

ORIGINAL ARTICLE 

Year : 2019  Volume
: 10
 Issue : 2  Page : 96103 

The effect of breast phantom, and geometry on dose distribution in breast brachytherapy using the strutadjusted volume implant and contura applicators
Maryam Papie^{1}, Sedigheh Sina^{2}, Reza Faghihi^{1}
^{1} Department of Nuclear Engineering, Shiraz University, Shiraz, Iran ^{2} Department of Nuclear Engineering; Radiation Research Center, Shiraz University, Shiraz, Iran
Date of Web Publication  9Sep2019 
Correspondence Address: Dr. Sedigheh Sina Department of Nuclear Engineering; Radiation Research Center, Shiraz University, Shiraz Iran
Source of Support: None, Conflict of Interest: None  Check 
DOI: 10.4103/jrcr.jrcr_16_19
Context: Accelerated partial breast irradiation using brachytherapy is a postlumpectomy treatment to reduce cancer recurrence and a choice for people in the earlystage breast cancer. Although accurate dosimetry is necessary to obtain successful clinical outcomes, the usual commercial treatment planning systems use a simple water phantom to simulate the patient. Hence, the precise attenuated radiation and also the scattering effects occurred in real situations may be dosimetrically ignored. Aims: The purpose of this study is to use Monte Carlo simulation to obtain the effect of phantom material and geometry corrections on dose distribution of the strutadjusted volume implant (SAVI) and Contura highdoserate brachytherapy applicators of breast cancer. Settings and Design: Contura with four lumens surrounding the central one and also SAVI with eight peripheral source channels are separately simulated into the breast phantoms. 192Ir high dose rate sources are located on dwell positions in each applicator. Subjects and Methods: The applicators were simulated inside three different phantom geometry and materials. The dose distribution and dosevolume histograms for each phantom were obtained for typical treatment. Gamma index evaluation is performed to examine the dose distribution according to the water phantom for each trial. Results: According to the results for SAVI and Contura applicators, breast material correction shows about 1% deviations from the calculations for water in most points of the breast. Maximum differences are not >3% that are found near the skin. Conclusion: Deviations from the water phantom in both SAVI and Contura treatments show good conformity especially in Contura and it seems that no serious dosimetric correction is necessary for simple water phantom. Although the results for SAVI were not observed with great deviations from water, areas with highgradient dose need to be precisely considered.
Keywords: Attenuationpartition based algorithm, breast cancer, breast composition, high dose rate brachytherapy
How to cite this article: Papie M, Sina S, Faghihi R. The effect of breast phantom, and geometry on dose distribution in breast brachytherapy using the strutadjusted volume implant and contura applicators. J Radiat Cancer Res 2019;10:96103 
How to cite this URL: Papie M, Sina S, Faghihi R. The effect of breast phantom, and geometry on dose distribution in breast brachytherapy using the strutadjusted volume implant and contura applicators. J Radiat Cancer Res [serial online] 2019 [cited 2019 Nov 18];10:96103. Available from: http://www.journalrcr.org/text.asp?2019/10/2/96/266119 
Introduction   
Breast cancer is one of the most common types of cancer with about 1 diagnosed case in 8 women.^{[1],[2]} There are some methods developed for cancer treatment that are especially depend on tumor size. Two centimeters or less breast lesion sizes categorize the disease as the earlystage breast cancer (Stage II or less). Breastconserving therapy is a postlumpectomy treatment to reduce cancer recurrence and a choice for people in the earlystage breast cancer that is recommended by the United States National Institute of Health. Good results and cosmetic outcome are made it as a preference option for patients, but it is only received by about 70% of them mainly because of limitations for who cannot access easily to the treatment facilities due to 6–7 weeks long course of radiation therapy.^{[1]} Accelerated partial breast irradiation (APBI) is an effective treatment for breast carcinomas. APBI utilizes some methods to deliver hypofractionated treatment with limited number of fractions that make it a preferred method for a large group of patients in the earlystage breast cancer. It is also presented by the cosmetic effect.^{[3],[4]}
The balloon catheter MammoSite is one of the designed methods that was rapidly accepted by physicians and patients. However, this kind of treatment is not suitable when the balloon surface is too close (<1 cm) to the skin. The dose optimization is very restricted due to the fixed symmetric geometry and the limited number of the source dwell positions.^{[5]} Contura is another balloon applicator similar to MammoSite that is used to deliver intracavity radiation in APBI treatments. Contura with four surrounding lumens for loading high doserate (HDR) sources from an after loader system can provide an assymetric dose distribution. It can give a more flexible shape of dose. This ability helps to spare normal tissues and skin. Each lumen is separated by 90° and it has 7 dwell positions to locate the sources into a 4–5 cm or 4.5–6 cm diameters balloon that is filled with saline to make contrast [Figure 1].^{[2]}  Figure 1: (a) Contura multilumen applicator,^{[3]} (b) strutadjusted volume implant with expanded struts^{[1]}
Click here to view 
The strutadjusted volume implant (SAVI) (Cianna Medical, Aliso Viejo, CA, USA) is an effective method of HDR brachytherapy treatment for earlystage breast carcinomas. SAVI place into a lumpectomy cavity through a single skin incision. It contains a central source channel, and also 6, 8, or 10 peripheral source channels which can be differentially loaded. Acceptable dose conformity to the lumpectomy cavity can be obtained by loading the channels with varying length of time. Unlike MammoSite, SAVI can be used effectively for the treatment of the regions near the skin or chest wall with good planning target volume (PTV) coverage and acceptable dose to the skin, chest wall, ribs, and lung.^{[5],[6]} [Figure 1] shows the Contura and SAVI applicators.
