Biological in-vivo measurement of dose distribution in patients' lymphocytes by gamma-H2AX immunofluorescence staining: 3D conformal- vs. step-and-shoot IMRT of the prostate gland

Individuals analyzed in this study were all males, with a median age of 71.4 years (range 51.1 - 83.6), and had an indication for irradiation of the prostate region. This selection was made, because the DNA damage level depends on the anatomic region [8]. Exclusion criteria were a prior radiation in the patients' medical history (so no exposition in advance could interfere with the test) or the additional radiation of lymphatic regions of the pelvis. For either treatment method (3D, SSIMRT), 20 patients were recruited. All patients gave their informed consent. The study was approved by the ethics committee of the University hospital of Heidelberg. The patients' treatment was not influenced by the study and indications for the different modalities were made clinically. Further patient data comparing 3D with SSIMRT is shown in Table 1. The body volume was calculated by the formula as it is published for male patients [9]:

The radiation was performed by a department's linear accelerator (Oncor, Siemens). Table 2 contents the technical parameters of the two irradiation modalities. To calibrate absolute doses to the investigated number of gamma-H2AX foci, blood of 5 volunteers was irradiated in-vitro for 3 independent measurements on different days. Utilization of volunteers was necessary because of intended test repetition, not suitable for patients. Inter-individual differences were considered by investigating 5 subjects. The venous blood was irradiated with doses of 0.02, 0.1, 0.5, 1 and 2 Gy by the same linear accelerator used for the irradiations of the patients. The object-to-focus distance was 1.58 m, the radiation field 10 × 5 cm. Radiation absorbing plates were stacked to a 20 cm tower to allow very low dosage; so the beam on time reaches the operating range of the linear accelerator after the stabilization phase. By varying the time of radiation, different doses were applied. Dose was measured by relative online dosimetry (DIN 6800-2) by using an ionization chamber (thimble 0,3 cm³, PTW, Freiburg, Germany).

7.5 ml of patient's blood were taken from a peripheral vein 10 min after the first fraction of the treatment. The blood circulation was given 10 minutes after fraction to mix the radiated lymphocytes with the rest that hadn't been exposed to radiation. Non-exposed controls were also taken before radiation.

The protocol of staining gamma-H2AX by indirect immunofluorescence is published in many papers and its purpose for detecting DNA DSB validated [10, 11, 12, 13, 14 and 15]. Lymphocytes were separated from the blood by layering 5 ml of heparinized, venous blood onto 3 ml of Ficoll and centrifuging at 2300 rpm for 20 min at 37°C. The lymphocytes were washed in 6 ml of PBS-buffer and centrifuged at 1500 rpm for 10 min (37°C). After aspirating the buffer, the cell-pellet was re-suspended in a 1:15 ratio. 200 μl of this suspension, containing about 300,000 lymphocytes, were spread onto a clean slide by means of the Cytospine Centrifuge at 22 rpm for 4 min (room temperature). Fixating the lymphocytes took 10 minutes (room temperature) in fixation buffer (3% paraformaldehyde, 2% sucrose in PBS). For all experiments, this step was performed 2 hours after finishing radiation to allow comparability between the samples. In order to allow the antibodies getting inside the nucleus, the cells were permeabilized for 4 min at 4°C (permeabilisation buffer: 20 mM HEPES (pH 7.4), 50 mM NaCl, 3 mM MgCl₂, 300 mM sucrose, and 0.5% Triton X-100). Samples were incubated with anti-gamma-H2AX antibody (Anti-Phospho-Histone-gamma-H2AX Monoclonal IgG-mouse-Antibody (# 05-636), Upstate, Charlottesville, VA) at a 1:500 dilution for 1 h, washed in PBS 4 times, and incubated with the secondary antibody (Fluoresceiniso-thiocyanat(FITC)-conjugate, Alexa Fluor 488 Goat-anti-mouse-IgG-conjugate, Molecular Probes, Eugene, OR) at a dilution of 1:200 for 0.5 h. Both incubations took place at 37°C. Cells were then washed in PBS four times at room temperature and mounted by using VECTASHIELD mounting medium including the nucleus stain DAPI (Vector Laboratories). Thus, the gamma-H2AX foci could be correlated with the nuclei.

