TOPLINE:

In a nationwide Swedish study of over 2 million girls and women followed up for more than 21 million person-years, the use of hormonal contraceptives was associated with a 24% increased risk for breast cancer, with both combined and progestin-only formulations associated with increased risk. The use of desogestrel-containing contraceptives was linked to a higher risk than the use of levonorgestrel-containing products, although the use of some formulations was not significantly associated with increased risk.

METHODOLOGY:

• Taking hormonal contraceptives is a recognized risk factor for breast cancer; although the absolute risk in any individual is modest, the population-level effect is substantial because of their widespread use. Estrogen’s role in promoting breast cancer — by stimulating epithelial proliferation and oncogene amplification — is well established. In contrast, the effects of progesterone and synthetic progestins, which are the primary components of many contraceptives, remain unclear.
• Researchers conducted a nationwide population-based cohort study involving 2,095,130 adolescent girls and women aged 13-49 years from 2006 to 2019. The cohort accrued 16,385 incident cases of breast cancer (median age at diagnosis, 45 years) and had a mean follow-up duration of 10.03 years.
• Exposure to contraceptives was ascertained from the Prescribed Drug Register, with contraceptives categorized as combined or progestin-only products and by hormonal formulation, including specific progestin type and route of administration.
• Researchers analyzed the association of ever use and duration of use of hormonal contraceptives with the risk for breast cancer.
• Overall, 1,284,613 individuals used hormonal contraceptives at any time during follow-up, corresponding to 12,356,854 person-years of follow-up and 8485 cases of breast cancer. Those who never used hormonal contraceptives corresponded to 8,663,992 person-years of follow-up and 7900 cancer cases.

TAKEAWAY:

• Ever use of any hormonal contraceptive was associated with an increased risk for breast cancer (hazard ratio [HR], 1.24), corresponding to 13 additional cases per 100,000 person-years and one additional case per 7752 users per year. The use of progestin-only formulations (HR, 1.21) was associated with a relatively greater risk than the use of combined formulations (HR, 1.12).
• When stratified by progestin type and route of administration, the risk for breast cancer was higher with oral desogestrel-only formulations (HR, 1.18), oral desogestrel-containing combined formulations (HR, 1.19), and implants containing etonogestrel (HR, 1.22) than with levonorgestrel-containing combined pills (HR, 1.09) and the 52-mg levonorgestrel intrauterine system (HR, 1.13). The use of medroxyprogesterone acetate injection, etonogestrel vaginal ring, or oral drospirenone-containing combined formulations was not associated with an increased risk.
• Longer duration of use was associated with a progressively increased risk; the HR was 1.21 for exposure of 1 to less than 5 years and 1.34 for 5-10 years. Among current users and current plus recent users, HRs were 1.17 and 1.21 for 1 to less than 5 years and 1.37 and 1.36 for 5-10 years, respectively. Among progestin-containing formulations, only oral desogestrel-only ones were associated with an increased risk after use for less than 1 year (HR, 1.08). When considering use for 5-10 years, all desogestrel‑containing formulations were associated with elevated risks: progestin-only pills (HR, 1.49), combined pills (HR, 1.48), and etonogestrel-containing implants (HR, 1.45). By contrast, the use of levonorgestrel-containing combined pills (HR, 1.20) and the 52-mg levonorgestrel intrauterine system (HR, 1.21) was associated with smaller increases in risk.
• Estrogen appeared to attenuate progestin‑associated risk: Each additional milligram of unopposed oral desogestrel was linked to a greater increase in the risk for breast cancer than the same milligram when combined with estrogen; unopposed lynestrenol and norethisterone showed modest per-milligram increases in risk, and ever use of desogestrel-containing combined formulations with 20 µg ethinylestradiol was associated with a higher risk than with ≥ 30 µg ethinylestradiol, with dose-response analyses mirroring this pattern.

IN PRACTICE:

“While the findings are robust, further research using causal inference methods and triangulation with other study designs is needed before clinical recommendations can be made,” the authors of the study wrote. “Although the relative risks are statistically significant, the absolute risk increase remains small and should be considered in the broader context of the well-established benefits of hormonal contraceptives.”

This study “confirms that the use of hormonal contraceptives is associated with a slight increase in the relative risk of breast cancer, but the magnitude of the absolute risk is very small — around one additional case per 7000-8000 users per year,” said Ramón Salazar, MD, PhD, in a statement from the Science Media Centre. “The study is methodologically sound, with long-term follow-up and detailed analysis by type of formulation.”

“The relevance of this study is that it helps to refine the personalized choice of method by identifying formulations with a slightly more favorable risk profile,” and “the decision should be shared between the patient and the healthcare professional, always taking into account age, family history, and individual preferences,” continued Salazar, who is head of medical oncology at the Catalan Institute of Oncology and associate professor of medicine at the University of Barcelona, Barcelona, Spain.

SOURCE:

The study, led by Fatemeh Hadizadeh, MD, PhD, Genetics and Pathology, SciLifeLab, Uppsala University, Uppsala, Sweden, was published online in JAMA Oncology.

LIMITATIONS:

Data on hormonal contraceptive use before mid-2005 were unavailable from the drug register, potentially attenuating associations through misclassification of previous users as nonusers. Register data reflected prescriptions redeemed rather than actual use, possibly biasing estimates toward null. Additionally, information on key confounders such as age at menarche, breastfeeding, and family history was unavailable.

DISCLOSURES:

The study was funded by the Swedish Cancer Society, the Sjöberg Foundation, and the Swedish Research Council. The authors reported having no relevant conflicts of interest.

This article was created using several editorial tools, including AI, as part of the process. Human editors reviewed this content before publication.