Predictors of Cardiac Sparing in Deep Inspiration Breath-Hold for Patients With Left Sided Breast Cancer

Mean Heart Dose Prediction Using Parameters of Single-Slice Computed Tomography and Body Mass Index: Machine Learning Approach for Radiotherapy of Left-Sided Breast Cancer of Asian Patients

Deep inspiration breath-hold (DIBH) is an excellent technique to reduce the incidental radiation received by the heart during radiotherapy in patients with breast cancer. However, DIBH is costly and time-consuming for patients and radiotherapy staff. In Asian countries, the use of DIBH is restricted due to the limited number of patients with a high mean heart dose (MHD) and the shortage of radiotherapy personnel and equipment compared to that in the USA. This study aimed to develop, evaluate, and compare the performance of ten machine learning algorithms for predicting MHD using a patient's body mass index and single-slice CT parameters to identify patients who may not require DIBH. Machine learning models were built and tested using a dataset containing 207 patients with left-sided breast cancer who were treated with field-in-field radiotherapy with free breathing. The average MHD was 251 cGy. Stratified repeated four-fold cross-validation was used to build models using 165 training data. The models were compared internally using their average performance metrics: F2 score, AUC, recall, accuracy, Cohen's kappa, and Matthews correlation coefficient. The final performance evaluation for each model was further externally analyzed using 42 unseen test data. The performance of each model was evaluated as a binary classifier by setting the cut-off value of MHD ≥ 300 cGy. The deep neural network (DNN) achieved the highest F2 score (78.9%). Most models successfully classified all patients with high MHD as true positive. This study indicates that the ten models, especially the DNN, might have the potential to identify patients who may not require DIBH.

The effect of respiratory capacity for dose sparing in left-sided breast cancer irradiation with active breathing coordinator technique

Background and aim

A subsequent cardiac toxicity is deemed to be dose-dependent for left-sided breast cancer irradiation. This study aims to demonstrate the effect of respiratory capacity for dose sparing when the deep inspiration breath hold with Active Breathing Coordinator technique (ABC-DIBH) is used in left-sided breast cancer irradiation.

Methods

74 left-sided breast cancer patients, who received whole breast or post-mastectomy chest wall radiotherapy with ABC-DIBH between 2020 and 2021 in our center, were retrospectively reviewed in this study. CT scans of free breath (FB) and ABC-DIBH were done for each patient, and two treatment plans with a prescription dose of 5000 cGy/25 Fr were designed separately. The dose to heart, left anterior descending artery (LAD) and lungs was compared between FB and ABC-DIBH. The correlation between individual parameters (dose to organs at risk (OARs) and minimum heart distance (MHD)) was analyzed, and the effect of respiratory capacity for dose sparing was assessed.

Results

The plans with ABC-DIBH achieved lower Dmean for heart (34.80%, P < 0.01) and LAD (29.33%, P < 0.01) than those with FB. Regression analysis revealed that both Dmean and D2 of heart were negatively correlated with MHD in the plans with FB and ABC-DIBH, which decreased with the increase in MHD by 37.8 cGy and 309.9 cGy per 1mm, respectively. Besides, a lower Dmean of heart was related to a larger volume of ipsilateral lung in plans with FB. With the increase in volume of ipsilateral lung, the linear correlation was getting weaker and weaker until the volume of ipsilateral lung reached 1700 cc. Meanwhile, a negative linear correlation between Dmean of LAD and MHD in plans with FB and ABC-DIBH was observed, whose slope was 162.5 and 135.9 cGy/mm, respectively. Furthermore, when the respiratory capacity of ABC-DIBH reached 1L, and the relative ratio (ABC-DIBH/FB) reached 3.6, patients could obtain the benefit of dose sparing. The larger difference in respiratory capacity had no significant effect in the larger difference of MHD, Dmean of heart and Dmean of LAD between FB and ABC-DIBH.

Conclusion

This study demonstrates the sufficiently good effect of ABC-DIBH when utilizing for cardiac sparing. It also reveals the correlations among individual parameters and the effect of respiratory capacity for dose sparing. This helps take optimal advantage of the ABC-DIBH technique and predict clinical benefits.

