As an extension to a previous study, a linear calibration curve covering doses from 0 to 10 Gy was constructed and evaluated in the present study using calyculin A-induced premature chromosome condensation (PCC) by scoring excess PCC objects. The main aim of this study was to assess the applicability of this PCC assay for doses below 2 Gy that are critical for triage categorization. Two separate blind tests involving a total of 6 doses were carried out; 4 out of 6 dose estimates were within the 95% confidence limits (95% CL) with the other 2 just outside. In addition, blood samples from five cancer patients undergoing external beam radiotherapy (RT) were also analyzed, and the results showed whole-body dose estimates statistically comparable to the dicentric chromosome assay (DCA) results. This is the first time that calyculin A-induced PCC was used to analyze clinical samples by scoring excess objects. Although dose estimates for the pre-RT patient samples were found to be significantly higher than the mean value for the healthy donors and were also significantly higher than those obtained using DCA, all these pre-treatment patients fell into the same category as those who may have received a low dose (<1 Gy) and do not require immediate medical care during emergency triage. Additionally, for radiological accidents with unknown exposure scenario, PCC objects and rings can be scored in parallel for the assessment of both low- and high-dose exposures. In conclusion, scoring excess objects using calyculin A-induced PCC is confirmed to be another potential biodosimetry tool in radiological emergency particularly in mass casualty scenarios, even though the data need to be interpreted with caution when cancer patients are among the casualties.

Ionizing radiation, such as X-rays used in diagnostic imaging and radiotherapy (RT) as well as gamma rays and neutron particles released from nuclear weapons, can result in a wide range of direct and indirect DNA damage [IAEA, 2011]. In large doses, radiation can cause serious tissue damage and increase the risk of developing cancer in later life [Stewart et al., 2012]. Even for lower doses, there is no suggested threshold dose for radiation-induced malignancy based on the stochastic nature of radiation carcinogenesis [Albert, 2013]. Therefore, it is critical to assess the exposure dose of the individuals as soon as possible in mass casualty radiation emergency cases. Biodosimetry, or the measurement of biological markers, such as dicentric chromosomes, translocations, micronuclei, and excess premature chromosome condensation (PCC) fragments, has proven to be a very important source of information in the evaluation of radiation overexposure, particularly when combined with clinical signs and symptoms as well as any available physical measurements [IAEA, 2011]. Dosimetric and radiological triage categorization results are essential in the support of medical and public health decision making [Ainsbury et al., 2014].

The dicentric chromosome assay (DCA) is the current gold standard method for biological dosimetry; however, it has several inherent limitations. For example, it requires well-trained scorers, and it is not accurate for high dose exposures over 5 Gy due to cell death, mitotic delay, and the saturation of dicentrics [IAEA, 2011; Pujol et al., 2014]. Calyculin A-induced PCC assay overcomes many of these limitations and has been widely used for the analysis of high dose exposures by scoring rings and excess fragments [Lamadrid et al., 2007; Guerrero-Carbajal et al., 2019; Puig et al., 2013; Romero et al., 2016] or by calculating the length ratio of the longest to the shortest chromosomes [Gotoh and Tanno, 2005; González et al., 2014] or the cell cycle progression index [Miura et al., 2014]. This highly efficient and up-scalable method is particularly advantageous when there is very limited availability of blood for analysis or when metaphase spreads cannot be obtained due to very high dose exposure. Recently, scoring the number of chromosomal objects in excess of 46 in calyculin A-induced PCC has been proposed as an easy and suitable biodosimetry method in the estimation of absorbed doses between 2 and 10 Gy [Sun et al., 2020b]. PCC objects can easily be confused with chromosomal fragments that are generated during the formation of chromosomes with multiple centromeres (e.g., dicentrics and tricentrics, etc.) and rings following exposure to high dose of radiation. They are identified as individual pieces of chromosome regardless of the shape and size, therefore eliminating the necessity to distinguish dicentrics, rings, minutes, and fragments from normal chromosomes by defining each of them as one object and the excess number of objects is used as the dosimetric endpoint (Fig. 1). In addition, when an exposure scenario is unknown, scoring objects and rings in parallel allows both high and low doses to be analyzed using the same sets of slides or digital images [Sun et al., 2020b].

Fig. 1.

One Giemsa-stained G2 phase PCC cell containing 55 objects with 9 excess above 46 (irradiated at 10 Gy). Every individual chromosomal piece regardless of shape and size is scored as one object. Numbers in red are placed next to the objects to assist with the understanding of the scoring. The blood donor was a healthy male aged in the range of 25–34 years.

Fig. 1.

One Giemsa-stained G2 phase PCC cell containing 55 objects with 9 excess above 46 (irradiated at 10 Gy). Every individual chromosomal piece regardless of shape and size is scored as one object. Numbers in red are placed next to the objects to assist with the understanding of the scoring. The blood donor was a healthy male aged in the range of 25–34 years.

Close modal

Scoring the numbers of total chromosomes at the G2 phase using calyculin A-induced PCC has been reported as a biodosimetry endpoint for low and high linear energy transfer (LET) radiation involving gamma rays [Gotoh et al., 2005] and carbon ion beam [Wang et al., 2007]. However, no specific assessment has been carried out for the suitability of this method at doses below 2 Gy; and not many cytogenetic assays have been used for the analysis of in vivo partial-body exposures [Darroudi et al., 1998; Hayata et al., 2001; Moquet et al., 2018, 2020]. The goals for triage dosimetry are to rapidly estimate the overexposure doses, to assign the patients into the correct categories, and to provide the information for timely medical treatment [IAEA, 2011]. There are three categories implemented in the MULTIBIODOSE emergency triage categorization software [Jaworska et al., 2015]: 1. Low exposure <1 Gy; 2. medium exposure 1–2 Gy; 3. high exposure >2 Gy. Therefore, dose estimation for overexposures below 2 Gy is crucial for triage categorization.

