METHOD FOR PREDICTING RADIOSENSITIVITY OF A CELL

Abstract

The present invention relates to a method for predicting the degree of radiosensitivity of a cell by determining the number of prompt double-strand breaks (prDSBs) and predicting the degree of radiosensitivity of the cell based on the number of prDSBs. The present invention relates furthermore to a method for predicting the degree of radiosensitivity of a cell by determining the number of thermally labile sugar lesion-dependent double-strand breaks (tlDSBs) and predicting the degree of radiosensitivity of the cell based on the number of tlDSBs.

Claims

1. An in vitro method for predicting the degree of radiosensitivity of a cell, comprising (a) irradiating a cell, (b) determining the number of prompt double-strand breaks (prDSBs) in the cell of step (a), and (c) using the number of prDSBs determined in step (b) to predict the degree of radiosensitivity of said cell, or an in vitro method for predicting the degree of radiosensitivity of a cell, comprising (a) irradiating a cell, (b) determining the number of thermally labile sugar lesion-dependent double-stand breaks (tlDSBs)) in the cell of step (a), and (c) using the number of tlDSBs determined in step (b) to predict the degree of radiosensitivity of said cell.

2. (canceled)

3. The method of claim 1, wherein the number of tlDSBs is determined by subtracting the number of prDSBs from the number of total double-strand breaks (tDSBs).

4. The method of claim 1, wherein the cell is a diseased cell, preferably a tumor cell, more preferably of epithelial origin, of mesenchymal origin, of hematopoietic origin, or of neuro-ectodermal origin, still more preferably, the cell is selected from a breast adenocarcinoma, sweat gland adenocarcinoma, salivary gland adenocarcinoma, skin squamous cell carcinoma, adenocarcinoma of the thyroid, lung, stomach, liver, pancreas, small intestine, colon, or prostate, transitional cell carcinoma of the bladder; adenocarcinoma of the kidney, testis or endometrium, fibrosarcoma, liposarcoma, osteosarcoma, chondrosarcoma, leiomyosarcoma, hemangiosarcoma, lymphoma, leukemia, astrocytoma, retinoblastoma, oligodendroglioma, schwannoma, melanoma, head and neck cancer, ovarian cancer, adenoid carcinoma, basal cell carcinoma, epidermoid carcinoma, meningioma, neurofibroma, glioblastoma, ependymoma, medulloblastoma, neuroblastoma, hepatoma, mesothelioma, brain cancer such as glioblastoma multiforme, hepatoma, lymphoma, myeloma, neuroblastoma, sarcoma, stomach cancer, thyroid cancer, non-melanoma skin cancer, non-small cell lung cancer, cervical cancer, or anal cancer, or preferably from the thyroid gland in case of the disease of Basedow or hyperthyroidism, from the pituitary gland in case of pituitary adenoma, from the meninges in case of a meningioma, from the skin with a non-cancerous skin disorder, particularly rosacea, poikiloderma of Civatte, angioma, telangiectasias, or psoriasis, or from the ankle in case of talalgia.

5. The method of claim 1, wherein the degree of radiosensitivity of said cell is predicted with respect to a reference.

6. The method of claim 5, wherein the reference is a diseased individual, diseased tissue or diseased cell or a plurality of diseased individuals, diseased tissues or diseased cells, wherein the disease may be a tumor or a disease, preferably wherein the reference is a tumor cell or a plurality of tumor cells, or wherein the reference is a normal individual, normal tissue or normal cell or a plurality of normal individuals, normal tissues or normal cells.

7. The method of claim 6, wherein the disease, preferably the tumor, is known to be treatable by radiotherapy.

8. The method of claim 7, wherein the radiation dose for treating the disease of the reference, preferably tumor, is known.

9. The method of claim 4 for determining the radiation dose for treating a diseased cell, preferably a tumor cell, in an individual, the method further comprising d) comparing the radiosensitivity of the diseased cell, preferably the tumor cell, with the radiosensitivity of the reference, and e) determining the radiation dose for treating the diseased cell, preferably the tumor cell.

10. The method of claim 1, wherein the cell is a normal cell, preferably an epithelial cell, such as a keratinocyte or a lens epithelial cell, a melanocyte, a cardiac myocyte, a chrondrocyte, an endothelial cell, a fibroblast, an osteoblast, a preadipocyte, a skeletal muscle cell, a smooth muscle cell, or a lymphocyte.

11. The method of claim 10, wherein the degree of radiosensitivity of said cell is predicted with respect to a reference.

12. The method of claim 4, wherein the reference is a normal individual, normal tissue or normal cell or a plurality of normal individuals, normal tissues or normal cells.

