Dose to the contralateral breast due to primary breast irradiation

The radiation dose received by the contralateral breast during primary breast irradiation is of concern because breast tissue is subject to cancer induction from low to moderate doses of radiation. In this paper the dose to the opposite breast has been studied in detail for common breast treatment techniques. Measurements have been made on 16 patients, a water phantom, a polystyrene phantom with cork inserts to simulate lung tissue, and a body-shaped phantom with wax breasts. Thermoluminescent dosimeters (TLD), ion chambers, diodes, and film have been used in the various configurations. The patient measurements have shown that there is a wide variation in the opposite breast dose received by patients, even when all are treated with, for example, tangential fields alone. Addition of more radiation fields, such as supraclavicular/axillary and internal mammary fields, may increase the dose to the opposite breast for a particular patient. Variations in the details of the technique such as what wedges are used, the use of blocks, and the orientation of the field edges are all important to the final dose received by the patient's contralateral breast. With the phantom measurements, it has been possible to determine the contributions to the opposite breast dose from each of the relevant factors. This makes it possible to explain the wide variation in patient dose measurements, and to make some relatively simple recommendations that will allow the reduction of the dose to the opposite breast from several hundred cGy to about 50 cGy for a typical treatment course dose of 5000 cGy.     

Radiation dose to conceptus resulting from tangential breast irradiation

PURPOSE: To estimate the radiation dose to the conceptus resulting from tangential breast irradiation. METHODS AND MATERIALS: Conceptus radiation doses were measured in anthropomorphic phantoms simulating the geometry of a pregnant woman at the first, second, and third trimesters of gestation. Medial and lateral field irradiations were generated using a 6-MV X-ray beam. Dose measurements were performed with thermoluminescent dosimeters. RESULTS: For a treatment course delivering 50 Gy to the tumor, conceptus dose at the first trimester of gestation was found to be 2.1-7.6 cGy, depending on the field size used and the distance between conceptus and primary irradiation field. The corresponding dose ranges to the conceptus during the second and third trimesters of gestation were 2.2-24.6 cGy and 2.2-58.6 cGy, respectively. Dose data and formulas are presented to estimate conceptus dose for individual patients undergoing breast radiotherapy during the entire pregnancy. CONCLUSIONS: This study may be of value in the management of pregnant women needing tangential breast irradiation, because it provides the required information to estimate conceptus dose.     

乳腺癌调强放射治疗和常规切线野治疗的三维剂量学研究

背景与目的:乳房保留治疗已在早期乳腺癌患者中逐渐推广应用,其中全乳根治性放疗的标准技术通常采用常规切线野技术。调强放射治疗(intensity-modulatedradiotherapy,IMRT)技术有望在保障相同疗效的同时进一步减少放疗并发症,提高生活质量。本研究利用三维计划系统评价全乳IMRT的剂量学优势与适应证。方法:选择10例接受保乳手术的Tis～2N0M0早期乳腺癌病例,利用三维治疗计划系统为每例患者设计两种全乳放射治疗计划,切线野常规计划与IMRT计划,处方剂量均为5000cGy。用剂量体积直方图(dosevolumehistograms,DVH)来比较各种计划中计划靶体积(planningtargetvolume,PTV)、危及器官(organsatrisks,OARs)的剂量学差异。结果:靶区覆盖率在两种计划中相似,分别为98.3%和97.7%。与常规计划比较,IMRT计划的PTV接受<95%处方剂量与>103%处方剂量的体积百分比之和(inhomogeneityindex,IHI)从29.9%减少到2.9%,PTV接受至少105%处方剂量照射的体积百分比(V105%)从28.2%减少到0.6%;IMRT计划改善IHI和减少V105%的平均值在PTV较大的患者中优势更明显。左侧患者中冠状动脉的最大剂量(Dmax)以及心脏的平均剂量(Dmean)分别从5057.1cGy减少到4832.9cGy和从629.8cGy到450.7cGy;右侧患者肝脏的Dmean从283.9cGy减少到172.0cGy;所有患者中同侧肺的Dmean、至少接受20Gy照射的体积百分比(V20)分别从925.2cGy减少到765.9cGy,从16.0%到15.3%,Dmean与V20的平均值在IMRT计划中减少的百分比在不同射野中心肺厚度(centrallungdistance,CLD)亚组中分别是14.7%与20.9%,7.0%与12.9%;对侧乳腺和对侧肺的Dmean也分别从75.4cGy减少到20.3cGy和从30.9cGy到16.1cGy。结论:全乳IMRT的剂量学优势主要在于保证靶区覆盖率的前提下,显著改善靶区的剂量分布均匀性并一定程度上降低OARs的受照剂量与容积。乳房体积和CLD较大的病例可以通过IMRT技术获得更好的剂量学结果。     

