Amifostine (AMF) has been shown to protect some normal tissues from acute effects of radiation therapy +/- chemotherapy. We enrolled 62 patients in a randomized study investigating the efficacy of AMF: 31 had concurrent chemoradiation for non-small cell lung cancer and 31 had the same treatment + AMF. AMF reduced the frequency and severity of esophagitis, pneumonitis, and neutropenic fever. We have tried to identify patients who get more benefit from AMF by checking their DNA repair capability of normal cells. It was hypothesized that DNA repair capacity from patients' lymphocytes damaged by bleomycin could predict their normal tissue sensitivity to chemoradiation and could be protected by AMF. Forty-six of the 62 patients provided pretreatment blood for assessment of mutagen sensitivity (MS) using a peripheral lymphocyte assay that infers DNA repair capacity from cellular damage remaining after in vitro mutagenic exposure and recovery. Bleomycin-induced chromosome breaks in 50 metaphases were counted and expressed as the mean number of breaks per cell. Patients with an average of more than one break/cell were deemed to exhibit the MS phenotype. Data analysis used Pearson's chi-square and Kaplan-Meier survival function estimates with Strata 8.2 statistical software. The Log-rank test was used to assess the equality of survival function using a P value of .05. Twelve patients (10 AMF, 2 control) exhibited the MS phenotype. The remaining 34 patients (13 AMF, 21 control) were considered to have normal DNA repair. There were no significant differences in overall survival, disease specific survival, or local control by MS. Those with high MS had shorter distant metastasis-free survival compared with low MS patients ( P = .029). There were no differences in severe esophagitis or neutropenic fever by MS. Both high and low MS patients from the control group developed severe lung fibrosis compared with five of 21 who had AMF ( P = .025). The incidence of grade 3/4 lung fibrosis was two of 10 with AMF compared with two of two in the control group ( P = .025) with higher MS. Higher MS was associated with shorter distant metastasis-free survival and more frequent grade 3/4 lung fibrosis. AMF reduced the incidence of grade 3/4 lung fibrosis among higher MS. These data suggest that MS might help identify subgroups of patients who could receive more benefit from AMF with respect to lung damage.