In the present study, a calibration curve for doses between 0 and 2 Gy was constructed, and the suitability for this curve to be combined with a previously published curve for doses between 2 and 10 Gy [Sun et al., 2020b] was assessed. This combined curve was validated using blind tests before it was used to evaluate the PCC method in comparison to DCA in clinical sample dose estimation. This in vivo study is part of the ongoing RTGene 2 project involving multiple organizations aimed at developing biomarkers of radiation response using longitudinal blood samples from cancer patients undergoing RT [Moquet et al., 2018]. The main objectives to carry out this study were to assess whether it is feasible to use calyculin A-induced PCC at 0–2 Gy for triage categorization; and whether the results obtained using this method are comparable to those obtained from the gold standard DCA method.

All chemicals and reagents used in this project were the same as those used in a previously published study [Sun et al., 2020b]. Peripheral blood lymphocyte isolation, irradiation, and PCC induction were performed and cells at G2 and M phases were scored as previously described [Sun et al., 2020b]. In brief, non-cycling (G0) blood lymphocytes were isolated using Histopaque® 1077, irradiated (for calibration curve construction and blind tests, but not for patient samples), and pre-cultured with a mitogen, phytohemagglutinin (PHA), for approximately 48 h to stimulate cell division. Following irradiation, cells were kept at 37° for 2 h before PHA stimulation to allow DNA repair. Calyculin A powder was reconstituted in DMSO and subsequently diluted to working concentration (50 nm) in complete RPMI1640 medium containing 20% (v/v) fetal bovine serum, 2 mml-glutamine, 100 units/ml penicillin, and 100 μg/mL streptomycin. Calyculin A was added into the cell suspension 30 min before harvest for PCC induction. Induced cells were finally harvested, fixed, and stained for visual analysis. Written informed consent and the approval of the West Midlands-Solihull Research Ethics Committee (REC 14/WM/1182) were obtained for the healthy donors.

In the present study, isolated lymphocytes were placed into 15-mL centrifuge tubes, positioned inside a 22-mm polystyrene block with 8-mm Perspex, and sham-exposed or exposed ex vivo to 0, 0.25, 0.5, 0.75, 1, and 2 Gy of 250-kVp X-rays (with a half-value layer of Cu/Al filtration). The X-ray set (Ago X-ray Ltd., Martock, UK) was calibrated to a dose rate of 0.5 Gy/min and dosimetry was performed with a calibrated reference ionization chamber for the exact exposure setup used. Exposures were always monitored using a calibrated UNIDOS E electrometer and “in-beam” monitor ionization chamber (all from PTW, Germany). Spatial dose uniformity was checked using Gafchromic EBT2 films (Vertec Scientific Ltd., UK). For different dose points, 200 (1 Gy), 400 (2 Gy), or 500 (0, 0.25, 0.5, and 0.75 Gy) cells were scored with more scored for 2 Gy than for 1 Gy. This is because 2 Gy was the overlapping dose between the 2 separate investigations (0–2 Gy and 2–10 Gy). The blood sample from a healthy donor (female, age range 18–25 years) without any known previous radiation exposure was used for analysis. As a condition of the ethical approval, the actual age was not disclosed. Like dicentrics, no statistically significant inter-personal difference is believed to exist among normal non-radiosensitive individuals; therefore, no biological replicates were considered necessary for this study, even though rigorous intercomparison may be needed to validate this observation. Further evidence has been referenced in a previously published study [Sun et al., 2019]. After appropriate statistical testing, the data for the above indicated irradiation doses were combined with the previously published data [Sun et al., 2020b] to generate a combined new curve covering 0–10 Gy.

Data from the previously published blind test [Sun et al., 2020b] were used to evaluate the combined calibration curve. 50 or 100 cells were scored for these three samples irradiated at higher doses (2–10 Gy). The blood sample from another donor (female; age range 45–54 years) was used for a fresh blind test with three additional doses. 200 cells were scored for each of these lower doses (0.2, 0.9, and 2.2 Gy).

To maintain confidentiality, coded blood samples from patients undergoing RT at Royal Marsden Hospital were sent overnight to the UK Health Security Agency (UKHSA) and were then processed in the same way as samples for the calibration curve construction without further exposure to radiation. Participants were all over the age of 18 years with (i) no previous RT, (ii) no concurrent chemotherapy, hormone, or biological therapy, and (iii) no chemotherapy, hormone, or biological therapy preceding RT by less than 4 weeks. IRAS258794/CCR5082 RTGene 2 was approved by Wales Research Ethics Committee 7 (19/WA/0147) and registered with ClinicalTrials.gov (NCT03809377). Clinical information including tumor type, gender, prescribed target dose, and number of fractions is provided in the Supplementary Table (for all online suppl. material, see https://doi.org/10.1159/000534656). Chemotherapy was only carried out for patient RTG-12, but not for the other 4 patients 4 weeks prior to RT. Age range for these five patients was 50–83 years. Two blood samples were analyzed for each patient: a pre-RT control (1) and a post-RT sample (3) taken before the last radiation fraction. (1) and (3) were used to label these samples in consistence with other studies of the RTGene project. The plan was to score 200 cells for each sample; however, due to the limited availability of blood, fewer cells were scored for some of the samples, i.e., one pre-RT sample: RTG-12(1), and two post-RT samples: RTG-12(3) and 13(3). In parallel, cultures for DCA were setup and processed using standard methods [IAEA, 2011]. Digital images generated using the Metafer4 slide scanning system (MetaSystems, Germany) were used for scoring.

