Initial clinical experience of postmastectomy intensity modulated proton therapy in patients with breast expanders with metallic ports

    Initial Clinical Experience of Postmastectomy Intensity Modulated Proton Therapy in Patients with Breast Expanders with Metallic Ports Robert W. Mutter, Nicholas B. Remmes, Mohamed MH Kahila, Kathy A. Hoeft, Deanna H. Pafundi, Yan Zhang, Kimberly S. Corbin, Sean S. Park, Elizabeth S. Yan, Valerie Lemaine, Judy C. Boughey, Chris J. Beltran PII: DOI: Reference:

S1879-8500(16)30299-5 doi: 10.1016/j.prro.2016.12.002 PRRO 712

To appear in:

Practical Radiation Oncology

Received date: Revised date: Accepted date:

12 October 2016 7 December 2016 12 December 2016

Please cite this article as: Mutter Robert W., Remmes Nicholas B., Kahila Mohamed MH, Hoeft Kathy A., Pafundi Deanna H., Zhang Yan, Corbin Kimberly S., Park Sean S., Yan Elizabeth S., Lemaine Valerie, Boughey Judy C., Beltran Chris J., Initial Clinical Experience of Postmastectomy Intensity Modulated Proton Therapy in Patients with Breast Expanders with Metallic Ports, Practical Radiation Oncology (2016), doi: 10.1016/j.prro.2016.12.002

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 Initial Clinical Experience of Postmastectomy Intensity Modulated Proton Therapy in Patients with Breast Expanders with Metallic Ports

RI P

Nicholas B. Remmes PhD*1

T

Robert W. Mutter MD*1

Mohamed MH Kahila MBBS1

SC

Kathy A. Hoeft BS1

NU

Deanna H. Pafundi PhD1 Yan Zhang PhD1

MA

Kimberly S. Corbin MD1 Sean S. Park MD, PhD1

ED

Elizabeth S. Yan MD1

PT

Valerie Lemaine MD2

CE

Judy C. Boughey MD3 Chris J. Beltran PhD1

1

AC

*RWM and NBR contributed equally to this work. Departments of Radiation Oncology, 2Division of Plastic and Reconstructive Surgery, 3

Department of Surgery, Mayo Clinic, Rochester, Minnesota. Reprints: Robert Mutter, Department of Radiation Oncology, 200 First St SW, Rochester, MN 55905; Phone: (507)284-8227; Fax: (507)284-0079; [email protected] Running title: Protons and expander reconstruction Key words: Protons, breast, expanders, reconstruction, mastectomy Conflicts of Interests: None Conflicts of Interest Notification There are no actual or potential conflicts of interest

ACCEPTED MANUSCRIPT Abstract

T

Purpose: The feasibility of proton post-mastectomy radiotherapy (PMRT) in patients

RI P

reconstructed with expanders has not been previously reported, limiting treatment options. We analyzed the dosimetric impact of the metallic port contained within expanders on intensity

SC

modulated proton therapy (IMPT) and report our techniques and quality control for treating patients in this setting.

NU

Methods and Materials: 12 patients with the same expander model underwent two-field IMPT

MA

as part of a prospective registry. All planning dosimetry was checked with an in-house graphic processing unit-based Monte-Carlo simulation. Proton ranges through the expander were validated using a sample implant. Dosimetric impact of setup metallic port position uncertainty

ED

was evaluated. Pre and post-treatment photographs were obtained and acute toxicity was

PT

graded using the Common Terminology Criteria for Adverse Events, version 4.0. Results: Nine patients had bilateral skin-sparing mastectomy with bilateral tissue expander

CE

reconstruction and three patients had unilateral skin-sparing mastectomy and reconstruction. The left side was treated in ten patients and the right side in two. Target coverage and normal

AC

tissue dose uncertainties resulting from the expander were small and clinically acceptable. The maximum physician-assessed acute radiation dermatitis was three in one patient, two in five patients and one in six patients. Conclusions: Postmastectomy IMPT in breast cancer patients with expanders is feasible and associated with favorable CTV coverage and normal tissue sparing, even when taking into account treatment uncertainties. Therefore, these patients should be eligible to participate in clinical trials studying the potential role of proton therapy in breast cancer. We caution, however, that institutions should carry out similar analyses of the physical properties and dosimetric impact of the particular expanders used in their practice prior to considering IMPT.

