Modeling a radiotherapy clinical procedure: total body
irradiation.




Abstract: 
Leukemia, nonHodgkin's lymphoma, and neuroblastoma patients
prior to bone marrow transplants may be subject to a clinical
radiotherapy procedure called total body irradiation (TBI). To mimic a
TBI procedure, we modified the Jones model of bone marrow radiation cell
kinetics by adding mutant and cancerous cell compartments. The modified
Jones model is mathematically described by a set of n + 4 differential
equations, where n is the number of mutations before a normal cell
becomes a cancerous cell. Assuming a standard TBI radiotherapy treatment
with a total dose of 1320 cGy fractionated over four days, two cases
were considered. In the first, repopulation and sublethal repair in the
different cell populations were not taken into account (model I). In
this case, the proposed modified Jones model could be solved in a closed
form. In the second, repopulation and sublethal repair were considered,
and thus, we found that the modified Jones model could only be solved
numerically (model II). After a numerical and graphical analysis, we
concluded that the expected results of TBI treatment can be mimicked
using model I. Model II can also be used, provided the cancer
repopulation factor is less than the normal cell repopulation factor.
However, model I has fewer free parameters compared to model II. In
either case, our results are in agreement that the standard dose
fractionated over four days, with two irradiations each day, provides
the needed conditioning treatment prior to bone marrow transplant.
Partial support for this research was supplied by the NIHRISE program,
the LSAMPPuerto Rico program, and the University of Puerto
RicoHumacao. [P R Health Sci J 2010;3:293298] Key words: Carcinogenesis mathematical model, Total body irradiation, Jones Model, Survival curves and cancerous cell populations Previo a un trasplante de medula osea, pacientes con leucemia, linfoma noHodgkin, y neuroblastoma podrian ser sometidos a un procedimiento en radioterapia llamado Radiacion Total del Cuerpo (RTC). Para simular RTC hemos modificado el modelo de Jones, anadiendo compartimientos virtuales para las poblaciones de celulas mutantes y cancerosas. Este modelo de Jones modificado es descrito matematicamente por un conjunto de n+4 ecuaciones diferenciales, donde n es el numero de mutaciones antes de que la celula normal se transforme en una celula cancerosa. Asumiendo un procedimiento comun de radioterapia RTC, con una dosis total de 1320 cGy fraccionada en cuatro dias, dos casos son considerados. Primero, el modelo I es definido cuando la repoblacion y reparacion del dano subletal no son tomados en cuenta. Se demuestra en este caso, que el modelo de Jones modificado puede ser resuelto en forma exacta. Segundo, si la repoblacion y reparacion del dano subletal son consideradas, entonces el modelo de Jones modificado puede ser solo resuelto numericamente (modelo II). Luego de un analisis grafico y numerico, podemos concluir que se pueden reproducir los resultados esperados para un tratamiento RTC usando el modelo I. El modelo II, puede tambien ser utilizado solo si el factor de repoblacion para las celulas cancerosas es menor que el factor de repoblacion de las celulas normales. Sin embargo, el modelo I tiene menos parametros libres que el modelo II. En cualquiera de los dos modelos estudiados, nuestros resultados sugieren que el tratamiento clinico usual, es decir el fraccionamiento de la dosis en cuatro dias con dos irradiaciones diarias, provee el acondicionamiento adecuado para el trasplante de medula osea. 


