

ORIGINAL ARTICLE 

Year : 2012  Volume
: 8
 Issue : 1  Page : 3439 

Conformal fields in prostate radiotherapy: A comparison between measurement, calculation and simulation
Seied R Mahdavi^{1}, Hamed Rezaeejam^{2}, Alireza Shirazi^{1}, Mohammad Hosntalab^{2}, Ahmad Mostaar^{3}, Mohsen Motamedi^{2}
^{1} Department of Medical Physics, Tehran University of Medical Sciences, Tehran, Iran ^{2} Department of Medical Radiation Engineering, Science and Research Branch, Islamic Azad University, Tehran, Iran ^{3} Department of Medical Physics, Shahid Beheshti University of Medical Sciences, Tehran, Iran
Date of Web Publication  19Apr2012 
Correspondence Address: Hamed Rezaeejam Department of Medical Radiation Engineering, Science and Research Branch, Islamic Azad University, Tehran Iran
Source of Support: All needed items and equipments were facilitated by Physics section of Radiotherapy Department, Conflict of Interest: None  Check 
DOI: 10.4103/09731482.95171
Aims: The objective of this study is to evaluate the accuracy of a treatment planning system (TPS) for calculating the dose distribution parameters in conformal fields (CF). Dosimetric parameters of CF's were compared between measurement, Monte Carlo simulation (MCNP4C) and TPS calculation. Materials and Methods: Field analyzer water phantom was used for obtaining percentage depth dose (PDD) curves and beam profiles (BP) of different conformal fields. MCNP4C was used to model conformal fields dose specification factors and head of linear accelerator varian model 2100C/D. Results: Results showed that the distance to agreement (DTA) and dose difference (DD) of our findings were well within the acceptance criteria of 3 mm and 3%, respectively. Conclusions: According to this study it can be revealed that TPS using equivalent tissue air ratio calculation method is still convenient for dose prediction in non small conformal fields normally used in prostate radiotherapy. It was also showed that, since there is a close correlation with Monte Carlo simulation, measurements and TPS, Monte Carlo can be further confirmed for implementation and calculation dose distribution in non standard and complex conformal irradiation field for treatment planning systems. Keywords: Conformal field, distance to agreement, dose difference, monte carlo, treatment planning system
How to cite this article: Mahdavi SR, Rezaeejam H, Shirazi A, Hosntalab M, Mostaar A, Motamedi M. Conformal fields in prostate radiotherapy: A comparison between measurement, calculation and simulation. J Can Res Ther 2012;8:349 
How to cite this URL: Mahdavi SR, Rezaeejam H, Shirazi A, Hosntalab M, Mostaar A, Motamedi M. Conformal fields in prostate radiotherapy: A comparison between measurement, calculation and simulation. J Can Res Ther [serial online] 2012 [cited 2019 Sep 21];8:349. Available from: http://www.cancerjournal.net/text.asp?2012/8/1/34/95171 
> Introduction   
The major goal of radiotherapy is to deliver the maximum dose to tumor cells and the minimum possible one to normal tissue. Since there is a close relationship between local tumor control and normal tissue damage, any approach to deliver an accurate dose to target volume makes a part of quality assurance program and physicists' task in cancers radiation treatment. ^{[1]}
Treatment planning system (TPS) is at the heart of the radiation therapy planning process and the functionality and quality of any TPS is dependent on the type of algorithms used in the different steps of the planning process. ^{[2]} In order to check TPS function, a controlling program is needed for accuracy of treatments. This type of programs intends to disclose current problems before any effects on treatment and can improve radiotherapy outcomes, raising tumor control rates as well as reducing complication and recurrence rates. ^{[3]} For this purpose, different radiotherapy techniques (e.g. conformal fields) Have been developed to improve cancer treatment which allow escalation of total dose and dose per fraction as well as minimizing the dose of normal tissues. ^{[4]} In addition, there are many reports for using Monte Carlo simulation (MCS) for dose distribution calculation in treatment planning system. Comparison between experimental measurements and MCS results was also reported as a convenient method to verify MCS and its translation into clinical treatment planning. ^{[5],[6],[7]}
Goal of this study is to evaluate the accuracy of equivalent tissue air ratio (ETAR) algorithm used by CorePLAN three dimensional TPS for conformal blocked fields in treatment of prostate cancer against dosimetric measurements and in comparison to MCS (MCNP4C).