Dosimetry of the breast for determining carcinogenic risk in mantle irradiation

Development of secondary malignancies following treatment of Hodgkin's disease with radiation (central axis midline dose of 3600-4500 cGy) is a recognized risk, and the incidence in breast cancer has been reported to increase by a factor of 4.3 (95% confidence level 2.0 to 8.4) for patients treated with mantle irradiation. Increased incidence of breast cancer has also been shown in atomic bomb survivors, women who underwent multiple fluoroscopic examinations, and women treated for postpartum mastitis. The dose response, however, for radiation dose above 1000 cGy is virtually unknown. Quantitative analysis of carcinogenesis after radiation is exacerbated by the large dose gradient across the breast (300-4200 cGy for midline doses of 4000 cGy), large individual variation in breast size and treatment field position. We have developed differential dose volume histograms calculated using a 3-dimensional (Eg TAR) algorithm as a potential tool for retrospective and prospective epidemiological evaluation. The breast volume forms a bimodal distribution with respect to dose and with further analysis other quantities, such as mean dose and integral dose, can be calculated from the histograms. Using the mean dose, the linear model for carcinogenesis predicts an increased incidence of secondary breast cancer by a factor of 11.6 and 9.6 for the left and right breast, respectively. The dose calculations has been corrected for inhomonogenities 3-dimensionally and test of the accuracy has been included.     

Multileaf field-in-field forward-planned intensity-modulated dose compensation for whole-breast irradiation is associated with reduced contralateral breast dose: a phantom model comparison.

PURPOSE: Static multileaf collimated field-in-field forward-planned intensity-modulated radiation treatment (FiF-IMRT) has been shown to improve dose homogeneity compared to conventional wedged fields. However, a direct comparison of the scattered dose to the contralateral breast resulting from wedged and FiF-IMRT plans remains to be documented. METHODS: The contralateral scattered breast dose was measured in a custom-designed anthropomorphic breast phantom in which 108 thermoluminescent dosimeters (TLDs) were volumetrically placed every 1-2cm. The target phantom breast was treated to a dose of 50Gy using three dose compensation techniques: No medial wedge and a 30-degree lateral wedge (M0-L30), 15-degree lateral and medial wedges (M15-L15), and FiF-IMRT. TLD measurements were compared using analysis of variance. RESULTS: For FiF-IMRT, the mean doses to the medial and lateral quadrants of the contralateral breast were 112cGy (range 65-226cGy) and 40cGy (range 18-91 cGy), respectively. The contralateral breast doses with FiF-IMRT were on average 65% and 82% of the doses obtained with the M15-L15 and M0-L30 techniques, respectively (p<0.001). Compared to the M15-L15 technique, the maximum dose reduction obtained with FiF-IMRT was 115cGy (range 13-115cGy). CONCLUSIONS: The dose to the contralateral breast is significantly reduced with FiF-IMRT compared to wedge-compensated techniques. Although long-term follow-up is needed to establish the clinical relevance of this finding, these results, along with the previously reported improvement in ipsilateral dose homogeneity, support the use of FiF-IMRT if resources permit.     