Development of deep learning chest X-ray model for cardiac dose prediction in left-sided breast cancer radiotherapy

Deep inspiration breath-hold (DIBH) is widely used to reduce the cardiac dose in left-sided breast cancer radiotherapy. This study aimed to develop a deep learning chest X-ray model for cardiac dose prediction to select patients with a potentially high risk of cardiac irradiation and need for DIBH radiotherapy. We used 103 pairs of anteroposterior and lateral chest X-ray data of left-sided breast cancer patients (training cohort: n = 59, validation cohort: n = 19, test cohort: n = 25). All patients underwent breast-conserving surgery followed by DIBH radiotherapy: the treatment plan consisted of three-dimensional, two opposing tangential radiation fields. The prescription dose of the planning target volume was 42.56 Gy in 16 fractions. A convolutional neural network-based regression model was developed to predict the mean heart dose (∆MHD) reduction between free-breathing (MHD_FB) and DIBH. The model performance is evaluated as a binary classifier by setting the cutoff value of ∆MHD > 1 Gy. The patient characteristics were as follows: the median (IQR) age was 52 (47–61) years, MHD_FB was 1.75 (1.14–2.47) Gy, and ∆MHD was 1.00 (0.52–1.64) Gy. The classification performance of the developed model showed a sensitivity of 85.7%, specificity of 90.9%, a positive predictive value of 92.3%, a negative predictive value of 83.3%, and a diagnostic accuracy of 88.0%. The AUC value of the ROC curve was 0.864. The proposed model could predict ∆MHD in breast radiotherapy, suggesting the potential of a classifier in which patients are more desirable for DIBH.

A Critical Overview of Predictors of Heart Sparing by Deep-Inspiration-Breath-Hold Irradiation in Left-Sided Breast Cancer Patients

Simple Summary

Adjuvant radiotherapy could damage the heart in left-sided breast cancer patients. The deep-inspiration-breath-hold technique may limit the heart exposure to radiation. As non-beneficiaries exist, there is some need to do an upfront cost-effective selection. Some easy-to-use anatomical predictors may help insiders in the treatment decision. The awareness of such findings may improve the efficiency of practitioners’ workflows.

Abstract

Radiotherapy represents an essential part of the therapeutic algorithm for breast cancer patients after conservative surgery. The treatment of left-sided tumors has been associated with a non-negligible risk of developing late-onset cardiovascular disease. The cardiac risk perception has especially increased over the last years due to the prolongation of patients’ survival owing to the advent of new drugs and an ever earlier cancer detection through screening programs. Improvements in radiation delivery techniques could reduce the treatment-related heart toxicity. The deep-inspiration-breath-hold (DIBH) irradiation is one of the most advanced treatment approaches, which requires specific technical equipment and uses inspiration to displace the heart from the tangential radiation fields. However, not all patients benefit from its use. Moreover, DIBH irradiation needs patient compliance and accurate training. Therefore, such a technique may be unjustifiably cumbersome and time-consuming as well as unnecessarily expensive from a mere healthcare cost point of view. Hence the need to early select only the true beneficiaries while tailoring more effective heart-sparing techniques for the others and streamlining the workflow, especially in high-volume radiation oncology departments. In this literature overview, we collected some possible predictors of cardiac dose sparing in DIBH irradiation for left breast treatment in an effort to provide an easy-to-consult summary of simple instruments to insiders for identifying patients actually benefitting from this technique. We critically reviewed the reliability and weaknesses of each retrieved finding, aiming to inspire new insights and discussions on this much-debated topic.

Preoperative spirometry and BMI in deep inspiration breath-hold radiotherapy: the early detection of cardiac and lung dose predictors without radiation exposure

Background

This study aimed to investigate preoperative spirometry and BMI as early predictors of the mean heart and lung dose (MHD, MLD) in deep inspiration breath-hold (DIBH) radiotherapy.

Methods

Left-sided breast cancer patients underwent breast-conserving surgery followed by DIBH radiotherapy enrolled. Patients who were not available for preoperative spirometry were excluded. One hundred eligible patients were performed free-breathing (FB-) CT and DIBH-CT for plan comparison. We completed the correlative and multivariate analysis to develop the linear regression models for dose prediction. The residuals were calculated to explore the unpreferable subgroups and compare the prediction accuracy.

