Physics Contribution
Dependence of Coronary 3-Dimensional Dose Maps on Coronary Topologies and Beam Set in Breast Radiation Therapy: A Study Based on CT Angiographies

https://doi.org/10.1016/j.ijrobp.2014.01.055Get rights and content

Purpose

In left-side breast radiation therapy (RT), doses to the left main (LM) and left anterior descending (LAD) coronary arteries are usually assessed after delineation by prior anatomic knowledge on the treatment planning computed tomography (CT) scan. In this study, dose sensitivity due to interindividual coronary topology variation was assessed, and hot spots were located.

Methods and Materials

Twenty-two detailed heart models, created from heart computed tomography angiographies, were fitted into a single representative female thorax. Two breast RT protocols were then simulated into a treatment planning system: the first protocol comprised tangential and tumoral bed beams (TGs_TB) at 50 + 16 Gy, the second protocol added internal mammary chain beams at 50 Gy to TGs_TB (TGs_TB_IMC). For the heart, the LAD, and the LM, several dose indicators were calculated: dose-volume histograms, mean dose (Dmean), minimal dose received by the most irradiated 2% of the volume (D2%), and 3-dimensional (3D) dose maps. Variations of these indicators with anatomies were studied.

Results

For the LM, the intermodel dispersion of Dmean and D2% was 10% and 11%, respectively, with TGs_TB and 40% and 80%, respectively, with TGs_TB_IMC. For the LAD, these dispersions were 19% (Dmean) and 49% (D2%) with TGs_TB and 35% (Dmean) and 76% (D2%) with TGs_TB_IMC. The 3D dose maps revealed that the internal mammary chain beams induced hot spots between 20 and 30 Gy on the LM and the proximal LAD for some coronary topologies. Without IMC beams, hot spots between 5 and 26 Gy are located on the middle and distal LAD.

Conclusions

Coronary dose distributions with hot spot location and dose level can change significantly depending on coronary topology, as highlighted by 3D coronary dose maps. In clinical practice, coronary imaging may be required for a relevant coronary dose assessment, especially in cases of internal mammary chain irradiation.

Introduction

Improvements in radiation therapy (RT) have increased patient life span and highlighted late effects, making cardiovascular disease a major concern 1, 2, 3, in particular after left-sided breast RT. For breast cancer, treatment of the internal mammary chain (IMC) is controversial because of the potential for increased cardiovascular disease risk 4, 5, 6, 7.

With older medical records, retrospective cardiovascular dosimetric studies are based on mathematical phantoms (8) or a computed tomography (CT) scan examination of a representative patient (9) because of the lack of anatomic information. With current medical records, CT allows 3-dimensional (3D) visualization of patient anatomy but does not permit precise visualization of coronary arteries because of limited resolution and contrast (10). In both case, the most exposed coronary segments (ie, the left anterior descending artery [LAD] and the left main coronary artery [LM]) can only be located by prior anatomic knowledge.

Coronary dosimetry is subject to several sources of uncertainty. Among these, anatomy, physiologic motion (breathing and heart beat), patient positioning, beam setup, and the dose calculation tool used should all be taken into account. Each uncertainty must be evaluated independently to assess its significance. In this study, the sensitivity of the coronary dose due to interindividual variation of coronary topology was studied in the context of left-sided breast RT. An approach using hybrid computational phantoms (HCPs) is proposed because they are appropriate for this type of systematic study in which a sole parameter (ie, the coronary topology) is changed. It aims to identify the order of magnitude of coronary dose uncertainties related to coronary topology. In clinical practice, it is of direct interest to assess the relevance of current coronary dosimetry methods. It will also be of interest for retrospective studies because it is among the uncertainties previously described.

Section snippets

Overview of the modeling process

In this study, rather than considering a generic coronary model based on anatomic charts or an atlas (11), different topologies were modeled using heart CT angiographies (CTAs). This imaging precisely displays coronary arteries because of its submillimetric slice thickness, intravenous dye, and cardiac cycle synchronization. Twenty-two heart CTAs of male and female patients treated for a Hodgkin lymphoma and diagnosed with coronary stenoses during their medical follow-up (12) were used. After a

Evaluation of the heart fitting in the representative thorax

The first indicator to assess the fitting of each heart model made from the CTAs on the original heart was the relative difference between their respective volumes. The representative patient had a heart of 461 cm3. The fitted heart models ranged between 440 cm3 and 473 cm3, with a mean of 460 cm3 and a relative SD of 0.5%. Fourteen of the 22 hearts had a volume within ±1% of the original heart.

