An evaluation of DNA double strand break formation and excreted guanine species post whole body PET/CT procedure

Abstract Ionizing radiation-induced oxidation and formation of deoxyribonucleic acid (DNA) double strand breaks (DSBs) are considered the exemplar of genetic lesions. Guanine bases are most prone to be oxidized when DNA and Ribonucleic acid (RNA) are damaged. The repair processes that are initiated to correct this damage release multiple oxidized guanine species into the urine. Hence, the excretion of guanine species can be related with the total repair process. Our study quantified the total DSBs formation and the amount of guanine species in urine to understand the DNA break and repair process after whole body (WB) exposure to 18F-FDG positron emission tomography/computed tomography (PET/CT). A total of 37 human participants were included with control and test groups and the average radiation dose was 27.50 ± 2.91 mSv. γ-H2AX foci assay in the collected blood samples was performed to assess the DSBs, and excreted guanine species in urine were analyzed by a competitive ELISA method. We observed a significant increase of DNA damage that correlated well with the increasing dose (p-value 0.009) and body weight (p-value 0.05). In the test group, excreted guanine species in urine sample significantly increased (from 24.29 ± 5.82 to 33.66 ± 7.20 mg/mmol creatinine). A minimum (r2 = 0.0488) correlation was observed between DSBs formation and excreted guanine species. A significant difference of DNA damage and 8-OHdG formation was seen in the test group compared to controls. Larger population studies are needed to confirm these observations, describe the fine-scale timing of changes in the biomarker levels after exposure, and further clarify any potential risks to patients from PET/CT procedures.


INTRODUCTION
There is always an increasing demand for oncologic diagnosis and management based on the results of positron emission tomography (PET)/computed tomography (CT) scanning. Hospitals and clinics are also installing more PET/CT scanners because of the versatility in its clinical applications [1]. PET uses a small amount of radioactive material called radiotracer which is detect by the scanner, and the acquired images are superimposed with X-ray-based CT scans for a better understanding of the a ected area of the body. As PET/CT is a combination of two di erent modalities, the e ective radiological dose is also a combination of the dose of PET and the dose of CT, resulting in an increased radiation exposure to the patients compared to single modality alone.
Currently, the use of whole body (WB) 2-[ 18 F] uoro-2 deoxy-D-glucose ( 18 F-FDG)-PET/CT is becoming increasingly popular in the diagnosis of abnormal metabolic activity, and other many types of disease processes [2]. As with any radiation exposure, the medical use of 18 F-FDG PET/CT raises concerns regarding the  possibility of radiation-induced toxicity [3,4]. In general, PET/CT involves intravenous injection of 10 mCi, which yields 350 MBq of 18 F-FDG and imaging initiated an hour later. The average yearly natural background radiation dose typically ranges from about 1.5 to 3.5 mSv/year, but a single PET/CT scan strategy can provide an e ective dose of 15 mSv, equivalent to 6.4 years of background radiation [5,6]. Ionizing radiation induced DNA double-strand breaks (DSBs) formation represents the most biologically deleterious type of lesion and results in persistent malfunctioning of cells, and is thought to induce both mutagenesis and carcinogenesis [7,8]. Reported epidemiological data from exposed human populations demonstrates that doses of 50-100 mSv in protracted exposure or 10-50 mSv acute exposure to ionizing radiation increases the risk of some cancers [9]. These epidemiological data cannot always provide a risk estimate below these levels [10], and are based at least partly on linear extrapolations from existing high-dose data [11][12][13]. With a linear, no-threshold hypothesis based on mechanistic considerations, low-dose hypersensitivity, delayed genomic instability and induced DNA repair mechanisms are important to measure but di cult to reconcile [14][15][16][17][18][19]. In vitro studies have been helpful in this regard, but human equivalence needs further exploration.
Our previous study observed that 18 F-FDG induced DSBs formation in isolated peripheral blood lymphocytes and chromosomal aberrations in a V79 cell line [20,21]. We also observed DNA breaks by comet and micronucleus assay a er exposure to PET/CT procedure [22]. In the present study, we extended that work through quantifying the DSBs formation in patients' blood samples by γ -H2AX assay, and assessed the total DNA RNA oxidative damage burden by quantifying urinary 8-hydroxy-2 ′ -deoxyguanosine (from DNA), 8hydroxyguanosine (from RNA) and 8-hydroxyguanine (DNA and RNA) levels. As DNA and RNA become oxidized due to radiation exposure, and guanine is most prone to oxidation, the repair processes initiated to correct the damage release multiple oxidized guanine species in urine. Hence by measuring the excretion of guanine species one can also quantify the total repair process. Therefore, we sought to ll the above gaps in knowledge by quantifying total DSBs formation and the amount of guanine species in urine to understand the DNA break and repair process a er WB 18 F-PET/CT.