> Materials and Methods   
Five conformal fields were planned for 20 prostate cancer cases and finally an average shape was selected for tailoring the blocks. Blocks were shaped according to the prostate anatomy for five fields (anterior, two posterior oblique and two laterals) corresponding to their beams eye view (BEV) and with small changes for their geometry to make MCS more accurate and feasible [Figure 1]. Conformal blocks were built from Cerrobend standard alloy (e.g. lead, bismuth, tin and copper) with minimum thickness of 7.0 cm for both 6 and 18 MV photon beams [Figure 2]. Dose distribution was calculated by CorePLAN (Seoul CandJ. Inc.). It is a three dimensional treatment planning system using basically two methods of equivalent tissue air ratio (ETAR) and collapsed cone convolution for dose distribution calculation.  Figure 1: Designed blocks in 3D treatment planning system for anterior field (A), oblique fields (B) and lateral fields (C)
Click here to view 
Measurements were performed within the radiation field from a Varian 2100C/D linear accelerator that uses two 6 and 18 MV X ray beams in addition to multiple electron beams. Beam profiles and percentage depth dose (PDD) measurements in water were made using the field analyzer water phantom with dimensions of 50×50×50 cm ^{3} (PTW, MP2 Freiburg Germany). Profiles were measured at two different depths of 5 and 15 cm for 6MV and 10 and 20 cm for 18 MV photon beams with an ion chamber (PTW 23323, Freiburg Germany) with 0.1 cc volume. All measurements were in accordance with TRS 398 protocol of international atomic energy agency (IAEA). ^{[8]}
MCNP4C is validated in this project as the calculation code for simulation of linear accelerator and measuring dosimetric parameters [Figure 3]. To obtain dosimeteric parameters (PDDs and profiles) by this code, a water phantom consisted of cylindrical cells with a radius of 1 cm and thickness of 0.2 cm along the central (Y) axis was defined for PDD in addition to cylindrical cells with a radius of 0.5 cm and a thickness of 0.2 cm along the X axis at two depths (5, 15 cm for 6 MV and 10, 20 cm for 18 MV in conformal fields) for the beam profiles.  Figure 3: Schematic configuration of Linac head and dosimetric phantom for simulation
Click here to view 
Transport cutoff and geometry splitting were changed to increase MC calculation speed. The cut off energy for photon and electron were respectively 0.01 MeV and 0.5 MeV. The number of browsed particles in photon mode was about 8×10 ^{8} . Statistical uncertainty of MC results was less than 1% for PDD and beam profile calculations.
Before operating the code for conformal fields, the accuracy of code was validated for open field (10×10 cm ^{2} ). All PDD and profile curves were normalized independently to the depth of dose maximum and the center of the field.
For comparison between measurement and calculations, two regions in profile curves were considered: (1) the flat and umbra (e.g. low dose gradient) regions and (2) the penumbra (high dose gradient) region. ^{[9]} In the low dose gradient areas. It is convenient to trace dose difference (DD), while in the high dose gradient a distance to agreement (DTA), in millimeter (mm), is a better parameter to express deviations between measurements and calculations. Differences between the measured and calculated curves were based on the measured central axis dose for PDD curves and to the field central axis dose for profiles: ^{[10],[11]}
Collimator scatter factors (Sc) were measured to assess the collimator effect on changing from open to blocked field. Determination of difference between conformal and open field Sc's were tested for significance. The difference of Sc factor between conformal field (CF) and open field (OF) was obtained from following equation:
All measurements were repeated three times and the mean error of measurements was lower than 0.5%.
> Results   
For validation of code, beam profiles were calculated in different depths for two 6 and 18 MV photon beams. Results were compared with measurements to validate the model and they were shown in [Figure 4]. Findings revealed that for beam profiles the dose differences in flat region and DTA in penumbra are well below accepted tolerance limit of 2% and 2 mm, respectively for all energies and depths according to the report of Vensselar et al. ^{[9]} For PDD curves, the local differences between measurement and MC calculation are less than 1.5% at descending part for both of two energies which is below 2% limitation and in good agreement with the previous results. ^{[10],[12],[13],[14]}  Figure 4: Comparison of MC and measured beam profi le and PDD curve for 6 and 18 MV photon beam at depth of 5 and 15 cm at open field (10×10 cm2). (A) Beam profi le at 6 MV photon beam and (B) Beam profi le at 18 MV photon beam and (C) PDD curve at 6MV photon beam. (D) PDD curve at18 MV photon beam
Click here to view 
Several factors may cause the difference of 1.5% between our measurements and MC calculation: uncertainty in manufacturing materials of the target and flattening filter, statistical fluctuation of MC results, and fluctuations in water phantom measurements are the most probable reasons. For the buildup region, local differences of down to 10% were seen that can be due to the highgradient dose distribution in this particular region, which may make ionization chamber measurements unreliable. Also, finite size of the ionization chamber, which perturbs the absorbed dose, may be another reason for the large local differences. ^{[15]}
Results of Monte Carlo code verification against measurement are shown in [Figure 4] for 6 and 18 MV beams. For beam profile curve in an open field of 10×10 cm ^{2} in SSD=100 cm. The relevant curves with depths of 5 cm was multiplied by a coefficient of 0.75 for displaying two curves in one graph.