The slides were viewed with an × 100 objective (fluorescence-microscope Laborlux S, Leica Microsystems CMS GmbH, Wetzlar, Germany). The spots inside the nucleus were counted by eye because of the possibility to focus manually through the whole nucleus by microscope to detect each focus in the 3D-room. All experiments were counted by one and the same, trained person. For each of the samples, 300 lymphocytes were analyzed within the patient samples with its heterogeneous dose-distribution. All nuclei were morphologically considered by eye (cell form and size) to be properly shaped and in G0/1-phase with haploid chromosome-set.

For every patient, gamma-H2AX foci of the lymphocytes were counted. For every count of gamma-H2AX foci per nucleus the averaged relative number of cells was calculated from 20 patients each group (3D and SSIMRT).

The calibration curve involved five subjects irradiated at six different doses in three independent measurements. Background foci levels were subtracted. As the relationship between dose application and irradiation induced gamma-H2AX foci formation is linear [4], a linear regression curve was generated, which implies the following general formula:

(Y = number of gamma-H2AX foci per nucleus, × = dose in Gy, m = gradient)

This linear regression curve was used to calculate an equivalent dose for every count of irradiation induced gamma-H2AX foci per nucleus in patients' lymphocytes. Background foci were subtracted again (controls before irradiation). In addition, the values of gamma-H2AX foci were converted into relative doses, whereas 100% corresponds to the given dose of 2.0 Gy (3D) and accordingly 2.17 Gy (SSIMRT). The calibration concerns only the single lymphocyte, irrespectively body site or blood flow.

In a further integral diagram, the relative number of lymphocytes with gamma-H2AX foci was plotted against the relative applied dose in %. Each point shows the cumulative number of lymphocytes exposed to a certain dose, or more. This visualization of distribution of radiated lymphocytes was defined as dose-lymphocyte-histogram (DLH).

The original dose-volume-histograms (DVH) were modified in order to compare them to our generated DLHs: in general, the volume percentage in the DVH refers to the contoured volume of the CT-scanned part of the body (aortic bifurcation to the thigh). The data was standardized by referring it to the individual's total body volume, allowing interpretation equivalent to the DLH. With the rule of proportion the values of the contoured volumes can transferred to values of total body volumes.

Formula:

The statistics were done by Sigma Plot 10.0^®. The level of significance was set at p < 0.05 using a Student's t-test.

The relation of dose and mean number of gamma-H2AX foci per nucleus (see also Figure 1) of all 5 subjects' lymphocytes follows the same characteristic without significant differences (p > 0.05), which confirms the absence of inter-individual differences [16]. The estimated regression line is used as a calibration curve (Figure 2) and its formula is:

(Y = number of gamma-H2AX foci per nucleus, × = dose in Gy)

For example, 0.5 Gy correlates with a mean number of gamma-H2AX foci per nucleus of 4.9, 1 Gy with 8.6 and 2 Gy with 16 foci, 2 hours after irradiation.

Related to investigated lymphocytes of 20 patients per group the mean number of gamma-H2AX foci per nucleus is 0.49 (3D) and 0.47 (SSIMRT) in the irradiated samples (Figure 3), while the non-irradiated control marks 0.06 (3D) and 0.05 (SSIMRT). The number of foci in the samples after irradiates were for all the patients larger than the number of foci in the non-irradiated control samples. The bars show significant difference between irradiated samples and the control (p ≤ 0.05). The mean number of gamma-H2AX foci in both radiation modalities is the same (p > 0.05).

The DLH is a cumulative histogram; each point shows the cumulated number of lymphocytes that has been exposed to a certain dose, or more (Figure 4). Background foci-levels have been subtracted, since they were also subtracted in the calibration line. The curves cross at about 20% of the described dose, while the SSIMRT curve lies above the 3D curve at lower doses and below it at higher doses. The significant difference is obvious between 40% and 90% of the delivered dose: here, the SSIMRT curve lies significantly below the 3D curve (p ≤ 0.05). There is no difference in relative number of lymphocytes, which get more than 95% of the applied dose. The percentage of lymphocytes exposed to more than 50% of the prescribed dose is 1.8% in 3D technique, compared to 0.9% in SSIMRT.

The curves' crossing point in the DVH takes place at just below 20% of the described dose, whereas the SSIMRT lies above the 3D at 0%-20% and significantly (p ≤ 0.05) below it between 30%-95% (Figure 5). The percentage of volume exposed to more than 50% of the prescribed dose is 1.7% in 3D technique, compared to 0.4% in SSIMRT.