Results

Among the parameters, vital capacity (VC) and BMI showed the strongest negative correlation with MHD (r = − 0.33) and MLD (r = − 0.34), respectively. They were also significant in multivariate analysis (P < 0.001). Elderly and less VC were independent predictors of increasing absolute residuals (AR). The VC model showed no significant difference in AR compared to the model using the CT parameter of lung volume in FB (LV-FB): median AR of the LV-FB model vs. the VC model was 0.12 vs. 0.11 Gy (P = 0.79). On the other hand, the median AR of the MLD model was 0.38 Gy, showing no specific subgroups of larger AR.

Conclusion

Preoperative spirometry and BMI are significant predictors of MHD and MLD, respectively. Although elderly and low-VC patients may have larger predictive variations, spirometry might be a substitute for LV-FB as a predictor of MHD.

Free breathing VMAT versus deep inspiration breath-hold 3D conformal radiation therapy for early stage left-sided breast cancer

The purpose of the in silico study was to compare free breathing volumetric modulated arc therapy (VMAT) to standard deep inspiration breath‐hold (DIBH) three‐dimensional conformal radiotherapy (3DCRT) and determine whether the former is a viable option for elderly patients with left‐sided early stage breast cancer. Data from 22 patients with early‐stage left breast carcinoma requiring breast‐only radiation therapy were used for this planning study. The robustness of VMAT plans when using the free breathing method was compared to that of standard 3DCRT plans using the DIBH method. The endpoints for evaluation were the target dose coverage as well as doses to the organs‐at‐risk. The free breathing VMAT plans produced a significantly higher mean dose to the heart and right breast than the DIBH‐3DCRT plans. Free breathing VMAT plans resulted in significantly better target coverage than did 3DCRT using DIBH. The external volume that received more than 40 Gy was significantly smaller in the VMAT plans. Free breathing VMAT is a viable alternative to DIBH 3DCRT in elderly patients with a limited life expectancy and in subjects who are unable to perform DIBH. The choice of treatment should be individualized, and all relevant risks ought to be considered.

Breast size and dose to cardiac substructures in adjuvant three-dimensional conformal radiotherapy compared to tangential intensity modulated radiotherapy

Background

The aim of the study was to quantify planned doses to the heart and specific cardiac substructures in free-breathing adjuvant three-dimensional radiation therapy (3D-CRT) and tangential intensity modulated radiotherapy (t-IMRT) for left-sided node-negative breast cancer, and to assess the differences in planned doses to organs at risk according to patients’ individual anatomy, including breast volume.

Patients and methods

In the study, the whole heart and cardiac substructures were delineated for 60 patients using cardiac atlas. For each patient, 3D-CRT and t-IMRT plans were generated. The prescribed dose was 42.72 Gy in 16 fractions. Patients were divided into groups with small, medium, and large clinical target volume (CTV). Calculated dose distributions were compared amongst the two techniques and the three different groups of CTV.

Results

Mean absorbed dose to the whole heart (MWHD) (1.9 vs. 2.1 Gy, P < 0.005), left anterior descending coronary artery mean dose (8.2 vs. 8.4 Gy, P < 0.005) and left ventricle (LV) mean dose (3.0 vs. 3.2, P < 0.005) were all significantly lower with 3D-CRT technique compared to t-IMRT. Apical (8.5 vs. 9.0, P < 0.005) and anterior LV walls (5.0 vs. 5.4 Gy, P < 0.005) received the highest mean dose (D_mean). MWHD and LV-D_mean increased with increasing CTV size regardless of the technique. Low MWHD values (< 2.5 Gy) were achieved in 44 (73.3%) and 41 (68.3%) patients for 3D-CRT and t-IMRT techniques, correspondingly.

Conclusions

Our study confirms a considerable range of the planned doses within the heart for adjuvant 3D-CRT or t-IMRT in node-negative breast cancer. We observed differences in heart dosimetric metrics between the three groups of CTV size, regardless of the radiotherapy planning technique.

Dose-sparing effect of deep inspiration breath hold technique on coronary artery and left ventricle segments in treatment of breast cancer

BACKGROUND AND PURPOSE

The risk of radiation-induced cardiac injury remains a challenging problem in the treatment of breast cancer. Certain cardiac structures receive higher doses than others, which results in variable frequencies of radiation-induced injuries across these structures. Radiation dose can be reduced using the deep inspiration breath hold (DIBH) technique. We aimed to investigate the dose reductions from DIBH in individual cardiac segments.

