The research activities of the Pathology Division of the Research Center for Innovative Oncology currently concentrate on the application of morphological study of cancer tissue to the clinical course of the patient (1) to elucidate the biological activity of cancer and establish new methods of treatment selection for patients and (2) to clarify the mechanism of cancer metastasis. The prognostic factors and clinicopathological characteristics of various cancers have been studied in collaboration with the Diagnostic Pathology Section of the Clinical Laboratory Division of the National Cancer Center Hospital East.
Cancer Microenvironment and Cancer Biology
The cancer microenvironment generated by cancer-stromal interaction plays an important role not only in carcinogenesis but also in cancer progression. Several studies have indicated that metastatic cancer cells can use the milieu of potentially metastatic organs, other cells, or matrices to achieve organ-specific metastases such as bone metastasis by prostate cancer. Among the growth factors stored in the bone matrix, prostate cancer cells use insulin-like growth factors (IGF) 1 and 2 for cell survival and growth in the bone tissue. The growth factors stored in bones are released only by mature osteoclasts that are induced by cancer cells. Using human adult bone (HAB) implanted in nonobese diabetic/severe combined immunodeficient (NOD/SCID) mice. (75), it was shown that the new bisphosphonate drug, YM592, accumulates in mature osteoclasts and induces their apoptosis, resulting in inhibition of IGF release from the bone matrix and growth of prostate cancer metastases in the bone.
Many growth factors and cytokines are immobilized by binding to the glycosaminoglycans on the extracellular matrix (ECM) and are stored in an inactive form in the cellular microenvironment. IGFs are bound to IGF-binding proteins (IGF- BPs) and become stable in blood and bones. Various types of kallikreins degrade IGF-BPs and activate IGFs. Kallikrein 11, produced by hormone-dependent breast cancer cells, can also degrade IGF-BP3, which is a major IGF-BP in serum IGFs, and can also activate IGFs resulting in promoting the progression of breast cancer. (76)
The ECM derived from the culture of the human colon cancer cell line, HT-29, on bovine serum is found to contain IGF-BP2 and IGF. Matrix metalloproteinase 7 (MMP7) can degrade IGF-BP2 and activate IGFs from ECM. Heparinase III that degrades heparin and releases heparin-binding growth factors from the ECM of HT29 cells can also inhibit bioactivated IGF in the ECM. These data indicate that MMP7 generates bioactivated IGF-2 by degrading the IGF-2/IGFBP-2 complex that is bound to the proteoglycan, heparan sulfate, in the ECM, resulting in IGF-2-induced signal transduction as a matricrine manner. (77)
Vascular endothelial growth factor (VEGF) production by cancer tissue is thought to play a crucial role in tumor angiogenesis, although VEGF is also produced in various normal tissues and organs. This indicates the presence of a tumor-specific angiogenic trigger around the tumor. VEGF produced by the normal human fibroblast cell line, VA-13, promotes tumor angiogenesis in vivo but does not stimulate endothelial cell growth in vitro. This cell line also produces connective tissue growth factor (CTGF) that binds and inactivates VEGF. MMP7 produced by human cancer cells degrades CTGF and bioactivates VEGF, resulting in switching the cancer-specific angiogenic trigger on. The presence of this tumor-specific angiogenic trigger in human colon cancer tissue has been confirmed. These data indicate that in the extracellular environment, latent state cancer stromal fibroblasts become a source of VEGF that can be used for urgent angiogenesis. (78)
Detection of circulating progenitor cells of human cancer stromal fibroblasts
Recent animal data have suggested that cancer-induced stroma consists of blood-borne fibroblasts as well as tissue-derived fibroblasts. In order to detect the circulating progenitor cells of human cancer stromal fibroblasts, the pulmonary venous blood of lung cancer patients was collected and cultured. The pulmonary venous blood of 34% of the lung cancer patients contained fibroblast precursor cells among the mononuclear cells. The cultured fibroblasts demonstrated a spindle shape and produced collagen type I but not endothelial cell markers. After fluorescence labeling, these mononuclear cells were transferred in a human lung tumor xenotransplanted mouse and were confirmed to migrate and differentiate to human cancer fibroblasts. A clinicopathological study also indicated that the presence of blood-borne fibroblasts in blood is significantly associated with the extent of central fibrosis in lung cancer. These data comprised direct evidence that blood in the vicinity of human lung cancer contains fibroblast progenitor cells that have the capacity to migrate into the cancer stroma. (79)
Evaluation of cancer morphology and establishment of an animal model
By processing of HE sections, it is often difficult to determine whether the artificial space around the cancer nest comprises artifacts or true intratumoral lymph vessel spaces. In order to precisely evaluate lymph vessel tumor emboli (LVTEs), a comparison between HE-stained and D2-40-immunostained breast cancer tissues was performed. The existence of intratumoral LVTEs was confirmed, and intratumoral LVTEs dentified by either HE staining or D2-40 immunostaining were accurate predictors of the outcome of patients with invasive ductal carcinoma (IDC) of the breast. (80)
ancreatic IDC shows continuous spread via neural routes. Nerve invasion in pancreatic IDC is frequently observed and reaches approximately 100%. To confirm the hypothesis that nerve invasion was a common invasive behavior of pancreatic IDC and the level of nerve invasion correlates with patient survival, a detailed histological analysis of nerve plexus invasion in pancreatic IDC and its association with the clinical impact was investigated. Tumor cells invading the nerve plexus showed a differentiated phenotype forming ducts in the perineurium and the perineural space. The extent of nerve invasion was a significant prognostic factor according to multivariate analysis. According to these data, a new animal model for evaluating the extent of nerve invasion was established using several human pancreatic cancer cell lines. The mechanism of nerve invasion by human pancreatic IDC will be elucidated using this model. (47)
Clinicopathological Studies on Various Cancers in Collaboration with the Diagnostic Pathology Section
Clinicopathological studies on gastrointestinal tract cancers (81) and lung tumors (67, 82, 83, 84) were performed in collaboration with the clinical divisions of the National Cancer Center Hospital East and other institutions. (85, 3)
● A. Ochiai ●
Table of Contents