Abstract
Background/Aim: To compare implant sparing irradiation with conventional radiotherapy (RT) using helical (H) and TomoDirect (TD) techniques in breast cancer patients undergoing immediate breast reconstruction (IBR). Patients and Methods: The dosimetric parameters of 40 patients with retropectoral implants receiving 50.4 Gy delivered in 28 fractions were analyzed. Three plans were created: H plan using conventional planning target volume (PTV) that included the chest wall, skin, and implant; TD plan using conventional PTV; and Hs plan using implant-sparing PTV. The H, TD, and Hs plans were compared for PTV doses, organ-at-risk (OAR) doses, and treatment times. Results: Dose distribution in the Hs plan was less homogeneous and uniform than that in the H and TD plans. The TD plan had lower lung, heart, contralateral breast, spinal cord, liver, and esophagus doses than the Hs plan. Compared to the Hs plan, the H plan had lower lung volume receiving 5Gy (V5) (39.1±3.9 vs. 41.2±3.9 Gy; p<0.001), higher V20 (12.3±1.3 vs. 11.5±2.6 Gy; p=0.02), and higher V30 (7.5±1.6 vs. 4.4±1.7 Gy; p<0.001). H plan outperformed Hs plan in heart dosimetric parameters except V20. The Hs plan had significantly lower mean implant doses (43.4±2.1 Gy) than the H plan (51.4±0.5 Gy; p<0.001) and the TD plan (51.9±0.6 Gy; p<0.001). Implementing an implant sparing technique for silicone dose reduction decreases lung doses. Conclusion: Conventional H and TD plans outperform the implant sparing helical plan dosimetrically. Because capsular contracture during RT is unpredictable, long-term clinical outcomes are required to determine whether silicon should be spared.
Post-mastectomy radiotherapy (RT) is the standard of care for breast cancer patients with high-risk disease and lymph node metastases, as clinical benefit has been demonstrated in previous studies with long-term follow-up (1, 2). In the past decade, the use of mastectomy with immediate breast reconstruction (IBR) with permanent silicone implants has increased, and as a result, its interaction with RT has become a topic of great interest (3, 4).
There is limited research on the potential impact of RT on breast implants, and the findings are inconsistent (5). However, RT may increase the capsular contracture of the implant, resulting in inferior cosmesis (6, 7). Furthermore, numerous studies have shown that ionizing irradiation can change the mechanical properties of silicone materials by increasing hardness and decreasing tear strength (8, 9). In addition to the potential for radiation-induced toxicity, the reconstructed breast could potentially hinder the delivery of RT, making it challenging to achieve adequate target coverage and effectively spare organs at risk (OARs).
There is no agreement on the clinical target volume (CTV) definition in patients undergoing IBR. The ‘European Society of Radiation & Oncology and Advisory Committee on Radiation Oncology Practice (ESTRO-ACROP) Consensus Guidelines’ recommendations for the CTV are based on the observation that the majority of local recurrences after mastectomy occur at skin and subcutaneous tissue, where the majority of residual glandular tissues and draining lymph nodes are located (10, 11). The ESTRO-ACROP consensus guidelines exclude the major pectoral muscle and ribs from the CTV of the thoracic wall and instead recommend limiting the CTV to the subcutaneous tissues and skin, particularly in patients with retropectoral implants, in order to limit the doses to the implant as well as to OARs.
Modern RT techniques, including field-in-field irradiation, intensity modulated RT (IMRT), volumetric arc therapy (VMAT), and helical tomotherapy (HT), improve target volume dose distribution and reduce dose exposure to OARs (12-15). Two modes of breast irradiation with HT have increased in popularity, helical mode (H) and TomoDirect (TD) (16). There have been limited studies on dosimetric evaluation of VMAT and HT plans in patients who are receiving implant-sparing post-mastectomy RT (17, 18). The dosimetric comparison of H and TD plans for post-mastectomy implant sparing RT utilizing HT has not been studied.
This study aimed to compare the dosimetric outcomes of two different modes of high-dose-rate brachytherapy (HT) and evaluate the effectiveness of implant sparing planning using a specific technique in patients with retropectoral implants. A comparison was made between the target volume, OAR doses, and implant doses measured in H and TD plans using standard CTV, and the H plan using CTV defined according to ESTRO-ACROP guidelines.
