In a recent article, Frolov et al. (2002) stated that only five somatic mutations in BRCA1 have been identified in sporadic ovarian carcinomas (Hosking et al., 1995;Merajver et al., 1995) and that only one case of clear cell carcinoma had been reported to be associated with BRCA1 mutations (Hartley et al., 2000). Previously, we reported another somatic BRCA1 mutation, 2723insGTCA, in a serous papillary ovarian cystadenocarcinoma, Grade 2, Stage III (Tworek et al., 1999). The mutation causes a frameshift and generates a premature stop at codon 903. Moreover, in our blinded, independent review of histopathology of ovarian cancers from the Gilda Radner Familial Ovarian Cancer Registry, we found three families with cases of clear cell ovarian carcinoma with germline BRCA1 mutations (Werness et al., 2000). The mutations were 2081insA, which causes a frameshift and generates a premature stop at codon 672; 5083del16, which causes a frameshift and generates a premature stop at codon 1661; and 2575delC, which causes a frameshift and generates a premature stop at codon 846. Interestingly, each individual with a clear cell tumor had a first-degree female relative with the family BRCA1 mutation who developed a serous ovarian carcinoma. Thus, either the clear cell or serous histopathologic subtype of ovarian tumors developed in carriers of the same BRCA1 mutation in each family. From this it can be inferred that within these families, BRCA1 mutation status was not the sole determinant of histopathologic subtypes and that acquired genetic or epigenetic differences were needed.