Pre-treatment predictors of cardiac dose exposure in left-sided breast cancer radiotherapy patients after breast conserving surgery

Objectives: This study aimed to identify high-risk factors for high cardiac radiation exposure, based on anatomical measurements taken from planning CT images of patients with left-sided breast cancer who underwent breast-conserving surgery and received radiotherapy. Methods: We retrospectively analyzed 45 patients with left-sided breast cancer who underwent whole-breast radiotherapy, either under free breathing (27/45) or deep inspiratory breath-holding (DIBH) (18/45), after breast-conserving surgery. Six anatomical parameters were measured from planning CT images, including treatment planning target volume (PTV), cardiopulmonary volume ratio (CVR), maximum cardiac margin distance, the relative distance between inferior boundaries of heart and PTV (DBIB(H2P)), axial cardiac contact distance, and para-sagittal cardiac contact distance (CCDps). Multiple linear regression analysis was performed using SPSS software to explore the correlation between the six parameters, body mass index (BMI), and the mean heart dose (MHD). Receiver operating characteristic (ROC) analysis was performed to evaluate the predictive power of the selected predictor of cardiac dose exposure. Results: Signi ﬁ cant correlations were observed between the MHD of patients and the CVR, DBIB(H2P), and CCDps parameters. Among them, the CVR was the most important predictor of cardiac dose exposure, with an area under the curve of 0.915 and a cut-o ﬀ value of 0.17. Conclusions: The study results indicated that CVR, DBIB(H2P), and CCDps are the primary parameters associated with the risk of cardiac dose exposure, with CVR being the most signi ﬁ cant predictor. Further prospective studies are required to determine whether these parameters can be used to identify patients who would bene ﬁ t from the DIBH technique.


Introduction
According to the GLOBOCAN 2020 database from the World Health Organization (WHO), there were 2.26 million new cases of breast cancer worldwide, making it the leading cancer type, surpassing lung cancer [1].Radiation therapy (RT) plays an essential role in the synergistic treatment of breast cancer, and its role in breast cancer management is rapidly evolving [2].
However, RT can lead to side effects, such as radiationrelated heart damage, which significantly affects the longterm survival of patients.Studies have demonstrated that RT for breast cancer increases the risk of developing ischemic heart disease [3].A study demonstrated that the rate of major coronary events increased by 7.4 % for every 1 Gy increment of the mean heart dose (MHD) [4].Patients with left-sided breast cancer who undergo RT experience a significant increase in cardiac toxicity compared to those with right-sided breast cancer [5,6].Patients with left-sided breast cancer who received RT after breast-conserving surgery had a 15-year incidence rate of cardiovascular events of 7.6 %, indicating a 1.1 % increase compared to patients who did not undergo RT [7].
The medical community has made significant efforts to reduce radiation dose to the heart through various techniques such as intensity-modulated radiation therapy (IMRT), hypofractionated radiation therapy, accelerated partial breast irradiation, proton therapy, and deep inspiratory breath-hold (DIBH) radiotherapy.DIBH radiotherapy has been proven effective in reducing cardiac dose; however, the benefits may vary among patients [7,8].Furthermore, inconsistent criteria for patient selection and additional resource requirements have hindered its widespread adoption in clinical practice [9].
The aim of this study was to identify high-risk factors for cardiac radiation exposure, based on anatomical measurements from planning CT images of patients with left-sided breast cancer after conserving surgery.By utilizing these predictors, physicians can assess the risk index of cardiac dose in patients and potentially establish guidelines for selecting those who would derive the greatest benefit from the DIBH technique.

Patients and data collection
We conducted a retrospective analysis on 45 patients with left-sided breast cancer who underwent radiotherapy after breast-conserving surgery (lumpectomy) at our hospital between January 2020 and September 2021.The study received approval from the institutional review board.The inclusion criteria were as follows: (1) age between 20 to 60 years, (2) no history of heart diseases, and (3) availability of clinical characteristic records.Of the 45 patients, 18 received DIBH radiotherapy and 27 received free-breathing (FB) radiotherapy.The clinical characteristics of the patients are summarized in Table 1.