Patients and Methods
Patients. Forty patients were enrolled in this study, 20 with left-sided tumors and 20 with right-sided tumors; all had received RT after implant reconstruction for locally advanced breast cancer (BC). The implants were all placed beneath the pectoral muscle. The median age of patients was 46 years (range=31-54 years). At the time of initial diagnosis, 18 patients had stage II BC and 22 had stage III BC.
Target volumes. All patients underwent a 2.5-mm slice thickness, free-breathing computed tomography (CT) scan in the supine position on a 10°-15° angle breast-tilting board with both arms elevated for treatment planning (15). The conventional CTV (CTVc) included the skin, subcutaneous tissues, pectoral major muscle, implant, and rib plane (19, 20). The CTV that spares the implant (CTVs) includes the skin, the subcutaneous tissues down to the implant, and the major pectoral muscle, thereby excluding the implant, chest wall muscles, and the rib plane (17, 18). A 5-mm expansion of the CTV in all directions around the tumor bed, excluding a 2-mm skin strip, yielded the planning target volume (PTV). Additionally, the lung and heart were excluded from the PTV. The PTVc is derived from CTVc, whereas PTVs is derived from CTVs.
The ipsilateral lung, contralateral breast and lung, heart, spinal cord, and liver were all included in the OARs. Excluding pericardial fat tissue, the heart was outlined from the pulmonary trunk to its farthest extent near the diaphragm.
Treatment planning. The target volumes were prescribed a total dose of 50.4 Gy, administered in fractions of 1.8 Gy. Three plans were generated in total: an H plan utilizing PTVc, a TD plan utilizing PTVc, and an Hs plan utilizing PTVs. The application of the TD technique is not suitable for implant sparing plans, as the tangential beam configuration employed in TD plans does not effectively spare the implants. The Hi-Art Tomotherapy system (TomoTherapy Inc., Madison, WI, USA), a helical fan-beam IMRT system with inverse planning software and a 6-MV photon beam for implant sparing (Hs), was used to generate the initial plan. Using the same CT images and conventionally delineated target volumes, two additional plans with H and TD modes were generated for each patient (Figure 1).
Dose distribution demonstrating 50% and 90% of prescribed dose for whole-breast in (A) helical and (B) TomoDirect plans using conventional clinical target volume, and (C) helical plan using implant sparing target volume for right-side breast, and same plans for left-side breast (D-F). The 50% isodose volume (blue area), 90% isodose volume (green area) and tumor-bed boost (orange area).
The H plans were devised for the TomoEdge Dynamic Jaws system, which is a component of the TomoHDA series. The collimator’s aperture measured 2.5 cm, its pitch was 0.25 mm, and its modulation factor was 5.0. Dose calculations were conducted utilizing the fine-dose calculation grid, which consisted of a 3 mm spacing in the craniocaudal direction over a 256×256 matrix in the axial plane of the original CT scan. During planning, the contralateral breast, hemibody, and posterior portion of the ipsilateral side of the body were blocked to prevent beamlets from passing through the virtual contour on the CT image, thereby reducing the dose to OARs. The TD plan was characterized by a jaw width measuring 5 cm, a pitch value of 0.25, and a modulation factor of 5.0.
The plan was designed to ensure that 95% of the PTV received at least 95% of the prescribed dose. More than 107% of the prescribed dose should be received in less than 1% of the volume. The dose constraints for OARs were as follows: (a) <10% and <30% of the heart volume may receive less than 25 Gy and 5 Gy, respectively; (b) <5% and <20% of the ipsilateral lung may receive more than 50 Gy and 15 Gy, respectively, with a mean dose of less than 12 Gy; and (c) the mean doses and <5% of the contralateral breast and lung should be limited to 3 Gy and 5 Gy, respectively. The maximum dose to the spinal cord must be less than 30 Gy.
The conformity indices (CI) and target homogeneity (HI) were compared. HI=[(D2-D98)/D50], where D2 and D98 (minimal doses to 2% and 98% of target volumes, respectively) were used as surrogates for maximum and minimum doses. A higher HI value indicates that the dose distribution is not uniform. CI=(VTref/VT)/(VTref/Vref), where VTref is the target volume covered by isodose, VT is the target volume, and Vref is the total volume covered by 95% of isodose. The CI value ranged from 0 to 1, with a value closer to 1 indicating better dose conformity to the PTV.
