Response of human tumor cell lines in vitro to fractionated irradiation

The surviving fraction of human tumor cell lines after 2 Gy (SF2) varies between 0.1 and 0.8. It has been postulated that differences in inherent radiosensitivity of tumor cells are a major determinant of radiation response in vivo. Assays of inherent radiosensitivity based on acute survival are being developed as predictors of tumor response which often assume that the same inherent radiosensitivity persists throughout a fractionated treatment. We have investigated the response of 2 human tumor cell lines (A549 and MCF7) with different inherent radiosensitivities to in vitro fractionated irradiation. A549 cells had an SF2 of 0.62 and a mean inactivation dose (D) of 3.07 Gy whereas MCF7 cells had an SF2 of 0.30 and a D of 1.52 Gy. Split dose repair capacity (at equal survival levels) was less for A549 than for MCF7 cells and recovery kinetics for both cell lines were substantially longer than those of rodent cell lines. Survival after 5 fractions of 2 Gy given 12 hr apart at 37 degrees C was near to that predicted from the acute survival curve, assuming complete repair and no proliferation. Acute survival of A549 cells which survived 5 fractions of 2 Gy given 12 hr apart was similar to the acute survival of unirradiated cells. When A549 cells were incubated at 22 degrees C between 5 fractions of 2 Gy given 12 hr apart, proliferation and split dose repair were substantially inhibited. These studies support the proposals to use in vitro inherent radiosensitivity assays for the prediction of in vivo response of tumors to fractionated treatment.     

Response of xenografts of human malignant gliomas and squamous cell carcinomas to fractionated irradiation.

The response of xenografts of five human malignant glioma cell lines and two human squamous cell carcinomas to fractionated irradiation was studied. For this, the tumors were transplanted into nude mice which had been further immunosuppressed by 6 Gy whole-body irradiation. Radiation was given as 30 fractions applied under normal blood flow conditions in two sessions per day over 15 days. Absolute and specific tumor growth delay after 48 Gy, and tumor control dose 50% (TCD50) were evaluated. Using local tumor control as experimental endpoint, four out of five malignant gliomas were more resistant to fractionated radiation therapy than the two squamous cell carcinomas. The TCD50s of these four gliomas ranged from 73 Gy to more than 120 Gy, whereas the TCD50s of the squamous cell carcinomas were 51 and 60 Gy. Absolute tumor growth delay correlated well with TCD50, but no correlation was obtained between specific growth delay and TCD50. The response of the human tumor xenografts in vivo did not correlate with the surviving fractions at 2 Gy of the same cell lines in vitro which have been previously obtained in our laboratory. The results suggest that the unique radioresistance observed in malignant gliomas in patients is at least in part reflected in human tumor xenografts. The lack of correlation between the surviving fraction at 2 Gy in vitro and the tumor response in vivo could be a consequence of an immune response by the host, a difference in cell radiation sensitivity between cell lines and xenografted tumors, or of differences of parameters such as hypoxic fraction, rate of repopulation, and cell cycle effects between the different tumor lines studied. It illustrates the difficulties which might be involved in the prediction of the response of individual tumors to radiation therapy based solely on the intrinsic radiosensitivity of the tumor cells as assayed by in vitro assays of colony formation.     

Response of a rat brain tumor to fractionated therapy with low doses of BCNU and irradiation

When intracerebral 9L tumors in male Fisher 344 rats were treated with daily fractions of BCNU and X-rays, 5 days per week for 2 weeks, the following were found: (1) there was a significant increase in the life-span of rats treated with the combination treatment schemes over that observed for rats treated with the single agent schemes, (2) the increase in Iife-span in the combination treatments was critically dependent on the time interval between the BCNU and X-ray treatments, (3) increasing the daily dose of BCNU increased the effectiveness of the combination treatments, and (4) the results of the optimum combination treatment were not critically dependent on the tumor size or growth state.     

Pre-treatment number of clonogenic cells and their radiosensitivity are major determinants of local tumour control after fractionated irradiation.

