Pattern of p53 gene mutations in breast cancers of women of the midwestern United States.
BACKGROUND: Mutation in the p53 gene is the most common genetic lesion in human cancers. The pattern of mutation in the p53 gene differs among cancers and may be a useful epidemiological tool for identification of factors contributing to carcinogenesis. PURPOSE: Our purpose was to determine if the pattern of p53 mutation in breast carcinomas in our population of women residing in the midwestern region of the United States is similar to the pattern of p53 mutation in breast cancers in patients from other regions of the United States and Europe and in other epithelial tumors. METHODS: With a technique we recently developed for the analysis of p53 mutations in genomic DNA from tumor cell clusters in touch preparations of solid tumors, we sequenced exons 5-9 and adjacent splice junctions of the gene in 44 breast cancers. Cells from each tumor were also stained with three monoclonal antibodies which recognize different epitopes of the p53 protein. RESULTS: We detected p53 mutations in 14 (32.6%) of 44 breast carcinomas. Only half of the mutations were missense changes. The other half included five microdeletions (three producing frame-shifts), one single-base substitution generating a stop codon, and one single-base substitution generating a splice junction abnormality. Nuclear expression of p53 antigen was present in eight of 44 cancers, including six with hemizygous missense mutations in the p53 gene. CONCLUSIONS: The pattern of p53 mutations in our breast cancer population differs from that reported in breast cancer populations by other investigators in which most p53 mutations were missense. Among 14 mutations in our population, at least five drastically altered the structure of p53, suggesting that a recessive mechanism of inactivation of the p53 gene may be more common than in other populations. IMPLICATIONS: Differences in the pattern of p53 mutation in breast cancers in Midwestern women and in breast cancers in other populations may reflect selection bias or small sample sizes currently available. However, our data are compatible with the possibility that an endogenous or exogenous factor influences p53 carcinogenesis in some women with breast cancer in the Midwest to a greater extent than in other regions of the United States and Europe.