radiotherapy by computational methods

Transcription

1 Evaluating 99m Tc Auger electrons for targeted tumor radiotherapy by computational methods Adriana Alexandre S. Tavares a) and João Manuel R. S. Tavares b) Faculdade de Engenharia da Universidade do Porto (FEUP) Rua Dr. Roberto Frias, S/N, , Porto, Portugal Abstract: Purpose: Technetium-99m ( 99m Tc) has been widely used as an imaging agent but only recently has been considered for therapeutic applications. This study aims to analyze the potential use of 99m Tc Auger electrons for targeted tumor radiotherapy, by evaluating the DNA damage and its probability of correct repair and by studying the cellular kinetics, following 99m Tc Auger electrons irradiation in comparison to Iodine-131 ( 131 I) beta minus particles and Astatine-211 ( 211 At) alpha particle irradiation. Methods: Computational models were used to estimate the yield of DNA damage (fast Monte Carlo damage algorithm), the probability of correct repair (Monte Carlo excision repair algorithm) and cell kinetic effects (virtual cell radiobiology algorithm) after irradiation with the selected particles. Results: The results obtained with the algorithms used suggested that 99m Tc CKMMX (all M-shell Coster-Kroning CK and super CK transitions) electrons and Auger MXY (all M-shell Auger transitions) have a therapeutic potential comparable to high linear energy transfer (LET) 211 At alpha particles and higher than 131 I beta minus particles. All the other 99m Tc electrons had a therapeutic potential similar to 131 I beta minus particles. 1

2 Conclusions: 99m Tc CKMMX electrons and Auger MXY presented a higher probability to induce apoptosis than 131 I beta minus particles and a probability similar to 211 At alpha particles. Based on the results here, 99m Tc CKMMX electrons and Auger MXY are useful electrons for targeted tumor radiotherapy. Keywords: targeted tumor radiotherapy; computational methods; I. Introduction The main characteristics of an ideal radionuclide for targeted tumor radiotherapy include 1 : (a) electrons emitted with energies lower than 40 kev; (b) photonic emission/electron emission ratio lower than 2; (c) half-life between 30 minutes and 10 days; (d) stable daughter nuclide or daughter nuclide with a half-life greater than 60 days; (e) amenable to radiolabeling; (f) economical preparation with high specific activity and radiochemical purity; and (g) efficient incorporation into a selective carrier molecule, which should be able to associate with the DNA complex for the time corresponding to the radionuclide half-life. Other requisites for successful targeted tumor radiotherapy have also been highlighted such as: (a) consecutive internal irradiations using Auger electron emitters must be possible; and (b) systemic radiation therapy must target the radiation homogeneously on a large proportion of all live cancerous cells. 2 Auger electrons have been recognized as potentially useful for targeted tumor radiotherapy, especially due to the Auger electron range (which is in the nanometer order), and the high ionization density of these electrons. 2-6 Previous studies also demonstrated that once Auger electron emitters are introduced close to DNA, the survival curves are similar to those obtained with high LET α particles. 5, 7 Despite these advantages, Auger emitters still have limitations including: long tracer retention times in blood flow and low penetration in certain tumor areas, which can lead to non-uniform doses absorbed in tumors. 4 It is well known that 99m Tc emits less than 1% Auger electrons per decay versus 3.7 to 19.9% for Iodine-125 ( 125 I), Iodine-123 ( 123 I) and Thallium-201 ( 201 Tl). Nevertheless, some potential 2

