Abstract LB-446: Genital talc use and ovarian cancer: Influence of histologic type and menopausal status on strength and dose response of the association

Introduction: An association between talc powder use in the genital area and ovarian cancer has been repeatedly observed but repeatedly challenged because of a “weak” overall association and apparent lack of dose-response. To determine whether these aspects of the association have been affected by the failure to consider, in a uniform fashion, histologic type of ovarian cancer, menopausal status, or family history, we examined data from our own case-control studies including a new series of 845 cases and 857 controls. Methods: We combined data from 3 phases of our case-control study of ovarian cancer that includes data on 2074 ovarian cancer cases and 2101 controls from eastern Massachusetts and New Hampshire recruited between 1992 and 2008. We used logistic regression to study the association between ovarian cancer and regular use of genital talc as well as total applications (estimated from frequency and duration of use), adjusted for study phase and center, age, parity, oral contraceptive use, tubal ligation, BMI, Jewish ethnicity, and family history of breast or ovarian cancer. Separate analyses were done for all ovarian cancer cases, non-mucinous invasive cases, and serous invasive cases. Results: The association of ovarian cancer with genital talc use [OR (and 95% CL)] was 1.30 (1.14, 1.50) for all cases, 1.37 (1.18, 1.59) for non-mucinous invasive cases, and 1.42 (1.17, 1.73) for invasive serous cases. A dose-response based on control quartiles for total applications was also present, p trend = 0.001, 0.0004, 0.001 for all, non-mucinous invasive, and invasive serous cases, respectively. However, the dose-response differed strikingly for pre- and postmenopausal women, p-heterogeneity = 0.07, 0.009, 0.002 for all, non-mucinous invasive, and invasive serous cases, respectively. The dose-response was more apparent for premenopausal women, and relatively flat for postmenopausal women. Further analyses showed that exclusion of cases with a possible familial disease further improved the dose-response, especially in premenopausal women with invasive serous cancer in whom the odds ratios (and 95% limits) were 2.22 (1.23, 4.02) for about 2000 to 8400 applications and 3.23 (1.52, 6.89) for women with greater than 8400 applications, compared to women with no use. Conclusion: The failure to take into consideration histologic type of ovarian cancer in a systematic fashion has led to both an under estimation of the overall effect in all women and the dose-response in premenopausal women, especially those without a family history. Repeating analyses in existing data sets may help clarify the still-debated association between talc and ovarian cancer. Citation Format: {Authors}. {Abstract title} [abstract]. In: Proceedings of the 102nd Annual Meeting of the American Association for Cancer Research; 2011 Apr 2-6; Orlando, FL. Philadelphia (PA): AACR; Cancer Res 2011;71(8 Suppl):Abstract nr LB-446. doi:10.1158/1538-7445.AM2011-LB-446