Statistics

The DoseEstimate software (version 5.1) is designed for calibration curve fitting and the calculation of doses using the statistical methods recommended by IAEA [Ainsbury and Lloyd, 2010]. This software was used to calculate the mean and standard error on aberrations per cell for each dose and to fit the combined curve. Furthermore, u test and the variance to mean ratio (Var/Mean) were used to determine whether the dispersion of excess objects followed a Poisson distribution. Values of u between ±1.96 are characteristic of a Poisson distribution [Papworth, 1975].

The two-sample t-test was used to test for difference between the 2 Gy data point in common to both the previous 0–10 Gy and the newly established 0–2 Gy calibration curves. The paired t-test was used to compare the dose estimates between the PCC and DCA methods before and after RT treatment. A one-sample t-test was carried out to compare the mean dose estimate of the donors with the dose estimates of the pre-RT patient samples.

Data used for the construction of the calibration curves generated by scoring excess objects as aberrations using calyculin A-induced PCC are shown in Table 1. The standard errors (SE) were adjusted for overdispersion for most dose points apart from 4, 6, 8, and 10 Gy. Overdispersion was also observed in one of the blind test samples (Y-2) (Table 2). Much higher degrees of overdispersion were observed in five patient samples (Table 3). For example, the highest u value for patient sample RTG-14(3) was 96.42, while for the healthy donor, the highest u value was 6.8.

Table 1.

Data generated for the construction of the calibration curve (Fig. 2) using calyculin A-induced PCC by scoring excess objects

Dose, GyCells scoredAberrationDistribution of excess objectsYield±SEVar/Mean ± SEu
01234567891011121314151617181920
1,500 98 1,410 82 0.065±0.008 1.1±0.036 2.72 
0.25 500 56 453 41 0.112±0.017 1.43±0.063 6.8 
0.5 500 89 431 53 13 0.178±0.022 1.39±0.063 6.14 
0.75 500 146 388 84 22 0.292±0.028 1.26±0.063 4.1 
200 87 140 35 23 0.435±0.055 1.24±0.1 2.39 
600 600 256 178 104 42 13 1±0.048 1.28±0.058 4.83 
200 584 17 23 47 50 28 19 2.92±0.14 1.12±0.1 1.2 
200 861 27 30 46 31 29 11 10 4.3±0.147 0.906±0.1 −0.936 
200 1,247 12 25 37 33 33 18 19 10 6.24±0.177 0.785±0.1 −2.15 
10 200 1,687 18 21 31 37 17 19 20 8.44±0.238 1.11±0.1 1.1 
Dose, GyCells scoredAberrationDistribution of excess objectsYield±SEVar/Mean ± SEu
01234567891011121314151617181920
1,500 98 1,410 82 0.065±0.008 1.1±0.036 2.72 
0.25 500 56 453 41 0.112±0.017 1.43±0.063 6.8 
0.5 500 89 431 53 13 0.178±0.022 1.39±0.063 6.14 
0.75 500 146 388 84 22 0.292±0.028 1.26±0.063 4.1 
200 87 140 35 23 0.435±0.055 1.24±0.1 2.39 
600 600 256 178 104 42 13 1±0.048 1.28±0.058 4.83 
200 584 17 23 47 50 28 19 2.92±0.14 1.12±0.1 1.2 
200 861 27 30 46 31 29 11 10 4.3±0.147 0.906±0.1 −0.936 
200 1,247 12 25 37 33 33 18 19 10 6.24±0.177 0.785±0.1 −2.15 
10 200 1,687 18 21 31 37 17 19 20 8.44±0.238 1.11±0.1 1.1 

Aberrations were scored as the total numbers of excess PCC objects at G2 and M phases.

SE, standard error. For 0 and 2 Gy, the numbers of cells scored at two dose ranges (0–2 Gy and 0–10 Gy) in two separate studies were combined to get 1,500 and 600 cells in total, respectively. No other dose was scored at two dose ranges.

Table 2.