ACCEPTED MANUSCRIPT Introduction Mastectomy is a common surgical procedure for the treatment of breast cancer, the

T

most common cancer amongst women. In patients with positive lymph nodes who undergo

RI P

mastectomy and axillary staging, post-mastectomy radiotherapy (PMRT) to the chest wall and regional lymph nodes reduces the risk of overall recurrence and breast cancer mortality.[1]

SC

Following mastectomy, immediate implant-based breast reconstruction is increasingly performed to reconstruct the breast mound.[2, 3] Breast reconstruction improves body image

NU

and cosmesis, and has other positive effects on psychosocial and emotional measures.[4-6]

MA

The most common approach to implant based reconstruction involves two stages. First, a tissue expander is placed at the time of the mastectomy. The expander is filled with fluid over several weeks to months by placing a needle through the skin to access a metallic port in the expander.

ED

Expansion enables gentle stretching of the skin, creating a pocket for the implant.[7, 8] The

PT

expander is subsequently exchanged for a permanent implant at a second operation several months later or autologous reconstruction can be performed at that time.

CE

Although the benefits of PMRT and breast reconstruction are established, the optimal integration of these therapies remains a subject of debate. Immediate breast reconstruction has

AC

been reported to lead to compromises in target volume coverage and normal tissue avoidance in radiotherapy treatment planning with traditional photon planning techniques, particularly among patients with left-sided tumors and when the internal mammary lymph nodes are targeted.[9, 10] Therefore, proton therapy may be an attractive alternative for radiotherapy delivery in patients with immediate reconstruction as proton therapy reduces the dose to the heart, lung, and uninvolved soft tissues while maintaining clinical target volume coverage.[11, 12] Proton therapy may be particularly useful in patients with bilateral reconstruction, which poses numerous challenges for photon delivery.[13] Early results of a small number of patients with implant reconstruction treated with protons have been reported and appear encouraging [11, 12]. However, despite the common use of two-staged reconstructive approaches in breast

ACCEPTED MANUSCRIPT cancer patients, no patients with expanders treated exclusively with protons have been reported in the literature to date. This is primarily a result of concern about the dosimetric impact of the

T

metallic port built into most expanders.[11] The fixed range of protons creates uncertainty in

RI P

delivered dose within the vicinity of high density materials such as those contained within the expander port, potentially impacting target coverage and normal tissue sparing.

SC

At present, the feasibility of proton therapy in breast cancer patients with expanders remains unknown, limiting reconstructive and radiotherapy treatment options in a substantial

NU

number of women. We carried out extensive analyses to determine the dosimetric impact of

MA

expanders on intensity modulated proton therapy (IMPT) PMRT. Here, we describe the details of this quality assurance process, which we recommend prior to the consideration of scanning beam proton therapy in this patient population. We also report initial planning and acute toxicity

CE

Methods and Materials

PT

ED

results in the first twelve patients treated at our institution with this approach.

Patients and Expanders

AC

All breast expanders were were Allergan Natrelle 133 Tissue Expanders. These breast expanders consist of a silicon rubber wall with a metallic injection port composed of a thin metal shell containing a magnetic core. The expanders are filled with saline through the injection port. In patients who receive neoadjuvant chemotherapy or who do not have indications for chemotherapy, tissue expansion takes place in the weeks prior to the CT simulation, which is scheduled 6 weeks after surgery. In patients receiving adjuvant chemotherapy, expansion may take place during that time. Patients are not simulated until after the final filling and the expander should remain in the same condition from the time of CT simulation until the completion of radiotherapy. For proton therapy, the expander should be full in order to optimize reproducibility.