3 advantages have been pointed out, including: a short half-life; a stable daughter nuclide; Auger electron energies between 0.9 and 15.4 kev; and good availability. Also 99m Tc is obtained economically, as it can be eluted and handled easily from a generator with high specific activity. Furthermore, its ideal characteristics for imaging can allow therapy monitoring and follow-up. 1-2, 8-10 These characteristics and the very small number of studies concerning 99m Tc Auger electrons (for a review see 11 ) motivated us to carry out this work to evaluate the usefulness of these specific Auger electrons for targeted tumor radiotherapy. We used cell radiobiology software 12 and two fast Monte Carlo simulators to: Study the radiobiological effects of 99m Tc Auger electrons by comparing these with other particles emitted by radionuclides currently used for systemic radiotherapy, namely, 131 I (beta minus emitter) and 211 At (alpha emitter). Evaluate the radiobiological effects of 99m Tc Auger electrons in comparison with 131 I beta minus and 211 At alpha particles on two different cell types (human fibroblasts and human intestinal crypt cells). II. Methods Three different computational simulators were used to study the therapeutic potential of 99m Tc Auger electrons This section explains the main principles of these simulators and points out the parameters adopted Fast Monte Carlo Damage Formation Simulator The Monte Carlo damage simulation (MCDS) algorithm is used to predict the types of DNA damage and their yield after irradiation. The model generates a random number of damage configurations expected within the DNA of one cell. This algorithm processes information in two main steps: 1) it randomly distributes, in a DNA segment, the expected amount of damage produced in a cell and 2) subdivides the distribution of damage in that section. The number and spatial distribution of damage configurations predicted by the MCDS algorithm are in reasonable agreement with those predicted by track-structure simulations. Furthermore, the MCDS allows the 3

4 collection of data from multiple irradiation scenarios within a few minutes on a common computer. These characteristics make the MCDS simulator useful for comparing 99m Tc electrons with other particles used for radiotherapy. 14 For a detailed description of the MCDS model as well as additional discussions on the validity and limitations of the model, see, for example, The classification scheme used by the MCDS to categorize DNA damage is based on the classification parameters proposed by Nikjoo et al. (1997), and it comprises essentially: (a) no damage; (b) single-strand breaks (SSB); (c) two strand-breaks on the same strand (SSB + ); (d) two or more strand-breaks on opposite strands separated by at least 10 base pairs (2SSB); (e) two strand-breaks on opposite strands with a separation not greater than 10 base pairs (double strand breaks - DSB); (f) DSB accompanied by one (or more) additional strand breaks within a 10-base pair separation (DSB + ); and (g) more than one DSB, whether within the 10-base pair separation or further apart (DSB ++ ). For further details see Fast Monte Carlo Excision Repair Simulator The Monte Carlo excision repair (MCER) algorithm is used to simulate repair outcomes such as correct repair, repair with a mutation and conversion into a DSB. This Monte Carlo simulation also calculates the formation and repair of damage within one cell. 13 The MCER algorithm starts using the MCDS algorithm to generate a random number of damage configurations expected within the DNA of one cell. Thereafter, the MCDS-generated damage configurations are superimposed over an actual nucleotide sequence or a random nucleotide sequence. Finally, the MCER model is used to simulate the repair, misrepair and aborted excision repair of damage within the entire genome or within a specific region of the DNA. The lesions forming a cluster are removed sequentially through repeated rounds of excision repair. Most DNA oxidative damage, including modified apurinic/apyrimidinic (AP) converted to strand breaks, require repair by base excision repair (BER). Two different types of BER processes have been observed in eukaryotic and prokaryotic cells: 1) excision and replacement of a single nucleotide, known as short-patch BER (SP-BER), which occurs in the majority of cases; and 2) replacement of 2 to 13 nucleotides, known as long-patch BER (LP-BER). Another enzymatically 4