Accurate dosimetry is necessary to avoid deviations in delivered dose from that prescribed for a successful treatment. Currently, water is used as the reference dosimetry phantom material in treatment planning systems. The American association of physicists in medicine task group No. 186 (TG186) is recommended using the modelbased dose calculation algorithms that take the exact determination of geometry, atomic ratios, and density of all materials including tissue and applicators into account to simulate the real treatment condition.^{[7]}
In this study, Monte Carlo simulation using MCNP5 code was employed to consider the effect of corrections on phantom material and geometry in dose distribution of Contura and SAVI HDR brachytherapy applicators.
Subjects and Methods   
Monte Carlo simulations with using MCNP5 code were performed to evaluate the dosimetry investigations. As shown in [Figure 2]a, Contura with four lumens surrounding the central one was simulated with seven dwell positions for^{192} Ir HDR sources on each lumen that were simulated in a water sphere as the salinefilled balloon (2.2 cm radius). SAVI with eight peripheral source channels was simulated in the breast phantom. Twelve sources were simulated inside the central lumen, and the eight peripheral lumens were loaded with 13 HDR^{192} Ir sources [Figure 2]b. In both applicators, all sources were simulated at the same time. Each source was composed of pure^{192} Ir cylindrical sources with 0.5 cm long and 0.017 radius. No wire was considered.  Figure 2: Simulated models of (a) Contura multilumen applicator, (b) strutadjusted volume implant with expanded struts
Click here to view 
The simulated phantom is included two cubes of water to introduce the homogeneous water bodies of the breast (a 12 × 12 × 12 cubic body) and chest (phantom1), as shown in [Figure 3]. In Contura treatment, a spherical part (with the radius of 3.2 cm) into the breast is assumed as the surrounding region with 1 cm margin from the balloon surface that is defined as the PTV_EVAL. PTV_EVAL for SAVI is included an ellipsoidshaped region consisting lumpectomy cavity that surrounds 1cm margin around the expanded struts of SAVI.^{[8]} Several corrections were performed on phantom geometry and content in three next followed steps of simulation. The first correction is included using the elemental composition (9.4% H, 61.9% C, 3.6% N, 24.5% O, 0.6% Ca) defined by the international commission on radiation units and measurements report 44 as the breast tissue instead of water (phantom2). In the next trial, an ellipsoid of air was considered to evaluate the effects of airfilled lung into the chest. Furthermore, a cavity of air into SAVI was defined in this simulation. This region is created by struts in the lumpectomy cavity rather than seroma^{[5]} (phantom3). All three phantoms for each treatment are shown in [Figure 3].  Figure 3: Simulated geometries for (up) Contura and (down) strutadjusted volume implant treatments. Phantom materials including (red) water, (yellow) air, and (violet) breast International commission on radiation units and measurements report 44 composition
Click here to view 
Dosimetric considerations were performed with using *f4 mesh tally data multiplied by appropriate energy absorption coefficients. With an assumption that 95% of prescription dose is delivered to the PTV boundary, normal values of data were used to investigations. 0.5 billion particle histories were performed to obtain data with <1% uncertainties. Also to have reliable measurements, the dose calculated using *f4 and f6 tallies were compared in spherical cells into the breast. The results were observed with good consistency.
