A dose-response relationship for the incidence of radiation-related heart disease


Session type:

Carolyn Taylor1, Sarah Darby1, Marianne Ewertz2, Paul McGale1, Anna Bennet3, Ulla Blom-Goldman4, Dorthe Bronnum5, Candace Correa6, David Cutter1, Giovanna Gagliardi4, Bruna Gigante3, Maj-Britt Jensen7, Andrew Nisbet8, Kazem Rahimi1, Per Hall3
1Oxford University, Oxford, UK, 2Odense University Hospital, Odense, Denmark, 3Karolinska Institutet, Stockholm, Sweden, 4Karolinska University Hospital, Stockholm, Sweden, 5Aalborg Hospital, Aarlborg, Denmark, 6H Lee Moffitt Cancer Centre, Florida, USA, 7Danish Breast Cancer Cooperative Group, Copenhagan, Denmark, 8Royal Surrey County Hospital and Surrey University, Surrey, UK


Randomised trials have shown that incidental radiation exposure of the heart during breast cancer radiotherapy can cause heart disease, but the magnitude of the risk is uncertain.


Population-based registries in Denmark and Sweden have been used to relate mean heart radiation dose to risk of developing a major coronary event (myocardial infarction, coronary revascularisation or death from ischaemic heart disease). 963 cases of heart disease and 1205 controls (matched for age, year of diagnosis and country) were identified in irradiated breast cancer patients. For both cases and controls, individual patient information on medical history prior to radiotherapy, and radiotherapy charts were obtained from hospital records. Each woman's radiotherapy was categorised according to regimen. Regimens were reconstructed using virtual simulation and computed tomography treatment planning. Mean heart dose was estimated using dose-volume histograms. For cases, the hospital cardiology or autopsy record was reviewed to confirm the event.


The estimated average mean heart dose was 5.4 Gray for cases and 4.5 Gray for controls. The risk of developing heart disease increased linearly with mean heart dose. On average, there was a 6% increase in heart disease risk per 1 Gray increase in mean heart dose (95% confidence interval 2-13%; 2p=0.0001). The risk per Gray was similar across all patients and tumour characteristics. Increases were apparent within 5 years of irradiation and continued beyond 20 years.


For women with no cardiac risk factors a mean heart dose of 1 Gray from breast cancer radiotherapy is likely to increase her absolute risk of an acute coronary event by around 1%, while for a mean heart dose of 2 Gray, the increase is likely to be around 2%. Absolute risk increases may be larger for women receiving higher heart doses, or who are already at increased cardiac risk prior to radiotherapy.