Dosimetric comparison and TCP-NTCP modelling for lung, heart, left anterior descending and right coronary artery in left sided breast cancer conventional and hypofractionated radiotherapy

The aim of this study was to evaluate the dose distribution, and also tumor control probability (TCP) and normal tissue complications probability (NTCP) models of left sided breast cancer females for 3D-CRT, 6 and 9 elds IMRT and hypofractionated tangential plans. Sixty left sided breast cancer females were included in this study. CT simulation images of the patients were imported on the treatment planning software (TiGRT, LinaTech, China), and the tangential treatment plans of the mentioned methods were done for each patient. The dosimetric evaluation, and TCP-NTCP models of stated modalities were done using Poisson Linear-Quadatric (PLQ) and Lyman-Kutcher-Burman (LKB) models on the MATLAB and R softwares. The mean (± SD) dose to ipsilateral lung, heart, LAD and RCA with/without internal mammary elds for 6FIMRT was lower compared to other modalities. Furthermore, V 20Gy for Ipsilateral lung and V 25Gy for heart, LAD and RCA of 6FIMRT was lower than other methods. In addition, the PTV dose coverage was higher for 9FIMRT and hypofractionated RT, while it may be lower for 3D-CRT among the studied methods. Although TCP values of 9 and 6eds and hypofractionated was not signicantly different, the TCPs of them were higher compared to 3D-CRT. However, the NTCP for ipsilateral lung, heart, LAD and RCA of 6FIMRT was lower than others.


Abstract Background
The aim of this study was to evaluate the dose distribution, and also tumor control probability (TCP) and normal tissue complications probability (NTCP) models of left sided breast cancer females for 3D-CRT, 6 and 9 elds IMRT and hypofractionated tangential plans.

Methods
Sixty left sided breast cancer females were included in this study. CT simulation images of the patients were imported on the treatment planning software (TiGRT, LinaTech, China), and the tangential treatment plans of the mentioned methods were done for each patient. The dosimetric evaluation, and TCP-NTCP models of stated modalities were done using Poisson Linear-Quadatric (PLQ) and Lyman-Kutcher-Burman (LKB) models on the MATLAB and R softwares.

Results
The mean (± SD) dose to ipsilateral lung, heart, LAD and RCA with/without internal mammary elds for 6FIMRT was lower compared to other modalities. Furthermore, V 20Gy for Ipsilateral lung and V 25Gy for heart, LAD and RCA of 6FIMRT was lower than other methods. In addition, the PTV dose coverage was higher for 9FIMRT and hypofractionated RT, while it may be lower for 3D-CRT among the studied methods. Although TCP values of 9 and 6 eds and hypofractionated was not signi cantly different, the TCPs of them were higher compared to 3D-CRT. However, the NTCP for ipsilateral lung, heart, LAD and RCA of 6FIMRT was lower than others.

Conclusion
6FIMRT is suitable choice for RT of breast cancer patients compared to other mentioned modalities, as a result of providing adequate PTV dose coverage and TCP, and also lower imposed dose and NTCP for OARs. Hypofractionated RT is a good alternative to reduce treatment time for the breast cancer patients.

Trial registration
This study was approved by the ethical board of Isfahan University of Medical Sciences, Isfahan, Iran (IR.MUI.MED.REC.1399.677). Background For high risk breast cancer patients, breast conserving surgery that followed by adjuvant radiotherapy (RT) is taken into account as the most common treatment method (1)(2)(3). Literatures have shown that, the number of breast cancer females have been increased during the recent two decades around the world (4-6). In this regard, a number of studies have shown that adjuvant RT of the mentioned patients may improve local control and survival of them (6-9). The RT of the patients is commonly done using two photon tangential beams, which irradiate the whole breast and the anterior part of thoracic volume (6, [10][11]. A large number of studies have been demonstrated that, RT of the patients may increase the normal tissue complications including, lung secondary cancer, and also heart morbidity and mortality which resulted from imposed radiation dose to cardiovascular structures (7,(12)(13)(14)(15)(16)(17). Zablotska and Neugut have reported that, the imposed radiation dose to lungs may pose to a moderate rise in complications for the mentioned organ (12). There are different RT modalities such as 3D-CRT and IMRT tangential plans which are commonly applied to treat mentioned patients. A host of studies have discussed about the imposed dose to the organ at risks (OARs) using different modalities (17)(18)(19)(20). Aznar et al. have found that, evaluation of the dose to the whole heart, arch and whole LAD is a suitable approach, because calculation of radiation dose to only one of these structures may lead to excessive imposed dose to them and also may increase the risk of cardiac complications (17). Zhang et al. have found that, IMRT may provide higher target dose coverage and dose uniformity compared to conformal RT for the left sided breast cancer patients (19).
Although many studies have been demonstrated that using tangential beams may provide enough dose distribution in the target volumes for the RT of the whole breast compared to other methods, data about depth dose distribution of above stated tangential methods and their complications on the normal tissues such as ipsilateral lung, heart and cardiovascular structures including, LAD, and also RCA are scarce. According to reading different studies and literatures, it would seem that, there is not an exact advantages and disadvantages of 3D-CRT, IMRT and hypofractionated RT with regard to their adverse effects to coronary arteries such as LAD and RCA. Therefore, this work aimed to evaluate the dose distribution, and also radiobiological models including tumor control probability (TCP) and normal tissue complications probability (NTCP) for target volume and normal tissues including ipsilateral lung, heart and cardiovascular structures namely, LAD and RCA in the above mentioned tangential plans.