Reducing scatter radiation to the contralateral breast with a mobile, conformal shield during breast cancer radiotherapy.

During a standard course of breast radiotherapy, the contralateral breast generally receives approximately 2.5 to 6.0 Gy of scattered radiation. Although most studies have not found an overall increase in metachronous contralateral breast cancers in patients undergoing radiotherapy, a cohort of younger women may be genetically more susceptible to radiation-induced breast cancers and may thus be adversely affected by the scattered radiation. We are attempting to develop a simple, convenient, effective mechanism for minimizing the scattered radiation to the contralateral breast during the process of clinical breast radiotherapy. We therefore designed a conformal, platform-based breast shield consisting of 2.5 cm of molded lead in a mobile counterweighted polystyrene casing. This shield was intended to serve as a physical barrier to prevent both low and high energy scattered photons from the medial and lateral tangential fields. We conducted a prospective trial of 20 women, each woman serving as her own control. Each woman received breast radiotherapy with and without shield, and an array of thermoluminescent dosimeters was positioned across the contralateral breast to evaluate the in vivo dosimetry and the impact of the breast shield on surface absorption of scattered radiation. We found that the use of the breast shield reduced the median dose of scattered radiation by approximately 60% (p < 0.0001). This represented a median dose reduction of approximately 300 cGy at the nipple. The shield was easily positioned and added < 1 min to daily setup time. We conclude that the use of this sort of surface barrier shielding technique was feasible, effective, and practical for clinical use. The degree of scatter reduction accomplished through the use of this breast shield may be biologically significant, especially for those patients with biologic or epidemiologic risk factors that may predispose them to the development of radiogenic breast cancers.     

Intensity modulated radiation therapy (IMRT) reduces the dose to the contralateral breast when compared to conventional tangential fields for primary breast irradiation

Summary Purpose To determine the dose received by the contralateral breast during primary breast irradiation using IMRT compared to conventional tangential field techniques. Methods and materials Between March 2003 and March 2004, 83 patients with breast carcinoma were treated using 6, 10, or mixed 6/18 MV photons (65 with tangential IMRT technique and 18 with 3-dimensional technique using tangential fields with wedges) for primary breast irradiation following breast-conserving surgery. Paired thermoluminescent dosimeters (TLDs) were placed on each patient’s contralateral breast, 4 and 8 cm from the center of the medial border of the tangential field. The TLDs were left on the patient during a single fraction and then measured 24 h afterwards. Results The mean dose delivered with photons to the primary breast for all patients was 4999 cGy (SD=52) with a mean single fraction dose of 199 cGy (SD=8). The mean percent of the prescribed dose to the contralateral breast measured at the 4- and 8-cm positions were 7.19% (SD=2.28) and 4.63% (SD=2.12), respectively, for patients treated with IMRT compared to 11.22% (SD=2.73) and 10.70% (SD=3.44), respectively, for the patients treated with conventional tangential field techniques. This represented a 36% and 57% reduction at the 4 and 8-cm contralateral positions, respectively, in the mean dose to the contralateral breast using IMRT compared to 3-D technique which was statistically significant (p<0.0005, <0.0005, respectively). Conclusion Primary breast irradiation with tangential IMRT technique significantly reduces the dose to the contralateral breast compared to conventional tangential field techniques. *Presented in part of the 46th Annual Meeting of the American Society for Therapeutic Radiology and Oncology, Atlanta, GA, October 2004.     