Dose estimation and distribution of excess PCC objects in the blind tests

Sample ID (blind tests)Cells scored, nAberrations, nActual dose, GyDose estimate, GyLower 95% CLUpper 95% CLDistribution of excess objectsYield±SEVar/Mean ± SEu
01234567891011121314
X-1 100 156 2.4 2.176±0.180 1.824 2.528 19 34 27 12           1.56±0.125 0.872±0.142 −0.905 
Y-1 50 370 9.2 10.550±0.552 9.473 11.64 7.4±0.385 0.965±0.202 −0.172 
Z-1 50 215 5.6 6.107±0.421 5.282 6.932    4.3±0.293 1.502±0.202 1.439 
X-2 200 52 0.2 0.311±0.047 0.219 0.403 154 41            0.26±0.036 1.014±0.099 0.144 
Y-2 200 121 0.9 0.806±0.052 0.705 0.907 118 60 14         0.605±0.055 1.593±0.100 5.94 
Z-2 200 352 2.2 2.463±0.135 2.198 2.728 44 54 40 40 16        1.76±0.094 1.161±0.100 1.604 
Sample ID (blind tests)Cells scored, nAberrations, nActual dose, GyDose estimate, GyLower 95% CLUpper 95% CLDistribution of excess objectsYield±SEVar/Mean ± SEu
01234567891011121314
X-1 100 156 2.4 2.176±0.180 1.824 2.528 19 34 27 12           1.56±0.125 0.872±0.142 −0.905 
Y-1 50 370 9.2 10.550±0.552 9.473 11.64 7.4±0.385 0.965±0.202 −0.172 
Z-1 50 215 5.6 6.107±0.421 5.282 6.932    4.3±0.293 1.502±0.202 1.439 
X-2 200 52 0.2 0.311±0.047 0.219 0.403 154 41            0.26±0.036 1.014±0.099 0.144 
Y-2 200 121 0.9 0.806±0.052 0.705 0.907 118 60 14         0.605±0.055 1.593±0.100 5.94 
Z-2 200 352 2.2 2.463±0.135 2.198 2.728 44 54 40 40 16        1.76±0.094 1.161±0.100 1.604 

Dose estimates from two separate blind tests with 6 doses were used to evaluate the linear calibration curve covering 0–10 Gy.

Four dose estimates were within the 95% CL and the other two (Y-1 and X-2) were just outside.

Table 3.

Post-RT distribution of excess PCC objects in cancer patients undergoing radiotherapy

Sample IDCells scoredAberrationDistribution of excess objectsVar/Mean ± SEu
0123456789101112131415161718192022...2730
RTG-12(3) 67 86 27 19                     1.837±0.173 4.836 
RTG-13(3) 86 98 46 23                     6.322±0.153 34.87 
RTG-14(3) 200 159 144 36                 10.640±0.100 96.42 
RTG-24(3) 200 125 141 35 10                    4.719±0.100 37.24 
RTG-25(3) 200 306 108 43 22             7.685±0.100 66.79 
Sample IDCells scoredAberrationDistribution of excess objectsVar/Mean ± SEu
0123456789101112131415161718192022...2730
RTG-12(3) 67 86 27 19                     1.837±0.173 4.836 
RTG-13(3) 86 98 46 23                     6.322±0.153 34.87 
RTG-14(3) 200 159 144 36                 10.640±0.100 96.42 
RTG-24(3) 200 125 141 35 10                    4.719±0.100 37.24 
RTG-25(3) 200 306 108 43 22             7.685±0.100 66.79 

Overdispersion was observed in all five samples indicating partial-body exposure.

The t-test showed no significant difference between the 2 Gy yields from the two calibration curves (p = 0.748), and as such, it was judged possible to combine the two sets of data from the lower dose (0–2 Gy) and higher dose (0–10 Gy) calibration curves. Both a linear and a linear-quadratic calibration curve covering doses from 0 to 10 Gy were constructed and evaluated, and a linear fit to the curve was considered better in terms of more accurate dose estimation, although the statistical difference between this and a linear-quadratic fit was negligible.

The combined calibration curve was fitted to a linear model: Y = 0.0433 (+/− 0.0182) + 0.6970 (+/− 0.0584)*D, in which Y represents the yield of excess objects and D the dose (Fig. 2). For this curve, the p value for goodness of fit was <0.0001, and the p values for coefficients (z-test) were p_A = 0.0446, and p_alpha <0.0001. The correlation coefficient of the curve was r = 0.9959. Calculated using this curve, 4 out of 6 dose estimates were within the 95% confidence limits (95% CL) with the other 2 just outside (Table 2) in the blind validation tests.

Fig. 2.

The combined calibration curve covering 0–10 Gy was fitted to a linear model: Y = 0.0433 (+/− 0.0182) + 0.6970 (+/− 0.0584)*D, in which Y represents the yield of excess objects and D the dose. Bars represent standard error.

Fig. 2.

The combined calibration curve covering 0–10 Gy was fitted to a linear model: Y = 0.0433 (+/− 0.0182) + 0.6970 (+/− 0.0584)*D, in which Y represents the yield of excess objects and D the dose. Bars represent standard error.

Close modal

Blood samples from five cancer patients were also assessed to give a dose estimate using this combined calibration curve. Results showed that the dose estimates for the pre-treatment patient samples were significantly higher than the mean value for the healthy donors (mean = 0.032 Gy, p = 0.035) and were also significantly (p = 0.048) higher than those obtained using DCA (Table 4). For the post-RT patient samples (Table 5), the t-test comparison for the dose estimates generated using PCC showed no significant difference to the DCA data (p = 0.406) overall within the group of 5 patient samples. Importantly, our results also showed that the whole-body dose estimates for all five cancer patients before RT were below 1 Gy; therefore, these patients all fell into the same triage category as those with low dose exposure and do not require urgent treatment (Table 4).

Table 4.