5 distinct repair pathway is nucleotide excision repair (NER). This last repair pathway, observed in eukaryotic cells, substitutes oligonucleotide fragments of 24 to 32 nucleotides in length. 13 The simulator results are presented in terms of three simplified repair scenarios due to the current uncertainties associated with the processing of radiation-induced damage by the BER and NER pathways. According to Semenenko et al. (2005), the simulator results correlate well with in vitro results from cell cultures, despite this simplification. 18 A detailed description of the MCER algorithm as well as additional discussions on the validity and limitations of the model can be found in 13, Virtual Cell Radiobiology Simulator Ionizing radiation frequently causes DSB and other DNA damages in less than one millisecond. Radiation induced damage is processed slowly via enzymatic repair and misrepair, which then determines the fate of the irradiated cell. As far as long time scales are concerned, cell cycle kinetics can influence and be influenced by the kinetics of damage processing. 19 DNA damage is a trigger for apoptosis, although cell membrane damage can also induce apoptosis. 15 The dose-response association, damage production and its repair mechanisms have been largely studied using radiobiological models that correlate the dose rate with cell response. Some of the existing models include: 1) the repair-misrepair model (RMR), 2) lethal-potentially lethal model (LPL) and 3) two-lesion kinetics (TLK). 19 The main disadvantage of the LPL model is its limits to correlate the biochemical processes of DSB with cell death. The RMR and linear quadratic models also have the same limitation. In order to overcome this limitation, the TLK model carries out an improved correlation between the biochemical processes of DSB and cell death by subdividing DSB into simple or complex DSBs. This subdivision is important, since simple and complex DSBs have different repair characteristics. 20 Therefore, simulations carried out with the virtual cell radiobiology simulator (VC) were performed using the TLK model Simulated Parameters 5

6 The 99m Tc spectrum of energies, presented in an AAPM (American Association of Physicists in Medicine) report in 1992, includes electron ranges from 2.05 nm to 251 μm and electron energies ranging between kev and 140 kev. 21 In the present study, all 99m Tc electrons were studied with the exception of Auger CK NNX (all N-shell CK and super CK transitions), due to its low energy (33 ev), which is below the lower energetic limit of the simulator (80 ev). In addition, alpha particles from 211 At and beta minus particles from 131 I were also studied. Further input details for the MCDS and MCER simulators can be found in Table I. Once the comparison between the MCDS and MCER results for 99m Tc electrons, 131 I beta particles and 211 At alpha particles was complete, the two best 99m Tc electrons (with the highest ability to induce DNA damage) were then used to study the kinetics of the cell after irradiation in two different cell types: 1) human fibroblasts (Tc the cell cycle time =0.900 hour and Tpot - potential doubling time=0.667 days) and 2) human intestinal crypt cells (Tc=1.000 hour and Tpot=1.625 days) The cell kinetic study (VC simulator) compared the two best 99m Tc electrons with 131 I beta particles and 211 At alpha particles. The number of DSBs and the percentage of complex DSBs were obtained from the MCDS simulator. These results were then applied as input parameters for the TLK model used on the VC simulator. Irradiation periods of 2, 6 and 24 hours (TCUT time allowed for repair after exposure), with total absorbed doses of 1, 1.5 and 2 Gy were studied using the VC simulator. Other parameters used on the VC simulator, specified in the TLK model input file, include: 1) DRM (damage repair model)=tlk; 2) CKM (cell kinetics model)=qeck (quasi-exponential cell kinetics model); 3) DNA (cell DNA content)=5.667d+09 base pair; 4) DSB (endogenous)=4.3349e-03 Gy -1 cell -1 ; 5) RHT (repair half-time)=xxx, XXX=0.25, 9 hours (simple DSBs are repaired faster than complex DSBs); 6) A0 (probability of correct repair)=aaa, AAA=0.95, 0.25 (simple DSBs are repaired more accurately than complex DSBs); 7) ETA (pairwise damage interaction rate)=2.5e-04 h -1 ; 8) PHI (probability of a misrejoined DSB being lethal)=0.005; 9) GAM (fraction of binarymisrepaired damages that are lethal)=0.25; 10) N0 (initial number of cells)=1000; 11) KAP (peak cell density)=1.0d+38 cells per cm 3 ; 12) VOL (tissue volume)=1 cm 3 ; 13) FRDL (fraction of residual that is lethal damage)=0.5; 14) ACUT (absolute residual-damage cutoff)=1.0d-09 expected number 6