Evaluation of standard tangential irradiation versus breast ring technique using three-dimensional tools in cancerous large breasts

AIMS AND BACKGROUND: In patients with large breasts, tangential irradiation after lumpectomy is technically difficult. Tangential plans of large breasts comprise a large volume of the lung, axilla and heart, as well as an important volume of the opposite breast and lung. METHODS: Instead of classical positioning of the breast, we encircled the breast with a plastic polyvinyl ring-shaped tube. All five left-sided breast cancer patients underwent CT scan planning with and without the plastic ring in conventional treatment positioning. Skin treatment portals at the midline and mid axilla were marked with radio-opaque markers. The scan volume encompassed the region from just below the diaphragm to the lung apex. Patients' plans were evaluated with regard to dose-volume coverage of the breast planning target volume, ipsilateral lung, contralateral lung, heart and contralateral breast. End points were maximum dose, minimum dose, and mean dose delivered to the target or critical structure. RESULTS: As expected, the average volume of the left lung treated above a tolerance dose of 2000 cGy was less than 10% in the ringed breast, whereas in the other method without the ring the corresponding lung volume was 30%. The cardiac volume receiving a dose over 4000 cGy was 2% in the ringed breast and 18% in the breast without a ring. CONCLUSIONS: With the use of a simple technique with a fairly low expenditure, acute skin reactions and late cardiac and lung morbidity can be minimized in patients with large breasts.     

The effect of a dynamic wedge in the medial tangential field upon the contralateral breast dose

The elevated incidence of breast cancer following irradiation of breast tissue has led to concern over the magnitude of the scattered radiation received by the uninvolved contralateral breast during radiation therapy for a primary breast lesion and the risk of an induced contralateral breast cancer. Some linear accelerators use a single dynamic (or universal) wedge that is mounted within the treatment head at an extended distance from the patient. Because of the combined effects of distance and shielding, the contralateral breast dose due to a medial tangent containing a dynamic wedge is expected to be less than that containing a conventional wedge. This paper presents contralateral breast dose (CBD) measurements performed on an anthropomorphic phantom with breast prostheses irradiated with 6 MV X rays from a linear accelerator equipped with a dynamic wedge. Doses were measured at 15 points within the contralateral breast prosthesis with thermoluminescent dosimeters. It was found that the contralateral breast dose per unit target breast dose decreases with the perpendicular distance from the posterior edge of the medial tangent to the dose measurement point and increases with effective wedge angle by factors ranging up to 2.8, in agreement with data presented earlier for a water phantom geometry. This dose elevation showed no statistically significant dependence (p less than 0.05) upon the perpendicular distance from the beam edge. Comparisons with data in the literature show that the contralateral breast dose increase by a dynamic wedge is typically only about half of that reported for a conventional wedge for the same wedge angle and distance from the beam.     

Radiation dose and contralateral breast cancer risk associated with megavoltage cone-beam computed tomographic image verification in breast radiation therapy

PurposeTo measure and compare organ doses from a standard tangential breast radiation therapy treatment (50 Gy delivered in 25 fractions) and a megavoltage cone-beam computed tomography (MV-CBCT), taken for weekly image verification, and assess the risk of radiation-induced contralateral breast cancer.Methods and MaterialsOrgan doses were measured with thermoluminescent dosimeters placed strategically within a female anthropomorphic phantom. The risk of radiation-induced secondary cancer of the contralateral breast was estimated from these values using excess absolute risk and excess relative risk models.ResultsThe effective dose from a MV-CBCT (8-monitor units) was 35.9 ± 0.2 mSv. Weekly MV-CBCT imaging verification contributes 0.5% and 17% to the total ipsilateral and contralateral breast dose, respectively. For a woman irradiated at age 50 years, the 10-year postirradiation excess relative risk was estimated to be 0.8 and 0.9 for treatment alone and treatment plus weekly MV-CBCT imaging, respectively. The 10-year postirradiation excess absolute risk was estimated to be 4.7 and 5.6 per 10,000 women-years.ConclusionsThe increased dose and consequent radiation-induced second cancer risk as calculated by this study introduced by the imaging verification protocols utilizing MV-CBCT in breast radiation therapy must be weighed against the benefits of more accurate treatment. As additional image verification becomes more common, it is important that data be collected in regard to long-term malignancy risk.     