Article Type:  Perspectiva general del procedimiento medico 
Subject: 
Radioterapia
(Investigacion cientifica) Radioterapia (Analisis de casos) Cancer (Investigacion cientifica) Cancer (Analisis de casos) Cancer (Cuidado y tratamiento) 
Authors: 
Esteban, Ernesto P. Garcia, Camille De La Rosa, Veronica 
Pub Date:  09/01/2010 
Publication:  Name: Puerto Rico Health Sciences Journal Publisher: Universidad de Puerto Rico, Recinto de Ciencias Medicas Language: Spanish Audience: Academic Format: Magazine/Journal Subject: Health Copyright: COPYRIGHT 2010 Universidad de Puerto Rico, Recinto de Ciencias Medicas ISSN: 07380658 
Issue:  Date: Sept, 2010 Source Volume: 29 Source Issue: 3 
Accession Number:  234999044 
Full Text: 
Radiotherapy is the use of Xrays, gamma rays, or electron or
proton beams to treat cancer. The aim of the treatment is to kill the
cancer cells, either directly or indirectly (i.e., by interfering with
cell reproduction). There are two main types of radiotherapy treatment:
external and internal radiotherapy. In external radiotherapy, a linear
accelerator is used to deliver the radiation dose. Internal radiotherapy
is where the source of radioactivity is put inside the human body so
that it can get close to the cancerous tumor. Cancer patients usually
undergo external radiotherapy in small doses; each dose is called a
fraction. Dose fractionating reduces toxicity to normal tissues. The
length of the radiotherapy treatment depends on the type of cancer. A
procedure called Total Body Irradiation (TBI) is considered a special
case in radiotherapy. This is because the treatment field is the entire
body, and thus, the irradiated volume is highly irregular in shape. TBI is used to treat leukemia, nonHodgkin's lymphoma, and neuroblastomas and as a preparatory regimen prior bone marrow transplant. Patients usually receive a total dose between 10 to 12 Gy in eight fractions over four days, with at least six hours between fractions. As pointed out by an anonymous referee, TBI is only part of the conditioning regimen in the bone marrow transplant process. The most common pretreatment conditioning is a combination of chemotherapy and TBI. Before TBI, very high doses of chemotherapy are given to render the patient in remission, virtually cancer free. In this case, the main purpose of TBI is to help wipe out the host's marrow as well as reduce the probability of rejection. However, in some forms of leukemia, TBI is used to kill the remaining cancerous cells, thereby allowing donor marrow to engraft. For this reason, the hypothetical clinical case to be considered was a leukemia patient in whom the chemotherapy treatment prior to TBI had killed 90% of the patient's cancer cell population. At this point, we posed three queries. First, is it possible to develop a quantitative model to reproduce the expected results of a TBI procedure? Second, are the supralethal doses used in TBI strong enough to suppress normal cells' sublethal repair and their repopulation among normal, mutated, and cancerous cells? Third, we wanted to investigate whether or not the standard delivery of the TBI dose over a period of four days, with six or eighteen hours between doses, is indeed ideal for the elimination of the remaining cancer cell populations. To explore possible answers for the proposed queries, we modified the Jones et al. model (12) of bone marrow radiation cell kinetics. Originally, in this study, continuously irradiated stemcell populations were modeled quantitatively by a threecompartmental formulation. However, Jones et al.'s cell kinetic model does not include possible mutations when ionizing radiation is acting on a cell population. Since DNA will not always be correctly repaired, the appearance of a mutation cannot be ignored. In fact, numerous studies have accumulated evidence of the causeeffect relationship between damage to DNA and the mutagenic effects of ionizing radiation. Therefore, in the proposed modified Jones model, it is assumed that n mutations will occur before a normal cell becomes a cancer cell. As a consequence, the number of compartments and the set of differential equations to mathematically describe the evolution in time of normal, injured, mutated, killed, and cancer cell populations increased from 3 to n + 4. When sublethal repair and repopulation are not taken into consideration, the set of n + 4 differential equations can be solved in a closed form. We explicitly found exact solutions for n = 1 and n = 2. However, when sublethal