A Systematic Review of Heart Dose in Breast Radiotherapy

A Systematic Review of Heart Dose in Breast Radiotherapy

Accepted Manuscript A systematic review of heart dose in breast radiotherapy Leah Drost, BSc(C), Caitlin Yee, BSc(C), Henry Lam, MLS, Liying Zhang, Ph...

1023KB Sizes 0 Downloads 39 Views

Accepted Manuscript A systematic review of heart dose in breast radiotherapy Leah Drost, BSc(C), Caitlin Yee, BSc(C), Henry Lam, MLS, Liying Zhang, PhD, Matt Wronski, PhD, Claire McCann, PhD, Justin Lee, MD, Danny Vesprini, MD, Eric Leung, MD, Edward Chow, MBBS PII:

S1526-8209(18)30177-0

DOI:

10.1016/j.clbc.2018.05.010

Reference:

CLBC 815

To appear in:

Clinical Breast Cancer

Received Date: 19 March 2018 Revised Date:

27 May 2018

Accepted Date: 30 May 2018

Please cite this article as: Drost L, Yee C, Lam H, Zhang L, Wronski M, McCann C, Lee J, Vesprini D, Leung E, Chow E, A systematic review of heart dose in breast radiotherapy, Clinical Breast Cancer (2018), doi: 10.1016/j.clbc.2018.05.010. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT

A systematic review of heart dose in breast radiotherapy Leah Drost BSc(C), Caitlin Yee BSc(C), Henry Lam MLS, Liying Zhang PhD, Matt Wronski PhD, Claire McCann PhD, Justin Lee MD, Danny Vesprini MD, Eric Leung MD, Edward Chow MBBS

M AN U

Correspondence: Dr. Edward Chow MBBS, MSc, PhD, FRCPC Department of Radiation Oncology Odette Cancer Centre, Sunnybrook Health Sciences Centre 2075 Bayview Avenue Toronto, Ontario Canada M4N 3M5 Phone: 416-480-4974 Fax: 416-480-6002 Email: [email protected]

SC

Running title: Review of heart dose in breast radiotherapy

RI PT

Department of Radiation Oncology, Odette Cancer Centre, Sunnybrook Health Sciences Centre, University of Toronto, Canada

Conflict of interest: None

AC C

EP

TE D

Acknowledgements: We thank the generous support of Bratty Family Fund, Michael and Karyn Goldstein Cancer Research Fund, Joey and Mary Furfari Cancer Research Fund, Pulenzas Cancer Research Fund, Joseph and Silvana Melara Cancer Research Fund, and Ofelia Cancer Research Fund.

ACCEPTED MANUSCRIPT May 26, 2018 A systematic review of heart dose in breast radiotherapy Conflict of Interest

AC C

EP

TE D

M AN U

SC

RI PT

None to declare.

1

ACCEPTED MANUSCRIPT May 26, 2018 Abstract Radiotherapy (RT) for breast cancer improves survival, but poses risk to the heart resulting from a linear relationship between RT dose and heart disease. This review presents studies worldwide

RI PT

reporting heart doses from whole breast RT after 2014 to update a previous systematic review (Taylor et al, Int J Radiat Oncol Biol Phys, 2015) in order to determine patterns of current heart dosimetry among varying RT regimens. Studies published between January 2014 and September

SC

2017 were included if they reported whole heart dose based on whole breast RT technique or treatment position and had a sample size of > 20 patients. Studies reporting brachytherapy,

M AN U

proton RT only, or boost to tumour bed were excluded. Among 99 studies, whole heart dose was reported by 231 regimens. Mean heart dose for left-sided breast cancer, reported by 84 studies (196 regimens), was 3.6 Gy, compared to a review of those previously reported (5.4 Gy). Regimens employing breathing control in any position had a significantly lower mean heart dose

TE D

(1.7 Gy) compared to regimens without breathing control (4.5 Gy) (p<0.0001). Mean heart dose varied significantly between continents (p<0.0001), with heterogeneity reported among countries within Europe (p=0.04) although not within other continents. On average, mean heart dose

EP

steadily decreased between 2014 (4.6 Gy) and 2017 (2.6 Gy) (p=0.003). Other heart dose parameters including mean dose to the left anterior descending artery (LAD) were reported by 80

AC C

left-sided regimens, and the mean LAD dose was 12.4 Gy.