Quantitative comparison with using dosevolume histogram (DVH) analysis with MATLAB R2014a software is employed to evaluate the data. In addition, the results are displayed as superimposed isodoses for all phantoms and also the percentage dose difference (DD) with corresponding gamma (γ) index plots. Gamma index is a tool to investigate the acceptability of dose calculations based on DD and distancetoagreement (DTA) distributions. DTA is defined as the distance of the nearest data point on a dose distribution that is found with the same dose on the other dose distribution, for example, both the measured and calculated dose distributions can be compared to investigate the conformity with using the Gamma index. For the spatial location of r_{m} and r_{c} on the measured and calculated distributions with dose points of D_{m} and D_{c}, acceptable data are limited to a γ≤1 if Γ function be defined as follows (with using equations 1–4):^{[9]}
[Figure 4] shows the geometric demonstration of the above equations. In this study, gamma index is considered for DD and DTA with the criteria of 2% and 2 mm, respectively. All the plots are provided by Surfer Version 12.0.626 (Surfer Mapping System, 2014, Golden Software, Inc, Colorado, USA).  Figure 4: Geometric display of dose distribution criteria for combined dose difference and distancetoagreement parameters, (a) twodimensional, and (b) onedimensional^{[9]}
Click here to view 
Results   
[Figure 5] shows the dose distributions that are obtained for point by point data sets.  Figure 5: Dose distributions (relative dose [%]) in four simulated breast phantoms with, (a) Contura; and (b) strutadjusted volume implant. Dimensions of the breast are presented in term of centimeter
Click here to view 
In addition to dose distributions, the point data values of percentage DD and γ index for both treatments can be graphically evaluated with using their plots in [Figure 6] and [Figure 7]. The figures show the plots of the percentage DD for each phantom (Phantom2 and Phantom3) related to the water one and also their corresponding γ distribution are presented subsequently.  Figure 6: (a and b) Percentage difference for phantom2 and 3 related, to the water (phantom1) (%) in Contura treatment and (c and d) Gamma index plots for the both
Click here to view 
 Figure 7: (a and b) Percentage difference for phantom2 and 3, related to the water (phantom1) (%) in strutadjusted volume implant treatment and (c and d) gamma index plots for the both
Click here to view 
Absolute values mean of percentage differences for phantoms 1 and 2 are obtained with 0.54 and 0.62 for Contura and also 0.44 and 0.75 for SAVI, respectively. In addition, standard deviations of 0.33 and 0.39 are found in Contura evaluations and data of SAVI show 0.36 and 0.48, respectively. In addition, data are presented with 0.0055 and 0.0065 standard deviation of the mean in Contura evaluations and also 0.006 and 0.008 for SAVI in phantoms 1 and 2, respectively.
As shown in [Figure 1], dose distributions for Contura treatment with both corrections show good consistence with phanom1. According to the percentage difference plots in [Figure 6], water breast show <3% overestimations at the skin and no significant differences is observed in the other regions. Deviations at the presence of lung into chest do not show more serious differences and only have a few greatest values (about 3% at the skin related to the phantom1). In both conformity examinations of dose distributions, Gamma index plots are found with γ <1 in all of the breast areas. Regions close to the sources show unacceptable values of γ that are restricted to the areas inside of the balloon.
Based on the results of SAVI simulations, breast material correction shows about 1% deviations from the calculations for water in most points of the breast. Maximum differences are not >3% that are found near the skin. In addition, Gamma index plots confirm the results. Dose distributions with about 3% decreased values are estimated at the skin and areas close to the sources. Almost conformity at all regions near the sources is declined by gamma index.
The results of DVH analysis performed for PTV and also total breast region in all phantoms are shown as superimposed curves for each applicator in [Figure 8] and [Figure 9]. Breast material correction does not show obvious different histograms in phantoms for both applicators, but air cavity into SAVI is caused explicit deviations in PTV DVHs from homogeneous water phantom. Based on the percentage difference plots, these deviations with up to 3% increased calculated values are found only in the air and not into the tissue.  Figure 8: Dosevolume histogram for planning target volume in high dose rate breast brachytherapy with contura
Click here to view 
 Figure 9: Dosevolume histogram for planning target volume in high dose rate breast brachytherapy with strutadjusted volume implant
Click here to view 
[Table 1] and [Table 2] represent volumes that received 90%, 95%, 100%, 150%, and 200% of the prescribed dose (V90, V95, V100, V150, V200) for both total breast and PTVEVAL regions in all three phantoms for Contura and SAVI simulations, respectively. Estimations are demonstrated as cc and the percentage of their volumes.  Table 1: Dosimetric parameters obtained from dosevolume histogram analysis for contura treatment
Click here to view 
 Table 2: Dosimetric parameters obtained from dosevolume histogram analysis for strutadjusted volume implant treatment
Click here to view 
According to [Table 1], the volume difference for V90, V95, V100, V150 and V200 of the total breast for both phantoms2 and 3 are determined by 0.5cc, 0.33cc, 0.34cc, 0.17cc, and 0cc less than phantom1 in Contura treatment. Estimations for SAVI show 0.17cc decreased V90 and V100 for phantom2 compared to phantom1 and also differences of −1.84cc, −2.08cc, −1.96cc, −1.87cc, and 1.56cc are observed in phantom3.
