Differential distribution of immune cells in breast invasive carcinoma vs. breast carcinoma and its significance in interpretation of immune surveillance.

Immune surveillance is a highly controversial subject in both the field of immunology and cancer biology. On one hand, in spite of extensive studies, there is no cancer specific antigens identified. Yet, the organisms do exert immune response to tumors. On the other hand, it is believed that immune surveillance suppresses tumorigenesis by eradicating mutated cells. However, it is also widely known that tumorigenesis is promoted by inflammation, which is in nature immune reaction. In the present study, we tried to find immune cells in early tumor lesions for the supportive or negative evidence of immune surveillance. We used immunohistochemistry to observe the localization and distribution of immune cells in the in situ carcinoma lesions and in the invasive cancer of breast. Interestingly, we did not see immune cells in either ductal carcinoma in situ (DCIS) or lobular carcinoma in situ (LCIS) of breast, the two basic supposed early cancer forms. In contrast, we observed extensive infiltration of immune cells in the invasive breast cancer, and close contact between immune cells and tumor cells. Based on these findings, we propose that the tumor antigens of breast cancer are not derived from the gene mutation or amplification such as HER2, but rather from misplacement of epithelial cells in the mesenchymal tissue. To avoid being targeted by the immune system, the carcinoma cells exert epithelial-mesenchymal transition (EMT). Therefore, immunosurveillance could be regarded as preventing the intrusion of epithelial cells to mesenchymal tissues, and EMT is a form of immune escape by the strategy of mimicry.