Methods
The study protocol of this work was approved by the ethical board of Isfahan University of Medical Sciences, Isfahan, Iran (IR.MUI.MED.REC.1399.677) according to the 1975 Helsinki declaration and its revision in 2000.
Sixty left sided breast cancer patients who underwent RT in Sayed-Al Shohada hospital, Isfahan, Iran were included in this study. CT simulation images of the patients were imported on the treatment planning software (TiGRT, LinaTech, China), and the treatment plans of studied modalities including; 3D-CRT, 6 elds (6FIMRT) and also 9 elds IMRT (9FIMRT) inverse planning, and also hypofractionated IMRT for each patient were done by three independent expert radiation oncologist using the mentioned TPS. The applied TPS was commissioned according to a Siemens Primus linac measured data (Siemens, Germany).
Treatment planning of the patients was performed using 6 MV photon beams of the stated linac according to patient's geometry and chest wall sizes. In this study, the chest wall surface, three levels of axilla and supraclavicular lymph nodes were de ned as CTV. Furthermore, the planning target volume (PTV) was de ned with 1 cm margin around the CTV. Moreover, the OARs were ipsilateral lung, heart, LAD and also RCA. The prescribed dose for the 3D-CRT and IMRT inverse plans was 2 Gy per fraction (25 fraction) for PTV with a total dose of 50 Gy, while the dose for the hypofractionated RT was 2.3 Gy per fraction (20 fraction) with the total prescribed dose of 46 Gy. The source to surface distance (SSD) was 100 cm, and the above stated plans was done in SSD set up for each patients. The treatment plan was done based on RTOG protocol and the whole chest wall was irradiated by 6MV tangential photon beams. Furthermore, the mean imposed radiation dose to the OARs was measured with and without internal mammary nodes for each patient. In addition, the volume of the heart, LAD and RCA irradiating more than 25 Gy (V 25Gy ), and also the volume of ipsilateral lung exposing more than 20 Gy (V 20Gy ) were compared for each patient among the mentioned modalities. Moreover, the maximum heart distance (MHD) was measured using beam eye view (BEV).

TCP_ NTCP modelling
The TCP and NTCP for each of the above stated situations were calculated using DVH data. In this study, the Poisson Linear-Quadatric (PLQ) model was employed for estimating the TCP according the following below (equation 1) (21,22): Where, N is the initial number of tumoral cells, and the p s (D) is the cell survival fraction after a dose D.
Where, the t is calculated based on the following formula (equation 3): In the above equation, "m" is the maximum inverse slope of dose-response curve, EUD is the equivalent uniform dose of the mentioned OARs (25)(26)(27), and also D 50 is the total imposed radiation dose to the studied normal tissues that may lead to force 50% complication to the organ.

