Comparison of conventional and advanced radiotherapy techniques for left-sided breast cancer after breast conserving surgery

Abstract

Whole breast radiotherapy (WBRT) after breast conserving surgery is the standard treatment to prevent recurrence and metastasis of early stage breast cancer. This study aims to compare seven WBRT techniques including conventional tangential, field-in-field (FIF), hybrid intensity-modulated radiotherapy (IMRT), IMRT, standard volumetric modulated arc therapy (STD-VMAT), non-coplanar VMAT (NC-VMAT) and multiple arc VMAT (MA-VMAT). Fifteen patients who were previously diagnosed with left-sided early stage breast cancer and treated in our clinic were selected for this study. WBRT plans were created for these patients and were evaluated based on target coverage and normal tissue toxicities. All techniques produced clinically acceptable WBRT plans. STD-VMAT delivered the lowest mean dose (1.1 ± 0.3 Gy) and the lowest maximum dose (7.3 ± 4.9 Gy) to contralateral breast, and the second lowest LAR (4.1 ± 1.4%) of secondary contralateral breast cancer. MA-VMAT delivered the lowest mean dose to lungs (4.9 ± 0.9 Gy) and heart (5.5 ± 1.2 Gy), exhibited the lowest LAR (1.7 ± 0.3%) of secondary lung cancer, NTCP (1.2 ± 0.2%) of pneumonitis, RCE (10.3 ± 2.7%), and LAR (3.9 ± 1.3%) of secondary contralateral breast cancer. NC-VMAT plans provided the most conformal target coverage, the lowest maximum lung dose (46.2 ± 4.1 Gy) and heart dose (41.1 ± 5.4 Gy), and the second lowest LAR (1.8 ± 0.4%) of secondary lung cancer and RCE (10.5 ± 2.8%). MA-VMAT and NC-VMAT could be the preferred techniques for early stage breast cancer patients after breast conserving surgery.

Introduction

Breast cancer has the highest incidence rate among women in the US other than skin cancer (www.cancer.org). Women who were diagnosed with early-stage breast cancer and had lumpectomy usually underwent whole breast radiation therapy (WBRT) after surgery, which could lower recurrence and metastasis rates and make lumpectomy as effective as mastectomy.¹

The current standard of care (SOC) for WBRT in the US is using parallel-opposed tangential photon fields to treat the whole ipsilateral breast and chest wall, plus additional photon and electron fields to treat supraclavicular, axillary and internal mammary nodes when necessary.² However, significant dose inhomogeneity can occur within the irradiated volume and can cause poor cosmetic outcomes, especially for women with large breasts.^3–5 Field-in-field (FIF) technique is used sometimes to improve dose homogeneity throughout the target volume.^{3, 6} Intensity-modulated radiation therapy (IMRT) has been used for WBRT and can improve dose conformity and homogeneity, reduce high dose to heart and lung at the expense of increasing overall low doses,⁷ and has been shown to decrease acute skin toxicity.⁸ Hybrid IMRT (combination of open tangential and IMRT beams) has been shown to have a good balance of plan complexity and dose coverage/OAR sparing.^9–11 Volumetric modulated arc therapy (VMAT) can achieve similar target coverage as IMRT, spare more normal tissues and can significant reduce treatment time.¹² Multiple arc VMAT (MA-VMAT) showed good feasibility and OAR sparing for WBRT.¹³ Non-coplanar VMAT (NC-VMAT) has been shown to improve OAR dosimetry for post-mastectomy breast cancer¹⁴ and partial breast cancer,^15–18 but has not been evaluated for WBRT.

The purpose of this study was to compare target coverage and risks of developing of radiogenic side effects for a sample of WBRT patients using various modalities, including SOC, FIF, hybrid IMRT, IMRT, VMAT, MA-VMAT and NC-VMAT. There have been multiple treatment planning studies of WBRT,^19–28 but most of them did not include hybrid IMRT in the comparison although hybrid IMRT had been recommended as the optimal technique for WBRT¹⁰; with the advance of inverse planning techniques, the differences in treatment plan outcomes should be evaluated among different VMAT techniques while none of the previous studies did so; in addition, the radiobiological metrics like normal tissue complication probability (NTCP) of pneumonitis, lifetime attributable risk (LAR) of second cancers, and risk of coronary events (RCE) should be evaluated and compared among various WBRT modalities because it has been shown inclusion of non-dosimetric factors can provide a more robust comparison of different radiotherapy techniques,²⁹ while most of the previous studies only performed dosimetric comparisons.