Simulation and treatment planning
Patients utilizing DIBH were required to possess the ability to hold their breath for at least 20 s, and chest breathing proficiency was necessary prior to CT simulation.During body film fixation and simulation CT scan, the patient's arms were positioned upward in the supine position.Of the total patients, 27 underwent thermoplastic membrane fixation using the FB technique, while 18 underwent fixation using the DIBH technique.CT scans were conducted using a Philips Brilliance Big Bore CT scanner (Philips Medical Systems, Cleveland, OH, USA) prior to RT, covering the area from the cricoid cartilage to 5 cm below the diaphragmatic region, with a slice thickness of 5 mm.During imaging and treatment, the Varian Real-time Position Management™ (RPM) system was utilized for monitoring DIBH.
Subsequently, the CT images were transferred to the Eclipse (Version 13.5) treatment planning system (Varian Medical Systems, Palo Alto, CA, USA) for dose calculation.The treatment volumes, encompassing the entire breast and tumor bed, were defined based on the consensus definitions provided by the Radiation Therapy Oncology Group (RTOG) breast cancer atlas for RT planning [10].To establish the planning target volume (PTV), a margin was added to the clinical target volume (CTV), with the margin determined by considering the setup errors observed in breast cancer RT patients at our hospital in 2019 and 2020.Organs at risk (OARs), such as the whole heart, ipsilateral lung, contralateral lung, and contralateral breast, were delineated according to the guidelines provided in the RTOG 1106 atlas.The left anterior descending (LAD) coronary artery was delineated as per the validated cardiac contouring atlas developed by Feng et al. in University of Michigan [11].
To ensure consistency in delineation, the same senior radiation oncologist performed the delineation.The prescribed dose to the target was 42.56 Gy delivered in 16 fractions, with the primary objective of achieving 95 % dose coverage of the PTV with 42.56 Gy.Furthermore, a dose boost of 10 Gy (2.0 Gy per fraction for 5 fractions) was administered to the tumor bed.For radiotherapy of breast cancers, various technologies such as 3D conformal radiation therapy, intensity-modulated radiation therapy (IMRT), and volumetric modulated arc therapy (VMAT) are utilized.Among these technologies, VMAT offers significant advantages, including cardiac avoidance, improved dose homogeneity and coverage, and reduced treatment time [12,13].Moreover, the reduced treatment time provides a notable advantage for patients utilizing the DIBH technology.Hence, we implemented the VMAT technology for breast cancer patients at our institute.
During the treatment process for breast cancer patients, VMAT technology with a 240°arc (ranging from 300°to 180°) was primarily utilized [12].The VMAT treatment plans were optimized by using the RapidArc technique executed with the AcurosXB algorithm and a direct aperture optimization approach.Subsequently, the VMAT treatments were administered on the Varian Novalis Linac using 6-MV photons.Each patient's optimal treatment plan was achieved by specifying objective functions tailored to their specific needs.Emphasis was placed on minimizing the volumes receiving doses of 20 Gy, 10 Gy, and 5 Gy (V20, V10, V5) in both the ipsilateral and contralateral lungs, as well as minimizing volumes of doses of 30 Gy, 25 Gy, and 2 Gy (V30, V25, V2) for the heart, while ensuring the MHD remained below 3 Gy.Treatment plans were optimized to achieve the prescription dose to the PTV with the lowest possible MHD.Evaluation of PTV coverage and doses to the OARs was performed using dose-volume histograms derived from the treatment plans.