7 of DNA damages per cell; 15) BGDR (average background absorbed dose rate on planet Earth)= E-07 Gy/h; 16) DCUT (dose cutoff)=0.01 Gy; 17) STOL (step-size tolerance)=0.01 Gy/h; 18) SAD (scaled absorbed dose)=rx1, RX1=1, 1.5, 2 Gy; 19) GF (growth fraction, if 0 (zero) all cells are quiescent, if 1 (one) all cells are cycling and if 0.5 the cell population is heterogeneous)=0, 0.5, 1. 12, Statistical Analysis The MCDS results are expressed as a percentage of damage. The MCER results are expressed as a probability of repair or number of cell cycles. The VC values are expressed as the number of lethal damages per cell, number of surviving cells, probability per cell and frequency per irradiated cell. The statistical significance was determined using either the Student t-test or ANOVA (p<0.01) for each group of irradiating agents. III. Results MCDS and MCER Results Results obtained by the MCDS simulator allowed an estimate to be made of the amount of DNA damage following irradiation with 99m Tc electrons, 131 I beta minus particles and 211 At alpha particles, as shown in Fig. 1. Results for the probability of correct repair, repair with a mutation and conversion into a DSB are presented in Fig. 2 and the number of repair cycles is presented in Fig. 3 (MCER simulator). Findings from the MCDS and MCER simulators showed that CKMMX electrons and Auger MXY were the best 99m Tc electrons for targeted tumor radiotherapy. Accordingly, these electrons were used for the study of cell kinetics after irradiation with the VC simulator VC Simulator The results of mutagenesis probability and induction of enhanced genetic instability (defined by the algorithm used as PGA PGH NCG=4.250E-06, with: PGA - probability a mutated gene induces genomic instability, PGH - probability that a randomly formed mutation hits a critical 7

8 gene and NCG - total number of target genes that must be damaged to induce genomic instability) after irradiation with different irradiating agents are shown in Fig. 4a. Statistical analysis showed significant differences between 99m Tc Auger MXY and 131 I beta minus particles (p=0.001, t-test) and also between 211 At alpha particles and 131 I beta minus particles (p=0.0006, t-test). The estimated number of lethal damages per cell due to mutations for all the irradiating agents is presented in Fig. 4b. Additionally, statistically significant differences were observed among the different irradiating agents (p<0.0001, ANOVA). Results for the neoplastic transformation (defined by the algorithm used as a function of dose and dose rate at t=42.0 days) per studied cell for different irradiating agents are presented in Fig. 4c. Once again, statistically significant differences were found among each irradiating agent per studied cell type (fibroblasts and intestinal crypt cells), p< (ANOVA). The estimated number of cells that survived irradiation when all cells were quiescent; when the cell population was heterogeneous (with quiescent cells and cells actively dividing/on cycle); and when all cells were actively dividing is presented in Figs. 5a, 5b and 5c, respectively. Statistically significant differences were observed between 131 I beta minus particles and the other types of radiation when all cells were quiescent (p<0.0001, ANOVA). However, no differences were found between 99m Tc Auger electrons and 211 At alpha particles or between fibroblasts and intestinal crypt cells for all the irradiating agents (Fig. 5a). In contrast, heterogeneous populations yielded statistically significant differences between different cell types for the same irradiating agent (p<0.0001, t-test). Differences among distinct types of irradiating agents in intestinal crypt cells (p<0.0001, ANOVA) and fibroblasts (p=0.0031, ANOVA) were also found in heterogeneous populations (Fig. 5b). For populations of cells actively dividing (Fig. 5c), statistically significant differences were found among distinct irradiating agents in intestinal crypt cells (p=0.0002, ANOVA), but no differences were found among distinct irradiating agents in fibroblasts (p=0.3014, ANOVA). A more detailed analysis showed that no differences were observed between both 99m Tc Auger electrons (CKMMX and Auger MXY, p=0.1406, t-test) or among each 99m Tc Auger electron under study and 211 At alpha particles (CKMMX, p= and Auger MXY, p= t-test). Nevertheless, 8

