Background
Breast cancer is the most common cancer in women, and 1–2% of these patients are diagnosed with synchronous bilateral breast cancer (SBBC) [
1]. While the incidence of SBBC is low, it presents significantly poorer overall survival than that of unilateral breast cancer [
2,
3]. There is no standard guideline for treating SBBC, and owing to an increasing demand for breast-conserving treatment in many cases, synchronous bilateral breast irradiation is commonly required.
Radiotherapy (RT) for unilateral breast cancer has used a tangential field with two-dimensional (2D) or three-dimensional conformal radiotherapy (3D-CRT). However, when the traditional field is applied to SBBC, overlapping of the RT fields could be inevitable, thereby compromising the target coverage. There is also a problem with the daily set-up of the patients, especially in the case of RT including regional node irradiation (RNI). Multiple isocenters are used in this method, so setting up the patient’s posture may often be inaccurate. Moreover, a larger treatment volume is required for SBBC, so the radiation dose to organs at risk (OARs), like the lung and heart, is considerable.
For treating such a complex target volume, recent trends have shown that intensity-modulated radiation therapy (IMRT) using helical tomotherapy or volumetric-modulated arc therapy (VMAT) are applied. When IMRT is used for SBBC, problems associated with the isocenter and junction can be solved. Because the irradiation beam in IMRT emits in many directions, a large volume of the lung and heart may be irradiated unnecessarily during bilateral breast treatment. Therefore, physicians have tried to reduce the dose to OARs by using hybrid-VMAT, limiting the beam direction, or using a static angle only in tomotherapy [
4‐
6].
However, there is still no standard for the best plan, and in most previous studies, only RT for bilateral breasts without RNI was considered. In addition, the optimal RT plan considering the morphologic variation of patients (e.g. large/small breasts or a funnel-like chest) has not been reported. In this study, we evaluated the optimal RT plan for SBBC, especially treatment plans including RNI, and analyzed the strategies considering the patients’ morphologic variations.
Discussion
In this study, the optimal RT plan for SBBC, especially when including regional LN, was investigated. The PTV is very large for treating the bilateral breasts and regional LN area. Thus, radiation exposure to OARs, like the lungs, increases compared to the PTV of unilateral breast cancer. A modified prescribed dose in a 3D plan for the right breast, considering the low dose distribution of the previous VMAT plan for the left breast and regional LN (i.e., the modified hybrid plan), was the best way to reduce the OAR dose while delivering the appropriate target volume coverage.
VMAT for the left breast and regional LN area improves target volume coverage and reduces treatment time and radiation dose to the lung, heart, and even LAD compared to the 3D plan [
11,
16,
17]. However, despite these dosimetric advantages of VMAT, treatment for SBBC is more complicated. When using the arc in multiple directions to treat both breasts, a larger volume of the bilateral lungs is inevitably exposed to irradiation. To solve this problem, Subramanian et al. proposed the hybrid–VMAT (h-VMAT) technique [
5]. The h-VMAT planning involves two steps. First, a field-in-field (FIF) forward planning setup with 80% of the prescription dose was planned for both breasts. The heart and lungs were spared using a high definition multi-leaf collimator. Second, the remaining 20% prescription dose for both the breasts was optimized using VMAT with three continuous arcs (arc length: 150°–210°) by keeping the dose delivered in a FIF arrangement as the base dose plan. With this method, the radiation dose for the lung was significantly decreased compared to that in the conventional VMAT plan. However, this strategy could be applied only for SBBC, but not including regional LN irradiation, and additional evaluation is needed.
The need to control and minimize the impact of breathing motion during IMRT for breast cancer has been investigated by several groups, and the breath-hold technique or breathing gating technique was identified as the most convenient and safe approach. As multileaf collimators without fully-opening jaws are used for VMAT, several set-up or positional errors could occur. Efforts are needed to reduce this gap, and Nicolini et al. suggested optimal VMAT planning for breast cancer [
18]. Alternative images were generated with an artificial expansion of 10 mm from the body in the breast region and additional PTV was contoured to this image. The two treatment plans, which are performed on original and alternative images respectively, were optimized using original and alternative images. The proposed planning strategy could represent a robust approach that could account for moderate changes in target or body volume during the course of breast radiotherapy and also account for residual intrafractional respiratory motion in VMAT.