Methods and Materials

Patient selection

Fifteen early stage left-sided breast cancer patients presenting for WBRT without nodal involvement after breast conserving surgery were included in this study. Computed tomography (CT) scans were obtained when patients were immobilized on a breast wing board with the left arm elevated above the head and free-breathing, and all CT data were anonymized³⁰ for this study. The target definitions were based on RTOG breast cancer Atlas and were approved by a radiation oncologist: clinical target volume (CTV) was defined as the ipsilateral breast with 5 mm skin extraction; planning target volume (PTV) was defined as CTV plus 7 mm expansion; PTV-Eval was based on PTV and defined to be limited anteriorly to exclude the part outside the patient and the first 5 mm of tissue under the skin, and posteriorly no deeper to the anterior surface of the ribs, and PTV-Eval was used in planning and for dose analysis. The contours of organs at risk (OARs) for each patient were approved by the same radiation oncologist and included lungs, whole heart and contralateral breast.

Treatment planning

The prescription dose for all patients was 50 Gy in 25 fractions. The following criteria were required for each treatment plan to be clinically acceptable: the volume of the PTV receiving at least 95% of the prescribed dose is greater than or equal to 95%; the volume of left lung receiving at least 20 Gy is less than 20%³¹; the volume of heart receiving at least 22.5 Gy is less than 20%.³² Maximum and mean doses for heart, lung and contralateral breast were constrained. All plans were generated in a commercial treatment planning system (TPS) (Pinnacle³ v9.8, Philips Medical Systems, Fitchburg, WI, USA).

SOC plans included two opposed tangential beams of 6, 10 or 15 MV energy depending on patient’s anatomy, and the tangential beam angles were determined by the fiducial markers placed on the skin and were usually around anterior midsternum and ipsilateral lower axilla. Collimator was rotated to shield the heart and lung, and dynamic wedges were used to minimize hotspots within PTV-Eval. FIF, hybrid IMRT and IMRT plans utilized the same beam energies as SOC plans. FIF plans used the same beam angles as SOC plans, and two to three subfields per beam were manually added using multi-leaf collimators (MLCs) to eliminate hotspots after the open field plan was created.⁶ Hybrid IMRT plans included a pair of open tangent fields and a pair of dynamic IMRT tangent fields, and 80% of prescription dose was delivered by open tangent beams and 20% of the prescription dose was delivered by IMRT beams.¹¹ IMRT plans were generated using the direct machine parameter optimization (DMPO) algorithm, and included seven coplanar beam equidistantly distributed in a sector of 180° that avoided direct exposure to the contralateral breast. All STD-VMAT, NC-VMAT and MA-VMAT plans used 6 MV beam energy and were generated using the SmartArc optimization algorithm. STD-VMAT plans utilized two coplanar partial arcs: the first arc was planned to be delivered counterclockwise (CCW) with starting and stopping gantry angles same as tangential fields, and the second arc was planned to be delivered clockwise (CW) over the same range of gantry angle. NC-VMAT plans utilized two partial arcs: the first arc was planned to be delivered CCW with starting and stopping gantry angles same as tangent fields and with 20° couch angle, and the second arc was planned to be delivered CW over the same range of gantry angle and with 340° couch angle. The collimator was rotated to align with the long axis of PTV in both arcs.³³ MA-VMAT plans consisted of six partial arcs (ARC01 to ARC06), each with 50°gantry rotations.¹³ ARC01 to ARC03 were delivered CW and ARC04 to ARC06 were delivered CCW. The starting angle of ARC01 and stopping angle of ARC03 were the same as SOC technique. The collimator was always rotated to align with the long axis of PTV in each arc.

