Eugene Chung, MD

In 2013, a publication in the New England Journal of Medicine renewed concerns over the exposure of the heart to radiation during treatment of breast cancer. While the benefits of radiation therapy outweigh the toxicity risks in the majority of patients, the data show a clear relationship between radiation dose and cardiac events, and therefore should be an essential part of the conversation with patients on cardiac risk after radiation therapy.

Radiation oncologist Eugene Chung, MD, University of Southern California Norris Cancer Center, discusses the current data on cardiac toxicity following radiation therapy for breast cancer, as well as how newer technologies, like brachytherapy, allow physicians to minimize cardiac dose.

 What are your concerns regarding cardiac dose with the delivery of whole breast radiation?

The data regarding cardiac toxicity following left-side breast radiation has been known for some time.  There have been several SEER and single institution analyses showing worse survival and increased number of cardiac events for left- versus right-sided disease.  Furthermore, the 2005 Oxford Meta-analysis showed excess cardiac deaths following radiation therapy, and the overall survival benefit of radiation therapy decreased over time due to the excess cardiac events. We therefore know that the long term cardiac risk associated with radiation therapy is real.  The issue with heart injury after radiation is thought to be a combination of both macrovascular and microvascular damage, leading to increased frequency and magnitude of events, respectively.  The increased macrovascular injury has been demonstrated by coronary angiograms showing increased stenosis in the regions of the LAD and RCA receiving high doses.  Microvascular injury has been shown through cardiac perfusion SPECT-CT scans, showing decreased perfusion following whole breast radiation within the radiation fields.

Some argue the study by Darby et al. in the New England Journal of Medicine in March 2013 is not relevant, as modern radiation delivery techniques reduced the risk of cardiac exposure compared to the time period of the study. What are your thoughts on this research and how it applies to current practice?

The Darby New England Journal publication from 2013 has probably been the biggest headline regarding increased cardiac injury.  Although there were several methodological issues with this study, especially with the estimates of low dose, they were able to show a clear dose-response relationship, with a 7.4% relative increase in major coronary events for every Gy increase in mean heart dose.  Interestingly, they found cardiac events happening within 5 years after radiation therapy, which is certainly earlier than what has been shown by other studies.   Despite some disagreement whether a threshold dose exists, the Darby publication provides convincing evidence that cardiac dose matters, and that there is potential for injury even at relatively low doses.

While there may be some disagreement of the dose response relationship at low doses and their conclusion that there is no threshold for cardiac events, I still believe this publication is highly relevant to our practice today.  The relationship between cardiac events and dose, especially at higher mean heart doses, is convincing.  This study helps drive the conversation on cardiac risk after radiation therapy, and many of the patients I see are fully aware of this publication. The study informs us of the potential risks associated with radiation, and helps us be cognizant of the heart location when we plan our treatment.

What parameters do you consider when measuring dose to the heart? What levels to do you consider to be acceptable?

I usually begin by looking at the mean heart dose, but also review the V5, V10, and the V20.   Although the RTOG 1005 constraints are a starting point for heart constraints, we try to achieve lower doses.  We contour the heart for all patients because there is still the possibility for radiation dose contribution to cardiac substructures even when treating right-sided breast cancer.

In general, we try to achieve a mean heart dose less than 1 Gy for right sided disease and less than 2 Gy for left sided disease.  For left-sided cancers I like to see the V5 < 20% and the V20 < 5%.  Although these are our typical goals, it’s important to be aware of the context in which we provide radiation.  By that I mean for DCIS, where there is no survival benefit with the addition of radiation therapy, we are very strict on our dose constraints, and push for mean doses lower than 2 Gy for left-sided disease.  On the other hand, for node positive disease where I’m treating the regional lymph nodes, including the internal mammary lymph nodes, the constraints might be a little looser.  Going back to the Darby New England Journal article, although there was a dose response relationship, with the 7.4% per Gy relative risk increase in major coronary events, the absolute increase in coronary events was relatively small, especially in patients who have no cardiac risk factors.  This is why we stress the importance of the clinical scenario rather than absolute constraints for any given patient.

There is interest in hypofractionated whole breast irradiation, primarily for the greater convenience for patients. But considering the increased dose per fraction with these accelerated fractionation schemes, are there additional concerns about long-term cardiac toxicity?

In our treatment plans, we do not allow the primary beam to go through the heart.  Therefore, even with hypofractionated whole breast radiation, our mean cardiac doses remain very low. If we did not specifically block out the heart from our treatment plans, I think the increased dose per fraction is something that should be more concerning.  The RTOG 1005 constraints are more stringent for the hypofractionated treatment arm, which I think demonstrates the increased concern with higher doses per fraction.  However, data beyond 10 years is limited to conclusively state that there is a deleterious association between hypofractionated whole breast radiation and long-term cardiac toxicity.

 What are your thoughts on the ability of brachytherapy to reduce cardiac dose?

I definitely think it’s promising with a fast dose fall-off for the properly selected patient. Lettmaier et al. compared the radiation exposure of whole breast radiation to multi-catheter brachytherapy and concluded that brachytherapy yields lower heart dose than WBI – in fact, it reduced the V10 by a factor of 4 and the V5 by a factor of 2. Especially with a multi-lumen catheter like SAVI, the potential to shape dose away from the heart allows for more tailored treatment plans.  As with any treatment technique, poor anatomy always presents a challenge, but the greater flexibility of brachytherapy may enable physicians to overcome these challenges while keeping dose to the heart as low as possible.

Dr. Chung is an Assistant Professor of Clinical Radiation Oncology at USC Norris Cancer Hospital and USC + Los Angeles County Medical Center. His past research includes studying dose response relationship for scan changes in breast cancer patients receiving radiation therapy.