In the treatment of breast cancer, a wealth of data from prospective clinical trials and meta-analyses has documented the benefits of radiation to prevent local-regional recurrence and improve survival. Accordingly, important quality indicators in breast cancer care include: (1) receipt of radiation to the breast after breast-conserving therapy for women under the age of 70 with invasive breast cancer, and (2) receipt of radiation to the chest wall and nodal basins after mastectomy for women with stage III breast cancer.
Dramatic Changes in Radiation Oncology
Despite the well-documented benefits of radiation, the heart is often an innocent bystander in the path of the radiation beam, resulting in incidental but potentially clinically relevant cardiac irradiation. It has long been known that incidental radiation to the heart has the potential to produce a host of cardiac complications, including ischemic heart disease, heart failure, pericarditis, valvular disease, conduction disease, and cardiac death. The work presented by Darby et al1 could easily be interpreted as providing yet another datapoint in this litany of radiation therapy perils, stoking fears among patients and provoking a defensive posture from radiation oncologists. (See pages 49 and 50 in this issue for a summary of this work and additional expert perspective from the oncology community).
Yet, radiation oncology has undoubtedly changed dramatically since the time that patients in the study by Darby et al were treated. In the past decade, the defining advance has been the transition from two-dimensional treatment planning to three-dimensional treatment planning. With two-dimensional treatment planning, radiation fields were designed using simple two-dimensional radiographs, which allowed for only a simplistic understanding of the anatomic structures in the path of the radiation beam.
For example, with two-dimensional radiographs, it could be determined whether a portion of the cardiac silhouette was included in the radiation beam, but the precise proportion of heart irradiated and the dose received by this portion of the heart could not be calculated with certainty. In contrast, with three-dimensional treatment planning, which is now standard in most practices, radiation fields are designed using computed tomography (CT) images.
As a result, the complex relationship between dose delivered and volume irradiated for any critical structure can be measured and potentially modified. The mean radiation dose delivered to the heart and the percent of the heart receiving a certain dose can be readily determined. Similar metrics can also be determined for specific cardiac structures, such as the left-anterior descending coronary artery, pericardium, and valves.
Conflicting Data
Despite the impressive descriptive cardiac dose-volume data now readily available, the clinical significance of these metrics has been unclear to date. Further, prior observational epidemiologic data regarding the significance of incidental heart radiation is conflicting.
For example, Giordano et al2 concluded that breast radiation delivered in the 1970s yielded an increased risk of cardiac death, but that breast radiation delivered in the 1980s and 1990s did not. Since relatively simple two-dimensional radiation techniques were still the norm in the 1980s and 1990s, the implication of these findings was that cardiac dose-volume metrics acquired from three-dimensional planning were unlikely to be relevant, since risks were already low with simple two-dimensional treatment planning. However, a study from Darby et al published in 2005,3 using the same population-based data source, reached a different conclusion, finding that left-sided breast radiation was associated with a late increase in cardiac death attributable to radiation. The implication of this study was that cardiac dose must be minimized.
It is in the setting of these two conflicting narratives on the relationship between breast radiation and cardiac risk that the current study from Darby et al provides novel, clinically relevant, and important data to assist radiation oncologists and their patients.1 Specifically, the finding that risk of cardiac events increased linearly with mean cardiac dose, without a threshold, strongly supports the assertion that incidental cardiac radiation during breast cancer treatment must be minimized.
Minimizing Cardiac Irradiation
Accompanying dissemination of three-dimensional treatment planning for breast cancer, several techniques have emerged to minimize incidental cardiac radiation. For example, a cardiac block can be used to block out the portion of the radiation beam that would have intersected with the heart. This approach works well for upper-quadrant tumors but runs the risk of creating a geographic miss in patients with lower-quadrant tumors.
Another technique is intensity-modulated radiation therapy, which uses sophisticated treatment planning algorithms and multiple radiation beams to bend the radiation dose around the heart. While this approach can be particularly helpful in minimizing high doses of radiation delivered to the heart, the amount of low-dose radiation given to the thorax, including the heart, lung, and contralateral breast, is often increased, with uncertain clinical ramifications.
