The influence of the p53 codon 72 polymorphism on ovarian carcinogenesis and prognosis

The frequency of dysfunction of the p53 tumor suppressor gene in cancer has made the concept of gene replacement therapy with wild-type p53 an attractive strategy. Codon 72 of the p53 gene is highly polymorphic with a reported arginine/proline allelotype frequency of 0.65/0.35 for Caucasians and a reversal of this ratio in African-Americans. Ovarian cancer is more common and less aggressive in Caucasians. The arginine and proline alleles have different biochemical properties. Thus, we have hypothesized that these alleles may also have different biologic propel-ties that could make one superior to the other for gene replacement therapy. To test this hypothesis in vivo, we investigated the prevalence of each allelotype in a population of 190 Midwestern American women with ovarian cancer and 52 healthy controls without a family history of cancer. We have found that: (i) the heterozygous arginine/proline allelotype is more common in probands with borderline cancers than in probands with invasive cancers (P = .0001) or healthy controls (P = .005); (2) despite a survival advantage (P = .006), probands homozygous for the arginine allele developed ovarian cancer at an earlier age (P = .01); (3) the frequency of tumor p53 mutations was independent of the germline p53 allelotype, but (4) when a loss of heterozygosity occurred in probands with invasive disease, the proline allele was lost preferentially (P = .002), and (5) any tumor which retained a proline allele was more prone to mutation (P = .04) than a tumor without a proline allele. Our results suggest that variation in the p53 codon 72 allelotype is an example of an intermediate risk polymorphism which interacts with epigenetic factors to play a role in ovarian carcinogenesis and may differentially influence cellular DNA repair and apoptotic pathways. These findings may have important ramifications in the choice of wild-type p53 genotype for gene replacement therapy of ovarian cancer.