13. The method of claim 1, wherein the cell is irradiated with ionizing radiation.

14. The method of claim 1, wherein the number of prDSBs or tlDSBs is determined using pulsed-field gel electrophoresis, preferably asymmetric field inversion gel electrophoresis, or the Comet assay.

15. The method of claim 1, wherein the number of prDSBs or tlDSBs is determined by the fraction of DNA released (FDR).

Description

FIGURES

[0072] FIG. 1: Representative flow cytometry histograms of the 15 cell lines used to generate the results described in the present invention, fitted with WinCycle to evaluate the distribution of the cells throughout the cell cycle. The calculated percentage of cells in G1, S and G2 phases of the cell cycle is summarized in Table 1.

[0073] FIG. 2: Panel A, Panel B und Panel C: Survival curves of the indicated cell lines obtained using exponentially growing cells and colony formation as endpoint. Results shown represent the mean and standard deviation calculated from 8 determinations in 2 experiments. The lines shown were fitted to the data points by eye. The different cell lines are allocated in the three panels aiming to maximize clarity. Broken lines in B and C show the response of HT144 and SQ20B cells and have been transferred from A to facilitate comparison.

[0074] FIG. 3: Yields of tDSBs measured using high temperature lysis and of prDSBs measured using low temperature lysis. Cells in the exponential phase of growth were embedded in agarose blocks, irradiated and processed immediately thereafter. Panel A: Representative images of gels stained with ethidium bromide after high temperature lysis for the indicated cell lines. Panel B: As in A after low temperature lysis. Panel C: Dose response curves measured after high temperature lysis (solid lines) and low temperature lysis (broken lines) in the different cell lines. Results shown represent the mean and standard deviation calculated from 6 determinations in 2 experiments. The lines shown are linear regressions through the measured points of each data set. The slopes of these lines are used in the comparison of the DSB-yields under different conditions and are also summarized in Table 1.

[0075] FIG. 4: Panel A: Compilation of all dose response curves obtained with the different cell lines using high temperature lysis. Panel B: Compilation of all dose response curves obtained with the different cell lines using low temperature lysis. Panel C: Compilation of all dose response curves obtained with the different cell lines specifically for the induction of tlDSBs. These results were obtained by subtracting from HTL-FDR values LTL-FDR values. Results have been compiled from the data shown in FIG. 3. The list reflects the sequence of the lines in the figures from top to bottom.

[0076] FIG. 5: Correlation between induction of prDSBs, represented here by the slope of the corresponding dose-response curve (FIGS. 3 and 4), as a function of the radiation dose required for a survival level of 37% (Panel A) and 10% (Panel B). Each circle represents one cell line and is labelled with its corresponding name. Panel C: As in panel B but for the yields of tDSBs (FIGS. 3 and 4).

[0077] FIG. 6: Correlation between inductions of prDSBs, represented here by the slope of the corresponding dose-response curve (FIGS. 3 and 4), as a function of the radiation dose required for a survival level of 1%. Each circle represents one cell line and is labelled with its corresponding name.

[0078] FIG. 7: Representative flow cytometry histograms depicting the changes of -H2AX fluorescent signal as a function of radiation dose for ten of the cell lines tested using this assay.

[0079] FIG. 8: Panel A: Normalized -H2AX signal intensity, measured by flow cytometry 1 h after IR, plotted versus radiation dose for the indicated ten cell lines. Results from three independent experiments are shown as mean values and standard deviations. Results are also normalized for the DNA content of each cell line. Panel B: Correlation between yields of -H2AX signal, i.e. slope of the corresponding dose-response curve in Panel A, as a function of radiation dose required for a survival level of 10% for each of the cell lines tested.

[0080] FIG. 9: Panel A: Representative flow cytometry histograms showing the signal of Histone H3 3meK9 in ten of the cell lines tested with this assay. Shown in the left of each individual panel are the histograms of cells incubated only with secondary antibody. Panel B: Bar plots representing the intensity of Histone H3 3meK9 signal, normalized to DNA content. The results are means from three independent determinations and the error bars represent standard deviations.

[0081] FIG. 10: Normalized (to the DNA content of each cell line) fluorescent signal measured by flow cytometry of histone H3-3meK9, a measure of heterochromatin, versus radiation dose required for a survival level of 10% for ten of the cell lines tested here.

[0082] FIG. 11: Panel A and Panel B: As in FIG. 9, but for Histone H3 acK9.

[0083] FIG. 12: As in FIG. 10, but for Histone H3 acK9.

