ABSTRACT
PURPOSE: Breast cancer radiotherapy can increase the risks of heart disease, lung cancer and oesophageal cancer. At present, the best dosimetric predictors of these risks are mean doses to the whole heart, lungs and oesophagus, respectively. We aimed to estimate typical doses to these organs and resulting risks from UK breast cancer radiotherapy. METHODS: A systematic review and meta-analysis was conducted of planned or delivered mean doses to the whole heart, lungs or oesophagus from UK breast cancer radiotherapy in studies published during 2015-2023. Average mean doses were summarised for combinations of laterality and clinical targets. Heart disease and lung cancer mortality risks were then estimated using established models. RESULTS: For whole heart, thirteen studies reported 2893 doses. Average mean doses were higher in left than in right-sided radiotherapy and increased with extent of clinical targets. For left-sided radiotherapy, average mean heart doses were: 2.0 Gy (range 1.2-8.0 Gy) breast/chest wall, 2.7 Gy (range 0.6-5.6 Gy) breast/chest wall with either axilla or supraclavicular nodes and 2.9 Gy (range 1.3-4.7 Gy) breast/chest wall with nodes including internal mammary. For right-sided radiotherapy, average mean heart doses were: 1.0 Gy (range 0.3-1.0 Gy) breast/chest wall and 1.2 Gy (range 1.0-1.4 Gy) breast/chest wall with either axilla or supraclavicular nodes. There were no whole heart dose estimates from right internal mammary radiotherapy. For whole lung, six studies reported 2230 doses. Average mean lung doses increased with extent of targets irradiated: 2.6 Gy (range 1.4-3.0 Gy) breast/chest wall, 3.0 Gy (range 0.9-5.1 Gy) breast/chest wall with either axilla or supraclavicular nodes and 7.1 Gy (range 6.7-10.0 Gy) breast/chest wall with nodes including internal mammary. For whole oesophagus, two studies reported 76 doses. Average mean oesophagus doses increased with extent of targets irradiated: 1.4 Gy (range 1.0-2.0 Gy) breast/chest wall with either axilla or supraclavicular nodes and 5.8 Gy (range 1.9-10.0 Gy) breast/chest wall with nodes including internal mammary. CONCLUSIONS: The typical doses to these organs may be combined with dose-response relationships to estimate radiation risks. Estimated 30-year absolute lung cancer mortality risks from modern UK breast cancer radiotherapy for patients irradiated when aged 50 years were 2-6% for long-term continuing smokers, and <1% for non-smokers. Estimated 30-year mortality risks for heart disease were <1%.
ABSTRACT
AIMS: Radiotherapy for Hodgkin lymphoma leads to the irradiation of organs at risk (OAR), which may confer excess risks of late effects. Comparative dosimetry studies show that proton beam therapy (PBT) may reduce OAR irradiation compared with photon radiotherapy, but PBT is more expensive and treatment capacity is limited. The purpose of this study is to inform the appropriateness of PBT for intermediate-stage Hodgkin lymphoma (ISHL). MATERIALS AND METHODS: A microsimulation model simulating the course of ISHL, background mortality and late effects was used to estimate comparative quality-adjusted life years (QALYs) lived and healthcare costs after consolidative pencil beam scanning PBT or volumetric modulated arc therapy (VMAT), both in deep-inspiration breath-hold. Outcomes were compared for 606 illustrative patients covering a spectrum of clinical presentations, varying by two age strata (20 and 40 years), both sexes, three smoking statuses (never, former and current) and 61 pairs of OAR radiation doses from a comparative planning study. Both undiscounted and discounted outcomes at 3.5% yearly discount were estimated. The maximum excess cost of PBT that might be considered cost-effective by the UK's National Institute for Health and Care Excellence was calculated. RESULTS: OAR doses, smoking status and discount rate had large impacts on QALYs gained with PBT. Current smokers benefited the most, averaging 0.605 undiscounted QALYs (range -0.341 to 2.171) and 0.146 discounted QALYs (range -0.067 to 0.686), whereas never smokers benefited the least, averaging 0.074 undiscounted QALYs (range -0.196 to 0.491) and 0.017 discounted QALYs (range -0.030 to 0.086). For the gain in discounted QALYs to be considered cost-effective, PBT would have to cost at most £4812 more than VMAT for current smokers and £645 more for never smokers. This is below preliminary National Health Service cost estimates of PBT over photon radiotherapy. CONCLUSION: In a UK setting, PBT for ISHL may not be considered cost-effective. However, the degree of unquantifiable uncertainty is substantial.