Exploitation of the chick embryo chorioallantoic membrane (CAM) as a platform for anti-metastatic drug testing.

The establishment of clinically relevant models for tumor metastasis and drug testing is a major challenge in cancer research. Here we report a physiologically relevant assay enabling quantitative analysis of metastatic capacity of tumor cells following implantation into the chorioallantoic membrane (CAM). Engraftment of as few as 103 non-small cell lung cancer (NSCLC) and prostate cancer (PCa) cell lines was sufficient for both primary tumor and metastasis formation. Standard 2D-imaging as well as 3D optical tomography imaging were used for the detection of fluorescent metastatic foci in the chick embryo. H2228- and H1975-initiated metastases were confirmed by genomic analysis. We quantified the inhibitory effect of docetaxel on LNCaP, and that of cisplatin on A549- and H1299-initiated metastatic growths. The CAM assay also mimicked the sensitivity of ALK-rearranged H2228 and EGFR-mutated H1975 NSCLC cells to tyrosine kinase inhibitors crizotinib and gefitinib respectively, as well as sensitivity of LNCaP cells to androgen-dependent enzalutamide therapy. The assay was suggested to reconstitute the bone metastatic tropism of PCa cells. We show that the CAM chick embryo model may be a powerful preclinical platform for testing and targeting of the metastatic capacity of cancer cells.

A prospective examination of circulating tumor cell profiles in non-small-cell lung cancer molecular subgroups.

BACKGROUND: We report the first study examining the clinical, numerical and biological properties of circulating tumor cells according to molecular subtypes of non-small-cell lung cancer. PATIENTS AND METHODS: 125 patients with treatment-naïve stage IIIb-IV NSCLC were prospectively recruited for CellSearch analysis. Anti-vimentin antibody was included for examination of CTCs to assess their mesenchymal character. Associations of total CTCs and vimentin-positive (vim +) CTCs with clinical characteristics, tumor genotype, and survival were assessed. RESULTS: 51/125 patients (40.8%) were total CTC+ and 26/125 (20.8%) were vim CTC+ at baseline. Multivariate analysis showed patients with ≥5 total CTCs had significantly reduced OS (HR 0.55, 95% CI 0.33-0.92, P = 0.022) but not PFS (HR 0.68, 95% CI 0.42-1.1, P = 0.118) compared to patients with <5 total CTCs. No OS difference was evident between vim+ CTC and vim-negative CTC patients overall (HR 1.24, 95% CI 0.67-2.28, P = 0.494), but after subdivision according to NSCLC driver mutation, we found an increase of vim+ CTCs in the EGFR-mutated subgroup (N = 21/94 patients; mean 1.24 vs 1.22 vim+ CTCs, P = 0.013), a reduction of total CTCs in the ALK-rearranged subgroup (N = 13/90 patients; mean 1.69 vs 5.82 total CTCs, P = 0.029), and a total absence of vim+ CTCs in KRAS-mutated adenocarcinomas (N = 19/78 patients; mean 0 vs 1.4 vim+ CTCs, P = 0.006). CONCLUSIONS: We validate that the baseline presence of ≥5 total CTCs in advanced NSCLC confers a poor prognosis. CTCs from EGFR-mutant NSCLC express epithelial-mesenchymal transition characteristics, not seen in CTCs from patients with KRAS-mutant adenocarcinoma.