ACCEPTED MANUSCRIPT

Immobilization and CT simulation

T

Patients are immobilized on a breast board in the supine position with arms up (Figure

RI P

1A). We do have an arm down immobilization available for patients with reduced shoulder range of motion who are unable to adequately abduct the shoulder for the time required to undergo

SC

proton therapy. With that arm down immobilization, a vac lok is used to support the ipsilateral arm in the akimbo position with couch-indexed hand grips and the contralateral arm is at the

NU

patient’s side during imaging and treatment (Figure 1B). However, we favor arms up as this

MA

positioning is suitable for both proton and photon planning and plan comparisons and reduces arm exposure from our two-beam IMPT approach (Supplementary Figure 1, described below. We use an aquaplast mask to reproduce the head and neck position at the time of simulation.

ED

CT simulation is done with 2 mm slices at 120 kVp, most commonly in free breathing. Deep

PT

inspiratory breath may also be used to displace the heart posteriorly and caudally away from the

Treatment planning

CE

CTV.

AC

For patients with new diagnoses of breast cancer and indications for PMRT, the CTV at our institution includes the chest wall, along with all 3 levels of the axilla, the supraclavicular lymph nodes, and the internal mammary lymph nodes. We do not routinely treat the chest wall alone in that setting. The chest wall contour is similar to the borders of the chest wall in the RTOG Breast Cancer Atlas but extends to, but not deeper than the anterior surface of the ribs and intercostal muscles, except in the vicinity of the internal mammary lymph nodes. The CTV excludes the first 3 mm of tissue under the skin. The target volume is similar regardless of whether the expander is located in a subpectoral or subcutaneous location. The 3 levels of the axilla generally follow the RTOG-endorsed consensus guidelines with a few exceptions. Caudally, level 2 and 3 extends to the obliteration of the fat space between the pectoralis major

ACCEPTED MANUSCRIPT and minor muscles and the cranial border of level II extends to the pectoralis minor muscle insertion on the coracoid.[14] In addition, both the medial and lateral supraclavicular lymph

T

nodes are included, similar to the Radiotherapy Comparative Effectiveness (RADCOMP)

RI P

Consortium Trial atlas.[15-17] The medial extent of the supraclavicular CTV extends no further than the medial border of the common carotid artery and does not abut the esophagus.[15,

SC

17]Finally, for elective treatments, the internal mammary lymph nodes are defined as a 4 mm medial and lateral expansion of the internal mammary vessels and extend from the cranial CT-

NU

slice of the 4th rib to the most caudal extent of the supraclavicular volume at the junction of the

MA

internal mammary vein with the brachiocephalic vein based on data from a lymph node mapping study we previously performed (Mutter et al. IJROBP in press).[14] In addition, based on this data, consideration is given to extending the CTV caudally to include the 4th intercostal space in

ED

patients with known internal mammary nodal disease at presentation elsewhere in the internal

PT

mammary chain, provided that normal tissue constraints can be achieved. The most commonly used beam arrangement is two multi-field optimized anterior-

CE

oblique beams arranged at 45 degrees. Multifield optimization enables the use of robust optimization tools in the planning system (Supplementary Figure 2). All doses are prescribed

AC

and reported in Gy RBE (1.1 x physical dose). The prescription is 50 Gy (RBE) in 25 fractions to the entire CTV. Target and normal tissue dose volume objectives are described in table 1. For treatment planning, setup uncertainty analyses simulating worst case scenarios of +/- 5 mm shifts in isocenter along each translation axis and +/- 3% beam range uncertainty are performed on the CTV and organs at risk. Planning objectives for coverage of the CTV (minimum D90% ≥ 45 Gy [RBE]) must be met in plans recalculated for each of these 8 setup and beam range uncertainty analyses. We do not routinely check simultaneous shifts in isocenter and range uncertainty in our daily practice as these increase computational time and case studies simulating simutaneous +/- 3 mm shifts in isocenter and +/-2% beam range uncertainty (which are more reflective of our day to day setup and calculated beam range uncertainties) showed

ACCEPTED MANUSCRIPT that the 5 mm shifts are more conservative.[18] In order to further evaluate the effects of setup uncertainty, rotation, and interfraction volume deformation verification CT scans are acquired

T

after the first week of treatment and at mid-treatment. For each verification CT scan, the proton

RI P

plan is cast onto the verification CT scan, replicating daily treatment setup by matching to sternum and anterior chest wall. After the heart, lung, and CTV contours are verified, the plan is

SC

recalculated using the CT dataset from the verification simulation and the resulting plan is then evaluated to determine, at the physician’s discretion, if changes in the patient’s anatomy warrant

NU

re-planning.

