Breast Cancer and Atrial Fibrillation – A Malignant Combination?

Atrial fibrillation (AF) and breast cancer are common conditions and are reported to complicate each other’s course. The incidence and prevalence of AF are increasing¹ with a lifetime risk of more than 1 in 3.² Breast cancer is the most common malignancy amongst American women and is the second leading cause of cancer-related death in women behind lung cancer.³ In 2018, there will be an estimated 266,120 new cases of breast cancer and 40,920 breast cancer deaths.³

Few studies have examined the relationship between AF and breast cancer. One case-referent study demonstrated increased prevalence of AF in patients admitted for breast cancer surgery compared to referents admitted for non-oncologic surgery (2.0% vs 0.6%).⁴ An analysis of two case-control studies reported that diagnosis of AF > 90 days prior to breast cancer diagnosis was associated with reduced incidence of breast cancer. Conversely, prospective cohort analysis in the same study found there was an increased risk of AF diagnosis in the first 3 months after breast cancer diagnosis but not thereafter.⁵ The Women’s Health Initiative, a prospective observational study of 93,676 post-menopausal participants, found that prevalent AF was associated with 19% higher rate of breast cancer (HR 1.19, 95%CI, 1.03 to 1.38).⁶ The Women’s Health Study, a prospective observational study of 34,691 participants 45 years and older, found an HR of 1.32, 95% CI, 0.99 to 1.77, associating new-onset AF with elevated risk of breast cancer.⁷

The reported bi-directional association between breast cancer and AF is not well understood. Advancing age is the most potent risk factor for AF;¹ breast cancer risk also increases with age, with two thirds of cases occurring in women age ≥55 years.⁸ Obesity is another prominent risk factor for both.^{9, 10} Inflammation may be a common process that elevates risk for both conditions.^{11, 12} Treatment of breast cancer may indirectly impact AF risk. Ionizing radiation increases risk of heart failure with preserved ejection fraction¹³ and ischemic heart disease in a dose dependent fashion,¹⁴ both of which are established risk factors for AF.^{15, 16} Chemotherapy may cause heart failure with reduced and preserved ejection fraction, ischemia, and hypertension,¹⁷ all risk factors for AF.^{15, 16, 18} Finally, increased medical care inherent with either diagnosis may contribute to detection biases for the contralateral condition.

In this issue of Heart Rhythm, D’Souza et al.¹⁹ retrospectively investigated four national Danish registries to study the incidence of AF in women with prevalent breast cancer compared to age- and sex-matched referents in a 1:3 ratio. In patients <60 years of age, breast cancer was associated with increased rates of AF during the first 6 months as well as during the 6-month – 3-year follow-up period. Patients >60 years of age had increased rates of AF only during the 6-month – 3-year follow-up period.

This study provides valuable insights from a large nationwide cohort with results generalizable to women of European ancestry. However, a few limitations that should be mentioned. First, follow-up was only 3 years, which may be too short a time for the cardiotoxic effects of breast cancer treatment to fully manifest. Second, unmeasured confounding from body mass index was not accounted for in the analysis. Third, the multivariable Cox regression model did not account for competing risk of death; the 3-year mortality risk was larger than the risk of AF in both groups. Of note, the cumulative risk of AF at 3 years (Figure 2) accounts for the competing risk of death but is not adjusted for covariates; in patients >60 years, it was higher in the background population than those with breast cancer. However, in multivariable-adjusted Cox model analyses, the hazard was higher among breast cancer patients. The reversal in the direction of association may be due to adjusting for covariates and not accounting for competing risk of death.

The clinical significance of the study is uncertain. Ultimately, the cumulative incidence at 3 years was low and similar for women with and without breast cancer (about 0.4% versus 0.2% before 60 years; 2.2% vs. 2.4% after 60 years). Broad-based monitoring for AF in women with breast cancer therefore is not warranted at this time.

The study does however suggest several substantive research questions. First, can the risk of AF in breast cancer patients be explained by shared risk factors? Is it a multiple-hit phenomenon in which a pro-inflammatory state and breast cancer treatments add insult to injury? A study design that accounts for multiple confounders is important in future endeavors. Are there certain chemotherapeutic regimens or cumulative radiation doses for which risk for AF rise? Lastly, are there subsets of women with breast cancer at sufficient risk for AF that more intensive monitoring may be merited? D’Souza et al.’s work is a contribution to the burgeoning field of cardio-oncology and provides support for further research into the potentially bidirectional relationship between cancer and AF.