Abstract Purpose: The purpose of our study is to (a) establish a reproducible mouse model of orthotopic pleural cancer that recapitulates the pathobiology of human pleural malignancy (including primary malignant pleural mesothelioma, lung and metastatic breast cancers), and (b) validate the use of bioluminescence imaging (BLI) and serum biomarker assessment to noninvasively monitor tumor progression and therapy response in preclinical therapeutic trials. Experimental Design and Results: Human mesothelioma and lung cancer cell lines with variable expression levels of the tumor-associated antigen, mesothelin, were retrovirally transduced to stably express GFP-Firefly Luciferase, to facilitate BLI. Orthotopic xenograft mice were established by direct tumor cell inoculation into the pleural space with > 95% engraftment in both athymic nude and SCID-beige mice. The pathobiologic resemblance of the xenograft tumors to human disease was confirmed by demonstrating: (a) visceral and parietal pleural tumor growth 58 pattern, mediastinal lymph node involvement, and diaphragmatic invasion by magnetic resonance imaging (MRI) and histology, (b) strong, sustained expression of tumor markers at advanced stage of disease by WT-1, Calretinin, Mesothelin, and TTF-1 immunohistochemistry, and (c) extensive tumor lymphangiogenesis by CD-34 and LYVE-1 immunofluorescence. Tumor burden correlated with BLI signal (total flux photons/s) and soluble mesothelin-related peptide, SMRP (Mesomark® ELISA assay) both in vitro and in vivo (Pearson's correlation r=0.998, p<0.0001 and r=0.89, p=0.04, respectively). We determined the accuracy of BLI and SMRP (range 0.4–23nM) in monitoring tumor progression by serial assessment of xenograft mice (n=36, with initial tumor burden of 1×103–1×106 cells). In the orthotopic pleural xenografts, BLI signal correlated with SMRP (r=0.56, p=0.0001); no such correlation was observed in subcutaneous flank or intraperitoneal xenografts. To further demonstrate the reliability of BLI and SMRP to noninvasively monitor therapeutic response, groups of pleural mesothelioma xenografts (n=28) were treated with either: (a) Cisplatin (4mg/kg IP weekly), (b) thoracic radiation (20Gy total in 5 fractions), (c) combination chemoradiation, or (d) no treatment. BLI, SMRP, and survival were followed serially. Therapy response was demonstrated by decreased BLI intensity in mice treated with chemoradiation compared to controls (3.7×107 vs 2.0×108 photons/s, respectively, p=0.04) as well as decreased SMRP (p=0.03) which correlated with improved survival by Kaplan-Meier analysis (64 vs 34.5 days, p=0.0001). Conclusions: Pleural involvement by solid tumors remains a significant problem in clinical oncology affecting nearly 250,000 patients annually in the U.S. with no validated preclinical model for therapeutic investigations. We have validated a clinically relevant, reproducible orthotopic model of pleural malignancy that recapitulates human disease and allows for noninvasive quantitative monitoring of pleural tumor burden, progression and therapeutic response. Our xenograft model provides a novel platform to investigate targeted therapies for pleural malignancies. Citation Information: Cancer Res 2009;69(23 Suppl):B27.
Read full abstract