MA

All patients are treated with multifield optimized pencil-beam scanning IMPT on a Hitachi PROBEAT-V proton therapy system (Hitachi, Tokyo, Japan). The maximum treatment field size is 30 cm x 40 cm. The spot size in air at the treatment beam energy varies depending on range

ED

shifter configuration, but is generally between about 5 mm to 10 mm (1-sigma) with spot

PT

symmetry better than 3% (major-to-minor). Spot position accuracy is approximately 1 mm. Energy layers are between 1 and 3 mm, and spot spacing is between 5 mm and 10 mm,

CE

depending on depth. Each field of a breast treatment is typically delivered in 2 to 4 minutes. Plans are constructed in the Eclipse (Varian Medical Systems, Palo Alto, CA) planning system

AC

using multi-field optimization. In order to treat the shallow depths required, a range shifter is used with a 4.5 cm water equivalent thickness. Treatment planning dosimetry for all plans is checked by an in-house GPU-based Monte Carlo system.[19] Surface dose was validated in a test case on a phantom with Markus chamber measurements and matched the Monte Carlo within 1.5%.

Expander analysis Prior to considering IMPT for patients with expanders, a sample implant was disassembled and the water-equivalent-thickness (WET) of its components were measured with a narrow 228.8 MeV pencil beam (figure 1 A-B). The magnet in the expander is thick enough

ACCEPTED MANUSCRIPT (approximately 2.4 mm) to produce substantial artifacts on a CT image. Normally, the artifacts produced by such metal implants would be contraindications for proton therapy due to the range

T

uncertainty they create. Fortunately, most of the artifact is in the immediate vicinity of the metal

RI P

and within the saline solution of the expander, rather than within target tissue. Based on our measurements, the expander wall and saline are close enough to water-equivalent material to

SC

be taken as water (HU=0) with negligible range uncertainty. The titanium casing of the injection port is also thin enough that setting it as water equivalent results in only about a 0.5 mm range

NU

difference. The magnet, however, has a WET of 13.2 mm, or a relative stopping power of 5.5,

MA

consistent with iron, and must be separately accounted for in treatment planning. The measurements of the expander were used to construct a contouring template in Eclipse. After matching a template to the magnetic and titanium casing visualized on the treatment planning

ED

images (figure 1C), the magnet template contour with stopping power override is copied onto

PT

the treatment planning image set. The rest of the expander is contoured and set to a relative stopping power of 1 (figure 1 D). The artifact outside of the expander is contoured and set to a

CE

relative stopping power representative of neighboring tissue (slightly less than 1). We estimate this keeps the range uncertainty through the expander to approximately 1mm, well under the

checks.

AC

3% additional margin we typically add to the beam range as part of our routine robustness

Port displacement relative to CTV is an additional concern. The diameter of the magnetic core of the port is 1.8 cm. In our photon and proton practice where immobilization and daily image-guidance localization to the chest wall bony anatomy are performed, >95% of fractions the expander port location is within 1 cm of its location on the planning CT. Therefore, to assess the impact of the port on CTV coverage in a hypothetical worst case scenario, we moved the port in the planning system 1 cm in various directions and assessed CTV coverage with the port in that location for all 25 fractions. In addition, to assess the impact of the port on dose to the

ACCEPTED MANUSCRIPT normal tissue under an additional worst-case scenario, dosimetry was calculated with a relative

T

stopping power override of 1 for the port.