Plan comparison metrics

Dosimetric parameters were evaluated for target, lungs, heart and contralateral breast. Dose homogeneity index (DHI)³⁴ and conformity index (CI)³⁵ were evaluated for PTV-Eval. Risks of radiogenic side effects were assessed: LAR was computed for secondary lung cancer and contralateral breast cancer using BEIR VII model³⁶ and organ equivalent dose (OED)³⁷. NTCP for pneumonitis was evaluated using Lyman-Kutcher-Burman (LKB) model^38–40. Dose-response model in Darby et al.⁴¹ was used to evaluate RCE for each patient and Reynolds risk model⁴² was used to calculate the baseline risk assuming medium risk type. More details of dose-risk models can be found in our previous study.¹⁴

The post hoc Tukey test was used to determine the statistical significance of the differences between two WBRT techniques. All statistical analyses were conducted with R software (version 3.2.3) and any difference was considered significant when p < 0.05.

Results

The axial dose distributions and DVHs for a typical WBRT patient are shown in Figs. 1 and 2. Table 1 lists PTV and OARs evaluation metrics for various WBRT techniques. The results of statistical tests are shown in Table 2 where the grey color indicates statistically significance (p values <0.05), e.g., SOC has significantly higher V₁₀₇% for PTV, D_mean, D_max, V₂₀ and NTCP for lung, V_22.5 and V₃₀ for heart, D_mean and V₅ for contralateral breast compared to IMRT, STD-VMAT, NC-VMAT and MA-VMAT plans; SOC has significantly higher D_max and V₁₀₇% for PTV compared to FIF and hybrid IMRT plans; STD-VMAT has significantly higher V₅ for lung and heart compared to other six WBRT techniques; no statistically significance shown in any comparison between FIF and hybrid IMRT.

Fig. 1.

Axial view of isodose distribution for SOC, FIF, Hybrid IMRT, IMRT, STD-VMAT, NC-VMAT and MA-VMAT plans for a typical WBRT patient. The red color wash represents the PTV-Eval.

Fig. 2.

DVHs for SOC, FIF, Hybrid IMRT, IMRT, STD-VMAT, NC-VMAT and MA-VMAT plans for a typical WBRT patient.

Table 1.

MU, PTV and OAR evaluation metrics (mean ± standard deviation). DHI values have more significant figures to show the difference among different techniques. SOC: standard of care; FIF: field in field; Hybrid: hybrid intensity-modulated radiation therapy; IMRT: intensity-modulated radiation therapy; STD-VMAT: standard volumetric modulated arc therapy; NC-VMAT: non-coplanar VMAT; MA-VMAT: multiple arc VMAT; MU: monitor unit; PTV: planning target volume; DHI: Dose homogeneity index; CL breast: contralateral breast. LAR: lifetime attributable risk; RCE: risk of coronary events; NTCP: normal tissue complication probability.

