Discussion

It is a known fact that BBC is a rare breast cancer disease. Treatment of BBC with external radiotherapy aiming for breast conservation is very challenging as the target volume is very large and complex and there is always a requirement for minimising dose to critical organs like heart and lungs. Many authors have reported the used of IMRT, VMAT and HT techniques in BBC patients and their dosimetric advantages in compared to conventional tangential 3DCRT and FIF techniques.^{Reference Wadasadawala, Visariya, Sarin, Upreti, Paul and Phurailatpam2–Reference Wadasadawala, Jain and Paul5,Reference Nicolini, Clivio, Fogliata, Vanetti and Cozzi7} There are very few papers reporting the direct dosimetric comparisons of VMAT and HT on BBC cancers.^{Reference Sun, Lin and Tong8–Reference Zedef, Şilem, Oguzhan, Fikri, Ayse and Deniz10} Our study reports dosimetric comparison of VMAT and HT of BBC with SIB for tumour bed and along with nodal irradiation. The cases we had used are real clinical cases already treated with HT. We have conducted this dosimetric study to gain confidence in use of VMAT on BBC in our clinic and evaluate the efficiency of VMAT with HT. The dose constraints used in this study are based on reported literatures.^{Reference Wadasadawala, Visariya, Sarin, Upreti, Paul and Phurailatpam2,Reference Wadasadawala, Jain and Paul5,Reference Valli, Cekani, Cima and Richetti19–Reference Choi, Munden and Erasmus22} As larger volume of lungs is irradiated in adjuvant radiotherapy for BBCs, lung doses should be specifically restricted so that there will be no possibility of late lung injury manifesting as radiation pneumonitis with resultant radiation fibrosis.^{Reference Choi, Munden and Erasmus22} Ipsilateral mean lung dose and V ₂₀ are considered the most important factors for pulmonary toxicity.^{Reference Valli, Cekani, Cima and Richetti19} Lee et al., reported that V ₂₀ is the greatest risk predictor and defined a dose–volume percentage constraint of V ₂₀ < 37% for the irradiated ipsilateral lung to maintain the incidence of mild symptomatic radiation pneumonitis below 20% as defined by the Common Terminology Criteria for Adverse Events (CTCAE) in a cohort of breast cancer patients who received hybrid IMRT technique.^{Reference Lee, Chao, Chang, Ting and Huang20} Similarly, Darby et al., reported that radiation-induced heart disease is related to both mean heart as well as high doses >30 Gy.^{Reference Darby, Cutter and Boerma21} We used V ₅ for heart and lung to compare the low dose spillage.^{Reference Wadasadawala, Jain and Paul5}

The second objective of our study was to compare potential differences between VMAT and HT dosimetrically. The results of our study show that dosimetric parameters of VMAT plans were satisfactorily comparable with those of HT plans. VMAT could achieve better homogeneity over HT plans for PTV_primary and PTV_SCF. There was no significant difference in dose conformity achieved for all the targets between the two techniques. For heart also, there was no significant difference in the dosimetric parameters achieved from both the techniques. However, VMAT did not result in sparing of lung, especially with respect to the low dose spillage (V ₅). Thereby, we could see significantly higher mean dose to lungs in VMAT compared to HT. Similarly, Lauche et al., also reported that though HT and VMAT are feasible techniques in cases of complex target volumes for breast and nodal irradiation, both cause a large low dose spillage. So, a careful follow-up regarding lung, heart, contra-lateral breast is warranted in both the techniques.^{Reference Lauche, Kirova and Fenoglietto6}