Regarding the results of beam profiles and PDD of MC to measurement, it is obvious that the average of total parameters (DTA and DD) are within the accepted criteria of DD=3% and DTA=3 mm. ^{[16],[17],[18],[19]}
While considering the beam profile curve in blocked fields and since these blocks create irregular fields, for the sake of simplicity, we divided beam profile curve into two left and right parts and then considered each part separately. [Figure 5]
For comparison between the results of TPS and measurement in conformal blocks, beam profile was calculated in two depth 5 and 15 cm for 6MV and depths 10 and 20 cm for 18 MV [Figure 6]. The profiles generally showed excellent agreement between TPS and measurement at all conformal fields. For the penumbra region (high dose gradient). The maximum DTA was 3.5 mm while the mean DTA was smaller than 3 mm at 6 and 18 MV and all fields [Table 1], For obtaining the mean DTA, the region of 80, 60, 40 and 20% of penumbra region in beam profile curve was calculated and their average with definition of "mean DTA" was declared and for the flat region (low dose gradient), the mean DDs were down to 2.5% for all depths and fields [Table 2].  Figure 6: Comparison of TPS and measured beam profile for 6 MV photon beam at depth of 5 and 15 cm and for 18 MV photon beam at depth of 10 and 20 cm. (A) Anterior filed and (B) Lateral field and (C) Oblique fi eld in 6MV photon beam. (D) Anterior fi eld and (E) Lateral field and (F) oblique field in 18 MV photon beam
Click here to view 
PDD was calculated in two nominal energies of 6 and 18 MV [Figure 7]. The mean DD between measurements and calculations were less than 2.5% for descending part of the curves for all blocked fields and energies [Table 3].  Figure 7: Comparison of TPS and measured PDD curves for 6 and18 MV photon beam. (A) Anterior field. (B) Lateral fi eld. (C) Oblique fi eld
Click here to view 
 Table 1: The results of comparing mean DTA of beam profi le curves between measurement and TPS for anterior, lateral and oblique blocks in two depths and standard conditions (**FS=10×10 cm^{2} , ***SSD=100 cm, ****NE=6,18 MV)
Click here to view 
 Table 2: The results of comparing the mean DD in flat regions of beam profile curves between measurement and TPS for anterior, lateral and oblique blocks in two depths and standard conditions (*FS=10×10 cm^{2}, **SSD=100 cm, ***NE=6, 18 MV)
Click here to view 
 Table 3: The obtained results out of comparing the mean DD in PDD curves of measurement with TPS for blocked fi elds at standard conditions (*FS=10×10 cm^{2}, **SSD=100 cm, ***NE=6,18 MV)
Click here to view 
For comparison between the results of TPS and MC in conformal fields, beam profile was calculated in two depths 5 and 15 cm for 6 MV and depths 10 and 20 cm for 18 MV [Figure 8]. In general, profiles showed excellent agreement between TPS and MC at all conformal fields. For the penumbra region (high dose gradient). The maximum DTA was 3.5 mm while the mean DTA was smaller than 3 mm for 6 and 18 MV and also for all fields [Table 4]. For the flat region (low dose gradient), the mean DD were below 2.5% for all depths and fields [Table 5].  Figure 8: Comparison of TPS and MC beam profi le for 6 MV photon beam at depth of 5 and 15 cm and for 18 MV photon beam at depth of 10 and 20 cm. (A) Anterior filed and (B) Lateral field and (C) Oblique field in 6MV photon beam. (D) Anterior field and (E) Lateral field and (F) oblique field in 18 MV photon beam
Click here to view 
 Table 4: The results of comparing the average DTA in beam profi le curves between MC and TPS for anterior, lateral and oblique blocks in two depths and standard conditions (**FS=10×10 cm^{2}, ***SSD=100 cm, ****NE=6,18 MV)
Click here to view 
 Table 5: The results of comparing the mean DD in flat regions of beam profile curves between MC code and TPS for anterior, lateral and oblique blocks in two depths and standard conditions (*FS=10×10 cm^{2}, **SSD=100 cm, ***NE=6, 18 MV)
Click here to view 
