Oral contraceptive use for the primary prevention of ovarian cancer.

OBJECTIVE: To estimate the overall balance of harms and benefits from the potential use of oral contraceptives (OCs) for the primary prevention of ovarian cancer DATA SOURCES: We searched PubMed®, Embase®, the Cochrane Database of Systematic Reviews, and ClinicalTrials.gov for English-language studies published from January 1990 to June 2012 that evaluated the potential benefits (reduction in ovarian, colorectal, and endometrial cancers) and harms (increase in breast and cervical cancer, and vascular complications) of OC use. REVIEW METHODS: Two investigators screened each abstract and full-text article for inclusion; the investigators abstracted data, and they performed quality ratings, applicability ratings, and evidence grading. Random-effects models were used to compute summary estimates of effects. A simulation model was used to estimate the effects of OC use on the overall balance of benefits and harms. RESULTS: We reviewed 55 studies relevant to ovarian cancer outcomes, 66 relevant to other cancers, and 50 relevant to vascular events. Ovarian cancer incidence was significantly reduced in OC users (OR [odds ratio], 0.73; 95% CI [confidence interval], 0.66 to 0.81), with greater reductions seen with longer duration of use. Breast cancer incidence was slightly but significantly increased in OC users (OR, 1.08; 95% CI, 1.00 to 1.17), with a significant reduction in risk as time since last use increased. The risk of cervical cancer was significantly increased in women with persistent human papillomavirus infection who used OCs, but heterogeneity prevented a formal meta-analysis. Incidences of both colorectal cancer (OR, 0.86; 95% CI, 0.79 to 0.95) and endometrial cancer (OR, 0.57; 95% CI, 0.43 to 0.76) were significantly reduced by OC use. The risk of vascular events was increased in current OC users compared with nonusers, although the increase in myocardial infarction was not statistically significant. The overall strength of evidence for ovarian cancer prevention was moderate to low, primarily because of the lack of randomized trials and inconsistent reporting of important characteristics of use, such as duration. The simulation model predicted that the combined increase in risk of breast and cervical cancers and vascular events was likely to be equivalent to or greater than the decreased risk in ovarian cancer, although the harm/benefit ratio was much more favorable when protection against endometrial and colorectal cancers was added, resulting in net gains in life expectancy of approximately 1 month. CONCLUSIONS: There is insufficient evidence to recommend for or against the use of OCs solely for the primary prevention of ovarian cancer. Although the net effects of the current patterns of OC use likely result in increased life expectancy when other noncontraceptive benefits are included, the harm/benefit ratio for ovarian cancer prevention alone is uncertain, particularly when the potential quality-of-life impact of breast cancer and vascular events are considered.

Duke Scholars

Author Gillian Denise Sanders Schmidler Population Health Sciences

Author Laura Jean Havrilesky Obstetrics and Gynecology, Gynecologic Oncology

Author Jennifer M. Gierisch Population Health Sciences

Author Patricia Gripka Moorman Family Medicine and Community Health, Prevention Research

Author Michaela Ann Dinan Population Health Sciences

Author Victor Hasselblad Biostatistics & Bioinformatics, Division of Biostatistics

Author Evan Robert Myers Obstetrics and Gynecology, Women's Community & Population He ...