repair and repopulation are considered, the corresponding set of differential equations can only be solved numerically. Methods The behavior of an irradiated stemcell population can be schematized (Figure 1). Namely, at any time t, there is a normal cell population, [N.sub.n] (t); an injured cell population [N.sub.i] (t); n different mutant cell populations [N.sub.mj] (t), (j = 1, 2,... n); a killed cell population [N.sub.k] (t); and a cancerous cell population [N.sub.c] (t). Each of [lambda's] parameters connecting compartments is associated with a biological process. The dotted line in Figure 1 indicates other possible mutations. In Table 1, the different biological processes associated with the parameters of [lambda] are explicitly given. [FIGURE 1 OMITTED] The number of mutations required for a normal cell to become a cancer cell has been estimated by Renan (3) to be between two and ten. However, Little (4) and Wheldon (5) argue that most of the essential features of a carcinogesis model can be captured with n = 2. Therefore, for n = 2 mutations, the dynamics of the carcinogenesis mathematical model schematized in Figure 1 can be described by a set of six differential equations. They are where D is the dose and [MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] are the mitosis factors of normal, mutated, and cancerous cells, respectively. Further, [F.sub.in] = 2  [N.sub.n]  [N.sub.i] and [F.sub.nn] = (1  [N.sub.n]  [N.sub.i]) [F.sub.in] represents dynamic factors that modify the normal cell repair ([F.sub.in] and proliferation rates ([F.sub.nn]). In an analogous manner, we have defined and [MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] to be the dynamic factors associated with the repopulation of both the first and second mutations and of cancerous cells, respectively. [dN.sub.n]/dt =  [[lambda].sub.nk][N.sub.n]dD/dt  [[lambda].sub.ni][N.sub.n]dD/dt + [[lambda].sub.in][F.sub.in][N.sub.i] + [[lambda].sub.nn][N.sub.n][M.sub.n][F.sub.nn], (1) [MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII], (2) [MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII], (3) [MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII], (4) [MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII], (5) [MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII], (6) Results Exact Solutions Equations 1  6 can be solved in an exact closed form when neither repopulation nor sublethal repair is considered. Thus, for n = 1 mutation, we obtain [MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII], (7) [MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII], (8) [MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII], (9) [MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII], (10) and in the above, [C.sub.1], [C.sub.2], [C.sub.3], and [C.sub.4] are constants, and [[lambda].sub.1] = [[lambda].sub.ni] + [[lambda].sub.nk], (11) [MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII], (12) [MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII], (13) [[lambda].sub.4] = [[lambda].sub.ck] + [[lambda].sub.1], (14) [[lambda].sub.5] = [[lambda].sub.ck] + [[lambda].sub.2], (15) [[lambda].sub.6] = [[lambda].sub.ck] + [[lambda].sub.3], (16) where [N.sub.k] can be obtained from [MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII]. For n = 2 mutations, [N.sub.n], [N.sub.i], and [MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] are the same as those for n = 1 mutation. The remaining solutions are [MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] (17) [MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII], (18) where, in the above, [[lambda].sub.1] and [[lambda].sub.2] have the same meaning, as in the case of n = 1 mutation. The remaining parameters, [[lambda].sub.3], [[lambda].sub.4], [[lambda].sub.5], [[lambda].sub.6], [[lambda].sub.7], and [[lambda].sub.8], are defined for n = 2 mutations, as follows [MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII], (19) [MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII], (20) [[lambda].sub.5] = [[lambda].sub.ck] + [[lambda].sub.1], (21) [[lambda].sub.6] = [[lambda].sub.ck] + [[lambda].sub.2], (22) [[lambda].sub.7] = [[lambda].sub.ck] + [[lambda].sub.3], (23) [[lambda].sub.8] = [[lambda].sub.ck] + [[lambda].sub.4], (24) notice that parameters [[lambda].sub.3], [[lambda].sub.4], [[lambda].sub.5], and [[lambda].sub.6] illustrated by equations 1922 are not the same as those defined for n = 1 mutation. The solution for [N.sub.k] can be obtained in the same way as