Sm_uary We established an in vitro cytokinesis-block micronucleus assay of human tumours for estimation of the proportion of cells undergoing mitosis (the dividing fraction, DF), the time for the number of nuclei to double and the radiosensitivity in terms of the micronucleus frequency, based on a concept described previously. Under certain conditions, the nuclear number doubling time (NNDT) Recently, the importance of an individualised approach to cancer therapy based on the biological characteristics of each patient's tumour has been increasingly stressed. In radiation therapy, the prediction of biological parameters such as the intrnsic radiosensitivity, proliferative activity (especially the potential doubling time: Tpo) and degree of hypoxia would be helpful in planning optimal treatment (Peters et al., 1988). Several assay systems have been developed to estimate these parameters (Begg et al., 1988;Peters et al., 1988;Brock et al., 1989; H6ckel et al., 1993), but newer and better assays are needed because all of the existing ones have some disadvantages. In addition, none of the assays allows two or more parameters to be estimated simultaneously.In our previous studies using xenografted human and mouse tumours , we developed a new method of estimating tumour radiosensitivity and proliferative activity, which involved determining the micronucleus (MN) frequency after irradiation, the fraction of tumour cells undergoing mitosis in vitro (the dividing fraction, DF) and the presumed time for the number of tumour cell nuclei to double in vitro (nuclear number doubling time, NNDT). The NNDT is considered to be similar to Tpot provided that the culture conditions do not alter karyokinesis. This method makes use of the cytokinesisblock MN assay (Fenech & Morley, 1985)
Materals and methodsTumour specimens All specimens were obtained at the time of major surgery and not by biopsy. A total of 73 solid tumours resected from patients without preoperative irradiation or chemotherapy were evaluated. Malignant ascites from a patient with pancreatic cancer was also investigated, but we could not obtain a result because of excessive contamination with nonmalignant cells. Since the first aim of this study was to determine the types of tumours suitable for this assay, a variety of lesions were studied irrespective of their suitability for post-operative radiotherapy. The tumours ranged from 50 mg to 4.0 g in weight, with a median weight of 850 mg.
Assai procedureThe method used was similar to that of previous studies with slight modifications. The current standard procedure was performed as follows. Tumour specimens were minced with scissors and treated at 3TC for 2 h with 1 mg ml-' collagenase/dispase (Boehringer, Mannheim, Germany) dissolved in phosphate-buffered saline. Then the resulting tumour cell suspension was filtered through a fine wire mesh and viable cells were counted using trypan blue. The proportion of viable cells ranged from 25% to more than 90%, and the cell yield ranged between 1 x 106 and 7 x 10' per gr...