Original Research

Cardiac dosimetry using a 3D conformal technique for adjuvant breast radiotherapy

Krishanthavathie Pillay, Sheynaz Bassa, Roy Lakier, Andrew Sarkin

Received: 15 Feb. 2025; Accepted: 06 Aug. 2025; Published: 25 Oct. 2025

Copyright: © 2025. The Authors. Licensee: AOSIS.
This work is licensed under the Creative Commons Attribution 4.0 International (CC BY 4.0) license (https://creativecommons.org/licenses/by/4.0/).

Abstract

Background: Radiation-induced cardiac toxicity is influenced by specific dose-volume parameters, including mean heart dose (MHD), volume of the heart receiving ≥ 25 Gy (V25), and the dose to the left anterior descending artery (LADA).

Aim: To evaluate the cardiac radiation dose parameters in patients receiving postmastectomy radiotherapy using three-dimensional conformal radiotherapy (3DCRT).

Setting: Department of Radiation Oncology, Steve Biko Academic Hospital.

Methods: A retrospective analysis was performed on treatment plans of 62 consecutive breast cancer patients who received adjuvant chest wall radiotherapy at Department of Radiation Oncology, Steve Biko Academic Hospital, which was selected for analysis. Dose-volume histograms were used to assess MHD and volume receiving 25 Gy (V25). The maximum point dose to the LADA was estimated from axial planning CT images. Fifty-seven patients received 50 Gy in 25 fractions; most received a supraclavicular field (SCF) and/or a scar boost. Analysis focused on the 27 left-sided cases.

Results: Among left-sided patients, 24 (88.9%) had an MHD > 2 Gy, and 13 (48.1%) had V25 exceeding 10%. The maximum LADA dose exceeded 6.7 Gy in 24 patients (88.9%). All right-sided patients remained within these constraints. There was a strong correlation between MHD, V25, and LADA dose (Spearman p > 0.8, p < 0.001). Inclusion of SCFs or boost did not significantly affect cardiac dose parameters.

Conclusion: Conventional 3DCRT techniques frequently exceed recommended cardiac dose constraints in left-sided chest wall radiotherapy.

Contribution: These findings underscore the need for advanced techniques, such as deep inspiratory breath hold, to reduce cardiac exposure and long-term cardiovascular risk.

Keywords: radiotherapy; breast cancer; cardiac toxicity; cardiac dosimetry; radiation-induced cardiac disease; conformal radiotherapy; breast radiotherapy; cardiac dose volume constraints.

Introduction

Breast cancer is the most common malignancy among women in South Africa.¹ It is estimated that over 50% of women with localised breast cancer will require adjuvant chest wall radiotherapy.² This includes patients with early-stage disease undergoing breast-conserving surgery, as well as those with advanced disease, T3 or T4 tumours, or pathological involvement of axillary lymph nodes.³

While radiotherapy plays a critical role in improving local control and survival, it can exacerbate age-related coronary artery disease (CAD) and cause direct myocardial damage and fibrosis, potentially leading to cardiomyopathy.⁴ Furthermore, many breast cancer patients receive systemic therapies such as anthracyclines, trastuzumab, and endocrine treatments (aromatase inhibitors and CDK4/6 inhibitors), which may further contribute to the risk of cardiac dysfunction.⁵ As breast cancer survival rates continue to improve, mitigating the risk of radiation-induced cardiac disease has become an increasingly important focus in radiation oncology.

Several radiotherapy dose-volume parameters have been associated with an elevated risk of myocardial injury and CAD, including mean heart dose (MHD), dose to the left anterior descending artery (LADA), and the volume of the heart receiving specified radiation doses.⁶ In this study, we retrospectively analysed the radiotherapy treatment plans of patients who received conventionally fractionated adjuvant breast radiotherapy using a three-dimensional conformal radiotherapy (3DCRT) technique. Special attention was given to cardiac dose exposure referencing Quantitative Analysis of Normal Tissue effects in the Clinic (QUANTEC) guidelines for acceptable dose volume parameters.⁷

Methods

This was a retrospective study evaluating cardiac radiation dose in patients who received post-mastectomy, adjuvant, conventionally fractionated radiotherapy at the Department of Radiation Oncology, Steve Biko Academic Hospital, between January 2015 and April 2016. Patients were excluded if they received palliative radiotherapy, underwent breast-conserving surgery, or had duplicate records. After applying these criteria, 57 patients with complete treatment plans were included for dosimetric analysis.