Study population and recruitment
Participants were recruited at the Institute of Nuclear Medicine & Molecular Imaging, AMRI Hospitals, Dhakuria, Kolkata, between February 2019 and November 2019. The Institutional Ethical Committee approved this study and all patients provided written informed consent. A total of 37 participants were recruited, including those who volunteered for the control group ( ve male, four female; mean age 40.11 ± 7.14, range 28-46) without PET/CT procedure, and 28 volunteer in the test group (14 male, 14 female; mean age 44.5 ± 10.48, range 22-59) who visited our center for cancer staging and evaluation of pyrexia of unknown origin (PUO) and underwent the PET/CT procedure. The characteristics of the study population are summarized in Supplementary Table S1. The average received dose by the test group was 27.50 ± 2.91 mSv.

Exclusion criteria
Participants with a history of smoking, diabetes, or cardiopulmonary disease were excluded. Also, volunteers with renal failure, those unwilling to provide written informed consent, those who had received contrast media during any other diagnostic investigation within the week prior to this study, or had received any treatment of chemotherapy/ radiotherapy and/or were pregnant were excluded from the study.

Research participants preparatory (pre-imaging) procedure
Patients were instructed to fast overnight, avoid any kind of physical or stress activity and drink one liter of water 1 h before the examination. An intravenous cannula was inserted in the patient's vein for an injection of 18 F-FDG before the examination. All patients were asked to sit in a post-injection waiting room and restrict their physical activity, including talking, to an absolute minimum.

PET/CT imaging protocol
One hour before the PET/CT procedure 8 MBq/kg 18 F-FDG was injected intravenously, patients were instructed to empty their bladder in the radioactive toilet before the scan in the GE Discovery 690 PET/CT scanner [23]. Patients were positioned inside the scanner for contrast-enhanced WB PET/CT examination, and were injected with Iohexol, Omnipaque 300 based on the patient body weight at dose of 84.57 ± 16.69 ml and ow rate of 2-4 ml/s (GE Healthcare; Shanghai co., Ltd. Shanghai, China) [24]. A scout scan (in which the patient rst undergoes an 'overview' or 'scout' scan procedure during which X-ray projection data are obtained to identify the axial extent of the CT and PET study), breath-holding lung CT, WB CT and PET (from head to mid-thigh) and low-dose cine CT (provides cross-sectional millisecond tomography that is used for the correction of motion due to diaphragmatic movement) were obtained in an arms-up position and the imaging parameters for WB CT were followed according to our previous study [22]. A er completion of CT, the attenuated corrected emission images were acquired for 2 min per bed position in a 3-dimensional mode for a total acquisition time of approximately 15-20 mins.

Internal dose estimation
Absorbed dose (DT) to a tissue or organ (T) resulting from the intravenous injection of radioactivity (A) of 18 F-FDG was estimated by using the dose coe cients as recommended by the International Commission on Radiological Protection (ICRP) in Publication 106 (ICRP 2007) for di erent organs and tissues of the adult MIRD phantom. The following equations were used: where Γ FDG T is dose coe cient recommended by ICRP inpublication 106 for variety of organs and tissue.
The e ective dose (E) from 18 F-FDG WB PET was calculated as follows: . where W T is Tissue weighting factor provided by ICRP Publication 103, and Γ FDG E is dose coe cient recommended by ICRP in publication 106 for the WB [25].

External dose estimation
To determine the radiation dose of patients resulting from CT scans, volume CT dose index (CTDI vol ) was used. The radiation dose CT component was estimated from dose length product (DLP). DLP was estimated from volumetric CTDI vol and the scan length for each patient. Values of CTDI vol and DLP were directly obtained from the display screen of the PET/CT operating console. DLP was calculated from CTDI vol as: The total e ective dose (ED) from the CT scan was estimated using the sex-speci c conversion factors k (mSv/mGy cm) [26], where:

Blood, urine sample collection and lymphocytes separation
Blood (in 4 ml heparin tube container) and urine (in 30 ml plastic container) samples were collected from the control group at only one single point of time. Blood samples from the test group were collected at three di erent time points: before PET/CT, immediately a er PET/CT and 120 min a er the PET/CT examination. Likewise, urine samples were collected at three di erent points of time: before PET/CT, 120 min a er PET/CT and at 24 h later or early morning void. The rst voided morning specimen (immediately a er waking) is particularly valuable because it provides a time average for biomarker concentrations that may occur during the hours of sleep (approximately eight hours) and also over the course of the day, as the composition and concentration of urine changes continuously [27].

