It is clear that breast cancer progression is associated with inactivation of a number of different recessive oncogenes. The most widely evaluated tumor suppressor gene, p53, is mutated in approximately 30-50% of sporadic breast cancers. Mutations usually occur early in malignant progression. Loss of heterozygosity (LOH) studies have identified numerous chromosomal regions where other recessive oncogenes relevant to breast cancer may be located.
Each LOH is seen in a varying proportion of breast cancers and may appear either early or late in progression. High-grade ductal carcinoma in situ (DCIS) and invasive carcinoma have similar genetic lesions, showing that aberrations can occur before invasive disease. Direct evidence that the same aberrations can be acquired later in progression comes from a study of multiple metastases from the same patient; other studies found that primary invasive cancers are characterized by marked intratumor heterogeneity for each lesion examined.
The model we propose to account for these results hypothesizes that multiple genetic lesions can accomplish each phenotype required for malignancy (ie., dysregulated proliferation, invasion, angiogenesis, etc.) and that, for a given tumor, at least one aberrant gene for each phenotypic change is stochastically selected. Biological heterogeneity of breast cancer results from the stochastic acquisition of various genetic aberrations. We further propose that the lymphocytic reaction in high-grade X I S may select for aggressive tumor subpopulations capable of escaping immune surveillance. Another aspect of tumor heterogeneity may be the multiple mechanisms employed by various tumors to escape immune surveillance.