MATERIALS AND METHODS

A dosimetric analysis was performed on left-sided breast cancer patients who underwent breast-conserving surgery and whole breast irradiation. Radiation doses to the cardiac structures were compared between the DIBH and free-breathing (FB) techniques and the dose reductions with DIBH were correlated to the lung expansion.

RESULTS

For the 75 patients included in our study, DIBH effectively reduced doses to the heart, left lung, left anterior descending coronary artery (LAD) and left ventricle (LV), but the degree of dose reductions was variable across different structures. The absolute dose reductions were greatest in the distal LAD (14.4 Gy) and apical LV (12.1 Gy) segments, compared with the other LAD (middle 9.7 Gy, proximal 1.6 Gy) and LV (anterior 5.3 Gy, lateral 2.9 Gy, septal 2.0 Gy, inferior 0.2 Gy) segments. Left lung expansion was significantly correlated with the dose reductions in the LAD (Spearman's rank correlation coefficient, ρ, 0.304) and LV (ρ, 0.420) segments.

CONCLUSIONS

Our study demonstrates the dose-sparing effects of DIBH in various cardiac structures, especially the distal LAD and apical LV segments. The large dose reductions seen in the distal LAD and apical LV segments could potentially translate into clinical benefit of reduced cardiac toxicity, as these structures have been previously shown to receive the highest doses and are associated with radiation-induced injury.

Dosimetric evaluation of deep inspiration breath hold for left-sided breast cancer: analysis of patient-specific parameters related to heart dose reduction

Deep inspiration breath hold (DIBH) is a common method used worldwide for reducing the radiation dose to the heart. However, few studies have reported on the relationship between dose reduction and patient-specific parameters. The aim of this study was to compare the reductions of heart dose and volume using DIBH with the dose/volume of free breathing (FB) for patients with left-sided breast cancer and to analyse patient-specific dose reduction parameters. A total of 85 Asian patients who underwent whole-breast radiotherapy after breast-conserving surgery were recruited. Treatment plans for FB and DIBH were retrospectively generated by using an automated breast planning tool with a two-field tangential intensity-modulated radiation therapy technique. The prescribed dose was 50 Gy in 25 fractions. The dosimetric parameters (e.g., mean dose and maximum dose) in heart and lung were extracted from the dose-volume histogram. The relationships between dose-volume data and patient-specific parameters, such as age, body mass index (BMI), and inspiratory volume, were analyzed. The mean heart doses for the FB and DIBH plans were 1.56 Gy and 0.75 Gy, respectively, a relative reduction of 47%. There were significant differences in all heart dosimetric parameters (p < 0.001). For patients with a high heart dose in the FB plan, a relative reduction of the mean heart dose correlated with inspiratory volume (r = 0.646). There was correlation between the relative reduction of mean heart dose and BMI (r = -0.248). We recommend considering the possible feasibility of DIBH in low BMI patients because the degree of benefit from DIBH varied with BMI.

Selection of patients with left breast cancer for IMRT with deep inspiration breath-hold technique

The deep inspiration breath-hold (DIBH) technique has been utilized to reduce the cardiac dose in left-sided breast cancer (BC) patients undergoing radiotherapy. Further investigation of the parameters for selecting which patients will benefit most from DIBH is essential. We performed dosimetric comparisons for 21 patients with left-sided BC who had both computed tomography (CT)-based free-breathing (FB) and DIBH plans. The doses to the heart and left anterior descending artery (LAD) and any reduction due to the DIBH technique were analysed. Based on CTFB plans, dosimetric analysis revealed that the irradiation doses to the heart and LAD were significantly correlated with the target volume, the ipsilateral lung volume (ILV) and the total lung volume (TLV). When patients had an ILV ≥ 950 cm3 or a TLV ≥ 2200 cm3, the irradiation doses to the heart and LAD were significantly decreased. Furthermore, the reduction in the mean heart dose (MHD) was correlated to the difference in lung volume between FB and DIBH. The difference in ILV between DIBH and FB of 1.8 indicated that the patients obtained more benefit from the DIBH technique. The data suggest that lung volume (ILV and TLV) measured on a CT-simulation scan and the difference between FB and DIBH could be utilized to help select patients for DIBH.