9 statistically significant differences were observed when comparing 131 I beta minus particles with the other particles studied (p<0.0001, t-test). Finally, statistically significant differences were also observed when comparing the same irradiating agent on the two distinct cell types (p<0.0001, t- test). IV. Discussion and Conclusions MCDS results showed that the percentage of simple and double strand breaks after irradiation was always higher for 99m Tc CKMMX electrons and Auger MXY than for 131 I beta minus particles and was similar to 211 At alpha particles. The same trend was observed for the percentage of complex single and double strand breaks. Furthermore, the remaining 99m Tc electrons obtained by internal conversion were less able to induce DNA damage, which correlates with Pomplun et al. (2006). 25 The results obtained with these conversion electrons were similar to 131 I beta minus particles. This may be explained by the higher tissue range of these 99m Tc electrons, whose behavior is similar to beta minus particles (low LET particles). The MCER outcome showed that the increased amount of DNA damage and its complexity hampers successful repair. Moreover, the probability of correct repair of single strand breaks is lower for 99m Tc CKMMX electrons and Auger MXY than for 131 I beta minus particles and is comparable to 211 At alpha particles. The probability of conversion to DSBs is also higher for 99m Tc electrons than for 131 I beta minus particles. These results were observed for all the repair processes studied, regardless of the repair route. In addition, a higher number of repair cell cycles had been correlated with prolonged repair times, which correlates with increased LET particles. Previous studies observed that complex damage repair by excision leads to an increased number of DSBs. 13 Accordingly, it is well known that DNA double strand breaks are frequently associated with apoptosis induction. 15 Therefore, the observed higher number of DNA double strands induced by 99m Tc CKMMX electron and Auger MXY (MCDS simulator) allied to its higher DNA single strand breaks conversion to double strand breaks (MCER simulator), suggest that the probability of apoptosis induction is likely to be higher for those electrons than for 131 I beta minus particles and comparable to 211 At alpha particles. 9

10 The mutagenesis and enhancement of genetic instability study showed that the best 99m Tc Auger electrons ( 99m Tc CKMMX electron and Auger MXY) had a higher probability of inducing mutagenesis and genetic instability than 131 I beta minus. However, 131 I beta minus particles were the most likely of all the irradiating agents studied to induce neoplastic transformation. Furthermore, the selected 99m Tc Auger electrons had a higher ability to induce lethal damage, due to mutations, than the other particles studied. These results suggest that the higher probability of induced mutagenesis and enhancement of genetic instability of the selected Auger electrons will potentially lead to cell death or benign mutations and not to neoplastic transformation. The findings obtained are consistent with in vitro studies conducted by Pedraza-López et al. (2000) and Ilknur et al. (2002) using lymphocytes The results showed that the various irradiating agents were equally effective at killing quiescent human fibroblast and intestinal crypt cells. In contrast significant differences were seen between irradiating agent and cell types when a mixed population of cycling and quiescent cells were irradiated. These observations highlight the influence of cell proliferation on the radiosensitivity of the cells. For heterogeneous populations, crypt cells were more radiosensitive than fibroblasts. Finally, for cell populations where all cells were actively dividing, the results also showed that the number of cells that survive irradiation was significantly lower for intestinal crypt cells when compared to fibroblasts. Nevertheless, no differences were observed among the distinct irradiating agents studied for actively dividing fibroblast populations, which suggests that cell response to irradiation is radiation type independent. This may be explained by the reduced radiosensitivity of this type of cell and its active proliferation state, which may compensate radiation induced damages by fast continuous cell duplication. Furthermore, intestinal crypt cells showed significant differences among all irradiating agents. This may mean that, due to its longer doubling time (39 hours versus 16 hours for fibroblasts), intestinal crypt cells were unable to compensate radiation induced damage by cell duplication. Häfliger and coworker s in vitro studies (2005) showed that 99m Tc induced double-strand breaks in DNA when decaying in its direct vicinity. 21, 28 In their paper, Häfliger and coworkers cited the Ftacnikova and Bohm (2000) study regarding theoretical calculations of energy deposition into 10