Statistical analysis. Statistical analysis was performed using IBM SPSS Statistics 22 (IBM Corp., Armonk, NY, USA) and MedCalc version 20.111 (MedCalc Software Ltd., Ostend, Belgium). Dn and Vn were calculated for the PTV and OARs. Vn represents the percentage of organ volume receiving ≥nGy and Dn represents the percentage of organ receiving n% of the prescribed dose. The comparative analysis involved evaluating the target volume doses, CI, HI, as well as the doses to OARs and treatment durations across different treatment plans. The variables were investigated using visual and analytical methods to determine whether or not they are normally distributed. The ANOVA test was used for the normally distributed groups, and the Wilcoxon signed-rank test was used for the groups that were not normally distributed. The correlation between mean silicone dose and mean lung and heart doses were investigated using Spearman correlation test. All p values reported are two-sided, and p<0.05 was considered statistically significant.
Results
Target volume doses. The median PTVc and PTVs were 915.4 cm3 (453.3-1,382.0 cm3) and 502.1 cm3 (227.6-869.0 cm3), respectively. Median implant volume was 379.7 cm3 (213.8-524.0 cm3). Figure 1 shows axial sections depicting the dose distributions for the H and TD plans with PTVc and Hs plan with PTVs in representative patients, respectively.
Table I summarizes the dosimetric parameters of target volumes for H, TD, and Hs plans. The target volume coverage criteria were met by all plans. The maximum PTV dose in the TD plan was significantly higher than that in the Hs plan, and the V107 of PTV in the Hs plan was significantly lower than those in the H and TD plans. When compared to the H and TD plans, the HI was significantly higher and the CI was significantly lower in the Hs plan, indicating poorer homogeneity and conformity of dose distribution for target volume.
Target volume doses according to helical (H), TomoDirect (TD) plan, and implant sparing helical plan (Hs).
Organs at risk doses. The OAR dosimetric data for the H, TD, and Hs plans are shown in Table II. All plans complied with OAR dose constraints. Ipsilateral lung V5 was significantly lower in the H plan than in the Hs plan (39.1 Gy vs. 41.2 Gy; p<0.001), whereas V20 (12.3 Gy vs. 11.5 Gy; p=0.02) and V30 (7.5 Gy vs. 4.4 Gy; p<0.001) were significantly lower in the Hs plan than in the H plan (Figure 2). However, there was no significant difference between H and Hs in terms of mean lung doses (8.4±0.7 Gy vs. 8.2±0.8 Gy; p=0.1). All dosimetric parameters according to dose volume parameters and mean lung doses in the TD plan were significantly lower than those measured in the Hs plan.
Organs at risk doses according to helical (H), TomoDirect (TD) plan, and implant sparing helical plan (Hs).
Box and whisker plot demonstrating ipsilateral lung doses according to dose volume parameters and mean lung doses across each plan (H plan: helical plan; TD plan: TomoDirect plan; Hs: implant sparing helical plan).
The heart V5 was significantly higher in Hs plan compared to H and TD plans in the entire cohort, and in patients with right- and left-sided tumors (Figure 3). The mean heart dose was significantly higher in the Hs plan (3.8±0.8 Gy) compared to the H plan (3.4 Gy±1.0 Gy; p<0.001) and TD plan (2.1±1.5 Gy; p<0.001) in the entire cohort. Similarly, the mean heart doses for patients with right-side tumors were significantly higher in Hs plan (3.2±0.6 Gy) compared to both the H (2.8±0.8 Gy; p<0.001) and TD plans (1.2±0.9 Gy; p<0.001). For patients with left-side tumors, mean heart doses were 4.4±0.5 Gy, 4.1±0.7 Gy and 2.9±1.5 Gy for Hs, H and TD plans, respectively (p=0.001).
Box and whisker plot demonstrating heart doses according to dose volume parameters and mean lung doses across each plan in (A) entire group, (B) right-side tumor group and (C) left-side tumor group.
The mean doses and V5 values of the contralateral lung were significantly higher in the Hs plan compared to both the H plan and TD plan (Table II). While comparing the Hs, TD, and H plans, it was observed that the contralateral breast V5 values were significantly higher in the Hs plan. However, the contralateral breast mean dose was significantly lower only in the TD plan compared to the Hs plan. No significant difference was found between the H and Hs plans in terms of contralateral breast mean dose. Furthermore, the TD and H plans demonstrated a significant decrease in doses to the spinal cord and esophagus when compared to the Hs plan. The hepatic doses exhibited a statistically significant decrease in the TD plan in comparison to the Hs plan. However, no statistically significant difference was observed between the H and Hs plans in terms of hepatic doses.
