Monitoring DNA Damage and Repair in Peripheral Blood Mononuclear Cells of Lung Cancer Radiotherapy Patients

Simple Summary Every patient responds to radiotherapy in individual manner. Some suffer severe side-effects because of normal tissue toxicity. Their radiosensitivity can be caused by inability of DNA repair system to fix radiation-induced damage. The γ-H2AX assay can detect such deficiency in untransformed primary cells (e.g., peripheral blood mononuclear cells, PBMC), over a period of only hours post ex-vivo irradiation. Earlier we have shown that the level and kinetics of decline (repair) of radiation-induced DNA damage detected by the assay is a measure of the cellular radiosensitivity. In this study, we applied the γ-H2AX assay to judge the radiosensitivity of lung cancer radiotherapy patients as normal or abnormal, based on kinetics of DNA damage repair. Considering the potential of the assay as a clinical biodosimeter, we also monitored DNA damage in serial samples of PBMC during the course of radiotherapy. This study opens an opportunity to monitor individual response to radiotherapy treatment. Abstract Thoracic radiotherapy (RT) is required for the curative management of inoperable lung cancer, however, treatment delivery is limited by normal tissue toxicity. Prior studies suggest that using radiation-induced DNA damage response (DDR) in peripheral blood mononuclear cells (PBMC) has potential to predict RT-associated toxicities. We collected PBMC from 38 patients enrolled on a prospective clinical trial who received definitive fractionated RT for non-small cell lung cancer. DDR was measured by automated counting of nuclear γ-H2AX foci in immunofluorescence images. Analysis of samples collected before, during and after RT demonstrated the induction of DNA damage in PBMC collected shortly after RT commenced, however, this damage repaired later. Radiation dose to the tumour and lung contributed to the in vivo induction of γ-H2AX foci. Aliquots of PBMC collected before treatment were also irradiated ex vivo, and γ-H2AX kinetics were analyzed. A trend for increasing of fraction of irreparable DNA damage in patients with higher toxicity grades was revealed. Slow DNA repair in three patients was associated with a combined dysphagia/cough toxicity and was confirmed by elevated in vivo RT-generated irreparable DNA damage. These results warrant inclusion of an assessment of DDR in PBMC in a panel of predictive biomarkers that would identify patients at a higher risk of toxicity.

. Frequency distribution histograms of Q-and R values based on the data in Table 3.  Table S3) and calculated (as described in Text S1) foci numbers in peripheral blood mononuclear cells (PBMC) at 1 h post first RT session. (A) Calculated relationship between the average foci number and the fraction of irradiated cells (solid line) and experimental data for individual patients (solid symbols). (B) Correlation of calculated (vertical axis) and experimental (horizontal axis) foci numbers. Symbols represent data for each patient, and solid line shows the linear regression. Pearson correlation coefficient r = 0.604, p < 0.01.     Table S2. The values of foci kinetics parameters obtained from nonlinear regression analysis and the ratio of the mean foci number at 24 and 1 h post first radiotherapy (RT) treatment.
* Q is the fraction of unrepairable DNA damage; R is repair rate, the fraction of fpc per hour. Data shaded grey indicate 5 patients that lacked γ-H2AX foci counts at 1 h post first RT due to logistic errors, sample loss or processing failures. These patients were excluded from calculation of 24 h/1 h ratio and from analysis presented in Table S3. Calculation of the average fpc number induced in PBMC by irradiation of blood during local RT is a complex process that requires the knowledge of a range of static and dynamic parameters. Static parameters are the average tissue dose (tumour and/or normal tissue(s)), the irradiated volume, the volume of blood in irradiated tissue (tumour and/or normal tissue(s)) and the total blood volume. Dynamic parameters are the blood flow rate (in tumour and normal tissue) and the dose rate (or detailed timing of individual beams delivering RT fraction). Accurate consideration of all these parameters requires development of a model of blood flow in tissues of interest and superposition of the treatment plan on this model. It is a complex task that is not subject of the present study, especially considering that accurate blood flow parameters might not be available. Therefore, here, we concentrated on calculation of a relationship of foci induction parameters that is not dependent on dynamic parameters and the value of which can also be obtained from experimental data.
First, we state that the total number of foci induced in all PBMC does not depend on dynamic parameters (blood flow and dose rate) and is proportional to the average dose and the number of PBMC in irradiated volume (which in turn is proportional to the fraction of blood, relative to the total blood volume, in irradiated volume). So, the total number of induced foci Ntot can be expressed as follows: where Y is the yield of foci per cell per Gy (approximately, 10-11 for PBMC fixed 1 h post irradiation), D is the dose of radiation, Vbir and Vb are the volume of blood in irradiated volume and total blood volume, respectively, and Ncells is the total number of PBMC in blood. Dynamic parameters (blood flow and dose rate) will affect the fraction of irradiated PBMC (f) and the average number of foci in irradiated subpopulation (Nav), and although we cannot calculate these values, we can establish the relationship between them based on Equation (S1). For fraction f, the number of irradiated cells will be fNcells, and we can obtain the average number of foci by dividing the total number of foci by the number of irradiated cells and substitution for Ntot from Equation (S1): As it follows from Equation (S2), there is an inverse proportional relationship between the average number of foci per cell and the fraction of irradiated cells.
Equation (S2) can be easy verified for extreme cases of instantaneous irradiation and indefinitely long irradiation. For instantaneous irradiation, there is no blood flow for the duration of irradiation, so all cells receive the dose D, thus, = . The fraction of irradiated cells equals in this case to the relative irradiated blood volume = ⁄ , so Equation (S2) converts to the same: For indefinite long irradiation (all blood/cells are mixed up and pass through irradiation volume many times), the fraction of irradiated cells f equals to 1. All cells receive the same dose Dc, and this dose is proportional to the relative time that cells are located in the irradiated volume. This relative time in turn is proportional to the relative volume of blood in the irradiation volume, so the dose is = ⁄ and = = ⁄ . Substitution of f = 1 into Equation (S2) results in the same expression: It can be shown that in a general case of tumour/tissue(s), the term is expressed as a sum of similar terms for each tissue that is subjected to irradiation. The following parameters can be used to calculate the relationship between the average foci number and the fraction of irradiated cells: (1) Blood volume-5000 mL (2) Pulmonary blood volume-500 mL (3) Average tumour volume-500 mL (4) Lung volume-3000 mL (5) Tumour dose-2 Gy (6) Lung dose-0.49 Gy (one fraction) Foci yield-15 foci/Gy/cell (although the reported yield of foci in PBMC varies in the range 9-12 foci/Gy/cell, for low doses up to 100 mGy, higher yield is reported).
We also assumed that the ratio of blood volume/tissue volume is similar for tumour and lung. Based on these parameters, the dose contributions are 49 and 33 mGy for lung and tumour, respectively. The calculated relationship between the average foci number and the fraction of irradiated cells is demonstrated in Figure S2A as a solid line. Data points show the results for individual patients presented in Table S3. Figure S2B demonstrates correlation between the average for each patient foci number (data from Table S3) and foci number calculated using Equation (S2) based on individual parameters (tumour volume and MLD) for each patient.