Discussion   
Data reported by Thrower et al. including 98%, 96% and 92% with using a commercially available Oncentra brachy treatment planning system (OncentraACE) and also 96%, 94%, and 89% with using a collapsed cone convolution algorithm regarded to V_{90}, V_{95}, and V_{100} are almost close to the corresponding obtained results for three phantoms in this study.^{[10]} PTV_EVAL coverage of Contura is greatly kept under the planning criteria following National Surgical Adjuvant Breast and Bowel Project (NSABP) guideline for multicatheter breast brachytherapy including V_{90 }>90%, V_{150 }<50cc, and V_{200 }<20cc.^{[11]}
The maximum percentage DD for Contura is observed with 2%–3.2% at the skin with <10% isodose coverage. However, γ index exhibits great conformity for shallow dose gradient in calculations for Contura. The larger γ values close to the sources are restricted into the balloon, and the limited exceeding values are found in some regions in PTV that are very closed to the balloon surface. The limited areas with insignificant overestimation for water phantom can conservatively show an acceptable calculated dose distribution.
The plots of dose distributions and percentage difference for Contura present good agreement with the water phantom. This is confirmed by the corresponding plot of SAVI after material correction, but the air cavity is made some differences. Deviations are restricted into the air and its underestimation in water phantom does not seem like a serious problem, but it is required that the other evaluations demonstrate that too. NSABP guideline for SAVI is also similar to Contura,^{[6]} but the results show unacceptable values for V_{150} and V_{200} in this simulation. In addition, Gamma index is completely declined at regions with highgradient dose so closed areas to the sourcetissue interfaces. It may be due to the fact that because of very near the sources and the metals inside tissue, chargedparticle equilibrium (CPE) is not valid^{[7]} and using fluxed averaged over a cell (MeV/cm^{2}) multiplied attenuation coefficient may give estimations with some inaccuracy. It is necessary to consider this issue accurately with emphasis on highdose gradient dosimetry.
Conclusion   
Attenuation and scattering effects in breast materials do not differ obviously from the water. Deviations from the water phantom in both SAVI and Contura treatments are found with less than uncertainty of Monte Carlo calculations. Because of good conformity for Contura it seems that no serious dosimetric correction is necessary for simple water phantom. Although the results for SAVI were not observed with great deviations from water, areas with highgradient dose need to be precisely considered.
Financial support and sponsorship
Nil.
Conflicts of interest
There are no conflicts of interest.
References   
1.  Njeh CF, Saunders MW, Langton CM. Accelerated Partial Breast Irradiation (APBI): A review of available techniques. Radiat Oncol 2010;5:90. 
2.  Pearson D, Williams EA. Dose modification factor analysis of multilumen balloon brachytherapy applicator with Monte Carlo simulation. J Appl Clin Med Phys 2013;15:54–62. 
3.  Arthur DW, Vicini FA, Todor DA, Julian TB, Cuttino LW, Mukhopadhyay ND. Contura multilumen balloon breast brachytherapy catheter: Comparative dosimetric findings of a phase 4 trial. Int J Radiat Oncol Biol Phys 2013;86:2649. 
4.  Rogers KL. Accounting for Heterogeneity Corrections in Breast Brachytherapy; 2017. 
5.  Koontz C. Monte Carlo Investigation on the Effect of Heterogeneities on Strut Adjusted Volume Implant (SAVI) Dosimetry; 2013. 
6.  Scanderbeg D, Yashar C, White G, Rice R, Pawlicki T. Evaluation of three APBI techniques under NSABP B39 guidelines. J Appl Clin Med Phys 2009;11:3021. 
7.  Beaulieu L, Carlsson Tedgren A, Carrier JF, Davis SD, Mourtada F, Rivard MJ, et al. Report of the task group 186 on modelbased dose calculation methods in brachytherapy beyond the TG43 formalism: Current status and recommendations for clinical implementation. Med Phys 2012;39:620836. 
8.  Gurdalli S, Kuske RR, Quiet CA. Dosimetric characteristics of SAVI: A new singleentry multicatheter breast brachytherapy device. Brachytherapy 2007;6:97. 
9.  Low DA, Harms WB, Mutic S, Purdy JA. A technique for the quantitative evaluation of dose distributions. Med Phys 1998;25:65661. 
10.  Thrower SL, Shaitelman SF, Bloom E, Salehpour M, Gifford K. Comparison of dose distributions with TG43 and collapsed cone convolution algorithms applied to accelerated partial breast irradiation patient plans. Int J Radiat Oncol Biol Phys 2016;95:15206. 
11.  Fisher ER, Dignam J, TanChiu E, Costantino J, Fisher B, Paik S, et al. Pathologic findings from the National Surgical Adjuvant Breast Project (NSABP) eightyear update of protocol B17: Intraductal carcinoma. Cancer 1999;86:42938. 
[Figure 1], [Figure 2], [Figure 3], [Figure 4], [Figure 5], [Figure 6], [Figure 7], [Figure 8], [Figure 9]
[Table 1], [Table 2]
