Long-term risks of secondary cancer for various whole and partial breast irradiation techniques
Introduction: For early stage breast cancer patients, non-breast cancer mortality including secondary cancers and cardiac events can overshadow the benefit of adjuvant radiotherapy. This study evaluates the excess risk of secondary cancer for various breast radiotherapy techniques including accelerated partial breast irradiation (APBI). Methods: Secondary cancers Lifetime Attributable Risks (LAR) were calculated using a modified BEIR-VII formalism to account for the specific survival of breast cancer patients. Those survivals were extracted from the SEER database. Doses scattered to various organs were measured into a Rando phantom with custom-made breast phantoms. Treatments delivered typical doses of brachytherapy APBI (34 Gy in 10 fractions), external beam APBI (38.5 Gy in 10 fractions) using 3D-conformal, Cyberknife stereotactic (CK), or VMAT, as well as whole breast irradiation (WBI) delivering 42.5 Gy in 16 fractions. Results: WBI resulted in the highest total LAR, with 4.3% excess risk of secondary cancer for a patient treated at age 50 years. Lung cancers accounted for 75–97% of secondary malignancies. For a typical early stage patient irradiated at 50, the excess risks of secondary lung cancer were 1.1% for multicatheter HDR, between 2.2% and 2.5% for 3D-CRT or CK, 3.5% for VMAT APBI, and 3.8% for WBI. Conclusions: APBI reduces the risk of secondary cancer 2–4 fold compared to WBI. These techniques are well suited for long-living early stage breast cancer patients. HDR brachytherapy and 3D-conformal APBI achieve mean lung doses between 1 and 1.5 Gy, which could serve as reference.
|, , , ,|
|Radiotherapy & Oncology|
|Organisation||Department of Radiation Oncology|
Hoekstra, N, Fleury, E, Merino Lara, T.R, van der Baan, P. (Peter), Bahnerth, A. (Andy), Struik, G. (Gerson), … Pignol, J.-P. (2018). Long-term risks of secondary cancer for various whole and partial breast irradiation techniques. Radiotherapy & Oncology. doi:10.1016/j.radonc.2018.05.032