Characteristics
Distribution

Measurement of anatomical parameters
In the field of breast RT, the evaluation of cardiac dosimetry has predominantly relied on the MHD as a constraining factor.A notable study conducted by Beaton et al. focused on patients receiving adjuvant breast/chest wall RT, revealing a remarkably low 10-year mortality rate attributed to radiation-induced cardiac death when the MHD remained below 3.3 Gy [14].Guided by these findings, we categorized our patient cohort into two groups: high-risk and low-risk, characterized by MHD values exceeding and falling below 3.3 Gy, respectively.Consequently, the ratio of patients with high vs. low cardiac dose exposure was 23:22 in this study.Numerous factors contribute to cardiac dose exposure in clinical practice.Given the ability of DIBH technology to induce positional changes in patients during treatment, our study focused primarily on the anatomical parameters derived from CT images.Specifically, we investigated seven key parameters that potentially correlate with cardiac dose exposure.These parameters encompassed: (1) patients' body mass index (BMI), (2) target PTV volume, (3) cardiopulmonary volume ratio (CVR), ( 4) maximum cardiac margin distance, (5) relative distance between the inferior boundaries of the heart and the PTV (DBIB(H2P)), ( 6) axial cardiac contact distance, and (7) para-sagittal cardiac contact distance (CCDps).By comprehensively examining these parameters, we aimed to gain insights into their associations with cardiac dose exposure.
Figure 1 illustrates the measurement process for obtaining the maximum cardiac margin distance.This involved the following steps: (1) identifying the slice where the heart is closest to the left, (2) connecting the two tangent points of the PTV on that slice to form a line, (3) drawing a parallel line through the outer boundary of the heart, and (4) measuring the vertical distance between the two parallel lines as the maximum cardiac margin distance.
Figure 2 demonstrates the measurement of DBIB(H2P) as the relative distance between the lower border of the heart and the lower border of the PTV along the coronary axis.The value is positive if the lower border of the heart is positioned above the lower border of the breast, and negative if it is not.
As outlined by Hiatt et al. [15], the measurement of cardiac contact distance (CCD) included the axial and para-sagittal planes, as shown in Figure 3.The axial cardiac contact distance was determined as the shortest linear distance between the medial and lateral points of contact between the cardiac silhouette and the chest wall in the axial plane of the CT scan.This measurement was taken at the level of the right hemidiaphragm dome.In the para-sagittal plane, CCDps was measured at the midpoint of the left hemithorax, which was established using the axial and coronal planes as references.CCDps was calculated as the linear  distance representing the direct contact between the heart and the chest wall in this para-sagittal plane.

Statistical analysis
The measured parameters were represented by their mean and standard deviation (SD) values.To examine their correlation with cardiac dose exposure, the IBM SPSS software (version 24.0) was utilized for analysis.Multiple linear regression analysis was conducted to identify the predictor of cardiac dose exposure.Additionally, receiver operating characteristic (ROC) analysis was performed to determine the predictive capability of the identified predictor.The 95 % confidence interval (CI) was calculated mainly by using the mathematical formula (i.e., mean ± standard deviation) method.Statistical significance was defined as a p-value of ≤0.05.

Results
Table 2 presents the values and results of multiple linear regression analysis for the mentioned parameters obtained from the 45 patients.It is evident that CVR, DBIB(H2P), and CCDps exhibited a significant correlation with the MHD, with p-values of 0.006, 0.004, and 0.033, respectively.However, no significant correlation was found between the MHD and the parameters of BMI, target PTV volume, maximum cardiac margin distance, and axial cardiac contact distance.
The multiple linear regression model yielded a linear equation describing the relationship between the MHD and the prediction parameters as follows: Y=524.78 × CVR + 45.50 × CCDps + 25.86 × DBIB(H2P) + 49.21, where Y represents the MHD.Among the measured parameters, CVR had the highest impact on the MHD, signifying its significance as the most influential predictor for cardiac dose exposure prediction.
As previously mentioned, patients with an MHD exceeding 3.3 Gy were categorized as the high-risk group for radiation-induced cardiac death.Accordingly, receiver operating characteristic (ROC) curve analysis was conducted, revealing that CVR exhibited an area under the curve (AUC) of 0.915 (95 % CI: 0.795-1.000) in predicting MHD greater than 3.3 Gy (i.e., high-risk group), with a cut-off value of 0.17 (Figure 4).Similarly, the AUC values for CCDps and DBIB(H2P) in predicting MHD greater than 3.3 Gy were 0.661 (95 % CI: 0.497-0.817)and 0.572 (95 % CI: 0.402-0.746),with a cut-off value of 1.30 and 1.75, respectively.The sensitivity, specificity, positive predictive value (PPV), and negative predictive values (NPV) of these key predictors in prediction of cardiac dose exposure were summarized in Table 3.
The exemplary measurements of three important parameters were obtained from both FB and DIBH scans of the same patient, as depicted in Figure 5. Noticeably, there were substantial changes in the positions of the chest, diaphragm, and heart following deep breath holding.Specifically, the patient's lung volume (Figure 5B), as well as the values of DBIB(H2P) and CCDps (Figure 5C), demonstrated significant variations between the two breathing states.
Figure 6 demonstrates a linear correlation between CVR and MHD.It can be observed that the CVR values for the  majority of patients with MHDs below 3.3 Gy were distributed within the range of 0.1-0.17.In contrast, for patients with MHDs exceeding 3.3 Gy, a substantial proportion of CVR values were higher than 0.17.
The key parameter values of CVR, DBIB(H2P), and CCDps and the values of heart dosimetric parameters of MHD and max LAD dose in FB and DIBH scan of the entire patient cohort were summarized in Table 4.It is observed that the mean CVR value for the DIBH cohort is only half of the value for the FB cohort.The mean values of CCDps and DBIB(H2P) for the DIBH cohort are also lower than those for the FB cohort.Consequently, it is evident that the heart and LAD doses are considerably lower for the DIBH cohort compared to the FB cohort.(The pink and green dots indicate the patients with mean heart doses of less than and greater than 3.3Gy, respectively.Dash line indicates the value of 0.17 of CVR.)

