uptake showed a positive correlation with the CT-based fibrosis index. Immunohistochemistry of human biopsy samples and the lungs of Nedd4-2^-/- mice showed a patchy expression of FAP in fibrotic lesions, preferentially in the transition zone to healthy lung parenchyma. Conclusion:⁶⁸Ga-FAPI PET/CT imaging is a promising new imaging modality for fILD and LC. Its potential clinical value for monitoring and therapy evaluation of fILD should be investigated in future studies.

Keywords: fibroblast activation protein; interstitial lung disease; lung cancer.

Graphical abstract

FIGURE 1.

Average (±SD) density-corrected SUV_max (A), SUV_mean (B), and TBRs for SUV_max (C) and SUV_mean (D) for tumor and fibrosis in 15 patients at 10, 60, and 180 min after application of ⁶⁸Ga-FAPI.

FIGURE 2.

(A) Representative maximum-intensity-projection PET images after injection of ⁶⁸Ga-FAPI in 75-y-old man with rheumatoid arthritis–associated ILD and non–small cell lung carcinoma (red arrow). Clinically, patient had significant decrease in forced vital capacity (from 70% to 38%) over the last 4 mo before FAPI PET/CT and complained about progressive weight loss and exertional dyspnea, which denotes progressive phenotype according to criteria of INBUILD study (45). We observed intensively ⁶⁸Ga-FAPI–positive pulmonary fibrosis in right middle lobe (blue arrow) and moderately ⁶⁸Ga-FAPI–positive pulmonary fibrosis in right lower lobe (yellow arrow). (B) Representative axial CT images and PET/CT images of intensively ⁶⁸Ga-FAPI–positive (middle lobe) and moderately ⁶⁸Ga-FAPI–positive (lower lobe) pulmonary fibrosis lesions of same patient. Intensively ⁶⁸Ga-FAPI–positive lesion in right middle lobe may be a correlate of increased fibrotic activity, leading to observed clinical progression of fILD in this case. p.i. = after injection.

FIGURE 3.

(A) Axial PET/CT image and corresponding time–activity curves showing uptake of ⁶⁸Ga-FAPI over time in aorta, left-sided LC lesion, and fibrotic area in right lung. (B) Comparison of average time to peak (±SD) of 3 fibrotic areas and 3 LC lesions as measured by dynamic PET imaging. TTP = time to peak.

FIGURE 4.

(A and B) Scatterplots of SUV_mean derived from ⁶⁸Ga-FAPI PET/CT and corresponding FIB indices (A) and GGO indices (B) in 75 lobe volumes of 15 patients with fILD and suspected LC. (C) Scatterplot of FIB indices and GGO indices of same lobe volumes.

FIGURE 5.

Hematoxylin and eosin staining (A), FAP immunohistochemistry (B), and α-SMA immunohistochemistry (C) of exemplary FAP-positive spot in fILD lesion of 71-y-old (at time of biopsy) man with IPF, who was diagnosed with small cell lung carcinoma after ⁶⁸Ga-FAPI PET/CT. High FAP expression (red arrows) and high α-SMA expression (green arrows) are widely inversely distributed in fibrotic tissue (×20; scale bars, 50 μm).