VMAT和IMRT在早期右侧乳腺癌保乳术后放射治疗中的比较剂量学研究

目的:比较三种放射治疗计划对右侧乳腺癌保乳术后放射治疗的剂量学差异。方法:选择11名早期右侧乳腺癌保乳术后的患者,采用处方剂量5000 cGy,分别设计2FIMRT、4FIMRT和2arcVMAT三种计划。使用剂量体积直方图(dose volume histogram,DVH)来分析PTV和OARs的每个评估指标。结果:VMAT和IMRT为所有患者实现了整个乳房的良好的剂量均匀性。VMAT计划比两个IMRT计划有更好的HI,并减少了V107。在乳腺体积>503 mm^3的患者中,2arcVMAT具有更好的HI。2FIMRT在OARs,包括心脏,左、右肺,对侧乳腺,食管,脊髓和肝脏,对比4FIMRT和2arcVMAT,受照射剂量更低,差异具有统计学意义。结论:这三种计划可以满足右侧乳腺癌术后放疗的临床剂量要求,综合靶区剂量分布及OARs进行评分,整体考虑使用2FIMRT可能比其他两种计划具有临床优势。     

A modified three-field technique for breast treatment

Radiation therapy of breast cancer commonly uses three high-energy photon fields. The supraclavicular field and the two tangential fields are often matched on the skin. An analysis of the dose distribution in the matchline region shows a dose of up to 200% of the prescribed dose over the width of 8 mm at 3 cm depth. This is a result of the beam divergence and the complex geometry. A modification of the treatment technique has reduced the dose to less than 140% of the prescribed dose over approximately 4 mm. The modified technique uses a shielding block which covers the caudad half of the supraclavicular field and makes the caudad edge vertical. The cephalad edges of the two tangential fields are defined by a shielding block, which hangs vertically from the treatment head regardless of gantry and collimator position. This block obstructs a small portion of the radiation field and defines a vertical field edge. These three vertical field edges are brought into coincidence by adjustment of couch turntable angle and longitudinal couch position. The nearly perfect geometric match in three dimensions achieved in this manner further improves the cosmetic results after radiation treatment of breast cancer.     

A comparison of different intensity modulation treatment techniques for tangential breast irradiation

Purpose: Several intensity modulation (IM) treatment techniques for tangential breast irradiation were evaluated in terms of dose uniformity in the treated breast volume, contralateral breast dose, and treatment irradiation time.

Methods and Materials: Contralateral breast dose was measured via TLD chips, and the dose uniformity was calculated on two anthropomorphic phantoms. IM was applied to all beams or to the lateral-medial (LM) beam only. The techniques evaluated include (a) IM via “step & shoot” multileaf collimator (MLC), (b) IM via intensity modulator (compensator), (c) virtual wedge, and (d) physical wedge. A dose optimization algorithm was used for the first two techniques.

Results: Collimator-generated IM techniques (MLC-IM and the virtual wedge) produced 50% (average) less contralateral breast dose than the conventional two-wedge technique. When the compensator or the physical wedge was used, contralateral breast dose was reduced 30% (average) by leaving the ML beam open.

Conclusion: The treatments generated by dose optimization algorithm and delivered via the compensator and MLC techniques offered superior dose uniformity. Single-beam IM techniques in general use less irradiation time without significant degradation of dose uniformity. The MLC-IM technique in this study required the longest treatment irradiation time, while the virtual wedge and compensator IM techniques required the least.     