Discussion
Radiation-induced heart damage is a critical complication associated with radiotherapy for breast cancer, primarily influenced by the level of heart irradiation exposure.This study investigated the predictive factors for cardiac dose exposure in breast cancer patients, utilizing anatomical parameters obtained from pre-treatment planning CT images.The parameters examined included CVR, DBIB(H2P), and CCDps, which were found to be significant in both FB and DIBH CT scans.Among these parameters, CVR demonstrated the strongest predictive capability, as evidenced by its high AUC value of 0.915.
To further demonstrate this finding, as shown in Table 5, we selected another eight patients with both FB and DIBH CT scans and calculated the means and SDs of the three parameters.Notably, substantial variations in the values of these parameters were observed between the two scan types.The DIBH technique, in comparison to FB scans, resulted in a significant increase in lung volume and induced anatomical shifts in the thoracic cavity, diaphragm, and heart.Consequently, these changes affected the values of the three parameters as well as the MHD.On average, the DIBH patients exhibited an approximate 30 % decrease in MHD compared to FB patients.These findings strongly support the effectiveness of the DIBH technique in mitigating cardiac dose exposure.
The study conducted by Smyth et al. [16] provided evidence of the efficacy of the DIBH technique in reducing both the MHD and the dose to the LAD coronary artery in patients with left-sided breast cancer.Similarly, several other studies [17][18][19][20] have reported favorable outcomes with regards to the reduction of cardiac dose through the implementation of DIBH.However, it is important to acknowledge that the technique does present certain limitations, including prolonged treatment time, reduced patient throughput, and increased workload for healthcare providers.A survey conducted among breast cancer patients treated at cancer centers affiliated with the European Organization for Research and Treatment of Cancer (EORTC) revealed that only 19 % of patients had adopted DIBH technology as part of their clinical practice [21].
While the implementation of DIBH technology offers notable advantages, its application involves a relatively intricate process.Consequently, it is crucial to carefully select appropriate patients for this technique prior to RT.To avoid the need for designing two separate RT plans, Tanna et al. [22] suggest the establishment of specific criteria of patient selection for DIBH.Wang et al. [9] have reported a direct correlation between the volume of the heart covered by the 50 % isodose line and the MHD, with an increase in MHD of 1 Gy per 1 cc increase in volume.Additionally, Soujanya et al. [23] predicted that reducing the volume of the heart covered by the 50 % isodose line by 6 cc could potentially result in a 20 % reduction in MHD.These findings emphasize the importance of carefully considering the volume of the heart exposed to radiation when aiming to minimize MHD and highlight the potential benefits of optimizing the DIBH technique for individual patients.
Nathalie et al. [24] conducted a study revealing a linear correlation between CCDps and the biologically equivalent uniform dose of the heart.However, they did not observe a significant correlation between axial cardiac contact distance and related dosimetric factors.In a separate study, Hiatt et al. [15] demonstrated that both axial cardiac contact distance and CCDps were associated with cardiac dose in both FB and DIBH conditions.Specifically, cardiac dose exposure increased when axial cardiac contact distance exceeded 5 cm and CCDps exceeded 2 cm.However, it is important to note that these studies primarily measured axial cardiac contact distance and CCDps under FB conditions.In contrast, our study assessed axial cardiac contact distance and CCDps in both FB and DIBH conditions.Our findings revealed a significant correlation between CCDps and the MHD, while axial cardiac contact distance did not exhibit a significant correlation with the MHD.These results align with the findings reported by Nathalie et al.Nonetheless, given the relatively small sample size in our study, we plan to validate these results with a larger sample size in the future study.The utilization of CVR measurements as a predictive tool for assessing the risk index of cardiac dose exposure was supported by the analysis of ROC curves.A CVR value exceeding 0.17 indicates a high risk of heart dose exposure.In cases where a patient's breath training meets the requirements for DIBH, we strongly recommend implementing DIBH to mitigate cardiac dose.Furthermore, if a significant discrepancy in CVR values is observed between FB and DIBH scans conducted on the same patient, it suggests that the patient would substantially benefit from the DIBH technique in terms of minimizing cardiac dose exposure.Thus, the variation in CVR between FB and DIBH scans can be employed as an indicator to evaluate the extent of benefit derived from the DIBH technique, facilitating the selection of patients who stand to gain the most from its implementation.
As this study was conducted retrospectively, we did not make any modifications or optimizations to the treatment plans based on the anatomical parameters strongly correlated with cardiac dose exposure.However, we intend to incorporate the findings from this study into our future treatment plan optimization processes to enhance heart protection measures.It is noteworthy that, thus far, we have not observed any occurrences of cardiac toxicity events during the long-term follow-up of these patients.