There was no major change in mean implant dose in H and TD plans, since the majority of implants were within the PTV. In contrast, the average implant doses exhibited a statistically significant decrease in the Hs plan when compared to both the H and TD plans. There was no statistically significant correlation between lung doses and implant doses in both the H and TD plans (Figure 4A and B). In the Hs plan, there was a significant and moderate correlation observed between the mean silicone dose and mean lung dose (Spearman=0.39, p=0.01; Figure 4C). A moderate and statistically significant correlation was observed between the mean heart and implant doses exclusively in the H plan (Spearman=0.34, p=0.03; Figure 4D). However, no significant correlation was observed between the heart and silicone doses in both the TD and Hs plans (Figure 4E and F).
Correlation between mean silicone dose and mean lung dose in (A) helical plan, (B) TomoDirect plan, and (C) implant sparing helical plan. Correlation between mean silicone dose and mean heart dose in (D) helical plan, (E) TomoDirect plan, and (F) implant sparing helical plan.
The median treatment duration for the H, TD, and Hs plans, respectively, was 5.4 min (range=4.1-8.1 min), 6.8 min (range=4.6-9.3 min), and 5.5 min (range=4.6-7.4 min). The treatment duration was significantly longer in the TD plan than in the Hs plan (p=0.001), but no significant difference was observed between the H and Hs plans (p=0.24).
Discussion
In the present study, we found that implant sparing irradiation leads to a notable reduction in silicone doses. However, it was found that this approach results in inferior homogeneity and conformity of radiation dose of target volume, in comparison to plans involving irradiation of the entire implant using TD and H plans. Although it is worth noting that all plans successfully adhered to the OARs dose constraints, it is evident that the TD plan exhibited significantly lower doses for the lungs, heart, contralateral breast, spinal cord, liver, and esophagus in comparison to the Hs plan. The utilization of a whole implant irradiation strategy in the H plan demonstrates advantages in terms of minimizing the impact on a limited number of OARs, as compared to the Hs plan. The implementation of an implant sparing technique for silicone dose reduction results solely in a decrease in mean lung doses. However, it should be noted that the treatment time was considerably longer for the TD technique in comparison to the H and Hs techniques.
While breast conserving management is widely accepted as the preferred approach for many breast cancer patients, the use of mastectomy with IBR using permanent implants has increased significantly in recent decades (3, 4). In spite of the fact that the implementation of IBR provides numerous benefits for patients, it is important to recognize that it may also pose certain difficulties in terms of RT planning. Previous research found that the skin and subcutaneous tissues anterior to the pectoralis muscles were the most common site for chest wall recurrence (21). The observed pattern of recurrence poses a potential challenge to the existing recommendation of including the anterior pleural surface, along with the ribs and intercostal muscles, as the border for the posterior chest wall (22, 23). While the target volume typically encompasses the skin and major pectoralis muscle, the occurrence of capsular contracture can be unpredictable. However, it is worth noting that preserving the implant and chest wall interface may potentially decrease fibrotic reactions, leading to improved adherence and fixation of the implants within the deeper layers. As a result, target volume definitions for patients undergoing IBR must be reconsidered in order to reduce the occurrence of capsular contracture while maintaining the effectiveness of RT in terms of locoregional control.
Multiple studies have demonstrated the difficulty of achieving adequate coverage of the chest wall and lymph node regions without increasing OAR doses (24, 25). However, limited number of studies assessed the dosimetric and clinical outcomes associated with implant sparing RT utilizing various techniques (17, 18, 26). Massabeau et al. (26) compared the field-in-field technique to HT plans in ten breast cancer patients who had retropectoral implants, with a prescribed dose of 50 Gy delivered in 25 fractions to the skin-implant space, as well as the supraclavicular, infraclavicular, and internal mammary nodes. The researchers concluded that HT has the potential to achieve complete coverage of the target volumes. The study conducted by Leonardi et al. (18) compared the dosimetric characteristics of conventional and HT plans using hypofractionated treatment regimen delivered 40.05 Gy in 15 fractions. The researchers did not find statistically significant difference in the coverage of the target volume between the two plans. Göksel et al. (17) found no statistically significant difference in target coverage among 16 patients who underwent implant sparing HT and VMAT plans for skin and subcutaneous tissue, including the axilla and supraclavicular lymph nodes. All dose parameters pertaining to target volume doses were determined to be acceptable in the current study. However, when comparing the Hs plan to the H and TD plans, the HI was found to be significantly higher, while the CI was significantly lower. These findings point to poorer homogeneity and conformity of target volume doses in the Hs plan, which could have a negative impact on cosmesis.
