Normal tissue radiosensitivity in breast cancer patients

Abstract

To investigate whether in vitro radiosensitivity of lymphocytes derived from a blood sample will predict late effects from radiotherapy in breast cancer patients. Blood samples were collected from consenting patients who had received radiotherapy for breast cancer. Lymphocytes were extracted and transformed by the Epstein-Barr virus. The resulting lymphoblastoid cell lines (LCLs) were assessed for in vitro radiosensitivity using a tetrazolium-based colorimetric assay (MTT). Survival curves were generated using doses of 0.5 to 2 Gy. For each analysis an LCL derived from an individual with the radiosensitive condition ataxia-telangiectasia (AT) was run as control. Some patients also consented to give skin biopsies from which fibroblast cultures were derived. Clonogenic assays were performed to generate survival curves using doses in the range of 0.5 to 4 Gy. Comparison was made with the data obtained from LCLs. Late effects of radiotherapy were assessed using the Radiation Therapy Oncology Group (RTOG) scoring scheme and compared with the in vitro radiosensitivity data. LCLs from 56 breast cancer patients were assessed for in vitro radiosensitivity. Surviving fraction to 2 Gy (SF2) generated from survival curves ranged from 0.04-0.35 with coefficient of variation for the whole group of 41%. None of the LCLs equalled the sensitivity of the AT line, but 16% showed equal or greater sensitivity to a line derived from an AT heterozygote. Comparison of LCL and fibroblast radiosensitivity showed reasonable correlation for 12 paired samples (r = 0.64). The majority of patients showed no or minimal effects after radiotherapy (Grade 0, 1 effects) but seven developed a Grade 2 reaction and four a Grade 3 or 4 reaction. Patients with a Grade 2-4 reaction were found to be more sensitive in vitro than those with a Grade 0-1 reaction (p < 0.02). The use of the MTT assay to assess LCL radiosensitivity has demonstrated considerable heterogeneity amongst the breast cancer population. The presence of a proportion of patients showing in vitro sensitivity but normal clinical response to radiotherapy would limit the usefulness of the assay for predictive purposes.