Conclusions
In conclusion, our findings demonstrate that CVR, DIBH(H2P), and CCDps are key parameters that exhibit significant correlations with the risk of cardiac dose exposure.Among these parameters, CVR emerges as the most influential predictor of cardiac dose exposure.A CVR value exceeding 0.17 indicates a relatively high heart risk index, highlighting the need to reduce CVR to mitigate cardiac dose.The implementation of DIBH technology proves effective in reducing CVR and subsequently minimizing cardiac dose exposure in patients who meet the necessary physical requirements for adopting this technique.Given the ease of obtaining CVR measurements from CT images, radiation oncology physicians can assess patients' cardiac exposure dose based on their CVR values prior to radiotherapy.This information can potentially guide the selection of patients who are at a high risk of cardiac dose exposure and would benefit from utilizing the DIBH technique.Thus, incorporating CVR assessment in clinical practice has the potential to optimize treatment planning and enhance the management of cardiac dose exposure in radiotherapy for improved patient outcomes.Further prospective studies are required to determine whether these parameters can be used to identify patients who would benefit from the DIBH technique.

Figure 1 :
Figure 1: Measurement of maximum cardiac margin distance (blue line) (PTV contoured in red, the parallel lines drawn in yellow).

Figure 2 :
Figure 2: Measurement of the relative distance between the inferior boundaries of the heart and the PTV (DBIB(H2P), blue line) (PTV is contoured in red, the yellow lines indicate the lower borders of the heart and the PTV, respectively).

Figure 3 :
Figure 3: Measurement of axial and para-sagittal cardiac contact distance (red lines).Landmarks for the measurement are at the apex of the right hemidiaphragm (yellow arrow) and at the left hemithorax.

Figure 6 :
Figure 6: The linear fitting of the relation between CVR and the MHD.(The pink and green dots indicate the patients with mean heart doses of less than and greater than 3.3Gy, respectively.Dash line indicates the value of 0.17 of CVR.)

Table  :
Regression analysis of the parameters of the patients (n=).
V-PTV, target PTV volume; CVR, cardiopulmonary volume ratio; CMD(max), maximum cardiac margin distance; DBIB(HP), the relative distance between the inferior boundaries of the heart and the PTV; CCDax, axial cardiac contact distance; CCDps, para-sagittal cardiac contact distance.

Table  :
Key parameter values for FB and DIBH patients (n=).

Table  :
Measurement of three important parameters of  patients.