PCC dose estimates for pre-RT patient samples with DCA results used in comparison

Patient samplesCalyculin A-induced PCCDCA
cells scored, nPCC objects, nyield±SEdose estimate (Gy)±SElower 95% CLupper 95% CLcells scored, ndicentrics, nyield±SEdose estimate (Gy)±SElower 95% CLupper 95% CL
RTG-12(1) 94 27 0.287±0.055 0.350±0.069 0.216 0.484 500 0±0.025 0.038 
RTG-13(1) 200 47 0.235±0.034 0.275±0.045 0.186 0.364 500 0±0.025 0.038 
RTG-14(1) 200 37 0.185±0.03 0.203±0.042 0.121 0.286 500 0.008±0.004 0.137±0.066 0.007 0.266 
RTG-24(1) 200 33 0.165±0.029 0.175±0.040 0.095 0.254 500 0±0.025 0.038 
RTG-25(1) 200 110 0.55±0.052 0.727±0.053 0.624 0.83 500 0±0.025 0.038 
Patient samplesCalyculin A-induced PCCDCA
cells scored, nPCC objects, nyield±SEdose estimate (Gy)±SElower 95% CLupper 95% CLcells scored, ndicentrics, nyield±SEdose estimate (Gy)±SElower 95% CLupper 95% CL
RTG-12(1) 94 27 0.287±0.055 0.350±0.069 0.216 0.484 500 0±0.025 0.038 
RTG-13(1) 200 47 0.235±0.034 0.275±0.045 0.186 0.364 500 0±0.025 0.038 
RTG-14(1) 200 37 0.185±0.03 0.203±0.042 0.121 0.286 500 0.008±0.004 0.137±0.066 0.007 0.266 
RTG-24(1) 200 33 0.165±0.029 0.175±0.040 0.095 0.254 500 0±0.025 0.038 
RTG-25(1) 200 110 0.55±0.052 0.727±0.053 0.624 0.83 500 0±0.025 0.038 

PCC dose estimates for pre-RT patient samples were significantly (p = 0.048) higher than those generated using DCA.

Table 5.

PCC dose estimates for post-RT patient samples with DCA results used in comparison

Patient samplesCalyculin A-induced PCCDCA
cells scored, nPCC objects, nyield±SEdose estimate (Gy)±SElower 95% CLupper 95% CLcells scored, ndicentrics, nyield±SEdose estimate (Gy)±SElower 95% CLupper 95% CL
RTG-12(3) 67 86 1.28±0.138 1.779±0.199 1.389 2.17 461 100 0.217±0.022 1.505±0.087 1.334 1.676 
RTG-13(3) 86 98 1.14±0.115 1.573±0.166 1.248 1.898 343 100 0.292±0.029 1.792±0.097 1.601 1.982 
RTG-14(3) 200 159 0.795±0.063 1.078±0.043 0.993 1.164 500 48 0.096±0.014 0.909±0.085 0.743 1.075 
RTG-24(3) 200 125 0.625±0.056 0.835±0.051 0.734 0.935 500 62 0.124±0.016 1.069±0.085 0.903 1.236 
RTG-25(3) 200 306 1.53±0.087 2.133±0.126 1.885 2.381 500 38 0.076±0.012 0.781±0.084 0.615 0.946 
Patient samplesCalyculin A-induced PCCDCA
cells scored, nPCC objects, nyield±SEdose estimate (Gy)±SElower 95% CLupper 95% CLcells scored, ndicentrics, nyield±SEdose estimate (Gy)±SElower 95% CLupper 95% CL
RTG-12(3) 67 86 1.28±0.138 1.779±0.199 1.389 2.17 461 100 0.217±0.022 1.505±0.087 1.334 1.676 
RTG-13(3) 86 98 1.14±0.115 1.573±0.166 1.248 1.898 343 100 0.292±0.029 1.792±0.097 1.601 1.982 
RTG-14(3) 200 159 0.795±0.063 1.078±0.043 0.993 1.164 500 48 0.096±0.014 0.909±0.085 0.743 1.075 
RTG-24(3) 200 125 0.625±0.056 0.835±0.051 0.734 0.935 500 62 0.124±0.016 1.069±0.085 0.903 1.236 
RTG-25(3) 200 306 1.53±0.087 2.133±0.126 1.885 2.381 500 38 0.076±0.012 0.781±0.084 0.615 0.946 

Dose estimates for post-RT patient samples generated using PCC were statistically (p = 0.406) comparable to the results obtained from DCA.

Radiological emergencies involving nuclear power plant accidents, the use of nuclear weapons, or terrorist attacks can result in mass casualty situations whereby a large number of individuals are exposed or are suspected to have been exposed. The estimation of the radiation dose of potentially exposed individuals using cytogenetic approaches can assist health workers to quickly triage those who require urgent medical treatment and/or monitoring for longer term health effects from those who are not at risk. As a well-established method in biodosimetry, DCA is more accurate for doses below 2 Gy than the calyculin A-induced PCC. PCC is therefore not preferred for this dose range, but rather it is suggested as an alternative method, especially when it is difficult to obtain sufficient number of cells to score for the elderly and those with pathological conditions [Gotoh and Durante, 2006; Hatzi et al., 2006; M'Kacher et al., 2023]. Using calyculin A, PCC can be induced with high efficiency in terms of a much higher number of cells to score in comparison to DCA using the same amount of blood [Sun et al., 2020a].

A high level of overdispersion was seen in all five cancer patient samples suggesting a partial-body nature of exposure even though overdispersion was also observed in some of the samples for calibration curve fitting and one in the blind evaluation tests. The cause for the overdispersion of excess objects in the blood of healthy volunteers is unclear at present. However, overdispersion for the distribution of excess acentrics among cells is common [Schmid and Bauchinger, 1980; Virsik and Harder, 1981]. Virsik and Harder [Virsik and Harder, 1981; Cornforth and Goodwin, 1991] suggested an aberration mechanism in which overdispersion of acentrics occurs when more than one acentric is formed simultaneously. Cornforth and Goodwin [Cornforth and Goodwin, 1991] suggested that overdispersion appears to be a general feature of high LET (e.g., 238Pu alpha-particle) radiation-induced PCC fragmentation, assuming that single particle traversals are capable of producing multiple fragments. As discussed below, the formation of excess PCC objects can result from multiple factors; therefore, it is highly likely that the aberration mechanism can result in simultaneous formation of more than one object in the cell and thus cause overdispersion. It should be noted that high-LET radiation is mainly referenced to assist with the explanation as X-rays are low-LET radiation.