Regional nodal irradiation in the conservative treatment of breast cancer

At this institution conservative treatment of breast cancer was begun in the 1960's. The following analysis represents our experience through 1984 with specific reference to the management of the regional lymph nodes. A total of 432 patients with clinical stage I and II breast cancer were treated between 1962 and 1984 with lumpectomy and radiation therapy. The breast was treated with tangential fields to a median dose of 4800 cGy and electron conedown to a total tumor bed dose of 6400 cGy. Axillary dissection was not routinely performed, particularly in the earlier years. More recently, axillary dissection has been used with increasing frequency if it was felt that the results of the dissection would influence systemic treatment. One hundred eighty-seven patients (43%) underwent axillary dissection (30% pathologically positive) and routinely received regional nodal irradiation (median dose 4600 cGy) to the internal mammary and supraclavicular lymph nodes. Two hundred forty-five patients (57%) did not undergo axillary dissection and routinely received regional nodal irradiation to the internal mammary, supraclavicular, and entire axillary regions to a total median dose of 4600 cGy. As of May 1989 with a median follow-up of 7.5 years, there have been a total of 12 nodal failures for an actuarial nodal control rate of 97% at 5 years and 96% at 10 years. The actuarial 5-year regional nodal control rate was the same for both the group of patients receiving regional RT alone without axillary dissection and the group of patients receiving axillary dissection and supraclavicular/internal mammary radiation. There has been minimal morbidity associated with this treatment policy. We conclude that regional nodal irradiation as described above, with or without axillary dissection, results in a high rate of regional nodal control and minimal treatment morbidity in patients undergoing conservative treatment of early stage breast cancer.     

Intensity-modulated tangential beam irradiation of the intact breast

Purpose: To evaluate the potential benefits of intensity modulated tangential beams in the irradiation of the intact breast.

Methods and Materials: Three-dimensional treatment planning was performed on five left and five right breasts using standard wedged and intensity modulated (IM) tangential beams. Optimal beam parameters were chosen using beams-eye-view display. For the standard plans, the optimal wedge angles were chosen based on dose distributions in the central plane calculated without inhomogeneity corrections, according to our standard protocol. Intensity-modulated plans were generated using an inverse planning algorithm and a standard set of target and critical structure optimization criteria. Plans were compared using multiple dose distributions and dose volume histograms for the planning target volume (PTV), ipsilateral lung, coronary arteries, and contralateral breast.

Results: Significant improvements in the doses to critical structures were achieved using intensity modulation. Compared with a standard-wedged plan prescribed to 46 Gy, the dose from the IM plan encompassing 20% of the coronary artery region decreased by 25% (from 36 to 27 Gy) for patients treated to the left breast; the mean dose to the contralateral breast decreased by 42% (from 1.2 to 0.7 Gy); the ipsilateral lung volume receiving more than 46 Gy decreased by 30% (from 10% to 7%); the volume of surrounding soft tissue receiving more than 46 Gy decreased by 31% (from 48% to 33%). Dose homogeneity within the target volume improved greatest in the superior and inferior regions of the breast (approximately 8%), although some decrease in the medial and lateral high-dose regions (approximately 4%) was also observed.

Conclusion: Intensity modulation with a standard tangential beam arrangement significantly reduces the dose to the coronary arteries, ipsilateral lung, contralateral breast, and surrounding soft tissues. Improvements in dose homogeneity throughout the target volume can also be achieved, particularly in the superior and inferior regions of the breast. It remains to be seen whether the dosimetric improvements achievable with IMRT will lead to significant clinical outcome improvements.     

Tangential breast irradiation: simple improvements.

A series of studies using phantom and in vivo dosimetry measurements with diodes and thermoluminescent dosimetry were undertaken to establish whether simple methods are available to obtain improvements in homogeneity of dose in the treated breast, and reductions of dose to tissues outside it, when using tangential 6MV photon fields. These studies confirmed that the use of an appropriate lung density correction factor in the planning process is likely to cause a reduction in the amount of wedge compensation predicted to be necessary to produce reasonable dose homogeneity in the central axis of the beam. This was shown to be of value in eliminating potential areas of overdosage in the chest wall medial and lateral to the breast mound, and also in reducing unwanted doses to the contralateral breast and lung. Further reductions in dose to contralateral breast were confirmed to occur when the majority, or all, of the wedge compensation predicted necessary is placed on the lateral field and when the lateral tangential field is angled anteriorly in such a way as to align the posterior beam edges. Skin does above, below, and over the breast mound were observed to depend critically upon angle of incidence of the beam, with its consequent effect upon electron build-up, and the position of the breast contour in the beam. Skin doses above and below the breast mound were shown to exceed over the mound itself. This differential effect was observed to increase with increasing wedge compensation. In some situations skin doses below the breast mound nearly reached prescribed dose within the breast. The design and use of simple lead attenuators to reduce these areas of high dosage is discussed.     