	SOC	FIF	Hybrid	IMRT	STD-VMAT	NC-VMAT	MA-VMAT
Average total MU	7343±962	5647±227	8673±1506	19300±2471	10263±1049	10397±1549	11523±1167
PTV
Dmean (Gy)	50.7±0.7	50.6±0.5	50.4±0.6	50.1±0.3	50.0 ±0.4	49.9±0.4	50.0±0.4
Dmax (Gy)	55.5±1.7	53.8±0.9	53.8±0.9	56.7±0.9	55.0±1.4	54.7±1.0	55.2±1.3
V107% (%)	0.1±0.1	0.0±0.1	0.0±0.1	0.0±0.0	0.0±0.0	0.0±0.0	0.0±0.0
CI	0.5±0.1	0.5±0.1	0.6±0.1	0.8±0.1	0.8±0.1	0.8±0.1	0.8±0.1
DHI	0.150±0.033	0.132±0.021	0.126±0.024	0.142±0.029	0.157±0.056	0.158±0.044	0.164±0.046
Total lung
Dmean (Gy)	6.7±1.0	6.4±0.8	6.1±1.0	5.9±0.9	6.0±0.9	5.4±1.1	4.9±0.9
Dmax (Gy)	53.3±1.7	51.4±1.4	50.3±1.5	48.3±1.5	47.9±2.3	46.2±4.1	50.8±2.8
V5	18.3±2.2	18.6±2.3	17.7±2.0	26.4±5.0	27.0±7.1	24.8±8.2	19.6±4.2
V10	14.8±2.0	15.0±2.2	14.4±2.0	15.1±3.1	14.3±2.7	12.8±3.6	12.5±3.3
V20	12.7±2.0	12.6±2.1	12.2±1.9	7.9±2.6	7.8±1.6	6.6±1.9	7.3±1.6
NTCP (%)	1.8±0.4	1.7±0.4	1.6±0.3	1.5±0.3	1.5±0.3	1.3±0.3	1.2±0.2
LAR (%)	2.2±0.4	2.2±0.4	2.0±0.3	2.0±0.3	2.0±0.4	1.8±0.4	1.7±0.3
Heart
Dmean (Gy)	9.6±3.7	8.1±3.7	8.1±2.8	7.4±1.3	7.8± 1.5	5.8±1.0	5.5±1.2
Dmax (Gy)	51.7±2.2	50.1±1.7	49.3±1.4	43.7±6.2	44.5±3.3	41.0±5.4	45.0±4.1
V5	25.3±10.1	25.7±9.4	23.9±8.7	48.3±14.0	53.2±8.5	30.5±9.0	22.1±9.0
V10	19.8±8.8	19.0±8.0	18.4±7.7	20.1±4.9	18.7±5.7	11.7±3.5	9.7±2.7
V22.5	16.6±8.0	14.9±6.9	14.7±6.6	4.7±3.2	6.2±2.7	4.3±1.9	5.0±1.8
V30	15.1±7.5	13.3±6.4	12.7±5.9	2.7±2.9	3.9±2.2	2.0±1.4	3.2±1.9
RCE (%)	12.4±3.5	11.5±3.1	11.6±3.1	11.4±3.3	11.6±3.2	10.5±2.8	10.3±2.7
CL breast
Dmean (Gy)	2.8±2.5	1.9±1.8	1.6±1.2	1.4±0.7	1.1±0.3	1.2±0.7	1.2±0.4
Dmax (Gy)	48.3±7.2	45.6±6.8	42.8±6.7	17.0±4.3	7.3±4.9	10.3±8.5	12.0±8.3
V5	4.2±2.6	3.8±3.0	4.0±3.0	0.2±0.3	0.3±0.7	0.2±0.5	0.1±0.1
LAR (%)	5.8±4.9	5.7±2.7	5.3±4.7	4.8±2.4	4.1±1.4	4.1±2.2	3.9± 1.3

Table 2.

Statistic comparison of seven WBRT techniques using post hoc Tukey test. Grey color indicates statistically significant (p values <0.05)

Variable	SOC vs FIF	SOC vs Hybrid	SOC vs IMRT	SOC vs STD-VMAT	SOC vs NC-VMAT	SOC vs MA-VMAT	FIF vs Hybrid	FIF vs IMRT	FIF vs STD-VMAT	FIF vs NC-VMAT	FIF vs MA-VMAT	Hybrid vs IMRT	Hybrid vs STD-VMAT	Hybrid vs NC-VMAT	Hybrid vs MA-VMAT	IMRT vs STD-VMAT	IMRT vs NC-VMAT	IMRT vs MA-VMAT	STD-VMAT vs NC-VMAT	STD-VMAT vs MA-VMAT	NC-VMAT vs MA-VMAT
PTV
Dmean
Dmax
V107% (%)
CI
DHI
Lung
Dmean
Dmax
V5
V10
V20
NTCP
LAR (%)
Heart
Dmean
Dmax
V5
V10
V22.5
V30
RCE (%)
CL breast
Dmean
Dmax
V5
LAR (%)

By combining the results from Table 1 and Table 2, we found that all seven WBRT techniques analyzed in this study meet clinical requirement of PTV coverage; SOC plans introduce significantly larger hot spots in PTV by showing the highest V_107%, and deliver relatively higher dose to OARs than inverse planning techniques (IMRT, STD-VMAT, NC-VMAT and MA-VMAT); FIF and hybrid IMRT plans exhibit better PTV coverage than SOC, deliver relatively lower dose to OARs than SOC, and both show relatively better dose homogeneity in PTV than the other five techniques; STD-VMAT plans provide the lowest D_mean and D_max for contralateral breast, and the second lowest LAR of secondary cancer in contralateral breast, but significantly increase the low dose cloud like V₅ for lung and heart; MA-VMAT plans show the lowest D_mean, V₁₀, NTCP and LAR for lung, the lowest V₅ and LAR for contralateral breast, and the lowest D_mean, V₅, V₁₀, and RCE for heart; NC-VMAT plans provide the most conformal target coverage, the lowest D_max, V₂₀ and the second lowest NTCP and LAR for lung, the lowest D_max, V_22.5, V₃₀ and the second lowest RCE for heart compared with other techniques.