EXPERIMENTS

Materials and Methods

Cell Culture and Irradiation

[0084] For experiments we employed: The human cervical epithelial carcinoma cell lines HeLa and C33A; the human melanoma cell line HT144; the human prostate epithelial cancer cell lines PC-3 and LnCap and the human colorectal carcinoma cell line HCT116. This group of cell lines was grown in Minimum Essential Medium (MEM), supplemented with 10% fetal bovine serum (FBS). In addition we employed: The human lung adrenal carcinoma cell line A549, and the human osteosarcoma cell line U2OS, which were grown in McCoy's 5A medium, supplemented with 10% FBS. We finally employed: Two human head and neck squamous carcinoma cell lines, SQ20B and SCC61, which were maintained in Dulbecco's Modified Eagle's Minimum Essential Medium (D-MEM) supplemented with 10% FBS. All cell lines were maintained at 37 C. in 5% CO.sub.2 and were used in the exponential phase of growth. Three human glioma cell lines (A7, Bogdahn 17, LN229) were grown in Minimum Essential Medium (MEM), supplemented with 15% fetal bovine serum (FBS) and 1% non-essential amino acids. The group of two human non-small cell lung carcinoma cell lines (H460, H520) was maintained in Roswell Park Memorial Institute (RPMI 1640) medium supplemented with 10% FBS.

[0085] Radiation exposures were carried out in parallel with multiple cell lines using both lysis protocols in two X-ray units (Precision X-ray, North Branford, Conn.) operated at 320 kV, 10 mA with a 1.65 mm Al filter. They were carried out on ice to prevent repair in experiments measuring DSB induction, and at room temperature in cell survival experiments.

Colony Formation Assay

[0086] Standard procedures were used to measure colony formation. For example, after exposure to different radiation doses, cells in exponential phase were trypsinized and plated into 60 mm tissue culture dishes at increasing numbers with increasing radiation exposure aiming for 20-200 colonies per dish. They were stained with crystal violet two weeks later and counted. Clones with more than approximately 50 cells were considered to originate from surviving cells.

Pulsed-Field Gel Electrophoresis

[0087] Induction of DSBs was measured by Asymmetric Field Inversion Gel Electrophoresis (AFIGE), a Pulsed-Field Gel Electrophoresis (PFGE) technique as previously described (29, 31-33), using, as appropriate, for the same pool of agarose blocks either high (50 C.) (HTL) or low (LTL) (4 C.) temperature during lysis. In this assay, the number of DSBs present in cells is indirectly measured by the fraction of DNA released (FDR) out of the well into the lane (38).

[0088] Briefly, cells were trypsinized, suspended in serum-free HEPES-buffered growth medium and mixed with 1% low-melting agarose (Bio-Rad). The agarose cell suspension was pipetted into glass tubes of 3 mm diameter and was allowed to solidify on ice before removing from the glass tube and cutting into 5 mm blocks (plugs), which were transferred for irradiation to tissue culture dishes containing serum-free growth medium.

[0089] After irradiation, plugs were transferred to lysis solution and lysed either at high (50 C.) (HTL) or low (LTL) (4 C.) temperatures. For HTL, plugs were lysed in a solution containing 10 mM Tris-HCl, 100 mM EDTA, pH 7.6, 2% N-lauryl (NLS) and 0.2 mg/ml protease added just before use, at 50 C. for 18 h. For LTL, plugs were first transferred into ESP buffer (0.5 M EDTA, pH 8.0, supplemented with 2% NLS and 1 mg/ml protease, both added just before use) for 24 h, and subsequently into a high-salt buffer (4 mM Tris, pH 7.5, 1.85 M NaCl, 0.15 M KCl, 5 mM MgCl.sub.2, 2 mM EDTA and 0.5% Triton-XI00 added just before use) for 16 h.

[0090] Electrophoresis was carried out after RNAase treatment in gels, prepared with 0.5% molecular biology grade agarose (Bio-Rad), at 8 C. for 40 h applying 50 V (1.25 V/cm) for 900 s in the forward direction and 200 V (5.00 V/cm) for 75 s in the reverse direction. Gels were stained with ethidium bromide and imaged in a fluor-imager (Typhoon, GE Healthcare). FDR was analyzed using ImageQuant 5.2 (GE Healthcare). Calculations of absolute numbers of DSBs (either tDSBs or prDSBs), when desired, were based on previously described calibrations (29, 37). For comparisons between cell lines, however, such absolute calculations of the numbers of induced DSBs are not required.

DSB Analysis Via Flow-Cytometry-Quantification of -H2AX

[0091] Samples containing 2-310.sup.6 cells for selected cell lines were suspended for 5 min on ice in phosphate buffered saline (PBS) containing 0.2% Triton X-100. Cells were fixed for 15 min in PBS containing 3% paraformaldehyde and 2% sucrose, and were incubated in PBG blocking buffer (0.5% BSA, 0.2% gelatin in PBS) overnight at 4 C. Cells were incubated for 1 h at RT with an antibody against -H2AX (GeneTex) diluted in PBG, and subsequently for 1 h with a secondary antibody conjugated with AlexaFluor647. DNA was stained with propidium iodide and samples were analyzed in a flow cytometer (Galios, Beckman-Coulter, USA).

