BACKGROUND We investigated the in-vitro radiosensitivity of peripheral blood lymphocytes with a special FISH/CISS-technique. METHODS From October 1993 through April 1996, a total number of 52 cancer patients was enrolled in the study. The tumor sites in these patients were: breast (n = 41), lung (n = 4), head and neck (n = 3) as well as prostate, bladder, rectal cancer and Hodgkin's disease (each n = 1). Twenty-six of them were examined prior to planned radiotherapy (prospective group) and 26 after radiotherapy (retrospective group). Three additional individuals (without cancer or radiotherapy) with proven ataxia telangiectasia (Louis-Bar syndrome, AT-homozygotes) were also investigated and their blood samples served as positive control for radiosensitivity. The clinical radiation response of normal tissue in radiotherapy patients was scored according to the WHO grading system for acute and according to the RTOG grading system for late effects. For to estimate the intrinsic radiosensitivity, blood samples were taken and irradiated in vitro with 0 (control) or 0.7 or 2 Gy with a 6 MV-linear accelerator, standard 48-hour lymphocyte cultures were prepared, chromosomes #1, #2 and #4 were simultaneously labeled with a FISH/CISS-technique and 200 to 1,000 metaphase spreads were scored for chromosomal aberrations. The radiation sensitivity of lymphocytes was expressed as the number of radiation-induced chromosomal breaks per mitosis after 0.7 Gy or 2 Gy corrected for the 0-Gy control value. RESULTS The frequency of chromosomal breaks/mitosis in the unirradiated control lymphocytes was 0.020 +/- 0.015 in prospective patients who had not yet received radiotherapy. It was significantly higher in retrospective patients (0.264 +/- 0.164 breaks/mitosis) as a result of the previous radiation exposure. The 3 AT-homozygotes showed also an increased number of spontaneous chromosomal breaks (0.084 +/- 0.016 breaks/mitosis), probably resulting from the chromosomal instability in this disease. This figure, however, was significantly lower than in retrospective patients. The number of radiation-induced breaks after in-vitro irradiation was comparable in lymphocytes of patients who showed no normal tissue reaction (n = 11) as compared to those with mild to moderate radiation reaction (n = 32, acute reactions Grade 1 to 2, late reactions Grade 0 to 2). In 9 patients with unexpected severe plus late Grade 3 to 4 reactions, however, a significantly higher number of radiation-induced chromosomal breaks was measured; the highest number was observed in a patient with a radiation myelitis. The 3 AT-homozygotes showed, as expected, an extreme radiosensitivity of their lymphocytes. The number of breaks/mitosis after 0.7 Gy in vitro irradiation of lymphocytes was 0.103 +/- 0.059 in patients with no normal tissue radiation reaction (n = 11), 0.122 +/- 0.146 in the group with mild to moderate radiation reactions Grade 1 to 2 (n = 32), 0.359 + 0.226 in patients with unexpected Grade 3 to 4 normal tissue reactions (n = 9) and 0.550 +/- 0.243 in the 3 AT-homozygotes (p < 0.01, t-test). The difference in lymphocyte radiosensitivity between these 4 groups was also detected after in-vitro irradiation with 2 Gy (0.484 +/- 0.132 vs. 0.535 +/- 0.228 vs. 0.926 +/- 0.349 vs. 1.423 +/- 0.072). CONCLUSIONS We found a significantly higher number of chromosomal breaks in lymphocytes of patients with severe or extreme radiation reaction of normal tissues as compared to patients with no or only mild to moderate radiation reactions. The radiosensitivity of lymphocytes in these radiosensitive patients was in the range between normal radiosensitivity and the radiosensitivity of AT-homozygotes. Detection of patients with severely enhanced intrinsic radiosensitivity might be possible with this method.