Blood is constantly circulating in the body, and only a small proportion of the blood cell population is exposed to the external beam in certain selected areas for the individual treatment fraction. It is possible that some cells are hit by the beam more than once and sustain heavy damage to the genetic material. Therefore, there are a very small number of heavily damaged lymphocytes randomly distributed in the blood of the partially exposed patients, which could partly explain the overdispersion of PCC objects in the patient samples. Similarly, overdispersion of dicentrics was also observed in DCA results. In addition, it is possible that damaged T-lymphocytes may reside in the lymph nodes in the treatment field during several fractions and then enter the circulation, which may subsequently contribute to overdispersion.

The background frequency of excess objects for calyculin A-induced PCC is higher than the background level of dicentrics in DCA. A background level of approximately 4–6% excess fragments has been reported [Balakrishnan et al., 2010; Puig et al., 2013; Sun et al., 2020b] for chemically induced PCC. In comparison, the spontaneous incidence of dicentrics is approximately 0–2 in 1,000 cells [IAEA, 2011]. Dicentric chromosomes are rare events as the result of mis-repaired DNA double-strand breaks [IAEA, 2011], while chemicals such as calyculin A, aphidicolin [Achkar et al., 2005], and bleomycin [Bolzán and Bianchi 2018] can all induce single- and double-strand breaks in DNA and may subsequently lead to the formation of PCC fragments. Exposure to environmental clastogens and aneugens (increasing with age) can also lead to the fragmentation of chromosomes [Alhmoud et al., 2020]. Hitherto, no population study has been carried out to assess the effects of other biological or environmental factors on PCC fragmentation, such as age, alcohol intake, smoking status, and occupational hazards. Clinical treatments (e.g., chemotherapy and CT scan) may also cause the formation of excess PCC fragments. As most mass casualty accidents will be caused by gamma rays, a future study assessing the effects of gamma radiation on the number of excess PCC objects would also be beneficial.

Importantly, the numbers of excess PCC objects for the pre-RT patient samples (Table 4) were found to be much higher than the mean value for the healthy donors. The higher pre-RT frequency of PCC objects in the cancer patients may be attributable to age (mean = 71 years). It is also possible that the patients involved in this study had been exposed to chemotherapy as well as repeated diagnostic radiology over 4 weeks prior to RT, such as CT scans and PET-CTs. Another plausible cause for this difference is that calyculin A induces chromosome damage at common fragile sites (CFSs) after perturbation of the replication dynamics [Achkar et al., 2005]. CFSs instability could be responsible for chromosome rearrangements and are frequently correlated with cancers [Glover et al., 2017; Ma et al., 2012]. Our results suggest that the CFSs of cancer patients may be more prone to calyculin A-induced breaks than healthy donors. It would be worthwhile to investigate the effect of calyculin A on the alteration of CFSs in cancer patients. Our analysis found that dose estimates for the post-RT cancer patient samples were statistically comparable to those from DCA. However, sample RTG-25(3) showed an unexpectedly higher dose estimate. The pre-RT sample for this patient, RTG-25(1), also showed a higher level of excess objects. Because this patient did not have any chemo/biological treatment 4 weeks prior to RT, it is possible that she may have genomic instability associated with CFSs, which manifested as an increased number of chromosomal breaks in the PCC-inducing procedure.

For triage dosimetry, the goal is to assign the patients with suspected overexposure into the appropriate category quickly and correctly to advice on medical interventions. In the present study, the whole-body dose estimates for all five cancer patients prior to RT were found to be below 1 Gy and thus can be allocated into the same triage category as those who have received a small dose but do not need urgent treatment. Therefore, calyculin A-induced PCC can potentially serve the purpose for triage categorization in mass casualty accidents or terrorist attacks, but further work will be needed. Even though it is beyond the scope of the present study, further information may need to be included in future studies with valid control samples, such as the type and stage of cancer, the type of irradiation facility, the dose used in each RT fraction as well as the gap between fractions as these may have significant impact on dose estimation.

In conclusion, the point of introducing this biodosimetry method is to eliminate the time-consuming identification of different types of aberrations so that the scoring can potentially be done by inexperienced workers in case of large-scale nuclear emergency. The simplicity in scoring may also enable the automation of the scoring procedure. Further work will be required to understand the issue of overdispersion as well as individual variability in background samples with age and other confounding factors taken into consideration. For unexposed cancer patients in similar circumstances, this assay may not be applicable to identify these individuals in radiation emergencies. However, in this exploratory study, it has been demonstrated that PCC is a valuable approach with the potential to complement or be used as an alternative to the DCA. Particularly, owing to its high induction efficiency for scorable cells, PCC can potentially be applied when the availability of blood is extremely limited or when the suspected overexposure is higher than 5 Gy and the DCA method may fail to produce sufficient metaphase spreads for analysis.

We thank all the blood donors at UKHSA, all the patients and staff who participated in the study from the Royal Marsden NHS Foundation Trust, Sutton, in particular, Dr. Susan Lalondrelle (endometrium), Dr. Shaista Hafeez (bladder), and Dr. Shree Bhide (head and neck) for recruiting patients into this study.