Analysis by Flow Cytometry of Euchromatin and Heterochromatin

[0092] Samples from selected cell lines were collected and fixed as described in the previous section. Antibodies against the tri-methylated and acetylated forms of Lysine 9 of Histone H3 (Abcam PLC) were then employed to stain and analyze the cells.

Results

[0093] To investigate possible correlations between cell radiosensitivity and induction of DSBs, 15 tumor cell lines were selected. In the tested 15 tumor cell lines, we determined radiosensitivity to killing using colony formation, induction of tDSBs using high temperature lysis (HTL) and of prDSBs using low temperature lysis (LTL). From these measurements the yields of tlDSBs can be estimated. Table 1 lists the cell lines employed, indicates their origins and shows their typical distribution throughout the cell cycle under the conditions used for experiments (see also FIG. 1 for representative flow cytometry data). It is evident that under the conditions employed, all cell lines show similar distributions throughout the cell cycle. Furthermore, Table 1 also summarizes quantitative aspects of results presented and discussed below.

TABLE-US-00001 TABLE 1 Alphabetical compilation of the cell lines tested, including information on their origin, cell cycle distribution, radiosensitivity to killing, as well as yields of tDSBs, prDSBs and tlDSBs. G1/S/G2, Slope of Cell-line Origin % HTL/Gy.sup.1 A549 human lung adrenal 52/31/17 0.0101 carcinoma A7 human glioma 49/36/15 0.0110 Bogdahn 17 human glioma 63/24/13 0.0106 C33A human cervical epithelial 57/31/12 0.0101 carcinoma HCT116 human colorectal 38/34/28 0.0099 carcinoma HeLa human cervical epithelial 48/39/13 0.0099 carcinoma HT144 human melanoma 50/35/15 0.0108 H460 human non-small lung 53/33/14 0.0099 carcinoma H520 human non-small lung 46/31/23 0.0106 carcinoma LnCap human prostate epithelial 56/24/20 0.0097 cancer LN229 human glioma 48/30/22 0.0104 PC3 human prostate epithelial 48/30/22 0.0101 cancer SCC61 human head-neck 43/35/22 0.0095 squamous carcinoma SQ20B human head-neck 52/33/15 0.0092 squamous carcinoma U2OS human osteosarcoma 39/41/20 0.0103 Slope of Slope of Cell-line LTL/Gy.sup.1 TLSL/Gy.sup.1 D37/Gy D10/Gy D1/Gy A549 0.0049 0.0052 4.1 5.9 9.6 A7 0.0067 0.0043 2.0 3.8 6.4 Bogdahn 17 0.0061 0.0045 2.8 5.3 8.2 C33A 0.0070 0.0041 0.9 3.1 5.2 HCT116 0.0035 0.0054 3.9 6.1 10.0 HeLa 0.0046 0.0053 2.3 4.4 7.2 HT144 0.0078 0.0030 0.9 1.8 3.3 H460 0.0048 0.0051 2.3 5.0 7.7 H520 0.0053 0.0054 2.2 3.9 7.3 LnCap 0.0044 0.0053 2.7 4.9 8.2 LN229 0.0051 0.0053 3.0 5.6 8.0 PC3 0.0054 0.0047 2.1 3.9 7.2 SCC61 0.0044 0.0051 1.9 4.2 6.1 SQ20B 0.0034 0.0058 4.2 6.8 10.8 U2OS 0.0061 0.0042 1.9 3.7 7.1

[0094] FIGS. 2A, 2B and 2C depict the survival curves of the group; they document a wide spectrum of radiosensitivities as required by the aims of the present study.

[0095] Despite wide fluctuations in radiosensitivity to killing, the tested cell lines show relatively small fluctuations in the levels of tDSBs measured immediately after exposure to IR (FIG. 3). This is better illustrated in FIG. 4A, which integrates in a single plot the calculated lines for DSB yields. The results are in line with previous reports (see Introduction) suggesting that tDSB induction is only rarely a robust predictor of cell radiosensitivity to killing.

[0096] Notably, when yields of prDSBs are determined by LTL, marked differences are detected among cell lines (FIG. 3C). This is again better illustrated in FIG. 4B that depicts prDSB yields for all cell lines in a single graph.