11 DNA According to Ftacnikova and Bohm (2000), the electrons with initial energies from 50 ev to 250 ev have the highest theoretical probability of inducing DNA DSB, because these electrons are able to produce clusters of inelastic interactions in a volume with a diameter of a few nm (which is characteristic of Auger emitters). 29 Based on that, Häfliger et al. listed the Auger electrons emitted by 99m Tc that are potentially the most interesting for targeted tumor radiotherapy: CK MMX electron, Auger MXY electron and CK NNX electron. 28 We used different computational methods to evaluate the 99m Tc electrons spectrum by comparing those with other particles used for radiotherapy. This represents a novel and faster method to evaluate and grade 99m Tc electrons for target tumor radiotherapy. All 99m Tc electrons were considered separately (except CK NNX electron due to simulator limitations) and their radiobiological effects evaluated. Our findings, obtained by means of three different computational simulators, provided evidence that 99m Tc CK MMX electron and Auger MXY electron are useful electrons for targeted tumor radiotherapy. 99m Tc Auger MXY and CKMMX electrons yield 1.1 and electrons per decay - the second and third highest yields of all 99m Tc Auger electrons, respectively. The uppermost yielding electrons are CKNNX with 1.98 electrons per decay, which was not evaluated due to previously explained simulator limitations. These yields are lower than those for 125 I, which yields 1.44 and 3.38 electrons per decay for CKMMX electrons and Auger MXY, respectively. 21 Nevertheless, this potential limitation may be overcome or compensated by the 99m Tc shorter half-life, as shown by previous studies. 1, 8, 10 Moreover, the 99m Tc electron irradiation results correlate with previous findings suggesting that the shorter half-life radionuclides reduce the dose fractioning to daughter cells and increase the absorbed doses per unit of time. 1-6 Higher DNA damage yields have been associated with Auger electrons due to their short range and LET quality. The high abundance of 99m Tc photons is an important factor that may influence the possible therapeutic outcome. Although photons emitted present a high tissue range and thus most energy will be deposited outside the target cell, their dosimetric implications may work as a limiting factor for this kind of target tumor radiotherapy. Nonetheless, these photons could facilitate therapy monitoring and the design of more selective and specific carriers. This may be challenging but would allow the delivery of radiation to a specific targeted cell. 11

12 Computational methods allow rapid and easy data collection. Nonetheless, some limitations have been pointed out, including modeling and evaluation based on current knowledge, which works as a mechanistic process. This disadvantage may underestimate or overestimate the results. Although the results obtained showed correlation with previous in vitro and other computational studies, which suggest that the simulators used may be useful for the characterization of different particles for targeted tumor radiotherapy, further comparison of 99m Tc electrons with other Auger and conversion electrons could provide extra information regarding the potential of 99m Tc as a therapeutic radionuclide. In summary, this study aimed to compare different irradiating agents using the same exposure conditions and controllable cell populations to clarify the potential usefulness of 99m Tc electrons for targeted tumor radiotherapy. An analysis of all the data obtained has led us to conclude that 99m Tc CKMMX electron and Auger MXY presents a higher probability to induce apoptosis than 131 I beta minus particles and a similar one to 211 At alpha particles. This characterizes 99m Tc CKMMX electron and Auger MXY as high LET particles and thus useful for targeted tumor radiotherapy. Acknowledgement The authors wish to thank Dr Robert Stewart (School of Health Sciences Purdue University, USA) for providing the simulator software packages used and for his kind technical assistance. References: a) Electronic adriana_tavares@msn.com b) Electronic tavares@fe.up.pt 1. P. Unak, "Targeted Tumor Radiotherapy," Brazilian Archives of Biology and Technology 45, (2002). 2. F. Buchegger, F. Perillo-Adamer, Y. Dupertuis and A. Delaloye, "Auger Radiation Targeted into DNA: A Therapy Perspective," European Journal of Nuclear Medicine and Molecular Imaging 33, (2006). 12