PDD was calculated for two nominal energies of 6 and 18 MV [Figure 9]. The mean of DDs between the measurements and calculations was less than 2.5% for descending part of the curves for all blocked fields and energies [Table 6].  Figure 9: Comparison of TPS and MC PDD curves for 6 and18 MV photon beam. (A) Anterior field. (B) Lateral fi eld. (C) Oblique field
Click here to view 
 Table 6: The obtained results out of comparing the mean DD for PDD curves with MC code and TPS for blocked fields at standard conditions (*FS=10×10 cm2, **SSD=100 cm, ***NE=6, 18 MV)
Click here to view 
Generally, many recommendations for acceptable levels of accuracy in treatment planning system have been published. ^{[16],[17],[18],[19],[20]} The specific parameters to be tested and their associated tolerances agreement to between 2 and 3% in low dose gradient regions and 2 3 mm in high dose gradients. In the UK, the guidelines in IPEM Report 81 ^{[16]} suggest an "ideal" agreement being to within 2% or 2 mm of the beam normalization value and "acceptable" levels for the same regions are 3% and/or 3 mm, respectively. Van Dyk et al,^{[20]} specified further limits for more complex situations, 4% or 4 mm within in homogeneities (but 3% in low dose gradient regions). Venselaar et al,^{[9]} suggested in complex geometry (wedge, homogeneity, asymmetry, block/MLC). agreement is 3% for low dose gradient region and 3 mm for high dose gradient region. Therefore, if the results are within the tolerance limits(less than 3% and 3 mm) there is no need for any corrective action, but if the results are more than the limits they will be reported as a significant difference.
> Discussion   
In this study, we have investigated the accuracy of dose calculation performed with equivalent TAR algorithm for the prostate conformal fields. Due to the complexity of photon and electron scattering and transport in blocked fields, ETAR algorithm, that used to employ certain approximations, may not be strong enough for dose calculation in the vicinity of the field borders. So, the meaningful difference reported by other researchers between the measurement and calculation can be from the algorithm failure.
However, our findings confirmed no significant difference between measurement, TPS and MC calculations for PDD and beam profiles. It was also showed that conformal blocks have no significant effect on the collimator scatter factor and there was no dramatic change when using a conformal block within a rectangular conventional limited field.
Monte Carlo calculation for megavoltage radiotherapy beams represents the next generation of dose calculation in the clinical environment. Our study was further emphasized on the accuracy and dosimetric potentials of the MCNP4C code for implementation in clinical treatment planning systems. However, the speed of simulations by computers should be considered for clinical practice. ^{[21]} The improved accuracy of these techniques will provide clinicians with better dose distributions in the presence of complex fields (e.g. conformal fields), heterogeneities, and small fields which exhibit electron disequilibrium and avoids the ionization chamber measurement and experimental setup.
> Acknowledgment   
We are greatful to radiotherapy personnel in Pars Hospital at Tehran, Iran for their assistance.
> References   
1.  Wambersie A. The role of the ICRU in quality assurance in radiation therapy. Int J Radiat Oncol Biol Phys 1984;10 Suppl 1:816. [PUBMED] 
2.  Andreo P, Cramb J, Fraass B.A, IonescuFarca F, Izewska J, Levin V, et al. Description of radiation treatment planning systems. Technical Report Series 430 Commissioning and quality assurance of computerized planning systems for radiation treatment of cancer, 1st edition. Vienna: Internatinal Atomic Energy Agency (IAEA); 2004. p. 1621. 
3.  Podgorsak EB, Andreo P, Evans MD, Hendry JH, Horton JL, Izewska J, et al. Quality assurance of externalbeam radiotherapy. Radiation oncology physics: A handbook for teachers and students. 1 ^{st} ed. Vienna : Internatinal Atomic Energy Agency (IAEA); 2005. p. 40748. 