OBJECTIVE: The response of tumours to fractionated radiotherapy is determined by many factors including repopulation, reoxygenation, the number of clonogenic cells, and their intrinsic radiosensitivity. However, after single radiation doses given under conditions of clamp hypoxia, the dose to control a tumour locally is dependent only on the number of clonogenic cells and their cellular radiosensitivity. Therefore, these parameters were investigated using local control after single doses given under hypoxia, to predict the outcome of fractionated irradiation. MATERIALS AND METHODS: Ten hSCC cell lines (FaDu, UT-SCC-15, UT-SCC-14, XF354, UT-SCC-5, UT-SCC-45, SAS, CAL-33, UT-SCC-8, and HSC-4) were transplanted subcutaneously into the right hind-leg of NMRI nude mice. At 7mm in diameter, tumours were irradiated either with graded single doses under clamp blood flow conditions (n=873) or with 30 graded fractions within 6 weeks (n=905) under ambient conditions. Local tumour control was determined 120 days after irradiation. Radiation response was quantified in terms of TCD(50), i.e. the dose required to control 50% of tumours locally. RESULTS: Ten tumour lines investigated showed a pronounced heterogeneity in both TCD(50(30fx/6w)) after fractionated irradiation and TCD(50(SDclamp)) after single dose irradiation. TCD(50(30fx/6w)) varied between 45Gy for UT-SCC-45 and 127Gy for SAS; TCD(50(SDclamp)) varied between 42Gy for UT-SCC-14 and 66Gy for CAL-33. Two tumours were excluded from further analysis due to immunogenicity or non-defined TCD(50). Linear regression analysis revealed a significant positive correlation between TCD(50(SDclamp)) and TCD(50(30fx/6w)) (R(2)=0.82, p=0.002). CONCLUSIONS: Significant association between TCD(50(SDclamp)) and TCD(50(30fx/6w)) suggests that the pre-treatment number of clonogenic tumour cells and their cellular radiosensitivity have a major impact on local control after fractionated radiotherapy.     

Effect of normobaric oxygen on tumor radiosensitivity: fractionated studies

The sensitizing ability of 100% normobaric oxygen was investigated in a mouse mammary carcinoma (CaNT) using a variety of fractionated regimens. Both regrowth delay and local control were used as assays of tumor response. With both assays, there was a similar and significant increase in radiosensitivity for all fractionated schedules. Enhancement ratios ranged from 1.24 to 1.45, the highest increase being observed with a 30 fraction schedule given in an overall time of 6 weeks. Thus, in CaNT tumors normobaric oxygen is a far more efficient radiosensitizer in fractionated treatments than the oxygen-mimetic compound misonidazole; an oxygen effect being observed at doses per fraction as low as 1.8 Gy. These results suggest strongly that normobaric gases could play an important role in the clinical management of tumors where hypoxia may limit the outcome of radiotherapy.     

Response of human tumour xenografts to fractionated X-irradiation

The response of two human tumour xenografts to single dose and fractionated X-rays has been tested using regrowth delay as the assay. The tumours were line transplanted cells from a moderately well-differentiated squamous carcinoma of the tonsillar fossa (XJ) and an undifferentiated carcinoma of the floor of the mouth (XR). Comparison of the dose response curves for single doses in air, clamped, or after misonidazole administration, led to estimates of the hypoxic fraction (approximately 15%) and the sensitizer enhancement ratio (less than or equal to 1.6). When 5 daily fractions were used, the effect of misonidazole (miso) was lost and reoxygenation appeared to be effective in both tumours. Comparison of single doses and 5 fractions in clamped tumours, and in those sensitized by miso, allowed the sparing effect of fractionation to be estimated. When analysed by the linear quadratic model the alpha/beta ratios were found to be in the range of 6.4-9.2 Gy and 6.8-16.0 Gy for the two tumours. These values are in good agreement with murine tumours (assayed in vivo or in vitro), with human tumour cells assayed in vitro, and with analyses of fractionated clinical data for skin cancer.     

