Background: Human xenograft tumor models established by transplantation of human tumor cell lines into immunodifficient mice have been routinely used for preclinical test of anticancer agents. But such tumor models have a relatively low transplantability and limited correlation with clinical findings. Recently, we have developed human non-small lung cancer (NSCLC) xenograft models by transplanting human fresh lung cancer tissues into nude mice, which have been employed for test of clinically used chemotherapeutical drugs and are under way for new drug chemosensitivity in vivo screening.

Methods: The fresh lung cancer samples were collected from local hospitals. The tumor fragments of 1–2 mm3 were subcutaneously implanted in the flanks of the Balb/c nude mice by trocar needle. Sixteen tumor fragments were grafted into four mice from one patient tumor tissue (passage 1). In the first passage, tumors derived from male patients were implanted into male mice, and tumors from women were inoculated into female mice. All therapeutic efficacy experiments used female mice. The drugs tested were clinically used drugs for NSCLC including cisplatin (4 mg/kg, q3d x 4), paclitaxel (15 mg/kg, q3d x 4), vinorelbine (2.5 mg/kg, q4d x 4), gemcitabine (120 mg/kg, q4d x 4), and the epidermal growth factor receptor (EGFR) inhibitor erlotinib (50 mg/kg/daily x 14, po).

Results: A total of 42 NSCLC samples were implanted, and 13 tumor models were established (tumor taking rate 31%) for the first passage. The tumor taking rates were higher in the second and third passages (80–100%). Cisplatin, paclitaxel, vinorelbine, and gemcitatbine produced tumor growth inhibition rates of 48–55% regardless of the EGFR mutation status. While erlotinib demonstrated a significant antitumor activity only in the tumors bearing EGFR, which were consistent with their clinical findings. The tumor xenografts' architecture, the cell and histopathological morphology from the three generations mirrored the original patient cancers.

Conclusions: These results suggest that human primary tumor xenograft models provide a unique renewable source of tumor material for test of novel anticancer agents. They may predict more relevant clinical response rates and higher correlation with clinical findings than use of xenograft models established from in vitro long-term cultured cancer cell lines, especially for test of target-oriented therapeutics in new drugs development programs.

Citation Information: Clin Cancer Res 2010;16(7 Suppl):B24