Fibroblast radiosensitivity in vitro and lung fibrosis in vivo: comparison between a fibrosis-prone and fibrosis-resistant mouse strain.

Radiation-induced pneumonitis and fibrosis in the lung after treatment to the thoracic cavity for malignant disease currently limit the maximum tolerated dose to that region. It has been suggested that heterogeneity in susceptibility to radiation-induced fibrosis exists in the population, implying that the lung tolerance dose is defined by a sensitive subset of the patient population. Studies of radiotherapy patients have indicated that the survival at 2 Gy (SF2) of cultured skin fibroblasts correlates with the incidence and severity of postirradiation damage in a number of tissues, suggesting that this assay may be a useful predictor of late tissue effects. The goal of the studies presented here was to determine if the radiosensitivity of fibroblasts in vitro isolated from mouse lungs was correlated with the severity of radiation-induced fibrosis in the lungs of two inbred strains of mice previously shown to differ markedly in their susceptibility to radiation-induced lung fibrosis: the C3Hf/Kam strain, classified as fibrosis-resistant, and the C57BL/6J strain, classified as fibrosis-prone. Quantitative measurements of lung fibrosis after irradiation were compared to SF2 values for fibroblasts of skin and lung cultured from each strain. Lung fibrosis was quantified, using computerized image analysis, as the percentage of fibrosis on Masson's Trichrome-stained lung sections from both strains after single doses of radiation to the thorax. For the measurements of SF2, fibroblasts plated at the second passage and grown to confluence were given single doses of radiation ranging from 0 to 6 Gy. Survival curves were constructed and SF2 values obtained from a linear-quadratic fit to the data. The radiosensitivity of fibroblasts from the lung and skin of SCID mice was determined and served as a positive control. The percentage of radiation-induced lung fibrosis was significantly different between the two strains, 5.1% and 0.2% in the C57 strain and C3H strain, respectively. Follow-up of long-term survivors (two mice) from the C3H strain did not change this conclusion. However, the lung fibroblast SF2 for the C57BL/6J strain (fibrosis-prone), 0.50 +/- 0.03, was not statistically different from the C3Hf/Kam strain (fibrosis-resistant), 0.55 +/- 0.07. These data indicate that in vitro radiosensitivity of lung fibroblasts as assayed by survival at 2 Gy does not correlate with the development of lung fibrosis in this mouse model. The SF2 for lung fibroblasts from SCID mice was 0.10. Similar SF2 values were obtained for both the C3Hf/Kam mouse lung and skin fibroblasts, 0.55 and 0.56, respectively, and C57BL/6J mouse lung and skin fibroblasts, 0.50 and 0.52, respectively, indicating that the radiosensitivity of fibroblasts isolated from lung and skin within a strain is the same.