Chemopreventive agent sulforaphane enhances radiosensitivity in human tumor cells

Sulforaphane (SFN), an isothiocyanate derived from broccoli and other cruciferous vegetables, is a positive regulator of phase II detoxification enzymes and is highly effective in protection against chemically induced cancers by inducing apoptosis and cell cycle arrest. Here, we report that SFN also enhances radiosensitivity in human tumor cells. Cell survival in HeLa human cervix carcinoma cells pretreated with SFN was significantly lower than in cells treated with radiation only. Constant‐field gel electrophoresis and a gamma‐H2AX foci assay showed marked inhibition of DSB repair in irradiated cells treated with SFN, while little inhibition was observed in cells with DMSO (control). In addition, immunofluorescence experiments revealed a significant delay in Rad51 (a key protein for homologous recombination repair) foci formation and disappearance in irradiated cells treated with SFN when compared to the cells with X‐irradiation alone. The dephosphorylation of DNA‐PKcs (a critical nonhomologous end joining protein) was also markedly delayed by SFN pretreatment in irradiated cells. These DSB repair inhibition data partially support the high apoptotic frequency of irradiated cells pretreated with SFN. Furthermore, the combined treatment of X‐rays and SFN (i.p. 300 μmol/kg) in the xenograft model with HeLa cells showed efficient inhibition of in vivo tumor growth. To the best of our knowledge, our study is the first report showing SFN‐enhanced radiosensitivity of tumor cells in vitro and in vivo, which opens the door for a multitude of clinical applications for chemoradiotherapy using SFN. © 2009 UICC     

Differential radiosensitization by the poly(ADP-ribose) transferase inhibitor 3-aminobenzamide in human tumor cells of varying radiosensitivity

Four newly-established human tumor cell lines, have been irradiated at dose rates of 150 and 3.2 cGy/min to compare their capacity to repair radiation damage. They included a neuroblastoma, a germ-cell carcinoma of the testis, a large cell carcinoma of the lung, and a carcinoma of the cervix. The four lines varied in their sensitivity to high dose-rate irradiation, with the neuroblastoma being most radiosensitive and the lung and cervix tumors the most radioresistant. The extent of dose sparing associated with lowering the dose rate to 3.2 cGy/min was similar in three of the lines but somewhat greater in the case of the cervix carcinoma cell line. The presence of non-toxic concentrations of the poly(ADP-ribose) transferase inhibitor, 3-aminobenzamide (3-AB), enhanced the response of 3 of the 4 tumors to irradiation; it failed to modify the sensitivity of a lung carcinoma cell line. The extent of sensitization was generally similar at high and low dose rate. Measurement of poly(ADP-ribose) transferase activity in control and irradiated cells showed the neuroblastoma cells to contain much higher initial levels than the other three lines but there were no significant differences in the extent of stimulation in enzyme levels after irradiation. Survival curves obtained at low dose-rate help define the initial slope of the acute curve and it appears that 3-AB may exert a differential effect among human tumors in modifying this component.     

Repair of fractionated radiation in plateau phase cultures of human tumor cells and human multicellular tumor spheroids

We have measured potentially lethal damage repair (PLDR) after fractionated radiation, delivered at 24-h intervals in density-inhibited plateau phase cultures of four human tumor cell lines derived from tumors of differing radiocurability (two melanoma and two breast). The repair of potentially lethal damage conferred significant radioresistance on the human melanoma cells but not on the breast carcinoma cells. We examined the effects of fractionated radiation on human tumor cells adapted to become multicellular tumor spheroids (MTS). MTS derived from a human neuroblastoma were "cured" by a total fractionated radiation dose about 50% of that required for MTS derived from a human melanoma.     

Response of the solid Guerin epitheliomas of rats to fractionated irradiation and a new 4-nitroimidazole.