Radiation dose and second breast cancer

Amongst 14,000 women with breast cancer treated between 1946 and 1982, 194 developed a second primary tumour in the contralateral breast more than one year after diagnosis of the first primary. The radiation dose to the contralateral breast was calculated for each member of this group and also for members of a control group matched for age, year of diagnosis and survival time. Comparison of the groups provides no evidence for radiation induced carcinogenesis on the contralateral breast in these patients.     

Initial clinical experience with moderate deep-inspiration breath hold using an active breathing control device in the treatment of patients with left-sided breast cancer using external beam radiation therapy

INTRODUCTION: We present our initial clinical experience using moderate deep-inspiration breath hold (mDIBH) with an active breathing control (ABC) device to reduce heart dose in the treatment of patients with early-stage, left-sided breast cancer using external beam radiation therapy (EBRT) limited to the whole breast. METHODS AND MATERIALS: Between February and August 2002, 5 patients with Stages I/II left-sided breast cancer received EBRT limited to the whole breast using an ABC device. After standard virtual simulation, patients with >2% of the heart receiving >30 Gy in free breathing were selected. All patients underwent a training session with the ABC apparatus to determine their ability to comfortably maintain mDIBH at 75% of the maximum inspiration capacity. Three patients received 45 Gy to the whole breast in 25 fractions, and 2 patients received 50.4 Gy in 28 fractions. For each of the medial and lateral tangential beams, radiation was delivered during 2 or 3 breath hold durations that ranged from 18 to 26 s. "Step-and-shoot" intensity modulation was employed to achieve uniform dose distribution. Open beam segments were purposely delivered over 2 breath hold sessions and captured on electronic portal images to allow intra- and interfraction setup error analysis. All electronic portal images of the tangential beams were analyzed off-line using an in-house treatment verification tool to assess the anteroposterior, craniocaudal, and rotational uncertainties. Corrections were applied if necessary. RESULTS: A comparison of treatment plans performed on breath-hold and free-breathing CTs showed that ABC treatments achieved a mean absolute reduction of 3.6% in heart volume receiving 30 Gy (heart V(30)) and 1.5% in the heart normal tissue complication probability. A total of 134 ABC treatment sessions were performed in the 5 patients. The average number of breath holds required per beam direction was 2.5 (4-6 per treatment) with a median duration of 22 s per breath hold (range: 10-26 s). Patients tolerated mDIBH well. The median treatment time was 18.2 min (range: 13-32 min), which was progressively shortened with increasing experience. A total of 509 portal images were analyzed. Combining measurements for all patients, the interfraction setup errors (1 SD) in the lateral and craniocaudal directions and in rotation were 2.4 mm, 3.2 mm, and 1 degrees, respectively, for the medial beam and 2.3 mm, 3.1 mm, and 1 degrees, respectively, for the lateral beam. For all patients, the intrafraction setup errors were about 1 mm and always less than 2 mm (1 SD). CONCLUSION: Reduction in heart V(30) can be achieved in patients with left-sided breast cancer using mDIBH assisted with an ABC device. With increasing experience, ABC treatments were streamlined and could be performed within a 15-min treatment slot. Our results suggest that mDIBH using an ABC device may provide one of the most promising methods of improving the efficacy of EBRT in patients with left-sided breast cancer, particularly when wide tangential beams are employed.Breast cancer; Breath hold; Radiation therapy; Intensity modulated radiation therapy