Radiotherapy technique

All patients were treated using a 3DCRT technique delivered on a linear accelerator (LINAC). Planning was performed on a CT simulator using 5 mm slice thickness. However, formal delineation of clinical target volume (CTV) and planning target volume (PTV) was not performed in accordance with modern International Commission for Radiation Units and Measurements (ICRU) definitions. Instead, field borders were defined based on anatomical landmarks, similar to conventional 2D planning, and treatment plans were reviewed and approved by a radiation oncologist.

Tangential photon fields were used to treat the chest wall. When indicated, a single anterior supraclavicular field (SCF) was added for regional nodal irradiation. Field matching between the tangents and SCF was performed manually using the central axis and independent jaw adjustments to minimise divergence at the junction. Most plans were constructed using a single isocentre and a source-to-axis distance (SAD) technique. A dual isocentre technique was employed only when necessary to accommodate non-coplanar field arrangements involving the SCF.

Photon beam energies of 6 MV – 10 MV were used, with selection based on chest wall separation to optimise dose homogeneity and minimise deep tissue exposure. Virtual wedges and field-in-field techniques were used as required to reduce hotspots. The dose to tangential fields was prescribed to the midpoint of the chest wall, typically at the midline between the tangents. The SCF dose was prescribed to an estimated depth sufficient for nodal coverage while avoiding hotspots > 107% of the prescribed dose.

A conventional fractionation regimen of 50 Gy in 25 daily fractions over 5 weeks was applied. In selected cases, a tumour bed boost was administered using en face electron beams (6 MeV – 9 MeV), prescribed to a depth of 1 cm encompassing the surgical scar with a 2 cm margin.

Dosimetric analysis

Dose normalisation followed departmental protocol, aiming for at least 95% coverage of the intended volume with the prescription dose and limiting hotspots to < 107%. Dose distribution was reviewed using planar isodose lines.

At the time of treatment, organ at risk (OAR) contouring was restricted to the heart and both lungs. For this study, the dose to the LADA was retrospectively contoured on non-contrast planning CTs by the principal investigator using the anatomical guidelines described by Feng et al.⁸ The following cardiac dosimetric parameters were extracted from the dose-volume histogram (DVH):

• mean heart dose
• volume of the heart receiving ≥ 25 Gy (V25)
• mean heart volume (MHV)
• maximum point dose to the LADA, determined on the first CT slice where the artery was clearly identifiable. The LADA was retrospectively contoured for all patients to facilitate this analysis.

Clinical data

Demographic and clinical characteristics, including age, smoking history, comorbidities, and prior systemic therapy, were recorded. Pathological features such as tumour stage and hormone receptor status were also documented. Treatment-related variables included field setup (tangents only or with SCF) and the use of a boost.

Statistical analysis

Descriptive statistics were used to summarise patient demographics, clinical characteristics and treatment parameters. Continuous variables such as age, weight and MHD were reported using means and standard deviations, while categorical variables (e.g. laterality, stage, co-morbidities, treatment factors) were summarised using frequencies and percentages.

Associations between categorical variables (e.g., tumour laterality vs. V25 ≥ 10%, MHD ≥ 2 Gy, and LADA Dmax ≥ 6.7 Gy) were assessed using Pearson’s chi-square test. Comparisons of means between two groups (e.g. MHD or LADA dose in patients with vs. without SCF fields or boost) were evaluated using two-sample independent t-tests assuming equal variances.

To evaluate the correlation between dosimetric parameters, Spearman’s rank correlation coefficients (ρ) were calculated for non-parametric association between MHD, V25, LADA Dmax and MHV. Correlation strength was interpreted using standard thresholds (e.g. ρ > 0.7 indicating strong correlation).

All statistical analyses were performed using Stata software (Stata version: BE 18.5), with a p-value of < 0.05 considered statistically significant.

Ethical considerations

The study was approved by the University of Pretoria MMED protocol review committee as well as the University of Pretoria Health Sciences Research Ethics Committee (protocol number: 423/2015). A waiver of informed consent was obtained from the latter.

Results

Patient characteristics

A total of 57 patients were included in the final analysis. The mean age was 50.0 years (s.d. 12.5; range 29–86), and among 47 patients with available data, the mean weight was 72.4 kg (s.d. 14.8; range 41–104 kg). Clinical stage distribution, based on AJCC 8th edition, included Stage IIA (n = 2, 3.5%), IIB (n = 14, 24.6%), IIIA (n = 25, 43.9%), and IIIB (n = 16, 28.1%).

