The difficulty of PMRT treatment planning has been shown mainly because of complex PTV geometry structures and uncertainly relative position of OARs and PTV. In our study, three techniques have been compared for PMRT patients and all methods provide an acceptable dose coverage to the target volume. While no specific technique could be superior in all criteria, the VMAT technique could significantly reduce IL and heart dose and achieve a good balance between planning target coverage and OARs sparing. However, IMRT and H-VMAT have lower dose in CB and CL. The importance of radiation-related occurrence rate in surrounding OARs should be illustrated.
The study from Darby et al. reported that reducing the mean heart dose 1 Gy could proportionally avoid a 7.4% rate of the risk of an acute coronary event . Hence, one of the most critical objectives in planning breast cases could be keeping the mean heart dose as low as possible. Jacobse et al. show that myocardial infarction (MI) rate linearly affected by mean heart dose and decreasing whole heart mean dose is excepted contribute better cardiovascular healthy . In multivariate analysis study of Chung et al including 1742 patients followed by 5–12 years indicated that mean heart dose increased 1 Gy, the risk of breast cancer treatment-related heart disease events significantly increased as adjusted hazard ratio: 1.23 . In addition, diabetes mellitus (DM) and history of heart disease (HTN) were found to be adverse risk factors as hazard ratios: 1.92 and 2.51 respectively. The dosimetry comparisons have been evaluated in different techniques over the last decades. Popescu et al. compared IMRT, 3DCRT, and VMAT planning techniques and reported mean heart dose of the VMAT technique is 10.9 Gy (range 9.2 to 11.0) . Xie et al. demonstrated nine left-sided PMRT patients including six advanced techniques and non-coplanar VMAT has the lowest mean heart dose as 7.4 ± 1.2 Gy . Kuo et al. investigated the plans using the VMAT technique achieved 95% prescription dose to 95% volume have 7.5 ± 1.1 Gy for left-sided breast cancer . Furthermore, their study based on 95% covering 95% of the prescription dose, our study shows the novel VMAT technique demonstrated that the mean heart dose is 5.0 ± 1.4 Gy for left-sided FB cases lower than all the reported study above with 95% covering 100% prescription dose condition.
However, the dose of VMAT technique for contralateral breast and contralateral lung significantly higher than IMRT and H-VMAT. The influence of dose to the contralateral breast in breast cancer has been illustrated by Stovall et al. The study shows that radiotherapy did not play a vital role in the development of second breast primary, but younger patients are more likely to have a long-term risk of developing breast cancer . McDuff SGR et al. evaluated the patient who younger than 45 years old and harbor certain rare ATM variants should be noticed the risk of developing contralateral breast cancer . However, inconsistency of published data limits the precise recommendations for contralateral breast cancer dose. Choi et al. propose that the number of 747 breast cancer patients with IMNs-PTV can significantly increase radiation exposure to the lungs, but the incidence and severity of radiation pneumonitis were minimal and acceptable . Chung et al. demonstrated that despite the large volume of lungs receiving a low dose, the incidence of grade 3 of radiation pneumonitis is significantly low . Wen et al. illustrated that study including 515 patients shows the risk of symptomatic radiation pneumonitis could be evaluated only by V20 and V30 .
The main limitation of this study for left breast cancer undergoing PMRT is not using deep inspiration breath hold (DIBH) to generate a planning image instead of using Free-breathing scanning technique. DIBH technique is highly recommended in Wiant et al. study including 25 patients, the mean heart dose using DIBH is 1.4 Gy and 3.0 Gy using free-breathing for whole breast cancer . The study from Kim et al involved 69 patients mean heart dose decrease from 3.7 Gy to 1.8 Gy, reduced around 51%  and 48 patients involved in the study of Dong et al. shows that heart D mean reduced from 5.4 Gy to 3.6 Gy . However, the DIBH technique has shown a longer beam-on time and treatment time which could lead to uncertainty of treatment delivery. Poeta et al demonstrated the 4 partial arcs split-VMAT plan could significantly reduce treatment delivery time from 8.9 mins to 5.4 mins compared to conventional treatment and each partial arc around 30 seconds which similar to our study (average partial arc approximately 27 seconds) . Base on their feedback from treatment delivery, most patients were able to comfortably hold their breath for more than 20s, the feasibility of our technique should be further analyzed. While CI (0.85 vs 0.61) and delivery time (2.6mins vs 5.4mins) of our VMAT technique is superior to using split-VMAT, our study shows more mean heart dose (5.0 Gy vs 2.9 Gy). The heart and other critical organs dose of our technique using DIBH could be evaluated in further study. In order to decrease the uncertainty of delivery, H-VMAT could be recommended when RHVTL is lower than 0.06. Tangential fields as a base plan with a larger margin could to some extend improve PTV coverage in terms of tumor motion during treatment.
The evaluation of the planning complexity of left-sided breast cancer has been discussed. The measurement of Heart Volume in Tangent Line and the ratio of HVTL and heart volume have been first proposed in this study. The unique VMAT technique could be more unaffected by a relative position of anatomy. But for a more conventional method like IMRT and 3DCRT, the relationship between RHVTL and mean heart dose have a significantly high value. So, in our institution, we generally use unique VMAT when the traditional techniques meet the difficulty achieving target coverage and normal tissue constraints. However, in order to obtain a more reliable pattern, more data will be needed for analysis and validation. The value of RHVTL exceeding 0.06 may not be the most appropriate cut-off value for the determination of personalized planning techniques. More advanced criteria should be further illustrated.