The research was conducted in compliance with internationally accepted ethical standards for research practice and reporting. Written informed consent and the approval of the West Midlands-Solihull Research Ethics Committee (REC 14/WM/1182) were obtained for the healthy donors. The clinical trial, IRAS258794/CCR5082 RTGene 2, was approved by the Wales Research Ethics Committee 7 (19/WA/0147) and registered with ClinicalTrials.gov (NCT03809377).

The views expressed are those of the authors and not necessarily those of the NIHR, the Department of Health or UKHSA. The authors have no relevant financial or non-financial interests to disclose. The authors have no competing interests to declare that are relevant to the content of this article.

We acknowledge NHS funding to the National Institute for Health Research (NIHR) Biomedical Research Center at the Royal Marsden and the Institute of Cancer Research (ICR). This work was partly supported by the NIHR Health Protection Research Unit (NIHR HPRU) in Chemical & Radiation Threats & Hazards at Imperial College London in partnership with UK Health Security Agency (UKHSA).

Elizabeth Ainsbury, Navita Somaiah, and Jayne Moquet contributed to the funding application and the setting up of the RTGene 2 project. Mingzhu Sun and Jayne Moquet conceived of the presented idea and performed the analytic calculations. Jayne Moquet supervised the findings of this work and carried out the experiments using DCA. Mingzhu Sun carried out the experiments using PCC and wrote the manuscript with support from all authors. Elizabeth Ainsbury verified the analytical methods and contributed to the interpretation of the results. Selvakumar Anbalagan, Harriet Steel, Aurore Sommer, and Lone Gothard contributed to patient recruitment, sample collection, and transport. David Lloyd provided critical review for the draft manuscript. Stephen Barnard provided technical assistance for the project. All authors provided critical feedback to help shape the research and analysis. All authors discussed the results and contributed to the final version of the manuscript.

All datasets on which the conclusions of the paper rely are available to editors, reviewers, and readers without unnecessary restriction wherever possible. All data generated or analyzed during this study are included in this article. Further inquiries can be directed to the corresponding author.