The second indicator was the comparison of Dmean and D2% between the heart models and the original

Discussion

The small volume difference and small dose indicator dispersion between the original heart of the representative thorax and the inserted heart models ensure correct scaling of the CTA-based heart models into the representative thorax, and thus the final models differ only by their coronary topology. Inclusion of the heart model into the single thorax was convenient with regard to heart location, volume, and doses (Dmean and D2%). P09, identified as an outlier regarding the whole heart Dmean and

Conclusions

We used a systematic modeling process to provide a single representative thorax anatomy with different coronary topologies, focusing on the dose sensitivity to individual LAD topology. Coronary topologies were based on heart CTAs because this allows precise visualization of the coronary arteries.

A significant variation of high doses (D2%) due to interindividual LAD topology variations has been shown. Only the 3D dose map provides a complete understanding of coronary dose sensitivity because of

References (30)

  • E. Konen et al.

    The prevalence and anatomical patterns of intramuscular coronary arteries. A coronary computed tomography angiographic study

    J Am Coll Cardiol

    (2007)
  • E.L. Lorenzen et al.

    Inter-observer variation in delineation of the heart and left anterior descending coronary artery in radiotherapy for breast cancer: A multi-centre study from denmark and the UK

    Radiother Oncol

    (2013)
  • S.B. Evans et al.

    Analysis of coronary artery dosimetry in the 3-dimensional era: Implications for organ-at-risk segmentation and dose tolerances in left-sided tangential breast radiation

    Pract Radiat Oncol

    (2013)
  • Scientific Committee 1-17. Second primary cancers and cardiovascular disease after radiation therapy (NCRP report No 170)

    (2011)
  • S.C. Darby et al.

    Risk of ischemic heart disease in women after radiotherapy for breast cancer

    N Engl J Med

    (2013)
  • Cited by (20)

    • Is radiation-induced arteriopathy in long-term breast cancer survivors an underdiagnosed situation?: Critical and pragmatic review of available literature

      2021, Radiotherapy and Oncology
      Citation Excerpt :

      Initially considered radiation-resistant and possibly sensitive to 30 Gy radiation after lymphoma, the heart clearly exhibits late RI toxicity, which was mostly described after the year 2000 [70–73]. Late coronary RIA is to date well documented in the literature in two situations: in earlier studies after mediastinal “large volume-low dose” RT for lymphoma-seminoma, and more recently “small volume-high dose” after thoracic parietal RT for BC [74,75]. While the risk of death from ischemic heart disease has substantially decreased over time with improvement in RT techniques, patient follow-up over several decades has always shown an increased risk of heart disease [76].

    • Cardiac Structure Doses in Women Irradiated for Breast Cancer in the Past and Their Use in Epidemiological Studies

      2019, Practical Radiation Oncology
      Citation Excerpt :

      Therefore, using these to derive quantitative dose-response relationships would be inappropriate. However, typical CT-scan doses consistently indicated whether a particular regimen typically gave a high, medium, or low dose to a segment (Tables E5 and E6), and our segment dose rankings were consistent with those in other publications.13,27–29 There was also consistency in the locations of hotspots within the ventricular myocardium and main coronary arteries for each regimen (Figs. 4 and 5).

    • Cardiotoxicity associated with radiotherapy in breast cancer: A question-based review with current literatures

      2018, Cancer Treatment Reviews
      Citation Excerpt :

      All the cardiac structures can be targeted by radiation. Although MHD has decreased to less than 5 Gy in the past decade [6,7], several heart substructures still appear to be particularly vulnerable to radiation especially for the left main (LM) and left anterior descending (LAD) coronary artery due to different situations like individual variety of anatomy, setup errors caused by breathing and heartbeat, interobserver variability of heart delineation, use of different dose calculation tools and RT fields including IMC [28,29]. Dosimetric studies have showed the importance of separated dose evaluation of LAD so that it can adjust RT strategy such as tangential angle to reduce the high doses of some hot spots [30,31].

    View all citing articles on Scopus

    Conflict of interest: none.

    View full text