The significance of low-dose spread, specifically lung V5, in predicting lung toxicity has been recognized in conjunction with conventional dosimetric factors, such as V20 and mean lung doses, ever since the advent of IMRT (27, 28). Previous studies had established that the use of the TD plan yields a substantial reduction in lung doses when compared to other modern irradiation techniques (13, 14, 16). Leonardi et al. (18) demonstrated that the implant sparing technique resulted in significantly lower doses to the lungs when compared to the conventional technique. In terms of doses delivered to the ipsilateral lung, Göksel et al. (17) discovered that HT outperformed VMAT. We discovered that the dosimetric parameters for the lungs in the TD plan were statistically significantly lower than the doses in the Hs plan. While lung V5 doses in the H plan were significantly lower than those in the Hs plan, lung V20 doses in the Hs plan were significantly lower. However, there were no statistically significant differences in mean lung doses between the H and Hs plans. Furthermore, we found that the doses received by the heart in the Hs plan were significantly higher than those in the H and TD plans. The heart doses in our study were lower than previous studies, most likely due to the inclusion of internal mammary lymph nodes in one study (18) and very strict heart dose constraints in our protocol compared to previous study (17).
Another important consideration with modern irradiation techniques, particularly in helical treatment, is the increased doses received by the contralateral breast and lung. This raises concerns about the possibility of secondary cancer development (29). According to the findings of this study, it is prudent to reduce the occurrence of low-dose spread. Both techniques demonstrated notably low V5 values and mean doses for the breast and lung in this study. Nonetheless, with the exception of contralateral breast mean doses in the H and Hs plans, the TD and H plans were more effective in reducing radiation exposure to the contralateral lung and breast.
Several studies have found that nearly 30% of patients require implant removal, while approximately 20% experience significant complications as a result of IBR and subsequent RT (6, 24, 29, 30). Capsular contracture is the most common complication associated with breast implants, with patients who have RT after IBR being more prone to developing this condition. As a result, limiting the therapeutic dosage given to the implant may reduce the risk of complications like capsular contracture while also improving the aesthetic outcome (31). Our findings show that using implant sparing RT resulted in a significant reduction in silicone doses when compared to plans using conventional CTV with H and TD techniques. Furthermore, we showed that decreasing implant doses results in a significant decrease in mean lung doses.
Study limitations. The dosimetric parameters of three distinct plans were evaluated in this study, but no assessment of radiation-related late toxicities or long-term cosmetic outcomes of these procedures was performed. Furthermore, our study was limited to assessing the chest wall with IBR. It is important to note, however, that the dosimetric evaluation of more complex treatment plans, such as those incorporating lymphatic field irradiation, may be investigated in future research. In addition to these limitations, the significance of our study lies in its ability to determine the potential benefits of implant sparing in breast cancer patients undergoing RT after IBR in comparison to two modern techniques using HT.
Conclusion
Our dosimetric results and a thorough literature review show that there is no optimal treatment plan for postoperative RT for breast cancer patients after IBR. Individuals prefer the H plan with standard target volumes because it focuses on coverage and uses smaller low dose volumes (5 Gy) to minimize radiation exposure to the ipsilateral lung and heart when compared to the implant sparing plan. This is especially important for avoiding long-term cardiac and pulmonary complications associated with RT. Furthermore, the TD plan outperformed the Hs plan in terms of reducing doses to the heart and ipsilateral lung while minimizing doses to the contralateral lung and breast. This is especially beneficial in preventing the development of secondary malignancies in younger patients with a family history. As a result, our study discovered that the implant sparing helical plan has no dosimetric advantage over other HT techniques. Furthermore, because capsular contracture is unpredictable, and because skin and subcutaneous tissue were irradiated in IBR patients, evaluation of long-term clinical outcomes is needed to determine whether silicon and the chest wall should be spared.
Footnotes
Authors’ Contributions
CO and RB conceptualized and designed this study. RB, YD and OCG collected and analyzed data. CO, AE, and OCG participated in the interpretation of results. OCG and AE drafted the article, CO revised the article. All Authors read and approved the article.
Funding
None.
Conflicts of Interest
The Authors declare no conflicts of interest in relation to this study.
- Received January 6, 2024.
- Revision received February 7, 2024.
- Accepted February 8, 2024.
- Copyright © 2024 The Author(s). Published by the International Institute of Anticancer Research.
This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY-NC-ND) 4.0 international license (https://creativecommons.org/licenses/by-nc-nd/4.0).