To investigate further the impact of distribution of both low and high doses, we had considered the third objective. It is to compare the IDs to the whole body and OARs from both the techniques VMAT and HT. Many authors had evaluated the IDs of OARs and whole bodies from different techniques (3DCRT, IMRT, VMAT and HT) for different disease sites.^{Reference Sharma, Gupta, Jalali, Master, Phurailatpam and Sarin23–Reference Patel, Drodge, Jacques, Warkentin, Powell and Chafe26} The IDs are found to be comparable with no statistically significant variation between the two techniques. Other authors also reported comparable IDs of HT with different techniques of IMRT and 3DCRT for different disease sites.^{Reference Sharma, Gupta, Jalali, Master, Phurailatpam and Sarin23,Reference Aoyama, Westerly and Mackie24} Yang et al., also evaluated IDs of OARs and whole body for post-operative whole pelvic radiation therapy (WPRT) of endometrial cancers and found out that IMRT and HT were more conformed thereby gave lower IDs to OARs compared to 3DCRT. However, for whole body, they observed that that IMRT plans gave lowest ID and there was no difference between HT and 3DCRT.^{Reference Yang, Xu, Jiang, Xie and Wang25} In our study, the IDs for whole body and heart are found to be comparable with no statistically significant variation between VMAT than HT. However, ID to lung is found to be significantly higher in VMAT. In another planning study on craniospinal irradiation for paediatric medulloblastoma, Patel et al., had found that VMAT may reduce the ID while providing comparable normal tissue sparing with HT.^{Reference Patel, Drodge, Jacques, Warkentin, Powell and Chafe26}

The fourth objective was to assess the treatment time efficiency. VMAT technique was found to reduce the treatment time and total MU delivered by 53 and 78%, respectively, compared to HT. Our finding is in accordance with other studies.^{Reference Pasquier, Cavillon, Lacornerie, Touzeau, Tresch and Lartigau27–Reference Kinhikar, Pawar, Mahantshetty, Murthy, Dheshpande and Shrivastava31} We had observed that VMAT can deliver physical dose to BBC which is comparable with HT in a reduced delivery time. The lower the delivery time, the reduced possibility of treatment being affected by intra-fractional movement of the patient. Therefore, VMAT has a potential in improving the treatment accuracy.

Luca Cozzi et al., had also reported that there is limited data for VMAT in breast cancer with short follow up.^{Reference Cozzi, Lohr and Fogliata32} Many authors had reported the feasibility of VMAT for breast cancer for technically difficult cases like BBC with nodal irradiations. They had also reported modest and compatible toxicity profiles from VMAT compared to other techniques.^{Reference Fiorentino, Mazzola and Naccarato3,Reference Lauche, Kirova and Fenoglietto6,Reference Nicolini, Clivio, Fogliata, Vanetti and Cozzi7}

Fiorentino et al., reported their experience of the use of VMAT for BBC with SIB of tumour bed without comparison with HT. They reported mean heart and lung dose of 8·3 ± 3·3 Gy and 11·8 ± 2·3 Gy, respectively. The average V ₅ and V ₂₀ total lungs reported by them were 78·9 ± 15·3 and 15·7 ± 5%, respectively.^{Reference Fiorentino, Mazzola and Naccarato3} With our VMAT plans, we could achieve dose parameters which are lower than their mean doses with the heart mean dose (D _mean = 4·9 ± 0·6 Gy) and lung mean dose (D _mean = 10·6 ± 1·3 Gy). The average V ₅ and V ₂₀ (Table 4) from our study are 78·1 ± 8·3 and 12 ± 3%, respectively. Dağ et al., also had reported that V ₅ in VMAT planning was on average 85%, whereas in HT, it was only 45·5%.^{Reference Zedef, Şilem, Oguzhan, Fikri, Ayse and Deniz10} The average V ₅ of lungs achieved by our HT plans is 43·4%. As the series reported by Kaidar-Person et al., entailed inclusion of internal mammary targets, very high heart mean doses to the tune of 20 Gy (range 13–28) have been reported with acceptable lung V ₂₀ and V ₅.^{Reference Kaidar-Person, Kostich and Zagar33} Similarly, Karthik Raj et al., reported mean dose, mean V ₅ and V ₂₀ to the heart as 4·7 Gy, 24·9 and 16·7%, respectively, using the 3DCRT employing mono-isocentre bitangential beam arrangement in six cases of BBC.^{Reference Mani, Basu and Bhuiyan34} The mean heart dose achieved from our HT and VMAT plans are comparable to mean heart dose achieved from 3DCRT by Mani et al. (Table 3). However, in their study, the mean V ₅ and V ₂₀ for the total lung were 36·6% and 16·7%, respectively. In our study, the mean V ₅ for HT and VMAT plan was 43·4% and 78·1%, respectively, which is comparatively higher. This is the result of the use of inverse plan IMRT.^{Reference Lauche, Kirova and Fenoglietto6} However, we could achieve mean total lung dose of 8·4 and 10·6 Gy with HT and VMAT plans, respectively, which is quite lower than their result of 15·7 Gy. Franco et al., has also reported the promising clinical feasibility of TomoDirect for bilateral breast irradiation.^{Reference Franco, Migliaccio and Torielli35} But we had not explored the used of TomoDirect in the current study. Dosimetric results of this study for HT plans are quite similar to the previous dosimetric study conducted by our group.^{Reference Wadasadawala, Visariya, Sarin, Upreti, Paul and Phurailatpam2,Reference Wadasadawala, Jain and Paul5}