Estimation of excreted guanine species
DNA/RNA oxidative damage was analyzed by the quantication of urinary 8-hydroxy-2 ′ -deoxyguanosine (from DNA), 8-hydroxyguanosine (RNA) and 8-hydroxyguanine (DNA and RNA) levels. The assay was performed by a specialized ELISA kit purchased from Cayman Chemical, USA. The protocol is based on the competitive analysis between oxidatively damaged guanine species and an 8-OHdG-acetylcholinesterase conjugate for a limited amount of DNA/RNA oxidative damage-detecting monoclonal antibody. Because the amount of tracer is held constant while the concentration of oxidatively damage guanine varies, the amount of tracer that can bind to the monoclonal antibody will be inversely proportional to the concentration of oxidatively damaged guanine in the well. The antibody-oxidatively damaged guanine complex binds to the goat polyclonal anti-mouse IgG that has been previously attached to the well. The product of this enzymatic reaction has a distinct yellow color and absorbs strongly at 412 nm. The assay has a detection range from 10.3-3,000 pg/ml and a sensitivity of approximately 30 pg/ml. The level was normalized with creatinine (Cayman Chemical) [28].

Statistical analysis
Statistical analysis was performed using Student's t-test for test vs control group comparisons. Reported data are represented as means ± SEM for three independent experiments. Linear regression analysis was performed using OriginPro 8 so ware. We considered di erences statistically signi cant at p < 0.05.

Quanti cation of DNA double strand breaks
The mean number of DSB foci in the control group, and for the test group at time points before PET/CT, immediately a er PET/CT and a er 2 h sample group, were 0.21 ± 0.15, 0.30 ± 0.20, 0.85 ± 0.33 and 0.67 ± 0.27 respectively (Fig. 1). A signi cant increase of mean γ H2AX foci formation observed with the total received dose in the treatment group a er immediate sampling (Fig. 2a) and was signicantly larger than the control group (p-value < 0.05). Inter-individual variation was also observed in γ -H2AX foci formation (Fig. 7a) at each point of time. The results also showed that the mean γ H2AX foci formation increased in relation to the body weight of the patients (Fig. 2b), but was inversely related to the patient's age ( Fig. 2c and  Fig. 3).

Estimation of excreted 8-Hydroxydeoxyguanosine quanti cation
In the nine control samples, the average level of normalized urinary 8-OHdG/creatinine (mg/mmol) was 21.44 ± 3.83 mg/mmol creatinine, whereas in the 28 treated samples the 8-OHdG level was signi cantly increased to 33.66 ± 7.20 mg/mmol creatinine; in the latter samples there was also a dose-response relationship. Before PET/CT and in the 2 h a er PET/CT, the 8-OHdG was 24.29 ± 5.82 and 27.27 ± 6.51 mg/mmol creatinine respectively (Fig. 4). Interindividual variation was also observed (Fig. 7b). The formation of 8-OHdG increased with the increasing of total received dose during PET/CT procedure (Fig. 5a) and was also increased in association with increasing body weight and age (Figs 5b and c). No signi cant relationship was observed between γ H2AX foci formation and 8-OHdG in the treated samples ( Fig. 8 and Supplementary Fig. Fig. S2).