Keywords: breast cancer, radiotherapy, heart dose, review

2

ACCEPTED MANUSCRIPT May 26, 2018 Introduction Many patients with breast cancer receive adjuvant radiotherapy (RT) for management of their disease, as RT has been well-documented to reduce rates of disease recurrence and mortality in

RI PT

early stage breast cancer; however, RT for breast cancer does not come without risk. It can include some incidental irradiation of important tissues in proximity to the affected breast,

including the heart (1-5). A previous landmark study by Darby et al. (1) found a dose-response

SC

relationship where an increased mean dose of RT to the heart correlated linearly with an

increased risk of a future adverse cardiac event, with no apparent threshold identified in this

M AN U

relationship.

There is evidence that administration of specific RT techniques correlates with varying heart doses: the use of breathing control and prone positioning, for example, are associated with lower heart doses and therefore less cardiac toxicity on average (4,6), whereas the use of

TE D

intensity-modulated RT (IMRT) has been correlated with a higher overall heart dose in recent studies (3,4). A wide range of heart doses has been reported in the last several decades. One systematic review by Taylor et al (5) remarked that cardiac doses have drastically varied in the

EP

past depending on specific RT techniques. Their review examining 149 studies reporting heart dose between 2003 and 2013 found a mean heart dose of 5.4 Gy among left-sided regimens.

AC C

However, there have been several advancements in RT techniques and heart contouring since this previous systematic review was published, potentially contributing to lower mean heart doses; as such, it was expected that the mean heart dose found in this review would be lower than that published earlier (5).

3

ACCEPTED MANUSCRIPT May 26, 2018 The purpose of the present systematic review was to document all studies reporting heart dose from whole breast RT between January 2014 and September 2017, in order to determine mean

RI PT

heart dose among varying whole breast RT regimens. Methods Search strategy

SC

Cochrane Central, Embase, and Medline databases were searched using the following methods: [breast cancer term] and [radiotherapy term] and [radiation injuries] and [dosimetry term] and

M AN U

[heart term]. Breast cancer terms included ‘breast neoplasm’, ‘breast cancer’, and ‘breast tumor or tumour’. Radiotherapy terms searched were ‘radiotherapy’, ‘irradiation’ and ‘radiation’; ‘radiation injuries’ was also searched. Dosimetry terms included were ‘radiation dosage’, ‘dose’, ‘dosage’, and ‘dosimetry’. Heart terms searched were ‘heart’, and ‘cardio*. The search included

Study selection

TE D

all studies published in the English language.

Studies were identified using the Preferred Reporting Items for Systematic Reviews and Meta-

EP

Analyses (PRISMA) guidelines (7). Studies published after January 1, 2014 until September 2017 and reporting whole heart dose based on RT technique or treatment position, including both

AC C

planned and/or delivered dose, were eligible. The search identified 12,329 studies and 1,196 duplicates were removed. Studies were screened by date, and 8,757 manuscripts were removed based on publication before December 31, 2013 as they would have been included in the previous systematic review by Taylor et al (5). Remaining titles were screened by two members of the study team and 1,410 publications were removed due to no radiation and/or incorrect site of radiation (no breast). Full-text articles were screened for eligibility, and 798 were removed due to failure to meet eligibility criteria. Sixty-nine studies in total were removed due to sample

4

ACCEPTED MANUSCRIPT May 26, 2018 size restriction (<20 patients). In total, 99 studies met inclusion criteria and were analyzed in this study (Figure 1). Eligibility was not affected depending on whether the RT plan was delivered. Studies

RI PT

employing brachytherapy, partial breast irradiation, proton or electron RT only, or boost to the tumour bed alone were excluded. Studies not explicitly reporting whole heart dosimetry were excluded, as were studies relying on models rather than patient contours. Only full-length articles

SC

were eligible for analysis. Data Collection and Analysis

M AN U

The following variables were collected from each regimen in the 99 studies: number of patients, RT technique, whole heart dose, treatment position, use of breathing control, laterality, publication year, study year range, country of publication, and target dose and fractionation. Other heart dose parameters including mean dose to the left anterior descending artery (LAD) as

TE D

well as dose-volume measurements including V5 (percentage of the heart volume receiving 5 Gy or more), V10, V20, V30 and V40 were recorded. In order to compare our results with the previous systematic review reporting on heart

EP

doses until 2013 (5), we calculated and used the arithmetic mean of all whole heart doses (henceforth referred to as ‘mean heart dose’) reported by each regimen in each study analyzed.