13 3. R. O'Donnell, "Nuclear Localizing Sequences: An Innovative Way to Improve Targeted Radiotherapy," The Journal of Nuclear Medicine 47, (2006). 4. S. Britz-Cunningham and J. Adelstein, "Molecular Targeting with Radionuclides: State of the Science," The Journal of Nuclear Medicine 44, (2003). 5. C. Boswell and M. Brechbiel, "Auger Electrons: Lethal, Low Energy, and Coming Soon to a Tumor Cell Nucleus Near You," The Journal of Nuclear Medicine 46, (2005). 6. G. Mariani, L. Bodel, S. Adelstein and A. Kassis, "Emerging Roles for Radiometabolic Therapy of Tumors Based on Auger Electron Emission," The Journal of Nuclear Medicine 41, (2000). 7. K. Sastry, "Biological Effects of the Auger Emiter Iodine-125: A Review. Report No.1 of AAPM Nuclear Medicine Task Group No.6," Medical Physics 19, (1992). 8. J. Humm and D. Chariton, "A New Calculational Method to Assess the Therapeutic Potential of Auger Electron Emission," International Journal of Radiation Oncology Biology Physics 17, (1989). 9. F. Marques, A. Paulo, M. Campello, S. Lacerda, R. Vitor, L. Gano, R. Delgado and I. Santos, "Radiopharmaceuticals for Targeted Radiotherapy," Radiation Protection Dosimetry 116, (2005). 10. J. O'Donoghue and T. Wheldon, "Targeted radiotherapy using Auger electron emitters," Physics in Medicine and Biology 41, (1996). 11. A. Tavares and J. Tavares, " 99m Tc Auger Electrons for Targeted Tumour Therapy: A Review," International Journal of Radiation Biology 86, (2010). 12. R. Stewart, "Computational Radiation Biology," (Purdue University, School of Health Sciences, 2004). 13. V. Semenenko, R. Stewart and E. Ackerman, "Monte Carlo Simulation of Base and Nucleotide Excision Repair of Clustered DNA Damage Sites. I. Model Properties and Predicted Trends," Radiation Research 164, (2005). 13

14 14. V. Semenenko and R. Stewart, "A Fast Monte Carlo Algorithm to Simulate the Spectrum of DNA Damages Formed by Ionizing Radiation," Radiation Research 161, (2004). 15. D. Carlson, R. Stewart, V. Semenenko and A. Sandison, "Combined Use of Monte Carlo DNA Damage Simulations and Deterministic Repair Models to Examine Putative Mechanisms of Cell Killing," Radiation Research 169, (2008). 16. H. Nikjoo, P. O'Neil, E. Wilson, D. Goodhead and M. Terrissol, "Computational modelling of low-energy electron-induced DNA damage by early physical and chemical events," International Journal of Radiation and Biology 71, (1997). 17. H. Nikjoo, P. O'Neil, E. Wilson and D. Goodhead, "Computational Approach for Determining the Spectrum of DNA Damage Induced by Ionizing Radiation," Radiation Research 156, (2001). 18. V. Semenenko and R. Stewart, "Monte Carlo Simulation of Base and Nucleotide Excision Repair of Clustered DNA Damage Sites. II. Comparations of Model Predictions to Measured Data," Radiation Research 164, (2005). 19. R. Sachs, P. Hahnfeld and D. Brenner, "The link between low-let dose-response relations and the underlying kinetics of damage production/repair/misrepair," International Journal of Radiation and Biology 72, (1997). 20. M. Guerrero, R. Stewart, J. Wang and X. Li, "Equivalence of linear-quadratric and twolesion kinetic models," Physics in Medicine and Biology 47, (2002). 21. R. Howell, "Radiation Spectra for Auger-Electron Emitting Radionuclides: Report No.2 of AAPM Nuclear Medicine Task Group No.6," Medical Physics 19, (1992). 22. R. Baserga, "The Cell Cycle and Cancer," in The Biochemistry of Disease - A Molecular Approach to Cell Pathology - Volume I, Vol. I, edited by E. Farber (Marcel Dekker, USA, 1971), pp R. Baserga, "Definig the Cycle," in Cell Biology - Organelle Structure and Function, edited by D. Sadava (Jone and Barlett Publishers, USA, 1993). 14