The transplantable Guerin epithelioma in Wistar rats was used to test the in vivo effectiveness of 1-(2-hydroxy-3-methoxypropyl)-2-methyl-4-nitroimidazole (P1) as a tumour-cell radiosensitizer after its oral administration at relatively low doses. The radiosensitizing ability of P1 was compared with that of metronidazole. The results indicate that P1 is less toxic than metronidazole, and greater concentrations of P1 in blood and tumour tissues are obtained for the same administered dose of the compounds. The radiosensitizing ability of P1, determined from tumour-regression rates and local-control percentage at 130 days, was higher than that of metronidazole.     

EGFR抑制剂与肿瘤细胞放射敏感性

表皮生长因子受体(EGFR)在多种上皮来源的恶性肿瘤中存在过表达,其表达水平与肿瘤的放射敏感性呈负相关.大量研究表明,阻断EGFR表达的靶向抑制剂能够增加肿瘤细胞的放射敏感性,提高放射治疗的疗效,其机制可能与抑制细胞增殖、诱导细胞凋亡、干扰细胞周期分布、延长DNA的损伤修复有关.因此,运用EGFR抑制剂可以增强肿瘤细胞对放射的敏感性.     

Response of the canine lung to fractionated irradiation: pathologic changes and isoeffect curves

Canine lungs were irradiated with a range of total doses given in 2, 3, or 4 Gy per fraction. Sequential histopathologic evaluations were done at 1, 3, 6, and 12 months. Pathologic changes in canine lungs were found to be similar to those found in other species demonstrating a clinically latent period, a pneumonitis phase, and late fibrosis and vascular damage. The relative impact of endothelial cell and pneumocyte injury on either early or late radiation injury of the lung is difficult to resolve. Therefore, it is not possible to define a target cell for lung injury at this time. The alpha/beta ratios determined in this study indicate that the target cell or cells associated with lung consolidation are slowly proliferating and represent late responding tissues. Lungs were evaluated histomorphometrically for alveolar air space and radiographically for alveolar consolidation at 6 months after irradiation. Alpha/beta ratios of 3 Gy and 4 Gy were calculated respectively. Both assays demonstrated an increasing effect on lung damage with increasing fraction size from 2 to 4 Gy. Application of the LQ model and use of alpha/beta ratios for calculation of dose adjustments remains theoretical. Clinical data are insufficient to define specific alpha/beta ratios for the various normal tissues at risk in radiation therapy. The data are sufficient to demonstrate the sparing effects of decreasing size of dose per fraction for late responding tissue. Results of this study suggest caution against the use of large doses per fraction for radiation therapy fields which include large lung volumes.     

BSO-induced reduction of glutathione levels increases the cellular radiosensitivity of drug-resistant human tumor cells.

Acquired resistance to cis-platinum and melphalan, in the human ovarian OAW42 tumor cell line, respectively, conferred a 3- and 1.5-fold decrease in photon sensitivity. Analysis of cell survival curves by the linear quadratic equation showed an accompanying 5- and 2-fold reduction in the magnitude of the initial slope (alpha). Treatment with the GSH depleting agent BSO restored the magnitude of alpha to a value similar to that of the parental line without evidence of dose modification in the high-dose region of the cell survival curve. This in conjunction with failure of alteration in GSH levels to affect parental OAW2 sensitivity and of the SER of BSO to reflect GSH levels suggest a possible GSH independent mechanism of action for BSO. If similar patterns occur in the clinic, the possibility exists of circumventing collateral resistance between chemotherapeutic agents and ionizing radiation, provided that tumor thiol levels can be preferentially depleted.     

Tumor radiocurability: relationship to intrinsic tumor heterogeneity and to the tumor bed effect