Discussion

We evaluated seven WBRT techniques for treating left-sided lumpectomy breast cancer patients. All seven techniques provide clinical acceptable dose coverage to the target volume. For the two forward planning techniques, FIF plans not only show better PTV coverage but also reduce OAR dose than SOC. Five inverse planning techniques show lower OAR doses than the two forward planning techniques. STD-VMAT plans spare contralateral breast well at the cost of larger low dose cloud for lung and heart. MA-VMAT plans show the most optimal OARs doses and minimum risk of developing late side effects among all techniques. NC-VMAT plans provide the most conformal PTV coverage and excellent sparing of lung and heart.

There are plenty of WBRT planning studies in the literature. Considering it is difficult to compare studies with different target definitions, we compared our study with previous work that had the same PTV delineation based on RTOG as ours (Table 3): Descovich et al.⁴³ concluded that hybrid IMRT can reduce the hot spot within PTV compared to FIF for left-sided breast cancer patients and can provide better coverage, which is consistent with our results; Jin et al.¹⁹ reported that tangential IMRT has the best balance of target coverage and normal tissue sparing compared with conventional tangential beams, FIF, multi-beam IMRT and VMAT for small breast size, while our study shows IMRT provides inferior target coverage or OAR sparing than VMAT especially the advanced VMAT techniques, which is mainly because the mean PTV volume in our study (910.2 ± 439.8 cc) is much larger than theirs (360.8 ± 149.1 cc). These results show that tangential IMRT may not be the optimal technique for all WBRT patients, and its application should be assessed based on patient’s anatomy; Schubert et al.²² and Haciislamoglu et al.²⁵ both concluded that Tomotherapy (TOMO) may reduce high doses to heart and lung at the cost of increased low dose cloud which may lead to an increased probability of radiogenic side effects, and Han et al.²¹ concluded that TOMO is recommended for WBRT compared to SOC, FIF, IMRT and VMAT since it exhibits lowest total LAR for OARs. Our study didn’t evaluate TOMO, but shows NC-VMAT and MA-VMAT have comparable or better sparing of OARs than TOMO, e.g. NC-VMAT and MA-VMAT deliver mean lung dose of 5.4 ± 1.1 Gy and 4.9 ± 0.9 Gy, respectively, and mean contralateral breast dose of 1.2 ± 0.7 Gy and 1.2 ± 0.4 Gy, respectively, while Haciislamoglu et al.²⁵ reported mean lung dose of 9.6 ± 2.0 Gy and mean contralateral breast dose of 3.1 ± 0.4 Gy for TOMO. This suggests that NC-VMAT and MA-VMAT could be used as good alternatives when TOMO System is not available; Zhang et al.²⁷ evaluated different IMRT techniques and recommended FIF-DMPO-IMRT because it can reduce doses to lungs and heart and decrease treatment time. Their FIF-DMPO-IMRT consists of 70~80% FIF and 20%~30% IMRT, which is similar to our hybrid IMRT technique that has 80% open tangent beams and 20% IMRT, but they required 95% volume of PTV to receive 100% of prescription dose which makes their dosimetric results not comparable to ours; Viren et al.²⁶ concluded that both tangential VMAT (tVMAT) with two dual arcs of 50°−60° and continuous VMAT (cVMAT) with a dual arc of 240° have improved DHI within PTV and better sparing of heart and ipsilateral lung tissues compared to FIF and tangential IMRT, and cVMAT provided the best target coverage at the cost of significantly increased dose to contralateral breast. In our study, STD-VMAT (dual arc of approximately 180°) also has improved DHI within PTV and better sparing of heart and lung compare to FIF, and STD-VMAT has much better sparing of contralateral breast (mean dose 1.1 Gy) than their cVMAT (mean dose 2.6 Gy) and is similar to their tVMAT (mean dose 1.2 Gy), which suggests that smaller arcs of VMAT can lower the dose to contralateral breast.

Table 3.

Comparison with previous WBRT planning studies that have the same PTV definition as ours.