15 24. UNSCEAR, "Report of the United Nations Scientific Committee on the Effects of Atomic Radiation to the General Assembly," edited by U. N. S. C. o. t. E. o. A. Radiation (2007). 25. E. Pomplun, M. Terrissol and E. Kümmerle, "Estimation of Radiation Weighting Factor for 99m Tc," Radiation Protection Dosimetry 122, (2006). 26. M. Pedraza-López, G. Ferro-Flores, M. Mendiola-Cruz and P. Moralez-Ramírez, "Assessment of Radiation-Induced DNA Damage Caused by the Incorporation of Tc-99m- Radiopharmaceuticals in Murine Lymphocytes Using Cell Gel Electrophoresis," Mutation Research 465, (2000). 27. A. Ílknur, E. Vardereli, B. Durak, Z. Gülbas, N. Basaran, M. Stokkel and E. Pauwels, "Labeling of Mixed Leukocytes with 99m Tc-HMPAO Causes Severe Chromosomal Aberrations in Lymphocytes," Journal of Nuclear Medicine 43, (2002). 28. P. Häfliger, N. Agorastos, B. Spingler, O. Georgiev, G. Viola and R. Alberto, "Induction of DNA-Double-Strand Breaks by Auger Electrons from 99m Tc Complexes with DNA-Binding Ligands," ChemBioChem 6, (2005). 29. S. Ftácniková and R. Böhm, "Monte Carlo Calculations of Energy Deposition in DNA for Auger Emitters," Radiation Protection Dosimetry 92, (2000). 15

16 TABLE CAPTION Table I. Input conditions for MCDS and MCER simulators. 16

17 TABLE I Energy Yield/Decay Particle [MeV] CK MMX Input conditions Auger MXY Auger LMM Auger LMX Auger LXY Auger KLL Auger KLX IC 1 M, N IC 2 K IC 3 K IC 2 L IC 3 L IC 2 M, N Beta I MCDS e MCER: Initial cell number = 1000 DMSO concentration = 0 (normal cell environment) MCER: Inhibition distance = 3 bp Probability of choosing a lesion from the first strand break (P1) = 0.5 Polymerase error rate for SP-BER=1.0-4 Polymerase error rate for LP-BER and NER = Probability of incorrect insertion opposite damaged base = 0.75 Probability of incorrect insertion of opposite base lost = 0.75 Alpha At

18 FIGURE CAPTIONS Fig. 1. a) Percentage of DNA radioinduced SSB and DSB after irradiation with 99m Tc electrons, 131 I beta minus particles and 211 At alpha particles (MCDS simulator). b) Percentage of two SSB on the same DNA segment (2SSB/DNA seg), two or more SSB on opposite DNA segments and separated by at least 10 base pairs (2 or +SSB DNA); and percentage of one DSB and one or more SSB separated by a maximum of 10 base pairs (DSB & 1 or + SSB) and one or more DSB separated by a maximum of 10 base pairs (1 or +DSB 10 bp), after irradiation with 99m Tc electrons, 131 I beta minus particles and 211 At alpha particles (MCDS simulator). c) Fraction of complex SSB and DSB DNA damages after irradiation with 99m Tc electrons, 131 I beta minus particles and 211 At alpha particles (MCDS simulator). Fig. 2. Probability of correct repair (p correct), repair with mutation (p mutation) and conversion to DSB (p conversion DSB) of DNA SSB, by SP-BER, LP-BER, SP-NER and LP-NER repair methods (MCER simulator). Fig. 3. Average number of repair cycles for all used repair methods and irradiating agents (MCER simulator). Fig. 4. a) Results from mutagenesis probability and induction of enhanced genetic instability simulations; b) average number of lethal mutations per cell; c) neoplastic transformation frequency in two different selected cell types after irradiation with distinct irradiating agents. Simulated doses of 1, 1.5 and 2 Gy during irradiation periods of 2, 6 and 24 hours (VC simulator). Fig. 5. Number of cells that survive irradiation with different irradiating agents a) in a quiescent cell population; b) in a heterogeneous cell population and; c) in cells actively dividing. Simulated doses of 1, 1.5 and 2 Gy during irradiation periods of 2, 6 and 24 hours (VC simulator). 18

19 FIGURES Figure 1 19

20 Figure 2 Figure 3 20

21 Figure 4 Figure 5 21