A novel in vivo model for predicting myelotoxicity of chemotherapeutic agents using IL-3/GM-CSF transgenic humanized mice.

Evaluating myelotoxicity is essential for ensuring the safety of novel drugs before they are approved for clinical applications. Although in vivo prediction of the maximum tolerated doses (MTDs) of anticancer drugs is usually performed in rodents, the results are not always applicable to clinical treatment because drugs may have different effects in human and rodent cells. Previously, we generated a human IL-3 and GM-CSF transgenic humanized mouse (hu-IL-3/GM Tg), in which human granulocytes effectively differentiated after hematopoietic stem cell transplantation. In this study, we established a novel in vivo preclinical evaluation model for predicting human myelotoxicity of anticancer drugs using these hu-IL-3/GM Tg mice. The myelotoxicity was investigated by kinetic flow cytometry of human or murine granulocytes and by colony-forming unit granulocyte/macrophage (CFU-GM) assays. In both in vivo and in vitro analyses, topotecan was more myelotoxic to human than murine granulocytes. In contrast, oxaliplatin was more myelotoxic to murine granulocytes. The level of myelotoxicity of paclitaxel treatment was comparable between human and mouse cells. These results demonstrate that our humanized mouse model can simultaneously evaluate myelotoxicity against human and mouse cells in vivo, and provides an effective preclinical tool for predicting appropriate doses of anticancer agents for clinical treatment.

Preclinical anticancer effects and toxicologic assessment of hepatic artery infusion of fine-powder cisplatin with lipiodol in vivo.

We conducted an in vivo study to evaluate the anticancer effect and toxicity of fine-powder cisplatin suspended in lipiodol (fCDDP/LPD suspension) after a single administration of three different doses to rats via the intrahepatic artery after transplantation of rat ascites hepatoma cells. The toxicity of the fCDDP/LPD suspension was also assessed in the same protocol in noncancer-bearing rats and the observed toxicologic changes were compared among groups administered saline (Sal), an aqueous solution of fCDDP (fCDDP/Sal solution), and LPD alone. In parallel with the toxicity test, plasma CDDP concentrations were compared between the fCDDP/LPD suspension and fCDDP/Sal solution. The mean weight of the tumors in the fCDDP/LPD suspension groups was significantly less than in the LPD-alone group. The pathologic changes in the liver observed in the fCDDP/LPD suspension group increased with dose, were more marked compared with those in the fCDDP/Sal solution and LPD-alone groups, and were reversible. No other toxicologic effects were observed. The concentration of CDDP in the plasma in the fCDDP/LPD suspension group was slightly lower than that in the fCDDP/Sal solution group. In conclusion, the results indicate that the fCDDP/LPD suspension has sufficient anticancer efficacy and tolerability for use in the clinical treatment of hepatocellular carcinoma.

Coexpression of HGF and c-Met/HGF receptor in human bone and soft tissue tumors.

To understand the interaction between hepatocyte growth factor (HGF) and its receptor c-Met on various bone and soft tissue tumors, their expressions were investigated by western blot analysis, immunohistochemistry and enzyme immunoassay. Western blot analysis revealed that c-Met protein was expressed in 21 (38.8%) of 54 tumors, which detailed to seven (25.9%) of 27 bone tumors and 14 (51.8%) of 27 soft tissue tumors. Most malignant fibrous histiocytomas (MFH) and all neurofibromas expressed c-Met protein. The highest expression of c-Met protein was seen in a case of biphasic synovial sarcoma, where its immunoreactivity was localized only on the epithelial component and not on the sarcomatous component. By enzyme immunoassay for HGF, all but one MFH showed HGF production and the mean level of HGF was the highest among the tumors investigated. Neurofibromas and osteosarcomas had the next highest mean levels of HGF production, respectively. Coexpression of HGF and c-Met was observed in 19 (35.2%) of 54 tumors and was frequently observed in neurofibroma, followed by MFH and synovial sarcoma. Although the mode of interaction between HGF and c-Met varies among the various bone and soft tissue tumors including MFH, their signaling system may play an important role in the development and progression of bone and soft tissue tumors.

c-Met expression of thyroid tissue with special reference to papillary carcinoma.

It has become clear that papillary carcinomas of the thyroid often express the receptor for c-Met/hepatocyte growth factor (HGF) receptor, but little is known about the role of the HGF and c-Met system in the pathogenesis of thyroid carcinoma. In this study, the expression of c-Met/HGF receptor was evaluated in thyroid tissue by western blot and immunohistochemistry, and compared with the concentration of HGF. Clinicopathological characteristics were also compared. Fifteen of 20 papillary carcinomas (75%) showed c-Met bands of 145 kDa. No or only a low frequency of c-Met expression was detected in healthy thyroid tissue (0/5), thyroiditis or Basedow's disease (0/2), adenomatous goiters (0/8), follicular adenomas (1/9, 11%) and undifferentiated carcinomas (0/2). These results were confirmed by immunohistochemistry, but a relatively higher frequency of c-Met expression was detected in adenomatous goiters (25%), follicular adenoma (44%) and papillary carcinoma (100%) using formalin-fixed and paraffin-embedded materials. A strong immunoreaction for c-Met was observed in the tumor cytoplasm of papillary carcinomas among the fibrous tissues situated at the periphery of the tumor. The densitometrically measured expression of c-Met had no relation to tumor stage in papillary carcinoma, but did correlate to the concentration of HGF in papillary carcinomas. In conclusion, in thyroid lesions, c-Met was highly expressed specifically in the cytoplasm of papillary carcinomas. c-Met expression was not related to the aggressiveness of the tumor but was related to the concentration of HGF, which was probably derived from the stroma. Also, the c-Met system might play a role in the pathogenesis of papillary carcinoma of the thyroid.

Expression of c-met/HGF receptor in human non-small cell lung carcinomas in vitro and in vivo and its prognostic significance.

The expression of c-met/HGF receptor was evaluated in non-small cell lung cancers (NSCLC) by western blot analysis of 11 established cell lines and 104 surgically resected tissues. All cancer cell lines (eight adenocarcinomas, two squamous cell carcinomas and a large cell carcinoma) showed strong c-met protein bands of 145 kDa and 170 kDa. Moreover, c-met protein was demonstrated in 34 (72.3%) of 47 surgically resected adenocarcinomas, 20 (38.5%) of 52 squamous cell carcinomas and 3 of 5 others, and the results were mostly confirmed immunohistochemically in formalin-fixed and paraffin-embedded tumors of the same case. Although squamous cell carcinomas showed relatively high c-met protein expression in established cell lines, more adenocarcinomas than squamous cell carcinomas showed c-met protein expression in the original cancers. Furthermore, two cell lines used in this study originated from primary cancers negative for c-met protein expression, suggesting that c-met protein expression might be influenced by cultivation. Furthermore, clinicopathological study revealed that NSCLC with c-met protein expression tended to be in a higher pathological tumor stage and to have a worse outcome than those without such expression. In conclusion, c-met protein is expressed in cell lines and primary tumors of NSCLC, and this phenomenon is probably closely related to the aggressive behavior or progression of NSCLC, especially of adenocarcinomas.