Reference	Num. of patients	Breast site	Techniques compared	Key findings
Descovich et al. (2010)	15	Left	FIF, hybrid IMRT	Hybrid IMRT is preferred since it can reduce hot spot, provide better coverage and require less planning time.
Schubert et al. (2011)	10	Left	SOC, FIF, tangential IMRT (2 Fields), TOMO, topotherapy	TOMO, topotherapy and tangential IMRT can reduce high dose to target and normal tissue; TOMO results in increased low doses to normal tissue.
Jin et al. (2013)	20	Left	SOC, FIF, tangential IMRT (2 Fields), IMRT (7 Fields), VMAT (starting and ending angles were same as tangential beam angles)	Tangential IMRT is recommended since it has improved DHI and reduced dose to heart and lung.
Viren et al. (2015)	10	Left	FIF, tangential IMRT (2 Fields), tangential VMAT (two dual arcs of 50°–60°), continuous VMAT (dual arc of 240°)	Both VMAT techniques show improved DHI and better sparing of heart and ipsilateral lung. Continuous VMAT provides best dose coverage at the cost of significantly increased dose to contralateral breast.
Haciislamoglu et al. (2015)	15	Left	SOC, FIF, 9-field IMRT, TOMO, VMAT (Starting and ending angles of the arcs were 10° posterior to tangential fields)	TOMO shown reduced high and mean doses to heart and lung at the cost of increased low dose cloud.
Han et al. (2016)	10	Left and right	SOC, FIF, IMRT (10 to 12 fields), VMAT (3–4 partial arcs spanned from 305° to 152° for the left, 60° to 214° for the right), TOMO	TOMO is recommended since it provides the lowest LAR for all surrounding OARs.
Zhang et al. (2018)	50	Left	5-field IMRT, 6-field IMRT, FIF-DMPO-IMRT	FIF-DMPO-IMRT is recommended due to reduced heart and lung doses and treatment time.
Our study	15	Left	SOC, FIF, hybrid, IMRT, VMAT (starting and ending angle were same as tangential beam angles), MA-VMAT, NC-VMAT	MA-VMAT and NC-VMAT are recommended due to reduced doses and risks for heart, lung and contralateral breast.

Jeulink et al.¹⁰ illustrated that hybrid IMRT is most optimal WBRT technique compared with full IMRT, STD-VMAT and MA-VMAT for the best reduction of mean and low OARs doses, while it is not the most optimal choice in our study. This is possibly because only two tangential IMRT fields are used for hybrid IMRT in our study whereas four tangential IMRT fields were used in theirs. Moreover, left anterior descending coronary artery (LAD) was contoured as an OAR for plan optimization in their study, which may further limit dose to the heart but may significantly increase workload for physicians and dosimetrists. Additionally, the PTV delineation is different since the boost PTV was included in their analysis. These results suggest that hybrid IMRT may not be the best choice for all WBRT patients and its application should be determined based on all clinical factors.

Among all WBRT techniques in our study, MA-VMAT and NC-VMAT have shown superior OARs sparing. Tsai et al.¹³ reported mean heart dose of 7.6 ± 1.4 Gy and lung dose of 5.6 ± 0.4 Gy for MA-VMAT, which were slightly higher than our mean heart dose (5.5 ± 1.2 Gy) and lung dose (4.9 ± 0.9 Gy). This can be explained by the fact that a slightly higher prescription dose (50.4 Gy delivered in 28 fractions) was used in their study, and different dose constrains were used, e.g. our mean heart dose limit is 7 Gy while their mean heart dose limit was 9 Gy. When multiple arcs were used and collimator angle was adjusted for each arc, treatment plans could be further optimized since more degrees of freedom were provided for MA-VMAT. Smyth et al.⁴⁴ summarized the recent advancement in non-coplanar radiotherapy and listed different techniques including static couch NC-VMAT, coronal VMAT, trajectory VMAT and dynamic wave arc etc. Our study utilizes static couch NC-VMAT, according to definitions in Smyth et al.⁴⁴, for WBRT for the first time and shows it can provide excellent sparing of lung and heart compared to other WBRT techniques, which demonstrates that OARs can be spared more by adjusting the couch angle to minimize direct irradiation.

