Role of integrins and other cell adhesion molecules in tumor progression and metastasis.

Despite rapid advances in our understanding of the biology of cell adhesion, the data available in the literature make it is difficult to propose one simple scheme in which cell adhesion molecules can be related to tumor growth and metastasis. This difficulty can be related to a number of factors. Some of the apparently conflicting experimental results that demonstrate both enhanced or diminished tumor cell adhesion during tumor progression may be attributed to the experimental systems used. Those studies that have injected tumor cells intravenously have, in general, shown that enhanced tumor cell adhesiveness correlates with metastatic ability. It should be recognized that this experimental approach bypasses many of the early stages of the metastatic cascade and is biased towards tumor cells with an enhanced ability to form aggregates with cells in the circulation and to adhere to distant vascular sites. On the other hand, studies that have implanted tumors into animals and allowed them to grow and metastasize (spontaneous metastasis) have generally demonstrated an inverse relationship between adhesive ability and the ability to metastasize. Another major obstacle in understanding the role of CAMs in metastasis is the well known problem of tumor heterogeneity and the phenotypic instability of metastatic cells over relatively short periods of time (141). The cells that make up a metastatic focus may thus be quite different from the tumor cells that originally formed the lesion. It is quite possible that the selective pressures that initially enable a cancer cell to form a metastatic lesion may be quite different than those that later favor rapid tissue growth. The major obstacle in making any sweeping generalizations about cell adhesion molecules and tumor progression, however, is that the process of successful metastasis is inherently complex, requiring tumor cells to possess decreased adhesive interactions with surrounding cells and extracellular matrix at some points in the cascade and increased adhesive interactions at other times. Based on the information available, the following scenario can be proposed. Using the schema shown in Figure 1, successful metastasis initially requires that normal cell-cell and cell-substratum adhesion be disrupted, causing release of neoplastic cells from the primary tumor (step 1). For epithelial tumors, down-regulation of cadherins and perhaps, integrins, appear to be involved. This loss of cell adhesion must be followed by migration of tumor cells into the vascular system (step 2), a step requiring efficient cell-substratum interactions. In melanomas, this step seems to require expression of the vitronectin receptor.(ABSTRACT TRUNCATED AT 400 WORDS)