[0097] The similarity in the induction of tDSBs among cell lines and the marked differences noted in the induction of prDSBs directly imply that the induction of tlDSBs will also be different in the different cell lines. We calculated therefore induction of tlDSBs by subtracting prDSBs from tDSBs and the results obtained are summarized in FIG. 4C. As anticipated, marked differences in tlDSB induction are observed among tested cell lines.

[0098] Direct visual inspection and comparison between prDSB yields and cell radiosensitivity to killing reveals that radioresistant cells show low induction of prDSBs. The opposite is true for tlDSBs. This is quantitatively illustrated in FIG. 5. Plotted in the figure for each cell line is the measured prDSB yields (slope of the dose response line, Table 1) as a function of radiation dose at which the survival of this cell line drops either to 37% (panel A) or to 10% (panel B). Results obtained by considering radiation doses corresponding to 1% cell survival are presented in FIG. 6. The values of the corresponding parameters are included in Table 1.

[0099] Notably, comparison of prDSB yields with radiosensitivity to killing quantitatively confirms an excellent correlation. In this set of cell lines, HT144, the most radiosensitive cell line, shows a slope in DSB induction curves of 0.0078 Gy.sup.1, which is over twofold higher than the slope of SQ20B cells, 0.0034 Gy.sup.1, the most radioresistant cell line. Evidently, high yields of prDSBs render cells radiosensitive to killing and vice-versa.

[0100] For comparison, FIG. 5C shows a plot similar to B but for tDSBs. It is evident that while a correlation between tDSB yields and radiation dose for 10% survival is observed, the available dynamic range and the statistical significance are by far not as robust as for prDSBs.

[0101] As outlined in the Introduction, analysis of -H2AX holds promise as predictor of radiosensitivity to killing. We wished to compare the predictive power of -H2AX, as a marker for DSB induction, with that of prDSBs-yields presented above. Therefore, we adapted a flow cytometry-based technology that allows analysis of integral -H.sub.2AX signal in a range of radiation doses similar to that used in PFGE.

[0102] Typical histograms of -H.sub.2AX signal obtained at different doses at 1 h postirradiation with the 15 cell lines under investigation are summarized in FIG. 7. From such data, the mean -H2AX intensity can be estimated for each radiation dose. Among the different possible ways of presentation of this dose response, most informative proved the one that plots as a function of radiation dose the normalized intensity of -H2AX signal, after correction for DNA content. This value is calculated by dividing the mean -H2AX signal intensity obtained for a given radiation dose by the mean -H2AX signal intensity calculated for the 0 Gy sample.

[0103] Results of normalized -H2AX signal intensity as a function of radiation dose are summarized in FIG. 8A for ten of the 15 cell lines utilized. Linear regression adequately fits the results obtained demonstrating that -H2AX signal saturation is not occurring in the range of doses examined.

[0104] The slopes of the dose response curves obtained show marked differences among cell lines, raising the potential of correlations with cell radiosensitivity to killing. FIG. 8B shows the slopes of the -H2AX dose response of each cell line, plotted as a function of the radiation dose required for 10% cell survival. Notably, the results show no correlation whatsoever between DSB induction quantitated as normalized -H2AX signal and cell radiosensitivity to killing. We note that -H2AX generates signals reflecting tDSBs (29), and that -H2AX generation is subject to a complex physiological regulation, which may disconnect signal intensity from the number of DSBs present (27).

[0105] We inquired whether the low induction of prDSBs measured in radioresistant cell lines correlates with aspects of chromatin organization. Therefore, we screened ten of the 15 cell lines with known markers of chromatin organization. Trimethylation of Histone-H3-Lys9 (H3-3meK9) is a widely accepted marker of condensed, heterochromatic organization (34)(35). We used flow cytometry to estimate integral H3-3meK9 signal, as a measure of heterochromatin content, in our panel of cell lines. Raw measurements of this analysis are shown in FIG. 9A. It is evident that quantitative differences exist among cell lines, indicated by fluctuations in mean signal intensity after correction for DNA content (FIG. 9B). However, when the DNA content-corrected H3-3meK9 signal is correlated with cell radiosensitivity to killing (10% survival) trends are apparent, but without statistical significance (FIG. 10).

[0106] As a complementary form of analysis we measured acetylation of the same Histone H3-Lys9, which is widely considered a measure of chromatin relaxation, i.e. of euchromatin (34, 35). We analyzed therefore again ten of the cell lines for H3-acK9 using flow cytometry. Raw measurements are summarized in FIG. 11A and their quantification (mean of signal intensity after correction for DNA content) in FIG. 11B. Plotting of the DNA content-corrected mean H3-acK9 signal against cell radiosensitivity to killing (10% survival; FIG. 12) shows again trends, but no statistically significant correlation.

[0107] We conclude that the chosen parameters of chromatin organization fail to correlate with cell radiosensitivity to killing and that no significance should be placed in the observed trends.

