In this study, we examined the relationship between c-kit expression and the malignant potential of pancreatic cancer cells. Thus, we demonstrated the influence of c-kit expression on proliferation and invasion of pancreatic cancer cell lines in vitro. In addition, we analyzed the relationship between KIT and SCF expression in pancreatic cancer tissues and examined clinical features and prognosis.
expression was examined in five pancreatic cancer cell lines. It was ascertained that two of the cell lines, PANC-1 and SW1990, expressed c-kit
mRNA and protein by RT-PCR and Western blot analysis. Based on those results, we defined these two pancreatic cancer cell lines as KIT-positive. Next, we stimulated those lines with SCF to investigate the differences in proliferative responses between KIT-positive and KIT-negative cell lines. In the two KIT-positive cell lines, proliferation was significantly enhanced by SCF at concentrations above 1 ng/mL. On the other hand, SCF did not enhance proliferation of KIT-negative cancer cell lines. Results from the invasion (Matrigel) assay were consistent with the proliferation assay. Namely, the invasive ability of KIT-positive pancreatic cancer cell lines was significantly enhanced by addition of SCF at concentrations above 1 ng/mL. The fact that KIT activation induces proliferation or invasion in vitro
was also reported in lung cancer and colorectal cancer [15
]. In this study, 1 ng/mL SCF significantly enhanced proliferation and invasion. Because the human in vivo
level of SCF in serum is between 1 ng/mL and 3 ng/mL [34
], serum SCF concentration may be high enough to stimulate the proliferation and invasion of pancreatic cancer cells in vivo
. Recently, it was shown that the existence of mast cells which showed immunoreactivity for SCF was ascertained in pancreatic cancer [31
]. Our studies indicate that KIT expression by pancreatic cancer cells may have functional significance. Thus, we propose that serum SCF and SCF secreted from mast cells could contribute to in vivo
tumor progression of KIT-positive pancreatic cancer.
In pancreatic cancer specimens, c-kit
receptor was reportedly expressed in 6.1% based on immunohistochemical techniques [28
]. On the other hand, two other studies from the same department reported that about 80% of pancreatic cancers expressed KIT [29
]. In the present study, KIT was expressed in 38% of pancreatic cancer specimens (16 of 42 patients) by immunohistochemistry. In the report which concluded that 6.1% of pancreatic cancer expressed KIT, antigen retrieval was not performed. Thus, the percentage expressing KIT may be erroneously low. As to the studies reporting about 80% KIT expression, antigen retrieval was done using urea. On the other hand, we used Target Retrieval Solution High pH (DAKO, Copenhagen, Denmark) for antigen retrieval according to the antibody manufacturer's instruction. The difference between the methods relating to antigen retrieval may contribute to the differences in KIT-positive ratio.
In this study, clinicopathological findings associated with venous system invasion were significantly more severe in the KIT-positive group than in the KIT-negative group. Interestingly, KIT activation has been linked to vascular endothelial growth factor (VEGF) expression in GIST [37
]. In addition, it was reported that VEGF expression was significantly correlated with venous invasion in gastric cancer [38
]. From these facts, we suppose that KIT induced VEGF expression may have relationship with venous system invasion in pancreatic cancer. As to prognosis, in small-cell lung cancer, patients with KIT expression had a significant tendency toward lower survival than did KIT-negative patients [14
]. Concerning pancreatic cancer, two studies reported that KIT-positive patients had a tendency toward lower survival than did c-kit-negative patients, but the differences were not significant [29
]. Those results were confirmed here, as the difference in prognosis between KIT-positive and KIT-negative patients was not significant.
With regard to expression of SCF, pancreatic cancer specimens were evaluated for simultaneous expression with KIT. Co-expression of KIT and SCF has been reported in small-cell lung cancer, and in those diseases, autocrine pathways are suggested [39
]. In our immunohistochemical study, 14 pancreatic cancer specimens co-expressed KIT and SCF, and the proportion of SCF-positive specimens was significantly larger in KIT-positive specimens than in KIT-negative ones. Patients co-expressing KIT and SCF had a tendency toward lower survival than others. Thus, autocrine pathways may play important roles in the evolution of malignant disease in the pancreas.
Recently, drugs which selectively inhibit tyrosine kinase receptors have been developed [24
]. The goal is tumor-specific disruption of cell proliferation and differentiation. Targeted receptors include epidermal growth factor receptor and vascular endothelial growth factor receptor [41
]. Because c-kit
tyrosine kinase receptor has an important role in tumor progression, the KIT inhibitor imatinib mesylate is being explored for therapeutic efficacy [27
]. Imatinib mesylate was initially designed to inhibit Bcr/Abl kinase which is present in 95% of patients with CML [43
]. Later, it was found that imatinib mesylate also inhibited the c-kit
tyrosine kinase receptor [44
]. Imatinib mesylate is used for the therapy of GIST and leukemia and contributes to improvement of the prognosis, response ratio and quality of life [45
]. In this study, we assessed the inhibitory effect of imatinib mesylate on proliferation and invasion of pancreatic cancer cells. In leukemia patients, imatinib mesylate plasma concentrations were in the range of 0.17 μM – 5.68 μM when 25 – 600 mg of the drug was administered per day [46
]. We administered 5 μM imatinib mesylate to pancreatic cancer cell lines, and the concentration was within the in vivo
range of plasma. In KIT-positive pancreatic cancer cell lines, 5 μM imatinib mesylate significantly inhibited SCF-enhanced proliferation to the level shown by nonstimulated control. Imatinib mesylate similarly reduced invasive potential. As to the therapy of pancreatic cancer, gemcitabine is used as the standard chemotherapy now, and therapies combining gemcitabine with other drugs have been explored [47
]. For KIT-positive pancreatic cancers, KIT inhibitors could block the ligand-dependent signaling pathway. Thus, we propose that a KIT inhibitor could be used with gemcitabine, or used as a second choice when the pancreatic cancer is resistant to gemcitabine.