• 2019-10
  • 2019-11
  • 2020-03
  • 2020-07
  • 2020-08
  • 2021-03
  • n br Fig Cohort selection


    Fig. 1. Cohort selection.
    anti-cancer treatment or concurrent medications (Table 1).
    3.2. Relationship of bisphosphonate use with cardiovascular disease incidence
    bisphosphonate users were free from any CVD events in the 36 months after cancer diagnosis, compared with 76.6% of those who did not use bisphosphonates (Fig. 2). The incidence of angina, atrial fibrillation, AUY922 failure, myocardial infarction and stroke 36 months after breast cancer diagnosis among bisphosphonate users was 6.3%, 4.1%, 4.5%, 1.1% and 3.4%, respectively, compared to an incidence of 11.8%, 6.2%, 6.9%, 3.1% and 6.5%, respectively, among
    Fig. 2. Kaplan-Meier plot for incident cardiovascular disease for bisphosphonate users and matched non-users.
    Baseline characteristics of stage 0e3 breast cancer patients in the SEER-Medicare database (2007e2010) according to bisphosphonate use.
    Income Quartiles N (%)
    Marital Status N (%)
    Comorbidities, N (%)
    Anticancer Treatment, N (%)
    IQR: interquartile range, SD: standard deviation, ACE-i: angiotensin converting enzyme inhibitor, ARB: angiotensin receptor blocker.
    bisphosphonate non-users.
    4. Discussion
    Previous studies have suggested that bisphosphonates may be cardioprotective in older adults and decrease risk for myocardial infarction [15]; however, data have been inconclusive [16,17]. Using population-based data, we found that bisphosphonate use was associated with significantly lower incident CVD events within 36 months of cancer diagnosis among older women with stage 0-III primary breast cancer.
    The overall reduction in risk of incident CVD events with bisphosphonate use in our study is comparable to other observa-tional studies. Among patients with rheumatoid arthritis, the adjusted risk of myocardial infarction was reduced by 28% (HR: 0.72, 95% CI: 0.54e0.96) with bisphosphonate use [9]. Similarly, in a retrospective cohort study of 7981 participants in a risedronate clinical trial, investigators reported a trend toward lower CVD mortality in those taking 2.5 mg of risedronate compared to the placebo group (Relative risk: 0.69, 95% CI: 0.49e0.99). The decrease in CVD mortality observed in the study was mostly due to the effect on the risk of stroke [10]. In another study, those using bisphosphonates after vertebral or hip fracture had a lower likeli-hood of suffering from stroke during a 2-year follow-up period (HR: 0.79, 95% CI: 0.66e0.99) [18]. Sing et al. evaluated associations between alendronate use and CV events among patients with hip fracture. The alendronate group was linked with significantly lower risk of 1-year cardiovascular mortality (HR 0.33; 95% CI, 0.17 to 0.65) and incident myocardial infarction (HR 0.55; 95% CI, 0.34 to
    0.69 to 1.01; p ΒΌ 0.065) [19]. Our data is consistent with the results of these studies and shows a homogenous effect across CVD events included in the composite outcome.
    The mechanism by which bisphosphonates may reduce incident CVD may by through their effect on cholesterol. Bisphosphonates affect cholesterol synthesis through the mevalonate pathway, a well-known pathway that is the target of cholesterol lowering drugs. Statins interfere the same pathway by inhibiting methyl-glutaryl-CoA reductase [20,21]. Nitrogen-containing bisphospho-nates (NCBPs) inhibit farnesyl-pyrophosphate synthesis and
    Table 2
    Comparison of incident CVD events among breast cancer survivors treated with and without bisphosphonates.
    Outcome HR 95% CI
    CVD: cardiovascular disease, CI: confidence interval HR: hazard ratio.  thereby disrupt the mevalonate pathway. Consequently, NCBPs and statins may share similar pharmacological effects. Similar to statins, bisphosphonates have been shown to decrease low-density lipo-protein (LDL) cholesterol levels by up to 5% and can raise high-density lipoprotein (HDL) cholesterol by 10e18% [22,23].
    Experimental data have also shown that bisphosphonates might limit vascular and valvular calcifications [24,25]. Several animal studies have found that etidronate, pamidronate and ibadronate prevent aortic calcification in uremic rats [26,27]. The Multi-Ethnic Study of Atherosclerosis (MESA) study found that NCBPs reduce cardiovascular calcification in older women, but more prevalent vascular calcification was observed in younger participants [28]. A similar effect was found among patients on long-term hemodialysis [29].
    Furthermore, bisphosphonates may play a pivotal role in car-dioprotection by modulating vascular smooth muscle cell (VSMC) proliferation and migration. A physiologic mechanism that may link osteoporosis and CVD involves monocyte-macrophage differ-entiation and function, so that modification of vascular calcification and modulation of serum cytokine levels may have a beneficial effect on both diseases [30e33]. Wu at al. showed that high doses of zoledronate suppress VSMC proliferation after 48 h in rat models [34]. It is well known that proliferation and migration of VSMCs from the media to the intima determine the development of intimal hyperplasia and atherosclerosis. Therefore, an inhibitory effect of bisphosphonates on this pathway could contribute to lowering CVD risk and might be considered as a potential target for future research.