1.
El Achkar E, Gerbault-Seureau M, Muleris M, Dutrillaux B, Debatisse M. Premature condensation induces breaks at the interface of early and late replicating chromosome bands bearing common fragile sites. Proc Natl Acad Sci U S A. 2005;102(50):18069–74.
2.
Ainsbury EA, Al-Hafidh J, Bajinskis A, Barnard S, Barquinero JF, Beinke C, et al. Inter- and intra-laboratory comparison of a multibiodosimetric approach to triage in a simulated, large scale radiation emergency. Int J Radiat Biol. 2014;90(2):193–202.
3.
Ainsbury EA, Lloyd DC. Dose estimation software for radiation biodosimetry. Health Phys. 2010;98(2):290–5.
4.
Albert JM. Radiation risk from CT: implications for cancer screening. AJR Am J Roentgenol. 2013;201(1):W81–7.
5.
Alhmoud JF, Woolley JF, Al Moustafa AE, Malki MI. DNA damage/repair management in cancers. Cancers. 2020;12(4):1050.
6.
Balakrishnan S, Shirsath K, Bhat N, Anjaria K. Biodosimetry for high dose accidental exposures by drug induced premature chromosome condensation (PCC) assay. Mutat Res. 2010;699(1-2):11–6.
7.
Bolzán AD, Bianchi MS. DNA and chromosome damage induced by bleomycin in mammalian cells: an update. Mutat Res Rev Mutat Res. 2018;775:51–62.
8.
Cornforth MN, Goodwin EH. The dose-dependent fragmentation of chromatin in human fibroblasts by 3.5-MeV a particles from 238 Pu: experimental and theoretical considerations pertaining to single-track effects. Radiat Res. 1991;127(1):64–74.
9.
Darroudi F, Natarajan AT, Bentvelzen PAJ, Heidt PJ, Van Rotterdam A, Zoetelief J, et al. Detection of total- and partial-body irradiation in a monkey model: a comparative study of chromosomal aberration, micronucleus and premature chromosome condensation assays. Int J Radiat Biol. 1998;74(2):207–15.
10.
Glover TW, Wilson TE, Arlt MF. Fragile sites in cancer: more than meets the eye. Nat Rev Cancer. 2017;17(8):489–501.
11.
González JE, Romero I, Gregoire E, Martin C, Lamadrid AI, Voisin P, et al. Biodosimetry estimation using the ratio of the longest:shortest length in the premature chromosome condensation (PCC) method applying autocapture and automatic image analysis. J Radiat Res. 2014;55(5):862–5.
12.
Gotoh E, Durante M. Chromosome condensation outside of mitosis: mechanisms and new tools. J Cell Physiol. 2006;209(2):297–304.
13.
Gotoh E, Tanno Y. Simple biodosimetry method for cases of high-dose radiation exposure using the ratio of the longest/shortest length of Giemsa-stained drug-induced prematurely condensed chromosomes (PCC). Int J Radiat Biol. 2005;81(5):379–85.
14.
Gotoh E, Tanno Y, Takakura K. Simple biodosimetry method for use in cases of high-dose radiation exposure that scores the chromosome number of Giemsa-stained drug-induced prematurely condensed chromosomes (PCC). Int J Radiat Biol. 2005;81(1):33–40.
15.
Guerrero-Carbajal C, Romero-Aguilera I, Arceo-Maldonado C, Gonzalez-Mesa JE, Cortina-Ramirez GE, Garcia-Lima O. Arceo-Maldonado C, Gonzalez-Mesa JE, Cortina-Ramirez GE, Garcia-Lima O: dose response of prematurely condensed chromosome rings after gamma irradiation. Int J Radiat Biol. 2019;95(5):607–10.
16.
Hatzi VI, Terzoudi GI, Paraskevopoulou C, Makropoulos V, Matthopoulos DP, Pantelias GE. The use of premature chromosome condensation to study in interphase cells the influence of environmental factors on human genetic material. Sci World J. 2006;6:1174–90.
17.
Hayata I, Kanda R, Minamihisamatsu M, Furukawa M, Sasaki MS. Cytogenetical dose estimation for 3 severely exposed patients in the JCO criticality accident in tokai-mura. J Radiat Res. 2001;42(Suppl l):S149–55.
18.
IAEA. Cytogenetic Dosimetry: Applications in Preparedness for and Response to Radiation Emergencies. 247International Atomic Energy Agency; 2011.
19.
Jaworska A, Ainsbury EA, Fattibene P, Lindholm C, Oestreicher U, Rothkamm K, et al. Operational guidance for radiation emergency response organisations in Europe for using biodosimetric tools developed in EU MULTIBIODOSE project. Radiat Prot Dosimetry. 2015;164(1-2):165–9.
20.
Lamadrid AI, García O, Delbos M, Voisin P, Roy L. PCC-ring induction in human lymphocytes exposed to gamma and neutron irradiation. J Radiat Res. 2007;48:1–6.
21.
M’Kacher R, Colicchio B, Junker S, El Maalouf E, Heidingsfelder L, Plesch A, et al. High resolution and automatable cytogenetic biodosimetry using in situ telomere and centromere hybridization for the accurate detection of DNA damage: an overview. Int J Mol Sci. 2023;24(6):5699.
22.
Ma K, Qiu L, Mrasek K, Zhang J, Liehr T, Quintana LG, et al. Common fragile sites: genomic hotspots of DNA damage and carcinogenesis. Int J Mol Sci. 2012;13(9):11974–99.
23.
Miura T, Nakata A, Kasai K, Nakano M, Abe Y, Tsushima E, et al. A novel parameter, cell-cycle progression index, for radiation dose absorbed estimation in the premature chromosome condensation assay. Radiat Prot Dosimetry. 2014;159(1-4):52–60.
24.
Moquet J, Higueras M, Donovan E, Boyle S, Barnard S, Bricknell C, et al. Dicentric dose estimates for patients undergoing radiotherapy in the RTGene study to assess blood dosimetric models and the new bayesian method for gradient exposure. Radiat Res. 2018;190(6):596–604.
25.
Moquet J, Rothkamm K, Barnard S, Ainsbury E. Radiation biomarkers in large scale human health effects studies. J Pers Med. 2020;10(4):155.
26.
Papworth D. Curve fitting by maximum likelihood. Appendix to paper by JRK Savage: radiation-induced chromosomal aberrations in plant Tradescantia: dose response curves. Radiat Bot. 1975;15:127–31.
27.
Puig R, Barrios L, Pujol M, Caballín MR, Barquinero JF. Suitability of scoring PCC rings and fragments for dose assessment after high-dose exposures to ionizing radiation. Mutat Res. 2013;757:1–7.
28.
Pujol M, Barquinero JF, Puig P, Puig R, Caballín MR, Barrios L. A new model of biodosimetry to integrate low and high doses. PLoS One. 2014;9(12):e114137.
29.
Romero I, Lamadrid AI, González JE, Mandina T, García O. Culture time and reagent minimization in the chemical PCC assay. Int J Radiat Biol. 2016;92(10):558–62.
30.
Schmid E, Bauchinger M. Analysis of primary processes in the foramtion of acentric fragments. Radiat Environ Biophys. 1980;17(2):143–9.
31.
Stewart F, Akleyev A, Hauer-Jensen M, Hendry J, Kleiman N, MacVittie T, et al. Icrp publication 118: ICRP statement on tissue reactions and early and late effects of radiation in normal tissues and organs–threshold doses for tissue reactions in a radiation protection context. Ann ICRP. 2012;41(1-2):1–322.
32.
Sun M, Moquet J, Barnard S, Lloyd D, Ainsbury E. Scoring rings in the cell fusion-induced premature chromosome condensation (PCC) assay for high dose radiation exposure estimation after gamma-ray exposure. Int J Radiat Biol. 2019;95(9):1259–67.
33.
Sun M, Moquet J, Barnard S, Lloyd D, Ainsbury E. A simplified calyculin A-induced premature chromosome condensation (PCC) protocol for the biodosimetric analysis of high-dose exposure to gamma radiation. Radiat Res. 2020a;193(6):560–8.
34.
Sun M, Moquet J, Lloyd D, Ainsbury E. A faster and easier biodosimetry method based on calyculin A-induced premature chromosome condensation (PCC) by scoring excess objects. J Radiol Prot. 2020b;40(3):892–905.
35.
Virsik RP, Harder D. Analysis of radiation-induced acentric fragments in human G0 lymphocytes. Radiat Environ Biophys. 1981;19(1):29–40.
36.
Wang ZZ, Li WJ, Zhi DJ, Jing XG, Wei W, Gao QX, et al. Biodosimetry estimate for high-LET irradiation. Radiat Environ Biophys. 2007;46(3):229–35.