The recent population-based case-control study by Darby et al provides meaningful insights of the risk of ischemic heart disease in patients with breast cancer receiving radiotherapy, based on dosimetric parameters. Estimation of dose-response relation for ischemic heart disease is an important tool for decreasing doses to the heart as low as reasonably achievable, without jeopardizing coverage of target volumes. Investigators found that the rate of major coronary events followed a nonthreshold linear relation with the mean cardiac dose increasing by 7.4% per Gy (95% confidence interval 2.9 to 14.5, p <0.001). The impact of radiation-induced cardiotoxicity is an important issue and the investigators should be congratulated for incorporating systemic agents or individual risk factors at time of radiotherapy. This is a fundamental step toward a more comprehensive approach in management of treatment-related toxicities.

To date, the impact of cardiotoxic drugs (e.g., taxanes, anthracyclines, or trastuzumab) on radiation-induced cardiac hazard remains unknown. As this is a case-control study, the impact of systemic agents does not confound any of the findings. However, the investigators themselves highlight that few women in this study were treated with anthracyclines and none with taxanes or trastuzumab. These agents are now being widely used as part of adjuvant therapy and could affect radiation-induced cardiotoxicity through additive or supra-additive effects. As a consequence, the relation that 1 Gy increases by 7.4% the rate of major coronary events could be underestimated in regard to today’s practice, and the iatrogenic impact of cardiac irradiation could be still greater.

Furthermore, this study gives full justification to the current developments from outdated irradiation techniques to modern irradiation methods that minimize dose to the heart and allow individual dosimetric optimization. From a technical point of view, irradiation techniques (energy used, type of particles) are important factors impacting on cardiac irradiation. The investigators had reported dosimetric details for these women in a previous report and highlighted that dose per fraction and field arrangement had drastically evolved over decades. Thus, patients treated before 1958 to 2001 benefited from mostly 2-dimensional techniques and low energy particles. In parallel, these technical changes were associated with dosimetric changes: cardiac doses from the cohort increased from the 1950s to the 1970s (mean heart dose for left-sided tumors: 5.1 and 10.5 Gy, respectively) and then reduced significantly from the 1980s (mean heart dose for left-sided tumors in the 1990s: 3.0 Gy). Combined with new irradiation methods, some particular irradiation techniques (e.g., breast radiotherapy in the lateral decubitus position) could help minimizing the dose to the heart, and thus the risk of treatment-related cardiac sequels.

Darby et al found that mean dose of radiation to the heart better predicted the rate of major coronary events than the mean dose to the left anterior descending coronary artery. Several studies suggest that parts of the heart that receive the highest radiation doses are the parts that are damaged. Correa et al have examined medical records of 961 patients with breast cancer treated with conventional tangential beam radiotherapy. They found no impact of lateralization (right vs left) on the 10-year risk of developing coronary artery disease. However, a statistically significant greater prevalence of stress test abnormalities was reported in patients who received left-sided irradiation (59% vs 8%, p = 0.001). Of 13 patients who received left-sided irradiation and underwent cardiac catheterization, coronary stenoses were revealed in 92%, mainly in the left anterior descending artery. Nilsson et al have studied distribution of coronary artery stenoses in patients who have received radiotherapy for breast cancer. The investigators found an increase of stenosis in mid and distal left anterior descending coronary artery and distal diagonal artery in irradiated left-sided breast cancer. There was a correlation between high-risk radiotherapy (left-sided irradiation and internal mammary chain irradiation) and coronary stenoses in dosimetric hotspots. The mean dose in the study by Darby et al was not significantly associated with the rate of major coronary events. Actually, the investigators estimated the mean radiation doses to the heart and the left anterior descending coronary artery using patients’ radiotherapy charts. Then, radiotherapy fields were reconstructed using a computed tomography scan of a woman with typical anatomy. It was the best possible method with the data available. However, the computed tomography–based simulation, even using a reproducible system as quoted by the investigators, contains a nonnegligible risk of error, and patients’ individual anatomy is by itself an important factor affecting cardiac irradiation.

Finally, it is useful to highlight that irradiation-induced cardiac dysfunction is not limited to coronary events. Beyond coronary toxicity, it will be important in the future to better examine the relation between dose distribution to specific cardiac structures and other major clinical end points such as congestive heart dysfunction or pericarditis.

Next step of cardiotoxicity research in patients with breast cancer will be to incorporate and, eventually, prospectively validate molecular and genomic predicative factors of cardiac toxicity.