mentioned before. Notice that it is easy to generalize the above results for n > 2. Interestingly, equation 10 suggests that a cancer cell population could be initiated with just one mutation. The cancer cell population will not depend only on [MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII], but also on the product of all the "decaying parameters" ([MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] for two mutations). There are also constraints on the values of the parameters of [lambda]. For example, for two mutations, [[lambda].sub.2] [[lambda].sub.1] [not equal to]0, [[lambda].sub.3] [[lambda].sub.1][not equal to]0, etc. Numerical Solutions Next, we numerically solved equations 16. To do so, we needed to assign numerical values to all of the parameters of [lambda] and to the repopulation dynamic factors for the normal, mutated, and cancerous cells. In Table 2, we listed the chosen numerical values for [lambda]'s parameters. The first five parameters in Table 2 are the resulting values for hematopoietic stem cells receiving 22 MeV of radiation, extrapolated from animal data to men by Morris et al. (6). The remaining values in Table 2 were estimated as follows. The [lambda]srate constants that mediate movement of cells between the normal, mutated, and cancerous compartments to the killed compartment were set to the same value. Also, it was assumed that [MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII], where [alpha], [beta], and [gamma] are dimensionless numbers. At any time t, we can expect that the mutant population is less than the population of the normal cells. Thus, for the plotting of Figures 27, we chose the following values: [alpha] = [beta] = 0.5. Usually, since the cancer cellcycle time of cancer cells is shorter than that of normal cells, it is expected that [gamma] > 1. However, TBI treatments affect cancer cells to a greater degree than they do normal cells; for that reason, we chose the following value: [gamma] = 0.5. Note that for more realistic calculations [alpha], [beta], and [gamma] could be timedependent. The dose rate depends on the cancer tumor, exposure time, and radiotherapy procedure. Clinical Case: Total Body Irradiation To test our proposed model, we considered a patient who had received (prior to TBI) a chemotherapy treatment that killed 90% of the original cancer cell population. As usual for TBI, the radiation dose was fractionated over several days. The total dose given to the cancer patient was 1320 cGy in 8 fractions over 4 days. Those fractions were 165 cGy delivered daily at 9:00 am and 3:00 pm. The dose rate was 14.2 cGy/min, and the exposure time was 11.6 min. Figures 24, show (for model I) the behavior of the survival curve [MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] and the killed/ cancer cell populations, respectively. [FIGURE 2 OMITTED] [FIGURE 3 OMITTED] [FIGURE 4 OMITTED] In Figures 57, we show a comparison between model I (purple and green) and model II (red and blue). As expected, Figures 57 demonstrate that the survival curve and the killed and cancer cell populations are greater when repopulation and sublethal repair are considered. [FIGURE 5 OMITTED] [FIGURE 6 OMITTED] [FIGURE 7 OMITTED] Conclusions In this study, we developed a modified Jones model with n = 2 mutations in order to ascertain the evolution in time of survival curves and cancerous and killed cell populations during a TBI clinical procedure. Two cases (model I and model II) were considered. First, in model I, we took neither repopulation nor sublethal repair into consideration and solved, in a closed form, the resulting set of differential equations. Second, in model II, repopulation and sublethal repair were considered, and the set of nonlinear differential equations was solved by implementing a computer code in MATHEMATICA 7.0 (www.wolframresearch.com). At the end of the first irradiation day, the number of killed cells in models I and II was about the same. In the following irradiation days, the killed cell population was higher in model II than in model I. The opposite happened with the survival curves. Both models I and II, predicted correctly that at the end of the TBI treatment the survival curves would reach zero. Regarding the cancer cell population, as expected, it was greater in model II than in model I for each of the four days of treatment. Notice also, that at the end of the TBI treatment, model I and model II both predicted the eradication of most of the cancer cell population. However, to