In a recent case-control study of epithelial ovarian cancer, Tong et al. 1 evaluated the possible association between cancer risk and the PROGINS allele of the progesterone receptor gene. This allele, a 306 bp Alu insertion in intron G of the gene, has been linked to ovarian cancer in several studies, 2-5 though others have not found increased risk. 6,7 Tong et al. 1 concluded that their data showed no difference in the distribution of the PROGINS allele in ovarian cancer patients and healthy control women. However, the data actually indicate that patients are significantly more likely than controls to carry 2 copies of the PROGINS allele (odds ratio Ο 8.0, p Ο½ 0.03). Such an excess prevalence of homozygous carriers among patients is consistent with a recessive role for the PROGINS allele in ovarian cancer. Thus, in contrast to the authors' conclusion, their data support a recessive role for this allele in ovarian cancer susceptibility.
The PROGINS variant has been speculated to encode an aberrantly spliced form of the progesterone receptor with an altered progesterone-binding domain. 8 If this is correct, the ovarian epithelial cells of PROGINS carriers would have decreased progesterone stimulation. There is mounting evidence that such progesterone stimulation may prevent malignant transformation of ovarian cells. 9,10 It is thus plausible that homozygous PROGINS carriers may have increased risk of the malignancy. Further large studies are needed to resolve this issue.