AC C

Descriptive statistical analyses including mean and range were employed to calculate mean heart dose among all studies, after stratification based on laterality of RT, year and country of publication, and treatment setup positions and techniques; the same statistics were employed to determine the mean LAD dose and averages among other dose-volume measurements. To compare our results with the previous systematic review that conducted most analyses using leftsided regimens only (5), the majority of our analyses to determine mean heart dose based on

5

ACCEPTED MANUSCRIPT May 26, 2018 various parameters (patient set-up and treatment technique, country and year of publication) were also conducted using only studies reporting left-sided regimens. To determine significant differences between varying treatment set-ups and positions, an

RI PT

Analysis of Variance model was used. Chi-squared test or Fisher exact test were used as

appropriate to determine heterogeneity among heart doses reported by varying countries and continents of publication. To determine trends in mean heart dose by year of publication, a linear

SC

regression analysis was employed to calculate slope and p-value. A p-value of <0.05 was considered significant for all analyses.

M AN U

Results

Among the 99 studies evaluated in this review, whole heart doses were reported in 231 regimens. A complete list of details as well as references for all studies can be found in Appendix A and Appendix B, respectively. Whole heart dose for left-sided breast cancer was reported by 84

TE D

studies (196 regimens). Left-sided regimens had a mean heart dose of 3.6 Gy. Thirteen studies reported left- and right-sided regimens with the results from each combined, with a mean heart dose of 2.8 Gy. Mean heart dose for right-sided regimens was 1.9 Gy. The mean heart dose

EP

among all regimens was 3.4 Gy (Table 1). Patient set-up for left-sided regimens

AC C

A variety of treatment set-ups were reported among studies describing left-sided regimens, including prone, supine and lateral decubitus. Breathing control was used in 65 regimens in total, resulting in a mean heart dose of 1.7 Gy; the 128 regimens not using breathing control reported an overall mean heart dose of 4.5 Gy (p<0.0001). Three regimens specified the use of breathing control for some patients, but did not specify which patients used breathing control; these studies were not included in the analysis of breathing control versus no breathing control. Regimens

6

ACCEPTED MANUSCRIPT May 26, 2018 using supine position with and without breathing control had a mean heart dose of 1.7 Gy and 4.7 Gy, respectively (p<0.0001). Comparing all treatment positions with or without breathing control, regimens employing the supine position had a mean heart dose of 3.7 Gy; those using

RI PT

prone position reported a mean heart dose of 2.3 Gy; and among the regimens employing the lateral decubitus position, the mean heart dose was 1.0 Gy. The differences between set-up positions were not statistically significant (p=0.07) (Table 2a).

SC

Treatment regimens that employed the use of tangent beams alone had a lower mean heart dose of 2.6 Gy among left-sided regimens when compared to regimens including dose to

M AN U

nodal areas (3.1 Gy) or additional boosted dose (4.8 Gy). Additionally, regimens employing IMRT or volumetric modulated arc therapy (VMAT) had higher mean heart doses on average (4.5 Gy and 5.0 Gy, respectively) (Table 2b). Country of publication

TE D

Studies from four continents were published: Europe, Australia, North America, and Asia; regimens from various countries among these were identified (Table 3). For this analysis, only whole heart doses from left-sided RT in the supine position without breath control were

EP

considered as this was the most common combination and was what the previous systematic review (5) published. The average for all 119 regimens in this category was 4.7 Gy. Mean heart

AC C

dose between the four continents varied significantly (p<0.0001). Among all 33 regimens reported from European studies, the mean heart dose was 3.8 Gy. There was significant heterogeneity within Europe, between individual countries (p=0.04). Among the 10 regimens published in Australia, the mean heart dose was 2.8 Gy. Overall, regimens in North America reported a mean heart dose of 2.9 Gy. Asian countries overall reported a mean heart dose of 6.2 Gy among 43 regimens.