Comorbidities included hypertension in 17 patients (29.8%), diabetes mellitus in 5 (8.8%), cardiac disease in 2 (3.5%), and a smoking history in 5 (8.8%). Oestrogen receptor (ER) positivity was noted in 36 patients (63.2%), all of whom were prescribed tamoxifen. Most patients (83%) had received prior chemotherapy (mean 6 cycles, s.d. 4.1), primarily anthracyclines (n = 51, 82%), followed by paclitaxel (n = 6, 10%) and docetaxel (n = 4, 6%).

Radiotherapy characteristics

Radiotherapy was delivered to the left chest wall in 27 patients (47.4%) and to the right in 30 (52.6%). All patients received 50 Gy in 25 fractions to the large fields. An SCF was used in 51 patients (89.5%), and a tumour bed boost in 52 patients (91.2%).

Supraclavicular field use was comparable across sides: 23/27 (85.2%) left-sided versus 28/30 (93.3%) right-sided (χ² = 1.00, p = 0.317). Similarly, boost administration was evenly distributed: 24/27 (88.9%) left-sided versus 28/30 (93.3%) right-sided (χ² = 0.35, p = 0.554). Boost doses varied: among left-sided patients, 10 Gy in 5 fractions (n = 21, 78%) and 16 Gy in 8 fractions (n = 4, 15%) were most common; right-sided patients received 10 Gy (n = 12, 40%), 14 Gy (n = 5, 16.7%), or 16 Gy (n = 11, 36.7%).

Cardiac dose parameters

The MHV was 538 cc (s.d. 105; range 320–752). Across the cohort, the MHD was 2.7 Gy (s.d. 2.9; range 0.02–11.8), the mean LADA dose was 1.6 Gy (s.d. 2.15; range 0–5.5), and mean heart V25 was 2.8% (s.d. 4.7; range 0–19.3%).

Cardiac doses were significantly higher in left-sided treatments:

• MHD: 4.7 Gy (s.d. 3.0) versus 0.9 Gy (s.d. 0.74), p < 0.001
• LADA Dmax: 33.5 Gy (s.d. 19.8) versus 0.4 Gy (s.d. 0.4), p < 0.001
• V25 ≥ 10%: 29.6% (8/27) versus 0% (0/30), p = 0.001

Subgroup analysis: Left-sided radiotherapy (n = 27)

• Supraclavicular field versus no SCF: No significant difference in MHD (464 cGy vs. 535 cGy, p = 0.67) or LADA Dmax (3380 cGy vs. 3206 cGy, p = 0.88) between the two groups.
• Boost versus no boost: No significant difference in MHD (443 cGy vs. 726 cGy, p = 0.13) or LADA Dmax (3279 cGy vs. 3953 cGy, p = 0.59) between the two groups.

Threshold exceedance analyses

• MHD ≥ 2 Gy: 88.9% of left-sided patients versus 3.3% of right-sided patients (p < 0.001).
• MHD ≥ 4 Gy: 48.1% left versus 3.3% right (p < 0.001).
• LADA Dmax ≥ 6.7 Gy: Observed in 88.9% of left-sided patients (24/27) and 0% right-sided patients (p < 0.001).
• V25 ≥ 10%: Found in 29.6% of left-sided patients and none of the right-sided group (p = 0.001).

Correlations between dose metrics

Spearman’s rank correlation test showed:

• MHD and V25: ρ = 0.8144, p < 0.001
• MHD and LADA Dmax: ρ = 0.8349, p < 0.001
• V25 and LADA Dmax: ρ = 0.8274, p < 0.001

These confirm a strong positive, statistically significant correlation. Mean heart volume showed no meaningful correlation with any cardiac dose parameter, suggesting heart size did not impact dose metrics in this cohort.

Discussion

This study was conducted in 2016, during which time all patients were treated with conventionally fractionated radiotherapy delivered using 3DCRT and in the absence of sufficient guidelines for cardiac organ constraints. At this time, the QUANTEC guidelines were applied as a guide to reduce cardiac toxicity, and the target constraint was the V25 < 10%. Hence, most cases were within these limits.