Discussion

[0108] The results provide, for the first time, evidence for an excellent correlation between prDSB-yields and cell radiosensitivity to killing and define a new parameter with strong predictive power. The results generate a basis for focusing on specific aspects of DSB induction for predicting radiosensitivity to killing at the expense of the more elaborate assays that are based on estimates of DSB repair capacity. While repair capacity certainly remains a key determinant of the cellular response to IR, our observations suggest that lethal events are generated with higher probability from prDSBs. We elaborate below that not only the predictive power of prDSB-yields is higher, but also their determination is far easier and more accurate than determination of repair capacity.

[0109] Defining surrogate predictors of cell radiosensitivity to killing is significant, as radiosensitivity of tumor cells is linked to tumor radiation response (10-12), and radiosensitivity of fibroblasts to late radiation effects (1-9) (see Introduction). Comparison with -H2AX- or tDSBs-yields demonstrates that under the conditions examined, the predictive power of prDSBs-yields for cell radiosensitivity to killing is superior.

[0110] The significance of our observations is further reinforced by the fact that determination of prDSB-yields using existing PFGE methods is straightforward and can be achieved with a high level of confidence. This is because the determination is based on an entire dose response curve with multiple dose points that is typically linear (FIG. 3). Since prDSBs yields are reflected, in a first approximation, by the slope of the resulting line, they can be accurately determined by linear regression. In addition, the approach relies on results obtained at high doses of radiation generating PFGE-signals that can be accurately measured. Yet, it predicts radiosensitivity at 37% and 10% survival levels (FIG. 5) that are achieved at doses well within the range of those routinely used in radiation oncology.

[0111] Previous work found relatively weak associations between tDSBs yields and radiosensitivity to killing (5, 10-12, 19) (see Introduction), an observation that is also supported by our results with tDSBs and -H2AX (FIGS. 5C and 8). Repair kinetics or residual DSBs, on the other hand, correlate with cell radiosensitivity to killing (6, 8, 9, 16, 23-26). Yet, accurate determination of the latter parameters is more demanding than measurement of prDSB-yields, as it requires maintenance of cells under conditions ensuring metabolic function equivalent to the in-vivo situation in order to maintain unchanged their repair potential. Even when this is achievable, very high radiation doses are required to obtain statistically significant differences in the number of unrepaired DSBs between cell lines with all assays that measure the physical presence of a DSBe.g. PFGE (7-12). Use of -H.sub.2AX based assays, however, ameliorates this concern (36).

[0112] Also the time point after radiation exposure at which residual DSB measurements are made is frequently debated, with different times after irradiation arbitrary chosen in different studies and actually showing different predictive power (19, 36). Measurement of prDSBs immediately after IR eliminates all these confounding factors and simplifies decisively the associated experimental protocol.

[0113] Importantly, this in-vitro method may be directly applied to biopsy material from an individual, obviating the tedious and time consuming step of establishing in vitro cultures. In this way, direct measurements may be possible using biopsy material, with the significant advantage of obtaining data actually reflecting the radiosensitivity of the tissue of origin (normal tissue or tumor), rather than that of cells selected by their ability to grow in vitro. There are intensive efforts at present along these lines in the field (36).

[0114] However, since biopsies may contain relatively few cells (10.sup.5 to 10.sup.6 cells for skin biopsies), the PFGE method may be employed after miniaturization that makes it feasible with fewer cells. Since key in the predictive power of the present assay is the selective measurement of prDSBs, methods may be applied providing this information on the basis of single-cell gel electrophoresis.

[0115] Why are tDSBs-yields or -H.sub.2AX signal a weak predictor of radiosensitivity to killing, while prDSB-yield appears so strongly predictive? It may be that lethal lesions arise predominantly from prDSBs, while tlDSBs are shunted with higher probability to error-free processing. Our attempts to link prDSB or tlDSB induction to salient features of chromatin organization did not prove informative (FIGS. 9-12). Certainly more work is required to address and possibly clarify this important issue.

[0116] We define prDSB-yields as a novel parameter with strong predictive power towards cell radiosensitivity to killing. We further define tlDSB-yields as a novel parameter with strong predictive power towards cell radiosensitivity to killing. We show that these parameters can easily and highly accurately be determined. The approach defined here offers tantalizing new possibilities for the development of predictive assays with direct and wide clinical applicability.