7

ACCEPTED MANUSCRIPT May 26, 2018 Year of publication Mean heart dose in all left-sided regimens significantly decreased with each subsequent year of publication (slope= -0.5884; p=0.003). In 2014, mean heart dose among the 36 regimens was 4.6

RI PT

Gy. In 2015, mean heart dose decreased from the year before to 3.9 Gy among 64 regimens; in 2016 it decreased further to 3.4 Gy among 46 regimens. Finally, in 2017, mean heart dose fell to 2.6 Gy in the 50 regimens reported until September 2017 (Table 4).

SC

Other heart dose parameters

A number of left-sided studies reported varying dose-volume measurements and mean dose to

M AN U

the LAD. LAD mean dose was reported by 80 regimens, with an average dose of 12.4 Gy. 53 regimens reported V5, with an average of 15.8%. Among the 48 regimens reporting V10, the average was 7.5%. 52 regimens reported V20, with a mean V20 of 5.1%. The average V30

Discussion

TE D

among 66 regimens was 3.6%, and the average V40 among 28 regimens was 3.9% (Table 5).

Whole heart dose is an important risk factor for serious cardiac complications, including ischemic heart disease and death from cardiac-related events (1-5). This review presents all heart

EP

doses reported after January 1, 2014 and up to September 2017 in order to update the literature

AC C

with a more current view of mean heart dose. Our systematic review of 99 studies found an overall mean heart dose of 3.6 Gy among left-sided regimens, although there was some variation depending on treatment set-up, use of breathing control, country and year of publication. The previous systematic review by Taylor et al. (5) examining studies from 2003 until December 31, 2013 found an overall mean heart dose of 5.4 Gy among all left-sided regimens. The mean heart dose of 3.6 Gy for left-sided regimens found in the present review was substantially lower. Our mean heart dose among right-sided regimens (1.9 Gy) compared to that of their review (3.3

8

ACCEPTED MANUSCRIPT May 26, 2018 Gy) was also lower. Despite the apparent decrease in mean heart dose in both left- and rightsided RT more recently as evidenced by the present review, there are several similar trends between the two reviews. Similar to the previous review, there was significant heterogeneity

RI PT

among countries in Europe where the lowest mean heart dose was 1.8 Gy and the highest was 9.5 Gy. However, in contrast to the previous review, our review found significant variation between the continents (p<0.0001) whereas previously, this heterogeneity was not statistically significant

SC

(p=0.1). In the previous review, two of the lowest mean heart doses were found in studies

employing breathing control in any position (1.3 Gy), and the lateral decubitus position (1.2 Gy).

M AN U

Similarly, the present study also found these to produce the two lowest mean heart doses of 1.7 Gy and 1.0 Gy respectively, although the difference between lateral decubitus and other treatment positions in the present review was not significant (p=0.07). The previous and present systematic reviews confirm results from several studies examining breathing control in breast RT

TE D

(8-12) that have found significantly lower mean heart doses when employing breathing control. Contrary to the previous review that found significantly higher overall mean heart doses in left-sided regimens published more recently (4.5 Gy from 2011-2013 compared to 3.6 Gy

EP

from 2003-2010) (5), our review demonstrated that on average, mean heart dose significantly decreased by year between 2014 and 2017 (p=0.003). A potential explanation for this decrease in

AC C

heart dose could be an improvement in RT technique or heart contouring over the years, but more likely, heart contouring may have improved with adherence to RTOG heart contouring guidelines after the landmark study published by Darby et al. in 2013 that found a linear correlation between whole heart RT dose and ischemic heart disease in breast cancer patients (1). Our finding of a significantly decreasing mean heart dose by year echoes that of a recent study published after the search period for the present review; this large study of 4688 patients (3)

9

ACCEPTED MANUSCRIPT May 26, 2018 demonstrated a significantly decreasing trend in mean heart dose for both left- and right-sided disease from 2012 to 2015. Additionally, in the present review, regimens employing IMRT, VMAT, and RT to nodal areas had higher mean heart doses than those using tangent beams

RI PT

alone. The use of IMRT in particular has recently been noted as a potential contributing factor to higher mean heart doses (3,4).

