Objective To evaluate the radiobiological effect of different irradiation fractionated regimens in human glioma cells ( BT 325 cell line). Methods The xenografts in Balb/c-nude mice were irradiated with different single and fractionated regimens. The single fraction dose was 10, 20, 30, 40 and 60 Gy, respectively. The fractionated regimens were 2 Gy × 5 fractions ( irradiated every day), and 3 Gy ×3 fractions (irradiated every other day), 3 Gy × 5 fractions (irradiated every day) and 4 Gy × 3 fractions (irradiated every other day), with total doses of 125 Gy, 114 Gy, 126 Gy and 112 Gy, respectively. The growth curve was used to evaluate the tumor doubling time. clonogenic assays was performed to draw the cell survival curve and analyze the radiobiological parameters with doses of 1, 2, 4, 6, 8 and 10 Gy. T1/2 was measured by comet assay. Results Tumor regression were not observed by single fraction irradiation, 2 Gy × 5 fractons and 3 Gy × 3 fractions irradiation regimens. The tumor regress was more significant with the increas of fraction dose. The 4 Gy × 3 fractionrs inhibited tumor more though not curing tumor. The cell doubling time of the BT 325 cell was 30. 16 h and the tumor doubling time of the xenograft was 43 days.When fitted with L-Q model ,α was 0. 36 Gy -1 and β was 0. 057 Gy -2. When fitted with the single-hit multitarget model, D0 was 1. 394 Gy, Dq was 2. 127 Gy and SF2 was 0. 714, respectively. The T1/2 was 9. 999min. Conclusions Glioma is a radioresistant tumor. Increase of the fraction dose improves recent effect.Further study is needed to control the tumor stem cells.

Objective To impmve the method of "modified comet assay" in predicting the radiosensitivitv of solid tumor. Methods A single cell suspension from biopsy sample was lmdlated on ice with a dose of 5 Gy.The microscope slide was spread with agarose,lysed for 50 minutes,rinsed 3 times rinse solution,and given electrophoresis for 20 minutes. After being stained with PI,cell images were collected through the microscope and analyzed with Lucia G software(Version 4.6).In order to check system/ background errors,every sample was made into control slide and irradiation slide.The end-points were cell DNA contents and tail moment. Results The factors influencing the results included:(1)Sample was iaulty tor the biopsv taken from mucosa and no tumor cells were contained. (2)The slides with a high backgmund ( induced by necrosis) disturbed the measurement of comet assay. (3) Setting lymphocytes as control to check svstem errors was very important. (4)To separately collect images of the normal tissue cells and tumor cells from the biopsy sample improved the conformity between the clinical obscrvation and the lab result. Conclusions To increase the correlation between comet assay and clinical response,it is very helpful to set double control for checking system/background errors and to collect images of the normal tissue cells and tumor cells through the microscope,respectively.

BACKGROUNDCells derived from lung cancer are biological heterogeneous and have different intrinsic radiosensitivity, it is a key question for us to investigate radiosensitive parameters for an individualized radiotherapy plan. The comet assay is a sensitive and facilitated method to detect single-cell DNA damage and repair, and the results from it have been proven to be so highly coincident with those from clonogenic assay by cell-line investigations that it has been considered as a promising method in predicting radiosensitivity. The study is designed to evaluate preliminarily the correlation between initial DNA damage detected by alkaline comet assay and the clinical endpoints.

METHODSBiopsy samples from 31 lung cancer patients by fibrous bronchial endoscopy were detected by alkaline comet assay from April, 2002 to November, 2002. The adjusted tail moment (R TM) was measured and thoracic local-region lesions were measured by computer tomography scan. Response rate (RR) and time to progression (TTP) for the local-region lesions were as clinical endpoints. SPSS 10.0 software was used to compare median R TM of different RR and TTP groups by Mann-Whitney U and Kruskal-Wallis H rank test, the correlations between R TM with RR and TTP were estimated by Spearman's rank test.

RESULTSThere were no statistic differences of median R TM among different pathological types with a median R TM of 0.98, 1.27 and 1.05 in squamous cell carcinoma group, adenocarcinoma group and small cell lung cancer group, respectively (Chi-square=0.347, P=0.84). Through a median follow-up of 10 months, a median R TM of 1.08 and 1.21 for squamous cell carcinoma group and small cell lung cancer group in RR≥50% group was greater than 0.88 and 0.91 in RR < 50% group; median R TM in TTP > 9-monthgroups stratified according to pathological type was greater than that in TTP≤9-month groups (1.26, 1.38 and 1.39 versus 0.71, 0.48 and 1.03 for squamous cell carcinoma group, adenocarcinoma group and small cell lung cancer group respectively), but the differences of R TM classified by RR or TTP were not statistically significant (U=63.5, P=0.58; U=71, P=0.057); the Spearman's coefficients of R TM with RR and TTP were -0.105 (P=0.57) and 0.38 (P=0.035). The coefficients of R TM with TTP was 0.47 for non-small cell lung cancer indicating a modest correlation (P=0.048) and 0.043 for small cell lung cancer (P=0.89).

CONCLUSIONSAlthough the results are confounded due to sampling and the greater background tail moments, Spearman's coefficient of R TM with TTP for non-small cell lung cancer indicates a modest positive correlation. The comet assay might be a promising method in predicting intrinsic radiosensitivity of lung cancer cells and techniques for sampling and assaying need to be further improved.