Many published papers have reported the reduction in the dose to heart and lung in left breast radiotherapy using deep inspiration breath-hold (DIBH) compared with FB.^{Reference Bruzzaniti, Abate and Pinnarò36–Reference Gaudino, Cima and Frapolli40} Dumane et al., had reported that VMAT combined with DIBH could reduce low dose to the OARs in the treatment of breast cancer patients with implant reconstruction receiving regional nodal irradiation.^{Reference Dumane, Saksornchai, Zhou, Hong, Powell and Ho39} However for HT, real-time motion management and breath control system is not available in the current time. In this study, we have not explored the role of motion management and breath control. Even though there is no report about the long-term clinical data from modern radiotherapy techniques on morbidity and mortality of cardiac structures, DIBH leads to a significant reduction in cardiac doses with as compared to FB plans. Hence, it is expected that cardiac morbidity will be reduced with the increased use of modern techniques including DIBH. Hong et al., had also reported average lower heart dose of 1·44 Gy in those patients treated with DIBH than those treated with FB.^{Reference Hong, Rahimy and Gross38} Similar dosimetric study was conducted by Gaudino et al. on synchronous bilateral breast cancer (SBBC) treated with adjuvant radiotherapy with DIBH. The authors reported a reduction in the maximum (19·2 Gy vs. 13·3 Gy) and the mean (6·5 Gy vs. 8·0 Gy) dose to the heart as compared to FB thus inferring improved OAR sparing from VMAT with DIBH in BBC adjuvant RT.^{Reference Gaudino, Cima and Frapolli40} VMAT with DIBH has got more potential in reducing dose to OARs compared to HT. However, it is important to note that the mean heart dose in the current study done on FB scans is still lower than the values reported by Gaudino et al. This suggests that the experience and skill of the planner plays an important role in the quality of plans.^{Reference Cilla, Ianiro and Romano41}

Sun et al., had also observed that different modern radiotherapy techniques (VMAT, IMRT and HT) could provide comparable good coverage dosimetrically to PTV while IMPT plans could provide the best dose coverage of target and sparing OARs in radiotherapy treatment of BBC. He suggested that VMAT and HT could be considered as a suboptimal technique for BBC patients as proton therapy being not commonly available technique and also more expensive.^{Reference Sun, Lin and Tong8} Cheng et al., also had concluded in their dosimetric study that because of longer treatment time in HT, there is a possibility of treatment uncertainty as patient discomfort will increase with longer treatment time. They had also observed that VMAT plans can deliver a plan with a better CI than did FIF and IMRT in shorter treatment time with acceptable doses to OARs.^{Reference Cheng, Chang, Shiau, Wang and Tsai9}