Much has emerged since then, including a better understanding of relevant end points. The MHD of < 26 Gy was specifically for pericarditis as the targeted endpoint. The dose to the LADA in this study was determined by the dose to the artery determined on the first non-contrast CT image on which the LADA could be identified. Studies have shown that the Dmax more than 6.7 Gy and, Dmean exceeding 2.7 Gy to the LADA were associated with a significantly increased risk of a major cardiac event.⁹ Perman et al. identified 134 cases of death due to ischaemic heart disease (IHD) in more than 2800 women treated for breast cancer.¹⁰ They determined that the Dmax to the LADA as well as the right coronary artery to be predictive of the risk of IHD along with the MHD. It is therefore important that the dose to multiple cardiac substructures be evaluated when analysing dose constraints and minimising risk.

The threshold for MHD is significantly higher in this study, as the end point selected was that of pericarditis. The dose constraint is substantially lower when evaluating the risk of IHD with a < 10% risk of ischaemic events when the MHD is below 2 Gy.¹¹

These constraints do not consider the impact of comorbid diseases, especially hypertension, which afflicts more than 8 million South Africans.¹² Added to this is the impact of menopausal status exacerbated by endocrine therapy as well as systemic therapies including doxorubicin, Herceptin and other targeted agents.⁴ Long-term follow-up to determine the onset of IHD and cardiac mortality is difficult in the South African patients, where follow-up is often limited by socio-economic factors and ease of access to healthcare facilities.

Since completion of the study, there have been publications that have better defined the dose constraints for patients receiving left-sided chest wall radiotherapy.^9,13 Alongside this is the availability of techniques such as deep inspiratory breath hold (DIBH) achieved using image-guided techniques. These techniques allow for predictable lung expansion and treatment during the inspiratory phase with maximal cardiac displacement.¹⁴

The department has since incorporated moderately fractionated radiotherapy for all adjuvant breast cancer treatments and the use of surface guidance imaging to facilitate DIBH. Furthermore, the dose constraints have been amended in keeping with an MHD < 2 Gy, the heart volume receiving 17 Gy of ≤ 10% and volume receiving 35 Gy of ≤ 5%.

Conclusion

This study demonstrates that conventional 3D conformal radiotherapy (3DCRT) for left-sided postmastectomy chest wall irradiation frequently exceeds established cardiac dose constraints, particularly for the MHD, V25, and maximum dose to the LADA. Despite adherence to standard planning techniques and field arrangements, a significant proportion of patients surpassed thresholds associated with increased risk of IHD. These findings highlight the limitations of 3DCRT in minimising cardiac exposure and underscore the clinical importance of incorporating modern radiotherapy techniques – such as deep inspiratory breath hold (DIBH) and cardiac substructure delineation – into routine practice. Adoption of these approaches, alongside updated dose constraints, is critical for reducing long-term cardiac morbidity and mortality in breast cancer survivors receiving left-sided radiotherapy. Future research should focus on long-term cardiac outcomes in resource-constrained settings and the integration of cardioprotective strategies in routine oncologic care.

Acknowledgements

This article is partially based on the K.P.’s thesis entitled ‘Cardiac Radiation Dosimetry using a 3D conformal technique for adjuvant breast cancer radiotherapy at the Steve Biko Academic Hospital’ towards the degree of MMed in the Department of Radiation Oncology, University of Pretoria, South Africa on 13 April 2018 with supervisors Prof. R. Lakier and Prof. A. Sarkin. It is available from the corresponding author, K.P., upon request.

The authors declare that they have no financial or personal relationships that may have inappropriately influenced them in writing this article. S.B. serves as an editorial board member of SA Journal of Oncology. The peer review process for this submission was handled independently, and S.B. had no involvement in the editorial decision-making process for this manuscript. S.B. has no other competing interests to declare.

K.P. was involved with protocol design, approval, data collection and article development. S.B. was involved in secondary data analysis and editing. R.L. was involved in the original conceptualisation and execution of the research protocol as part of an MMed degree. R.L. served as a supervisor for the project, providing oversight and assisting with the interpretation and write up of the thesis used as a basis for this article. A.S. was a co-supervisor of the research project.

This research received no specific grant from any funding agency in the public, commercial or not-for-profit sectors.

The data that support the findings of this study are not openly available due to privacy and ethical restrictions, and are available from the corresponding author, K.P., upon request.

The views and opinions expressed in this article are those of the authors and are the product of professional research. The article does not necessarily reflect the official policy or position of any affiliated institution, funder, agency or that of the publisher. The authors are responsible for this article’s results, findings and content.

References