There is a lack of clinical outcome data of radiogenic late effects for advanced WBRT techniques, but our calculated RCE and LARs for SOC WBRT show good agreement with clinical data for breast cancer patients who went through SOC WBRT: Taylor et al.⁴⁵ reported the annual risk of developing radiogenic lung cancer and contralateral breast cancer was 0.2% and 0.36%, respectively, and our calculated annual risk is 0.22% and 0.38% for SOC; Hooning et al.⁴⁶ reported the annual cardiac toxicity was 1.19% whereas our estimation is 1.23% for SOC. Based on these good agreements, we expect our estimated radiogenic risks values for advanced WBRT techniques to be reasonable. Prospective clinical studies can validate our calculations and further illustrate the benefit of advanced WBRT techniques.

Cardiac toxicity is a major concern for breast cancer patients who received radiotherapy, especially for left-sided breast cancer patients.⁴¹ Our study shows comparable mean heart dose for FIF, IMRT and VMAT as those reported by Viren et al.²⁶. However, our study shows higher mean heart dose for conventional SOC, FIF, IMRT and VMAT than those in Jin et al.¹⁹, which is possibly because their extra dose constrains on coronary artery has further limited heart dose, and the larger breast size in our study could inevitably lead to larger fields that will induce higher heart dose in order to provide enough PTV coverage. Furthermore, all patients were treated in supine position in our study, while literature^{47, 48} has shown that breast irradiation in prone position may result in lower risk of cardiac toxicity and improve dose homogeneity within PTV compared to standard irradiation in supine position. Further studies are needed to evaluate advanced WBRT techniques in prone treatment position.

In order to further reduce heart irradiation, deep inspiration breath hold (DIBH) has been implemented for WBRT and studies have shown that DIBH can minimize irradiation of heart without compromise target coverage for most left-sided breast cancer patient^49–55. However, not all patients could benefit from it, e.g. Dell’Oro et al.⁵⁶ recently reported that DIBH may not be recommended for some patients due to little dosimetric benefit. Several studies^{54, 56, 57} reported criteria of selecting breast cancer patients for DIBH, including patient’s age, ability to hold breath for a specific duration of time, total lung volume, in-field heart volume, sternal excursion etc. In our study, the fifteen patients were not selected for DIBH in the clinic mainly due to the limited dosimetric benefit for them. The benefit of various WBRT techniques for DIBH patients will be investigated by our group in the near future.

One limitation of our study is that we did not compare the WBRT techniques in terms of planning or treatment time, quality assurance (QA) workload, and patient’s comfort. Among the seven WBRT techniques analyzed in this study, the delivery efficiency may be a concern for non-coplanar VMAT plans,^{44, 58} and the rotation of gantry and couch between beams can take considerable amount of time. The delivery time for non-coplanar plans can be shorten by implementing the automated machine transitions between beams, e.g., Liang et al.¹⁸ illustrated that coronal VMAT which had dynamic patient couch rotation can be delivered in 4.5 minutes for a 3.85 Gy fraction for accelerated partial breast irradiation. Moreover, IMRT, STD-VMAT, NC-VMAT and MA-VMAT require more MUs than SOC, FIF and hybrid IMRT, as shown in Table 1. The increased number of MUs will inevitably result in longer treatment time and increased leakage radiation which may increase patient’s discomfort and risk of developing radiogenic side effects. The planning time for the seven WBRT techniques was comparable. Contouring for one patient took 60 minutes on average, and treatment planning for a patient using one technique took another 45 minutes on average, which is comparable to the literature,⁷ except for MA-VMAT which needed slightly longer planning time (about 60 minutes) due to increased small fields that required longer optimization time to achieve the objectives. Hybrid IMRT, IMRT, STD-VMAT, NC-VMAT and MA-VMAT all require QA of dosimetric output for every treatment plan,⁵⁹ and NC-VMAT requires extra specific QA procedures on couth rotation which may further increase the workload of patient’s specific QA.⁶⁰

Conclusions

Seven WBRT techniques were evaluated in this study. Among them, NC-VMAT was evaluated for WBRT for the first time and showed excellent sparing of lung and heart, STD-VMAT could reduce dose to the contralateral breast, and MA-VMAT showed the best OAR sparing. MA-VMAT and NC-VMAT might be the appropriate WBRT techniques for early stage breast cancer patients after breast conserving surgery.