Epidemiologic differences between women with borderline ovarian tumors and women with epithelial ovarian cancer

Objective. The aim of this study was to study the relationship between borderline ovarian tumors (BLOT) and epithelial ovarian cancer (EOC) by comparing the epidemiologic features of women with BLOT with those of women with EOC of similar histology. Material and Methods. The epidemiologic features of 32 women with serous and mucinous BLOT were compared with those of 273 women with primary serous or mucinous EOC. We included all women with the documented respective histologic diagnoses admitted to Roswell Park Cancer Institute between 1982. and 1996 who returned a self-administered epidemiologic questionnaire which contained 44 items pertaining to reproductive, contraceptive, medical, social, dietary, occupational, and family histories of cancer. Individual variables between both groups were compared using the Student t test, chi(2) analysis, the Mantel-Haenszel test, and the Wilcoxon nonparametric test. Two-tailed P < 0.05 was considered significant. Results. The response rate to the questionnaire was 63% in the BLOT group acid 60% in the EOC group. There was no significant difference between the two groups in geographic location, race, education, income, smoking, marital status, age at first pregnancy, age at first birth, history of hysterectomy, history of infertility, history of tubal surgery, use of hormone replacement therapy, or history of diaphragm or intrauterine contraceptive device use. There were no significant differences in family history of malignancy between women with BLOT and those with EOC. Women with BLOT were significantly younger than those with EOC (mean age 47 +/- 14.0 versus 56 +/- 13.7, P < 0.01). There was an apparent difference in oral contraceptive pill use between both groups. However, when we adjusted for age by stratification this difference was not significant (P = 0.089). Conclusions. The epidemiologic features of women with BLOT are similar to those of women with EOC with the exception of an earlier age of onset. These findings might be consistent with one etiology for both conditions, (C) 1999 Academic Press.