4.  Huq S, Mayles P. Conformal radiotherapy. Transition from 2 D radiotherapy to 3 D conformal and intensity modulated radiotherapy. 1 ^{st} ed. Vienna: International Atomic Energy Agency (IAEA); 2008. p. 121. 
5.  Juste B, Miró R, Campayo JM, Díez S, Verdú G. Comparison of Experimental 3D dose curves in a heterogeneous phantom with results obtained by MCNP5 simulation and those extracted from a commercial treatment planning system. Appl Radiat Isot 2010;68:9137. 
6.  Chow JC, Jiang R, Leung MK. Dosimetry of oblique tangential photon beams calculated by superposition/convolution algorithms: A Monte Carlo evaluation. J Appl Clin Med Phys 2010;12:3424. 
7.  Juste B, Miró R, Campayo JM, Díez S, Verdú G. Radiotherapy treatment planning based on Monte Carlo techniques. Nucl Instrum Meth Phys Res A 2010;619:2527. 
8.  Andreo P, Burns DT, Hohlfeld K, Hug MS, Kanai T, Laitano F, et al. Code of Practice for high energy photon beams. Absorbed dose determination in external beam radiotherapy: An International Code of Practice for Dosimetry Based on Standards of Absorbed Dose to Water, 1 ^{st} ed. Vienna: Internatinal Atomic Energy Agency (IAEA); 2000. p. 6681. 
9.  Venselaar J, Welleweerd H, Mijnheer B. Tolerances for photon beam dose calculations using a treatment planning system. Radiother Oncol 2001;60:191201. 
10.  Bragg CM, Conway J. Dosimetric verification of the anisotropic algorithm for radiotherapy treatment planning. Radiother Oncol 2006;81:31523. 
11.  Mijnheer B, Olszewska A, Fiorino C, Hartmann G, Knoos T, Rosenwald JC, et al. Quality assurance of treatment planning systems practical examples for non IMRT photon beams. 1 ^{st} ed. Brussels: ESTRO: 2004. p. 1120. 
12.  Sheikh Bagheri D, Rogers DW. Sensitivity of megavoltage photon beam Monte Carlo simulations to electron beam and other parameters. Med Phys 2002;29:37990. 
13.  Hartmann Siantar CL, Walling RS, Daly TP, Faddegon B, Albright N, Bergstrom P, et al. Description and dosimetric verification of the PEREGRINE Monte Carlo dose calculation system for photon beams incident on a water phantom. Med Phys 2001;28:132237. 
14.  Verhaegen F, Seuntjens J. Monte Carlo modeling of external radiotherapy photon beams. Phys Med Biol 2003;48:10764. [PUBMED] [FULLTEXT] 
15.  Siebers JV, Keall PJ, Libby B, Mohan R. Comparison of EGS4 and MCNP4B Monte Carlo codes for generation of photon phase space distributions for a Varian 2100C. Phys Med Biol 1999;44:300926. 
16.  Mayles WP, Lake R, McKenzie A, Macaulay EM, Morgan HM, Jordan TJ, et al. Physics aspects of quality assurance in radiotherapy. York : The Institute of Physics and Engineering in Medicine (IPEM); 1999. 
17.  Fraass B, Doppke K, Hunt M, Kutcher G, Starkschall G, Stern R, et al. American Association of Physicists in Medicine Radiation Therapy Committee Task Group 53 : Quality assurance for clinical radiotherapy planning. Med Phys 1998;25:1773829. 
18.  Yu CX, Wong JW. Implementation of the ETAR method for 3d inhomogeneity correction using FTT. Med Phys 1993;20:62732. 
19.  Arnfield MR, Siantar CH, Siebers J, Garmon P, Cox L, Mohan R. The impact of electron transport on the accuracy of computed dose. Med Phys 2000;27:126674. 
20.  Van Dyk J, Barnett RB, Cygler JE, Shragge PC. Commissioning and quality assurance of treatment planning computers. Int J Radiat Oncol Biol Phys 1993;26:26173. 
21.  Mesbahi A, Mahdavi SR, Allahverdi M. Comparison of the different computer speed in calculating of Co 60 depth dose by MCNP4A and MCNP4B Monte Carlo codes. J Babol Univ Med Sci 2004;6:711. 
[Figure 1], [Figure 2], [Figure 3], [Figure 4], [Figure 5], [Figure 6], [Figure 7], [Figure 8], [Figure 9]
[Table 1], [Table 2], [Table 3], [Table 4], [Table 5], [Table 6]