reach this result with model II, we had to set [gamma] = 0.5, i.e., the cancer cell repopulation factor would be less than the normal cell repopulation factor during a TBI procedure. In this regard, model I is better than model II because it has fewer free parameters. Finally, although it is not shown here, we modified the initial conditions to those of a TBI treatment and solved the respective differential equations that define models I and II. We can confirm that, qualitatively for all trials, the same conclusions were reached and held true. In summary, the three queries posed at the beginning of this paper are now answered. First, a quantitative model to mimic a TBI clinical procedure has been developed and hypothetically tested. As far as know, the proposed quantitative model is new in the literature. It can be easily modified for use in other TBI treatments with different protocols, i.e., different doses, days fractionated, or number of irradiations per day. Second, it seems that during a TBI procedure, the repopulation and sublethal repair of normal, mutated, and cancerous cell populations is nearly suppressed. Third, the simulation of the TBI procedure proved that the standard clinical treatment for bone marrow transplants is correct in terms of the number of fractionated doses, the number of days, and the exposure time. Acknowledgements This research was partially supported by an NIHRISE grant and the University of Puerto RicoHumacao. We also are thankful for the support given by PRSLAMP. References (1.) Jones TD, Morris MD, Young RW. A Mathematical Model for RadiationInduced Myelopoiesis. Radiat Res 1991; 128: 258266. (2.) Jones TD, Morris MD, Young RW, Kehlet RA. A cellkinetics model for radiationinduced myelopoiesis. Exp Hematol 1993; 21: 816822. (3.) Renan MJ. How many mutations are required for tumorigenesis? Implications for human cancer data. Mol Carcinog 1993; 7: 139146. (4.) Little MP. Generalisations of the twomutation and classsical multistage models of carcinogenesis fitted to the Japanese atomic bomb survivor data. J Radiol Prot 1996; 16: 724. (5.) Wheldon EG, Lindsay KA, Wheldon TE. The doseresponse relationship for cancer incidence in a twostage radiation carcinogenesis model incorporating cellular repopulation. Int J Radiat Biol 2000; 76: 699710. (6.) Morris MD, Jones TD, Young RW. Estimation of coefficients in a model of radiationinduced myelopoiesis from mortality data for mice following xray exposure. Radiat Res 1991; 128: 267275. Ernesto P. Esteban, PhD *; Camille Garcia ([dagger]); Veronica De La Rosa ([double dagger]) * Laboratory of Theoretical Physics ([dagger]) Department of Physics; ([double dagger]) Department of Physics; ([double dagger]) Department of Biology, University of Puerto Rico Address correspondence to: Ernesto P. Esteban Ph D, Department of Physics, University of Puerto Rico, PO Box 10100, Humacao, PR 00792. Tel/Fax:(787) 8527810 * Email: ernesto.esteban@upr.edu Table 1. The biological processes associated with the parameters of [lambda] as seen in Figure 1. Parameter Cellular Process [[lambda].sub.ni] Sublethal Damage [[lambda].sub.in] Repair of Sublethal Damage [[lambda].sub.nk] 1 hit killing [[lambda].sub.ik] 2 hit killing [[lambda].sub.nn] Normal Cell Repopulation [MATHEMATICAL EXPRESSION NOT Mutant Cell Repopulation REPRODUCIBLE IN ASCII] [MATHEMATICAL EXPRESSION NOT No DNA Repair or Incomplete Repair REPRODUCIBLE IN ASCII] [MATHEMATICAL EXPRESSION NOT n Mutation REPRODUCIBLE IN ASCII] [MATHEMATICAL EXPRESSION NOT Mutant Cell [right arrow] Cancer Cell REPRODUCIBLE IN ASCII] [[lambda].sub.cc] Cancer Cell Repopulation Table 2. [lambda]'s numerical parameter values. Parameters [[lambda].sub.ni] 6.4 x 103 cGy 1 [[lambda].sub.in] 6.0 x 103 cGy 1 [[lambda].sub.nk] 2.3 x 103 cGy 1 [[lambda].sub.ik] 7.0 x 102 cGy 1 [[lambda].sub.nn] 2.2 x 104 min 1 [MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] 6.41 x 103 cGy 1 [MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] 2.31 x 103 cGy 1 [MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] 6.42 x 103 cGy 1 [MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] 6.43 x 103 cGy 1 [[lambda].sub.ck] 2.31 x 103 cGy 1 [MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] 2.32 x 103 cGy 1 [MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] 2.2 x 104 min 1 [MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] 2.2 x 104 min 1 [[lambda].sub.cc] 2.2 x 104 min 1 
Gale Copyright:  Copyright 2010 Gale, Cengage Learning. All rights reserved. 