There is evidence that heart dose parameters other than whole heart dose could have an

SC

effect on future cardiac health. Several studies have found a correlation between RT exposure to the coronary arteries and future coronary artery stenosis (13-15). A study by Correa et al. (13)

M AN U

reported that abnormalities most often occurred in the anterior heart, specifically the LAD, due to its position and the greater risk of inclusion in tangential RT fields. This study also noted that most abnormalities relating to RT-associated coronary damage occurred more than a decade after RT. Due to increasing evidence that the LAD is an important structure in RT-related heart

TE D

abnormalities, many of the regimens included in the present review reported mean LAD dose. Overall, the mean LAD dose (12.4 Gy) observed in this review among 80 left-sided regimens was substantially higher than the mean heart dose (3.6 Gy). Although the difference between

EP

LAD dose and heart dose is striking, this is likely due to the fact that the LAD is located in the anterior region of the heart where it is exposed to the tangential fields used in whole breast RT.

AC C

There is a wide range (1.9 to 40.8 Gy), and therefore considerable heterogeneity in LAD doses reported likely due to its smaller volume and inherent contouring variability; the LAD is a notoriously difficult structure to contour (14). More controlled research is needed to determine actual and acceptable RT dose to this structure. Other heart dose parameters have also been found to be a risk factor for cardiac events; one study showed that the left ventricle V5 was a better predictor of acute cardiac events than mean heart dose (16). More research is needed to

10

ACCEPTED MANUSCRIPT May 26, 2018 determine which indicator of heart dose from breast RT is the most important for determining cardiac toxicity and morbidity. A limitation of the present review was the necessity to exclude any unpublished heart

RI PT

dose; there is a potential, as with any systematic review, for publication bias. Volume of the heart as well as adherence to certain heart contouring guidelines were not reported uniformly by all studies, so the present analyses did not consider these variables which could contribute to

SC

variation in the mean heart dose observed. Additionally, our result of the lowest mean heart dose found in regimens using lateral decubitus is based on only 5 published regimens and therefore

M AN U

statistical significance was unachievable; more studies are needed examining this position and its effect on heart dose before conclusions on its effectiveness to decrease heart dose can be made. We followed the same methodology of the previous systematic review (5), reporting on heart doses using the mean heart dose as reported by each regimen in each study analyzed; for this

TE D

reason, we did not account for individual variation in sample size among the various regimens. Despite these limitations, the present review demonstrates trends similar to the previous review, but with a lower overall mean heart dose in recent years.

EP

Conclusions

The present systematic review found a lower mean heart dose of 3.6 Gy among left-sided

AC C

regimens for breast RT when compared to previously reported mean heart doses. Within the study period of 2014-2017, there was a significant decreasing trend in heart dose. The lowest doses were found in studies employing breathing control.

11

ACCEPTED MANUSCRIPT May 26, 2018 Acknowledgements

AC C

EP

TE D

M AN U

SC

RI PT

We thank the generous support of Bratty Family Fund, Michael and Karyn Goldstein Cancer Research Fund, Joey and Mary Furfari Cancer Research Fund, Pulenzas Cancer Research Fund, Joseph and Silvana Melara Cancer Research Fund, and Ofelia Cancer Research Fund.

12

ACCEPTED MANUSCRIPT May 26, 2018 References 1. Darby SC, Ewertz M, McGale P, et al. Risk of ischemic heart disease in women after radiotherapy for breast cancer. N Engl J Med 2013;368(11):987-98.

RI PT

2. Darby SC, McGale P, Taylor CW, et al. Long-term mortality from heart disease and lung cancer after radiotherapy for early breast cancer: prospective cohort study of about 300,000 women in US SEER cancer registries. Lancet Oncol 2005;6(8):557-65.

SC

3. Pierce LJ, Feng M, Griffith KA, et al. Recent time trends and predictors of heart dose from breast radiotherapy in a large quality consortium of radiation oncology practices. Int

M AN U

J Radiat Oncol Biol Phys 2017 Dec 1;99(5):1154-1161.