A third approach is the use of deep inspiration breath hold, in which delivery of radiation is synchronized with the patient’s breathing cycle. During deep inspiration, the lingula inflates, displacing the heart inferiorly and posteriorly, outside of the path of the radiation beam. At The University of Texas MD Anderson Cancer Center, we routinely use deep inspiration breath hold to minimize cardiac exposure in patients receiving radiation for left-sided breast cancer.
Deep Inspiration Breath Hold
Figs. 1 and 2 depict examples of patients recently treated at our institution in the breast-conserving and postmastectomy settings, respectively. Panels 1A and 2A demonstrate cardiac exposure that would have resulted from treatment in free breathing. It can be observed that a portion of the heart is irradiated in each of these treatment plans. Nevertheless, the mean heart dose is relatively low, at 1.8 Gy for Fig. 1A and 2.8 Gy for Fig. 2A. In comparison, the mean heart dose was 6.6 Gy for women with left-sided breast cancers included in the Darby et al study.1
The significantly lower mean heart doses achievable in our practice even without advanced technologies such as deep inspiration breath hold illustrates that with careful three-dimensional radiation treatment planning alone, mean heart dose, and accordingly cardiac risk, may be considerably lower than that reported by Darby et al. This fact provides reassurance to patients and their physicians that the cardiac risks reported by Darby et al do not necessarily have to be recapitulated in currently treated patients provided that care is taken in designing radiation treatment plans.
Figs. 1B and 2B on page 52 illustrate the added benefit derived from implementing deep inspiration breath hold. As can be readily observed, displacement of the heart with deep inspiration breath hold helps to minimize incidental cardiac irradiation. Accordingly, mean cardiac doses with deep inspiration breath hold were 0.8 Gy in Fig. 1B and 1.3 Gy in Fig. 2B.
It is important to note that there are no data from prospective randomized trials indicating a clinically measurable benefit from implementation of deep inspiration breath hold in clinical practice. However, if the model presented by Darby et al is accurate, one could infer that deep inspiration breath hold served to reduce the cardiac event risk by 7% for patient 1 and 11% for patient 2. When considering that over 100,000 women receive radiation for breast cancer yearly in the United States, this small incremental individual benefit from deep inspiration breath hold would be likely to exert a significant population-wide benefit if uniformly adopted.
Barriers to Implementation
Despite the benefits of deep inspiration breath hold, it is not necessarily commonly used in routine community practice across the United States. There are several reasons for this.
First, implementation of deep inspiration breath hold typically requires a capital investment to acquire the necessary technology to obtain a gated simulation CT scan and subsequently to synchronize delivery of radiation to the respiratory cycle. Second, implementation of deep inspiration breath hold requires increased expertise on the part of radiation therapists and physicists, which may not be readily available.
Third, deep inspiration breath hold increases daily treatment time per patient by approximately 33%, and thus may not be feasible in centers already operating at capacity. Fourth, deep inspiration breath hold is not currently recognized by any Common Procedural Terminology (CPT) code, and thus, the additional staff effort and machine time may not be economically feasible for many practices.
In Summary
The findings from Darby et al go beyond simply adding to the literature regarding late cardiovascular effects of radiation. The study provides the scientific rationale to support ongoing efforts to develop and disseminate technologies that will reduce incidental cardiac irradiation. These efforts should ultimately improve the therapeutic ratio of radiation treatment, thus easing our patients’ fears and illustrating the gains that can be realized through implementation of advanced radiation oncology technologies. ■
Dr. Smith is Assistant Professor of Radiation Oncology, The University of Texas MD
Anderson Cancer Center, Houston.
Disclosure: Dr. Smith receives research funding from Varian Medical Systems and is the recipient of an ASCO Career Development Award.
References
1. Darby SC, Ewertz M, McGale P, et al: Risk of ischemic heart disease in women after radiotherapy for breast cancer. N Engl J Med 368:987-998, 2013.
2. Giordano SH, Kuo YF, Freeman JL, et al: Risk of cardiac death after adjuvant radiotherapy for breast cancer. J Natl Cancer Inst 97:419-424, 2005.
3. Darby SC, McGale P, Taylor CW, et al: Long-term mortality from heart disease and lung cancer after radiotherapy for early breast cancer: Prospective cohort study of about 300,000 women in US SEER cancer registries. Lancet Oncol 6:557-565, 2005.