Here, we conducted a comparative study for the optimal RT plan for SBBC including regional LN irradiation. Among the previous studies on SBBC, Seppälä et al. and Boman et al. reported RT for SBBC including RNI [
19,
20]. However, they only included axillary LNs and/or internal mammary LNs, whereas we included supraclavicular LNs; this resulted in a significant increase in lung dose. We implemented a hybrid plan using both VMAT and 3D-CRT instead of VMAT for this large target volume. Specifically, we used a 3D-tangential plan considering the widespread but low-dose area generated by previous VMAT plans on the left side. The dose coverage for the PTV was comparable to that of the VMAT-only plan, and the dose to OAR was significantly reduced. Especially for the lung, VMAT plans generally produced a larger volume in which a low dose of radiation is distributed, compared to that of a tangential field [
11]. Considering the mean lung dose (MLD) of 6–16 Gy in previous SBBC studies [
4,
9,
19,
21], our results were comparable to previous reports, despite the inclusion of the regional LN area such as the internal mammary chain and supraclavicular area.
In breast cancer, evidence is accumulating that RT can increase the risk of heart disease [
22], and an increased risk of stenosis in the LAD for left-sided RT compared to right-sided RT has been reported [
23]. Contemporary techniques usually deliver lower mean doses to the heart than they did in the past, but some parts of the heart may still receive high doses including the LAD, which is located near the left breast and may receive a high dose in RT for left breast cancer. Although Darby et al. suggested that increasing the mean heart dose could increase the incidence of ischemic heart disease [
22,
24], specific thresholds have been defined for neither the dose in the heart nor the dose in the LAD, according to several authors [
25]. Since no threshold doses for the heart and LAD are available and the clinical effect of low doses is not completely clear, we think the best clinical practice would be to keep the dose in the heart and LAD as low as achievable. In this study, the mean dose and V
xGy of the heart and LAD were also highest in the bilateral VMAT plan. Hybrid techniques using VMAT +3D-CRT can reduce the heart dose, and additional methods such as deep inspiration breath holding can reduce the radiation exposure to cardiac structures.
Although the hybrid plans showed shorter beam on time than that of the VMAT-only plan, hybrid plans would consume more treatment time due to the movement of the isocenter. Moreover, the setup during treatment could result in some errors involving factors such as quality of treatment and patients’ satisfaction. However, previous studies reported the feasibility of the two-isocenter technique in treating SBBC [
20]. The two-isocenter technique induced 2–5 mm of errors but no clinically significant change of dose coverage. The studies also emphasized that the two-isocenter VMAT technique could reduce the mean dose for the lung and heart better than the single isocenter technique.
There are several limitations in this study. First, we only considered the regional LN in the left side. For bilateral regional LN, the VMAT-plan may be the best, as the AP field for supraclavicular LN in 3D CRT generates a large amount of upper lung dose. To counter this issue further evaluation is needed based on the two-isocenter VMAT suggested by Boman et al. [
18]. Second, this study was conducted using an Asian population, and the criterion for classifying large breasts was extremely small when compared to those for Caucasian populations; therefore, further studies based on Caucasian populations are needed. In addition, only the beam delivery time was calculated, and it was the longest when the VMAT-only plan was performed. In actual patient treatment, however, two isocenters are used in the hybrid plan, which may lead to a longer patient set-up time. Therefore, further evaluation is needed to confirm whether hybrid plans are more efficient than the VMAT-only plan in clinical use. Moreover, since VMAT plans can be adjusted according to the physician’s intention, better plans could be generated by modifying the geometry of the beams.
Conclusions
The modified hybrid plan using VMAT + modified 3D-CRT is best when considering both PTV coverage and protection of OARs. To identify the clinical efficacy of the modified hybrid plan in terms of oncologic outcomes and treatment toxicities, advanced long-term follow-up studies with a large number of patients are needed. In addition, there is no standard guideline for the RT of SBBC including RNI, so it is also necessary to determine whether there are additional RT strategies beyond the method presented in this study.
To reach a compromise between dosimetric and therapeutic efficiency, and to improve the treatment of patients with SBBC, further studies of the optimal RT planning method, with a larger number of participants, should be performed.