REFERENCES

[0117] (1) Little J B, Nove J, Strong L C, Nichols W W. Survival of Human Diploid Skin Fibroblasts from Normal Individuals after X-irradiation. Int J Radiat Biol. 1988; 54:899-910. [0118] (2) Loeffler J S, Harris J R, Dahlberg W K, Little J B. In Vitro Radiosensitivity of Human Diploid Fibroblasts Derived from Women with Unusually Sensitive Clinical Responses to Definitive Radiation Therapy for Breast Cancer. Radiat Res. 1990; 121:227-31. [0119] (3) Burnet N G, Nyman J, Turesson I, Wurm R, Yarnold J R, Peacock J H. The relationship between cellular radiation sensitivity and tissue response may provide the basis for individualising radiotherapy schedules. Radiother Oncol. 1994; 33:228-38. [0120] (4) Burnet N G, Nyman J, Turesson I, Wurm R, Yarnold J R, Peacock J H. Prediction of normal-tissue tolerance to radiotherapy from in-vitro cellular radiation sensitivity. Lancet. 1992; 339:1570-1. [0121] (5) Wurm R, Burnet N G, Duggal N, Yarnold J R, Peacock J H. Cellular radiosensitivity and dna damage in primary human fibroblasts. Int J Radiat Oncol Biol Phys. 1994; 30:625-33. [0122] (6) Kiltie A E, Orton C J, Ryan A J, Roberts S A, Marples B, Davidson S E, et al. A correlation between residual DNA double-strand breaks and clonogenic measurements of radiosensitivity in fibroblasts from preradiotherapy cervix cancer patients. Int J Radiat Oncol Biol Phys. 1997; 39:1137-44. [0123] (7) Sarkaria J N, Bush C, Eady J J, Peacock J H, Steel G G, Yarnold J R. Comparison between Pulsed-Field Gel Electrophoresis and the Comet Assay as Predictive Assays for Radiosensitivity in Fibroblasts. Radiat Res. 1998; 150:17-22. [0124] (8) Dikomey E, Brammer I, Johansen J, Bentzen S M, Overgaard J. Relationship between DNA double-strand breaks, cell killing, and fibrosis studied in confluent skin fibroblasts derived from breast cancer patients. Int J Radiat Oncol Biol Phys. 2000; 46:481-90. [0125] (9) Kasten-Pisula U, Tastan H, Dikomey E. Huge differences in cellular radiosensitivity due to only very small variations in double-strand break repair capacity. Int J Radiat Biol. 2005; 81:409-19. [0126] (10) McMillan T J, Cassoni A M, Edwards S, Holmes A, Peacock J H. The Relationship of DNA Double-strand Break Induction to Radiosensitivity in Human Tumour Cell Lines. Int J Radiat Biol. 1990; 58:427-38. [0127] (11) Ruiz de Almodovar J M, Nunez M I, McMillan T J, Olea N, Mort C, Villalobos M, et al. Initial radiation-induced DNA damage in human tumour cell lines: a correlation with intrinsic cellular radiosensitivity. Br J Cancer. 1994; 69:457-62. [0128] (12) Woudstra C, Driessen C, Konings AWT, Kampinga H H. DNA damage induction and tumour cell radiosensitivity: PFGE and halo measurements. Int J Radiat Biol. 1998; 73:495-502. [0129] (13) Brock W A, Tucker S L, Geara F B, Turesson I, Wike J, Nyman J, et al. Fibroblast radiosensitivity versus acute and late normal skin responses in patients treated for breast cancer. Int J Radiat Oncol Biol Phys. 1995; 32:1371-9. [0130] (14) Geara F B, Peters L J, Ang K K, Wike J L, Brock W A. Prospective comparison of in vitro normal cell radiosensitivity and normal tissue reactions in radiotherapy patients. Int J Radiat Biol. 1993; 27:1173-9. [0131] (15) Johansen J, Bentzen S M, Overgaard J, Overgaard M. Relationship between the in vitro radiosensitivity of skin fibroblasts and the expression of subcutaneous fibrosis, telangiectasia, and skin erythema after radiotherapy. Radiother Oncol. 1996; 40:101-9. [0132] (16) Russell N S, Arlett C F, Bartelink H, Begg A C. Use of fluorescence in situ hybridization to determine the relationship between chromosome aberrations and cell survival in eight human fibroblast strains. Int J Radiat Biol. 1995; 68:185-96. [0133] (17) Schipler A, Iliakis G. DNA double-strand-break complexity levels and their possible contributions to the probability for error-prone processing and repair pathway choice. Nucleic Acids Res. 2013; 41:7589-605. [0134] (18) Radford I R. The Dose-response for Low-LET Radiation-induced DNA Double-strand Breakage: Methods of Measurement and Implications for Radiation Action Models. Int J Radiat Biol. 1988; 54:1-11. [0135] (19) E I-Awady R A, Dikomey E, Dahm-Daphi J. Radiosensitivity of human tumour cells is correlated with the induction but not with the repair of DNA double-strand breaks. Br J Cancer. 