4. Hong JC, Rahimy E, Gross CP, et al. Radiation dose and cardiac risk in breast cancer treatment: An analysis of modern radiation therapy including community settings. Pract Radiat Oncol 2017 Jul 20 [cited 2017 Sep 10]. Available from:

print]

TE D

http://www.sciencedirect.com/science/article/pii/S1879850017301996 [Epub ahead of

5. Taylor CW, Wang Z, Macauley E, et al. Exposure of the heart in breast cancer radiation

EP

therapy: A systematic review of heart doses published during 2003 to 2013. Int J Radiat Oncol Biol Phys 2015;93(4):845-53.

AC C

6. Formenti SC, DeWyngaert JK, Jozsef G, Goldberg JD. Prone vs. supine positioning for breast cancer radiotherapy. JAMA 2012;308(9):861-3.

7. Preferred Reporting Items for Systematic Reviews and Meta- Analyses. Website. Available at: http://www.prisma-statement.org/. Accessed Nov 27, 2017.

13

ACCEPTED MANUSCRIPT May 26, 2018 8. Pierce LJ, Feng M, Griffith KA, et al. Recent time trends and predictors of heart dose from breast radiation therapy in a large quality consortium of radiation oncology practices. Int J Radiat Oncol Biol Phys 2017;99(5):1154-1161.

RI PT

9. Lin A, Sharieff W, Juhasz J, Whelan T, Kim D. The benefit of deep inspiration breath hold: evaluating cardiac radiation exposure in patients after mastectomy and after breastconserving surgery. Breast Cancer 2017;24:86-91.

SC

10. McIntosh A, Shoushtari AN, Benedict SH, Read PW, Wijesooriya K. Quantifying the reproducibility of heart position during treatment and corresponding delivered heart dose

M AN U

in voluntary deep inhalation breath hold for left breast cancer patients treated with external beam radiotherapy. Int J Radiat Oncol Biol Phys 2011;81(4):e569-e576. 11. Borst GB, Sonke J, den Hollander S, et al. Clinical results of image-guided deep inspiration breath hold breast irradiation. Int J Radiat Oncol Biol Phys 2010;78(5):1345-

TE D

1351.

12. Stranzl H, Zurl B. Postoperative irradiation of left-sided breast cancer patients and cardiac toxicity. Does deep inspiration breath-hold (DIBH) technique protect the heart?

EP

Strahlenther Onkol 2008;184(7):354-8.

13. Correa CR, Litt HI, Hwang W, Ferrari VA, Solin LJ, Harris EE. Coronary artery findings

AC C

after left-sided compared with right-sided radiation treatment for early-stage breast cancer. J Clin Oncol 2007;21:3031-3037.

14. Duane F, Asnar MC, Bartlett F, et al. A cardiac contouring atlas for radiotherapy. Radiother Oncol 2017;122:416-422.

15. Nilsson G, Holmberg L, Garmo H, et al. Distribution of coronary artery stenosis after radiation for breast cancer. J Clin Oncol 2012;30(4):380-386.

14

ACCEPTED MANUSCRIPT May 26, 2018 16. van den Bogaard VA, Ta BD, van der Schaaf A, et al. Validation and modification of a prediction model for acute cardiac events in patients with breast cancer treated with

AC C

EP

TE D

M AN U

SC

Clin Oncol 2017;35(11):1171-1178.

RI PT

radiotherapy based on three-dimensional dose distributions to cardiac substructures. J

15

ACCEPTED MANUSCRIPT

Table 1. Studies reporting whole heart doses from whole breast RT regimens published between Jan 2014 to September 2017 Mean heart dose** (Gy)

AC C

EP

TE D

M AN U

SC

RI PT

Number of Number of RT Laterality studies regimens Average Range Left sided 84 196 3.6 0.1-18.7 Right sided 20 32 1.9 0.2-8.8 Left and Right combined 13 27 2.8 0.1-9.0 Did not specify 3 8 5.8 1.1-10.4 Total 99* 231 3.4 0.1-18.7 *Some studies reported doses for both left- and right-sided regimens in addition to the overall left- and right-sided combined **Mean heart dose defined as the arithmetic mean of median whole heart doses reported by studies