Recent in vitro clonogenic assays of cells from human head and neck carcinomas recurring after fractionated irradiation showed that the mean radiation resistance was significantly greater than that of cells from unirradiated head and neck cancers. Analysis of these data using the linear-quadratic formalism of radiation inactivation indicates that such increased resistance was associated with a large decrease in the parameter describing cell killing by one-hit processes, while the parameter describing cell inactivation by two-hit processes was unchanged. These data suggest that head and neck carcinomas contained intrinsically radioresistant tumor subpopulations that were selected for during the fractionated radiation therapy. Consequently, we have generated a model of how the phenomenon of clonal intraneoplastic diversity might be related to the selection of radioresistant sub-populations. There is, however, an additional major modulating factor which may impact upon the proposed selection process, known as the tumor bed effect. This effect refers to the fact that solid tumors grow at slower rates and to decreased carrying capacities in radiation-damaged normal tissue stroma. We hypothesize that the tumor bed effect is a reflection of an altered, hostile microenvironment within which competition among neoplastic subpopulations for limited resources is increased. This then produces a situation leading to extinction of minority subpopulations within heterogeneous neoplasms, regardless of whether such populations are radioresistant. The final result of these selection/extinction processes will be dependent upon total dose, dose per fraction, overall time, and the initial partial volume percentages of radioresistant subpopulations.     

The ESTRO Regaud lecture. Inherent radiosensitivity of tumor and normal tissue cells as a predictor of human tumor response

This lecture reviews the history and current status of attempts to correlate the outcome of radiation therapy with estimates of tumor cell radiosensitivity, made by either histologic assessment of tumors irradiated in vivo or by assays for the survival of tumor cells irradiated in vitro. Genetically determined variability in the radiosensitivity of normal tissue cells from different individuals is also discussed in terms of its effect on the determination of "tolerance" doses and the possible correlation between tumor cell and normal tissue cell radiosensitivity in a given patient. It is concluded that measurement of both tumor cell and normal tissue cell radiosensitivities, along with other radiobiologically-based assays, will improve our ability to predict treatment outcome and provide the basis for further therapeutic advances in radiation oncology.     

Radioresistant and repair proficient cells may determine radiocurability in human tumors

Radiosensitivity and repair of radiation damage was studied in 29 human tumor cell lines. Nine tumor lines appeared radioresistant (RAD+) and were associated with radioincurable tumors. When the ability to repair potentially lethal damage was studied in plateau phase cultures, some tumor cell lines derived from radioincurable tumors were more efficient in this process (PLDR+). The results suggest that the presence of radioresistant or repair proficient cells may be responsible for local failure following radiotherapy. Advancements in cell culture and molecular biology may lead to predictive assays useful to clinicians.     

Recovery from sublethal damage during fractionated irradiation of human FaDu SCC.

BACKGROUND AND PURPOSE: The present study addresses whether recovery of sublethal damage in tumours may change during fractionated irradiation in FaDu human squamous cell carcinoma and whether such an effect might contribute to the pronounced time factor of fractionated irradiation previously found in this tumour. PATIENTS AND METHODS: FaDu tumours were transplanted s.c. into the right hind leg of NMRI nu/nu mice. Single doses or 2, 4, and 8 equal fractions in 3.5 days were applied in previously unirradiated tumours and after priming with 18 fractions of 3 Gy in 18 or 36 days. All irradiations were given under clamp hypoxic conditions. Experimental endpoints were tumour control dose 50% (TCD50) and alpha/ beta values without and after priming. RESULTS: Without priming TCD50 increased with increasing number of fractions from 38.8 Gy (95% CI 35;45) after single dose irradiation to 54.0 Gy (42;57) after 8 fractions. No increase in TCD50 when given in 1, 2, 4, or 8 fractions in 3.5 days was found after priming with 18 3-Gy fractions in 18 and 36 days. After priming with 18 fractions in 18 days TCD50 remained constant at 25 Gy and after priming with 18 fractions in 36 days at 42 Gy. The alpha/beta ratio without priming was 68 Gy (42;127). After fractionated irradiation with 18 3-Gy fractions in 18 and 36 days the alpha/beta ratio increased to 317 Gy (38; infinity) and to infinite, respectively. CONCLUSIONS: Our results indicate that clonogenic cells in FaDu tumours lose entirely their capacity to recover from sublethal radiation damage during fractionated irradiation. Therefore, an increased repair capacity as an explanation for the pronounced time factor of fractionated irradiation in this tumour can be ruled out.