DISCUSSION
Imaging procedures like CT, single-photon emission CT (SPECT) and PET is highly useful diagnostic modalities in the health sector, but their widespread use has also raised concerns about possible adverse e ects of radiation [29]. Our previous studies, both in vitro and in vivo, have revealed that a signi cant amount of DSBs is generated a er contrastenhanced PET/CT examination [20,21,22]. Another study by Prasad et al. [30] reported that conventional 18 F-FDG-based PET/CT scanning induces DNA damage and chromosomal aberrations in blood lymphocytes. There are no such studies that investigated the in uence of 18 F-FDG PET/CT scanning and DNA DSBs on urinary excretion of 8-OHdG concentration levels. In the present study we quanti ed the DSB generation by γ H2AX method, which is very sensitive biomarker, and quanti ed the formation of 8-OHdG. The control group samples showed a low but measurable level of foci formation, because DSBs can be caused naturally by a variety of factors like reactive oxygen species (ROS), metabolic processes, de cient repair, programmed biological processes and other causes [31]. The number of γ H2AX foci increased markedly in the test group a er PET/CT scanning. We also observed that the immediately-collected sample in the test group showed a higher frequency of foci formation than the sample collected 2 h later, possibly re ecting host repair response mechanisms that rapidly reduced the detectable foci formation [32].
Since WB PET/CT is followed by considerable amount of radiation dose, risk-bene t ratios should be carefully weighed prior to every investigation. In our in vivo study, the mean γ H2AX foci increased proportionally with the increasing total received dose, and the accumulation of even a small amount of mis-repaired DNA damage may be carcinogenic. Previously many studies have also reported the relationships between radiation exposure by PET/CT and formation of γ H2AX in lymphocytes [30,33,34]. Prasad et al. reported that, compared to conventional CT, the 18 F-FDG PET/CT scanning instrument delivers higher doses to patients, inducing γ H2AX foci in blood lymphocytes [30]. Cheezum et al. [33] reported the appearance of γ H2AX foci in 101 female patients who had undergone coronary CT angiography, and a very recent study by Schumann et al. [34] showed that even at very low absorbed doses to the blood of less than 3 mGy, the number of γ H2AX in the blood was signi cantly increased compared to baseline. So, in a general point of view, the dose of up to 32 mSv per study adding to the background radiation is non-negligible [35]. Therefore, PET/CT scanning protocols should be optimized for reducing doses and their associated cancer risks.
In the study protocol, we included some baseline parameters like total received dose, body weight and age to check any associated relationship with enhancement of DNA damage. The observed data clearly shows a direct relationship between total received dose and the number of foci formation. Patients body weight is also one of the important dependable factor with the formation of γ H2AX foci, because it is directly proportional to the total received dose. In a general PET/CT procedure the activity of 18 F prepared for a patient was determined by his or her body weight [36] so that, the total received dose also changes with the administrated activity: this is why we saw a signi cant (r 2 = 0.3556) relationship (Fig. 6) between the body weight and total received dose. Interestingly we also observed a trend for increasing formation of H2AX foci with increasing age in the control group, but not in the test group ( Supplementary Fig. Fig. S1). Age-related DNA breaks formation is well corroborated by many authors [37,38].
Several studies have reported an association between increased 8-OHdG and carcinogenesis in various types of malignant tumors [39,40]. Laboratory measurement of urinary 8-OHdG's o ers some advantages also in this regard, as in addition to being noninvasively measured in urine with high stability, it re ects the overall level of oxidative DNA damage and repair from all cells in the organism [41]. The rst voided morning urine sample is very important in response to nullify the other activity which are directly associated in elevation of 8-OHdG. Over the course of the day, the composition and concentration of urine changes continuously. The rst voided morning specimen (upon waking) is particularly valuable because it provides a time average for biomarker concentrations that may occur during the hours of sleep (approximately 8 hours) [42]. In contrast, a late morning or a ernoon urine sample may re ect dietary, physical and environmental (i.e. tobacco smoke, pollution) exposures. According to Rodrigues et al. and Abusoglu et al. [28,43] 8-OHdG levels can be high in some stressful situations including smoking, aging, lack of or extreme exercise and occupational exposure to chemicals. With the increasing of DNA DSBs in patients sample we also observed the elevation of 8-OHdG from the control group, and we observed that the early morning samples show the highest elevation, whereas the 2 h sample did not show any signi cant di erence. These results may suggest that the products of repaired DNA in the form of 8-OHdG are excreted into the urine without any further metabolic processing, and also the presence of the modi ed nucleoside (8-OHdG) in urine represents the primary repair product of oxidative DNA damage in vivo, presumably nucleotide excision repair [44]. Therefore, determination of the level of 8-OHdG directly re ects the oxidative damage of cellular DNA.
Levels of 8-OHdG also depend on certain baseline parameters such as total dose, body weight and age. We observed that the total dose and body weight were signi cantly associated with the increment of 8-OHdG, which has been previously associated with radiation induced oxidative damage [45], whereas body weight directly relates to the total dose distribution. The positive association of increasing age with the formation of 8-OHdG is plausible because oxidative stress is known to increase during aging [46]. Recent studies suggest that age-associated functional losses are due to the accumulation of ROS-and nitrogen species-induced damages as per the oxidative stress theory [47].
In this study we observed a weak correlation between mean γ H2AX formation and the level of 8-OHdG in urine samples (Fig. 8). A stronger relationship may be possible only when factors such as lifestyle habits (e.g. tobacco smoking), metabolic activity and other types of internal and external stressor and epigenetic factors are considered [48,49].
In summary, the study observed a signi cant amount of DNA damage a er WB 18 F-FDG PET/CT, which corroborates previous observations. Additionally, our study showed that the level of 8-OHdG in urine sample also increased with the DNA damage, as an integrated, WB indicator of oxidative stress and DNA repair processes. Larger population studies are needed to con rm these observations, describe the ne-scale timing of changes in the biomarker levels a er exposure, and further clarify any potential risks to patients from PET/CT procedures.