Discussion
For female breast cancer patients, RT is taken into account as the most common treatment choice, and to the best of our knowledge tangential plans are widely applied for these patients. A number of studies have discussed about the imposed radiation dose to OARs such as ipsilateral lung, heart and LAD (17)(18)(19)(20), but it is considered that there is a limited data about the imposed radiation dose to other organs such as LAD and RCA for 3D-CRT, IMRT and hypofractionated modalities. In addition, it would seem that, data about the evaluation of TCP and NTCP models using MATLAB and R programs for the target volumes and OARs among the mentioned method are scarce. Therefore, the aim of this study was to evaluate the imposed dose and TCP-NTCP models for the target volumes and OARs including; ipsilateral lung, heart, LAD and RCA using MATLAB and R programs with and without internal mammary nodes.
Based on our ndings, the mean (± SD) imposed dose (with/ without internal mammary elds) to ipsilateral lung, heart, LAD and RCA for 3D-CRT was signi cantly higher compared to 6 and 9 elds IMRT inverse plans (p < 0.005). Table 1 illustrates that, the mean (± SD) imposed radiation dose to the studied OARs for 6FIMRT was signi cantly lower than 3D-CRT (p < 0.005). In addition, the mean (± SD) dose of 6FIMRT for the discussed OARs was lower compared to 9FIMRT and hypofractionated IMRT, but it was not signi cant (p > 0.005). Furthermore, the V 20Gy for Ipsilateral lung and the V 25Gy for heart, LAD and RCA of 6 and 9 elds and hypofractionated (with/ without internal mammary elds) was signi cantly lower compared to 3D-CRT (p < 0.005), while there were not any signi cant differences for the V 20Gy of ipsilateral lung and the V 25Gy of heart, LAD and RCA among 6 and 9 elds with hypofractionated RT (p > 0.005).
Moreover, it was found that the mean (± SD) imposed dose to LAD was higher compared to heart and RCA for the all studied modalities, while the imposed dose to RCA was lower than heart and LAD among the mentioned RT methods (Table 1), which is highly due to their different anatomical features. Furthermore, it is considered that the minimum dose of heart, LAD and RCA was not signi cantly different for the 3D-CRT, IMRT and hypofractionated RT (p > 0.005).
Our ndings are in an agreement with Taylor et al. who stated that the imposed RT dose to the anterior part of heart including LAD is higher than whole heart (28). Recently, Gocer et al. have performed a dosimetric study on OARs of breast cancer patients including; heart, LAD, left circum ex coronary artery, right and left ventricles using tangential beams of 3D-CRT. In their study, they concluded that, the highest mean imposed dose was for LAD among heart and its coronary arteries for left-sided breast cancer patients (20), which is in line with our study. In addition, in their study, they have reported that the highest maximum radiation dose was for heart for left sided breast cancer patients (20). Moreover, the results of our study are similar to Rudat et al. who mentioned that the tangential IMRT plans may signi cantly reduce the dose-volumes of the ipsilateral lung and heart compared to tangential 3D-CRT for post mastectomy breast cancer patients (10). Aznar et al. have focused on the imposed dose to heart, arch and also whole LAD in respiration-adapted RT of leftsided breast cancers. In this study they found that, the radiation dose to arch and whole LAD was different for some of their patients, and thus it is demonstrated that the assessment of the dose to whole heart as well as to the whole LAD is crucial for the stated patients (17).
Based on Table 2, TCP models of MATLAB and R programs showed that, the TCP of the stated target volumes (with/ without internal mammary elds) for 9FIMRT and hypofractionated RT was higher than 6 elds and 3D-CRT, but it was not signi cant (p > 0.005). Whereas, the NTCP of the Ipsilateral lung for 6FIMRT was lower compared to the 9 elds and hypofractionated RT (p > 0.005). Also, the NTCP for 3D-CRT was signi cantly higher than other methods (p < 0.005). Li et al. have found that the TCP of conventional RT and IMRT is more than 90%, but the NTCP of lung for IMRT is less than conventional method (29), which is similar to our results. Furthermore, our ndings showed that the NTCP of heart and its coronary arteries including LAD and RCA for 6FIMRT was lower compared to 3D-CRT, 9FIMRT and hypofractionated method. Moreover, our data showed that the hypofractionated RT may decrease the treatment time, which is in line with Li et al. who mentioned that their used hypofractionated technique may reduce the RT time for their breast cancer patients (29).
According to results of our study, 6FIMRT not only may provide suitable PTV dose coverage, but also may impose lower complication probabilities to OARs compared to 9FIMRT, hypofractionated RT and 3D-CRT. Nevertheless, hypofractionated method may be a good alternative to reduce breast cancer treatment time.

Conclusion
In this study, the dose distribution and TCP-NTCP models of 3D-CRT, 6 and 9 elds IMRT inverse plans and hypofractionated IMRT are evaluated for left sided breast cancer patients. Based on our ndings, 6F IMRT inverse planning is well worth treating left sided breast cancer patients, due to the enough dose coverage for PTV, suitable TCP for target volumes and lower NTCP for ipsilateral lung, heart, LAD and also RCA compared to others. However, as a result of lower treatment time for the hypofractionated IMRT, using this modality is suggested.