2003; 89:593-601. [0136] (20) Okayasu R, Iliakis G. Linear DNA elution dose response curves obtained in CHO cells with non-unwinding filter elution after appropriate selection of the lysis conditions. Int J Radiat Biol. 1989; 55:569-81. [0137] (21) Okayasu R, Iliakis G. The shape of DNA elution dose response curves under non-denaturing conditions: Int J Radiat Biol. 1992; 61:455-63. [0138] (22) Dahm-Daphi J, E I-Awady R A, Dikomey E. Reply: Tumour and normal cells differ in the induction and repair of DNA double-strand breaks. Br J Cancer. 2004; 90:556-. [0139] (23) Koch U, Hhne K, von Neubeck C, Thames H D, Yaromina A, Dahm-Daphi J, et al. Residual H2AX foci predict local tumour control after radiotherapy. Radiother Oncol. 2013; 108:434-9. [0140] (24) Menegakis A, Yaromina A, Eicheler W, Drfler A, Beuthien-Baumann B, Thames H D, et al. Prediction of clonogenic cell survival curves based on the number of residual DNA double strand breaks measured by gH2AX staining. Int J Radiat Biol. 2009; 85:1032-41. [0141] (25) Banath J, Klokov D, MacPhail S, Banuelos C, Olive P. Residual gammaH2AX foci as an indication of lethal DNA lesions. BMC Cancer. 2010; 10:4. [0142] (26) Olive P L, Banath J P, Keyes M. Residual g H2AX after irradiation of human lymphocytes and monocytes in vitro and its relation to late effects after prostate brachytherapy. Radiother Oncol. 2008; 86:336-46. [0143] (27) Kinner A, Wu W, Staudt C, Iliakis G. g-H2AX in recognition and signaling of DNA double-strand breaks in the context of chromatin. Nucleic Acids Res. 2008; 36:5678-94. [0144] (28) Singh S K, Bencsik-Theilen A, Mladenov E, Jakob B, Taucher-Scholz G, Iliakis G. Reduced contribution of thermally labile sugar lesions to DNA double strand break formation after exposure to heavy ions. Radiat Oncol. 2013; 8:77. [0145] (29) Singh S K, Wang M, Staudt C, Iliakis G. Post-irradiation chemical processing of DNA damage generates double-strand breaks in cells already engaged in repair. Nucleic Acids Res. 2011; 39:8416-29. [0146] (30) Singh S K, Wu W, Stuschke M, Bockisch A, Iliakis G. Reduced Contribution of Thermally-Labile Sugar Lesions to DNA Double-Strand Break Formation after Exposure to Neutrons. Radiat Res. 2012; 178:581-90. [0147] (31) Singh S K, Wu W, Wu W, Wang M, Iliakis G. Extensive Repair of DNA Double-Strand Breaks in Cells Deficient in the DNA-P K Dependent Pathway of NHEJ after Exclusion of Heat-Labile Sites. Radiat Res. 2009; 172:152-64. [0148] (32) Stenerlw B, Karlsson K H, Cooper B, Rydberg B. Measurement of prompt DNA double-strand breaks in mammalian cells without including heat-labile sites: Results for cells deficient in nonhomologous end joining. Radiat Res. 2003; 159:502-10. [0149] (33) Rydberg B. Radiation-induced heat-labile sites that convert into DNA double-strand breaks. Radiat Res. 2000; 153:805-12. [0150] (34) Kakarougkas A, Ismail A, Klement K, Goodarzi A A, Conrad S, Freire R, et al. Opposing roles for 53BP1 during homologous recombination. Nucleic Acids Res. 2013; 41:9719-31. [0151] (35) Grewal S I S, Jia S. Heterochromatin revisited. Nat Rev Genet. 2007; 8:35-46. [0152] (36) Menegakis A, von Neubeck C, Yaromina A, Thames H, Hering S, Hennenlotter J, et al. H2AX assay in ex vivo irradiated tumour specimens: A novel method to determine tumour radiation sensitivity in patient-derived material. Radiother Oncol. 2015; in press. [0153] (37) Iliakis G E, Cicilioni O, Metzger L. Measurement of DNA double-strand breaks in CHO cells at various stages of the cell cycle using pulsed field gel electrophoresis: calibration by means of 125I decay. Int J Radiat Biol. 1991 February; 59(2):343-57. [0154] (38) DiBiase S J, Zeng Z C, Chen R, Hyslop T, Curran W J Jr, Iliakis G. DNA-dependent protein kinase stimulates an independently active, nonhomologous, end-joining apparatus. Cancer Res. 2000 Mar. 1; 60(5):1245-53.