Advanced Breast Cancer at Diagnosis Linked to 10% Higher CVD Risk

A statistically significant increase of 10% in terms of the odds of having prevalent cardiovascular disease (CVD) was observed among patients (n = 19,292) with locally advanced or metastatic breast cancer at diagnosis (OR, 1.10; 95% CI, 1.03-1.17; P = .007), according to data from a case-control study including patients with stage I or II and stage III to IV disease published in JAMA Network Open.1

Data from the retrospective study, which used propensity score–matched, multivariable-adjusted models in the primary analysis population, also showed that ORs were directionally consistent when locally advanced (OR, 1.09; 95% CI, 1.02-1.17; P = .02) and metastatic (OR, 1.20; 95% CI, 0.94-1.54; P = .15) disease were examined separately among patients with all receptor subtypes. Furthermore, the association between locally advanced or metastatic breast cancer at diagnosis and prevalent CVD was observed in the hormone receptor (HR)–positive population (OR, 1.11; 95% CI, 1.03-1.19; P = .006), but not in those with HR-negative disease (OR, 1.02; 95% CI, 0.86-1.21; P = .83).

“We would theorize that if you have a more aggressive cancer, [such as] HR-negative and HER2-positive cancers, they tend to grow more quickly [and] there’s less of an opportunity for CVD—which is a component of the aggressiveness of the cancer, but certainly not the main driver—to have an impact,” Kevin T. Nead, MD, MPhil, said in an interview with OncLive®. “For a slower growing cancer, it might [sit] for a bit before it comes to awareness or is picked up on a mammogram; you may have a longer period of time for the CVD to create that immunosuppressive state and for that cancer to develop a bit more quickly because it’s not being surveilled by the immune system in the same way.”

In the interview, Nead highlighted additional findings from the study and their potential implications. Nead is an assistant professor in the Department of Epidemiology and Department of Breast Radiation Oncology at The University of Texas MD Anderson Cancer Center in Houston.

OncLive: What emerging evidence suggests that CVD may play a direct causal role in the etiology and progression of cancer?

Nead: There are two different types of evidence. One is [evidence from] population-based studies looking at patients and cancer registry data or electronic health record data showing that individuals who have CVD are more likely to be diagnosed with cancer. Historically, we have attributed that largely to shared risk factors [with] the most obvious being smoking. But some of these recent population-based studies have shown that even when you account or adjust for all these shared risk factors, you still see that increased risk of cancer among individuals who have CVD.

Two, because these data are there, people have been trying to look at more mechanistic models such as mouse models. If you take a mouse model of cancer, [such as] a breast cancer mouse model, and induce a cardiovascular event in those mice, you have more cancer growth and more cancer spread after that cardiovascular event. It appears to be secondary to a cardiovascular event induced suppression of the immune system, which leaves the immune system not as adept at surveilling the body and doing things [such as] getting rid of those abnormal cells or keeping them in check from spreading to other places.

What was the rationale for conducting this study?

We set up this study to test a hypothesis that was proposed and examined in a paper that came out a couple of years ago [that] took a mouse model of breast cancer and induced a cardiovascular event. They showed that the mice who had that cardiovascular event induced had larger primary breast cancers and were more likely to [experience a] spread of that breast cancer.

[Therefore], we tried to set up a dataset in patients that essentially tested that same hypothesis. We took individuals who had been undergoing breast cancer screening, so they weren’t people who had never been screened for breast cancer and were showing up with something advanced, [and] identified people who did and did not have CVD. Then we looked at whether the individuals with CVD were presenting at initial breast cancer diagnosis with more advanced cancers. [We were] trying to test that same hypothesis.

What were the findings of the study and how are they significant?

When individuals had that diagnosis [of cancer] and got staged, [we] looked at the size of the tumor, spread to lymph nodes, [and] spread to areas outside the lymph nodes. The individuals who had CVD prior to their diagnosis of breast cancer had more advanced cancer at diagnosis.

There’s one finding from the study that I would define as hypothesis generating. We looked at different subtypes of breast cancer and most of this effect [with breast cancer and CVD] appears to be driven by the more indolent types of breast cancer, [such as] the HR-positive, HER2-negative [cancers], the ones that grow more slowly. One thing that we are trying to investigate further is whether this effect of CVD on cancer may be stronger in the patients who have cancers that tend to stick around for a bit longer before they present with them, so there’s more of an opportunity for CVD to have an effect on the cancer itself.

What are the next steps for this research?

One thing I’m always cautious about is that this is a retrospective analysis—it doesn’t prove causality—and it’s in one cancer. We would like to look at this in some additional cancers, looking at cancers that tend to be more aggressive and cancers that tend to be less aggressive, [to] see if we see this difference between more aggressively growing cancers and less aggressively growing cancers. If we see this same pattern in additional cancers, it’s going to give us more confidence that what we’re seeing is the truth, and it’s not some bias or confounding that we couldn’t completely account for. Once we establish that, then we can think about next steps.

Should personalized screening approaches be advised for patients with CVD?

I don’t believe that this study alone can allow us to make that recommendation. However, if you look at screening recommendations for breast cancer right now—[there’s] the US Preventative Task Force and the American Cancer Society, for example—it’s not all the same. There is already variation and uncertainty in what age we should be screening [in and] how frequently we should be screening. Given that uncertainty, it’s reasonable to take into account some of these factors, even if they’re not 100% proven. For a patient who you may be considering screening more frequently or [at] a younger age, if they have CVD, it’s reasonable to take that into account when you’re trying to decide which guidelines to follow as far as age to start and how frequently we’re going to do mammograms.

What’s most important for doctors to know regarding the shared risk factors?

The thing that is important to state in all of this, and which can get lost in cancer research and when we talk about cancer, is that prevention [via] taking steps in [terms of] lifestyle factors is the best approach that we can take. It’s much better for a patient to never have cancer than to have cancer and have an amazing drug to treat it. Our goal as physicians and health care providers should be trying to help set patients up so that they don’t get cancer in the first place because that’s clearly the best outcome.

This study reinforces that a heart-healthy lifestyle is benefiting both cardiovascular health and [likely] the risk of being diagnosed with cancer and also potentially how aggressive that cancer is at diagnosis. At the end of the day, it comes down to prevention and helping people lead a lifestyle that’s going to help keep them from having cancer in the first place. A heart-healthy lifestyle can go a long way. Both of the most common causes of death in the US are CVD and cancer.

Reference

1. Angelov I, Haas AM, Brock E, et al. Cardiovascular disease and breast cancer stage at diagnosis. JAMA Netw Open. 2025;8(1):e2452890. doi:10.1001/jamanetworkopen.2024.52890