Association of levonorgestrel-releasing intrauterine device with gynecologic and breast cancers: a national cohort study in Sweden.

BACKGROUND

The levonorgestrel-releasing intrauterine device (LNG-IUD) is widely used for the treatment of menorrhagia, dysmenorrhea, and for contraception. However, the association between the use of LNG-IUD and the risk of site-specific gynecologic and breast cancers remains inconclusive.

OBJECTIVE

We aim to address this knowledge gap by investigating whether the use of LNG-IUD is associated with a significant risk of site-specific gynecologic and breast cancers. This will be achieved by accessing the nationwide Swedish Registers, with consideration given to the influence and potential interaction of family history of cancer.

STUDY

DESIGN

A total of 514,719 women aged 18 to 50 years who have used LNG-IUD between July 2005 and December 2018 were identified from the Swedish Prescribed Drug Register and randomly matched with 1,544,157 comparisons who did not use LNG-IUD at a ratio of 1:3. The propensity score was calculated and matched among women who used LNG-IUD and the matched comparisons. The follow-up period started from the date of the first prescription of LNG-IUD for users as well as for their matched comparisons and ended at the date of diagnosis of gynecologic and breast cancers, date of death from any cause, and the end of the study period, whichever came first. The Cox proportional hazard model with a competing risk analysis was used to calculate hazard ratios (HRs) and 95% confidence intervals (CIs). Additive interaction was calculated as the relative excess risk for interaction, while multiplicative interaction was calculated by including a product term in the regression model.

RESULTS

The use of LNG-IUD was associated with a 13% higher risk of breast cancer (adjusted HR, 1.13; 95% CI, 1.10-1.17), a 33% lower risk of endometrial cancer (adjusted HR, 0.67; 95% CI, 0.56-0.80), a 14% lower risk of ovarian cancer (adjusted HR, 0.86; 95% CI, 0.75-0.99), and a 9% reduced risk of cervical cancer (adjusted HR, 0.91; 95% CI, 0.84-0.99) compared to women who did not use LNG-IUD. A significant additive interaction between LNG-IUD use and family history of cancer was observed in breast cancer, indicating a relative 19% excess risk for interaction (P<.002), and 1.63 additional cases per 10,000 person-years.

CONCLUSION

The risk of gynecologic and breast cancers exhibits a site-specific effect among LNG-IUD users. It is important to note that the observed effect is small for breast cancer and the results are limited by the observational study design. Clinical recommendations regarding the use of LNG-IUD should carefully weigh its potential benefits and risks. Close monitoring is advisable for the potential development of breast cancer, particularly among women with a family history of breast cancer.