ACCEPTED MANUSCRIPT

Table 2a. Mean heart doses from left whole breast RT with varying patient setup positions Number of regimens 65

Set-up Breath control

Average

Range

p-value

1.7

0.4-4.8

<0.0001

128

4.5

0.1-18.7

Supine (with breath control)

63

1.7

0.4-4.8

Supine (without breath control)

119

4.7

0.1-18.7

Supine

185

3.7

0.1-18.7

Prone

6

2.3

0.7-4.2

Lateral decubitus

5

1.0

RI PT

No breath control

<0.0001

SC

0.0692

0.5-1.5

M AN U

Overall average is 3.6Gy

Table 2b. Mean heart doses from left whole breast RT with varying RT techniques Number of regimens

Mean

Range

3D conformal

20

3.3

0.5-14

IMRT

36

4.5

0.9-18.7

Tangents alone

82

2.6

0.4-8.7

17

3.1

0.5-7.1

14

4.8

0.24-7.7

8

5.9

1.4-11.8

17

5

0.1-12.4

Tangents plus nodal

Tomotherapy VMAT

EP

Tangents with boost

TE D

Technique

AC C

Other 2 1.35 0.8-1.9 IMRT, intensity modulated radiotherapy; VMAT, volumetric modulated arc therapy Overall average is 3.6Gy

ACCEPTED MANUSCRIPT

Range

3.5 3.1 3.1 4.6 2.9 1.8 3 4.4 3 2.6 9.5 3.6 4.4 1.9 3.8

n/a 2.9-3.3 n/a 2.3-12.9 n/a n/a 1.9-4.5 n/a n/a n/a 5.0-14.0 2.5-4.7 n/a 1.8-2.1 1.8-14.0

2.8

1.8-3.9

2.8 3.2 2.9

1.2-11.9 1.3-8.0 1.2-11.9

6.6 5.1 6.4 4.5 6.2

2.8-11.0 0.1-18.7 4.1-12.9 2.8-6.1 0.1-18.7

AC C

M AN U

SC

Average

TE D

10 p=0.8** 25 8 33 p=0.7** 22 7 12 2 43

EP

Continent EUROPE Denmark Finland France Germany Hungary Ireland Italy Netherlands Norway Poland Spain Sweden Turkey UK All Europe AUSTRALIA Australia NORTH AMERICA USA Canada All North America ASIA China India Korea Taiwan All Asia

Number of regimens p<0.0001* p=0.04** 1 2 1 8 1 1 6 1 1 1 2 4 1 3 33

RI PT

Table 3. Mean heart doses from left whole breast RT in the supine position without breathing control according to country; overall average is 4.7 Gy.

P values demonstrate heterogeneity between and/or within continents *between all continents **within respective continent

ACCEPTED MANUSCRIPT

Table 4. Mean heart doses (Gy) from left whole breast RT by year Number of Average Range p-value regimens 2014 36 4.6 0.24-14.0 0.003* 2015 64 3.9 0.4-18.7 2016 46 3.4 0.7-12.9 2017 50 2.6 0.1-7.1 Total 196 3.6 0.1-18.7 *Slope is -0.5884 (SE=0.1935) P value demonstrates significant reduction in mean heart dose between years

AC C

EP

TE D

M AN U

SC

RI PT

Year

ACCEPTED MANUSCRIPT

80 53 48 52 66 28

12.4 15.8 7.5 5.1 3.6 3.9

1.9-40.8 0.66-90.1 0-48.18 0-23.6 0-20.3 0-13.9

AC C

EP

TE D

M AN U

SC

LAD mean (Gy) V5 (%) V10 (%) V20 (%) V30 (%) V40 (%)

RI PT

Table 5. Other heart dose parameters reported in left-sided regimens Number of Parameter Average Range regimens

ACCEPTED MANUSCRIPT May 17, 2018 Figure 1. PRISMA flow diagram

11,133 publications after duplicates

RI PT

12329 publications identified through database searching

SC

8,757 publications removed after screening for date

798 publications removed because of ineligibility criteria

69 publications removed because of sample size <20

EP

TE D

966 full-text articles assessed for eligibility

M AN U

1,410 publications removed for no breast cancer radiation

AC C

99 studies included in qualitative and quantitative analysis

1