RI P

Treatment Delivery

All patients are aligned daily using a kV 2D/3D imaging system and a robotic couch

SC

allowing for 6 degrees-of-freedom (6DoF) matching and positioning for each fraction. The sternum and chest wall (i.e. the anterior and lateral portions of the ribs within the CTV) are

NU

matched to better than 2 mm. In addition, the humeral head, clavicle, acromioclavicular joint and

MA

lower cervical and upper thoracic spine are assessed. Repositioning of the patients is considered if any of these areas are off by > 3 mm after matching to the sternum and chest wall. We prioritize matching to the sternum and chest wall over the metallic port. However, the

ED

location of the metallic port is assessed and gross displacements (i.e. > 1 cm) prompt

PT

repositioning. Optical surface imaging with a 3-camera AlignRT (VisionRT) system is used to monitor intrafraction motion (including respiratory motion) and to verify surface positioning after

CE

alignment to bony anatomy. Chest wall respiratory motion is small relative to isocenter shifts used in our routine robustness analyses.[18] Deviations greater than 5 mm in any location

AC

(matching the +/- 5 mm shifts in isocenter along each translation axis used in setup uncertainty analyses) trigger repositioning of the patient prior to continuation of treatment. One patient was treated in deep inspiratory breath hold in order to displace the heart away from the chest wall and internal mammary lymph node CTV using a combination of marker based tracking with visual feedback to the patient to set the gate, and optical surface imaging for monitoring consistency.

Toxicity Assessment Acute toxicity was graded using the Common Terminology Criteria for Adverse Events, version 4. Patients were also consented to have photographs taken to document skin toxicity.

ACCEPTED MANUSCRIPT

T

Results

RI P

Patients

Twelve patients with tissue expanders were treated with adjuvant proton PMRT between

SC

June 2015 and May 2016. The median patient age was 46 (range 32-63). Nine patients had bilateral skin-sparing mastectomy with bilateral reconstruction and three patients had unilateral

NU

skin-sparing mastectomy and reconstruction. In eight patients the treated expander was placed

MA

subcutaneously (anterior to the pectoralis muscle) and in four patients the expander was subpectoral. The left side was treated in ten patients and the right side in two. All patients received 50 Gy (RBE) in 25 fractions to the entire CTV. Three patients had non-biopsy proven

ED

suspicious appearing internal mammary lymphadenopathy at presentation but had excellent

PT

pathologic responses to preoperative chemotherapy in the breast and axilla and no gross residual disease in the internal mammary chain at simulation. Therefore no nodal boosts were

CE

administered. Five patients were clinical stage two, six patients were clinical stage three, and one patient was clinical stage four due to a suspicious contralateral internal mammary node but

AC

treated with curative intent by including the internal mammary chains bilaterally. Eight of twelve patients received preoperative chemotherapy. No patients received concurrent or postoperative chemotherapy. In ten patients, immobilization was with both arms up (Figure 1A). In two patients, both arms were down with the ipsilateral arm in an akimbo position (Figure 1B).

Dosimetry Average planned target coverage and normal tissue doses for the twelve patients are shown in table 2. In order to assess the potential impact of expander inter-fraction port movement on the dose to the normal tissues under a potential worst case scenario we assessed the dose to the heart and ipsilateral lung if the port was completely removed from the

ACCEPTED MANUSCRIPT plan by considering it water-equivalent material. The average doses to heart (mean 1 Gy [RBE] vs 0.9 Gy [RBE]) and ipsilateral lung (V20 Gy [RBE] 13.5% vs 13.3%) were slightly higher with

T

the port HU set to 0, compared to the base plan (figure 2A, 2B, 2C). These absolute differences

RI P

were small relative to setup and range uncertainty analyses (table 3) and lower than that achieved with photon plans in our practice (data not shown). Similarly, figure 2D, 2E, and 2F

SC

demonstrates clinically acceptable interfraction variation in CTV coverage even with a large, hypothetical 1 cm port displacement in the superior direction. As shown in figure 2E, a

NU

significant portion of the very small cold spot produced by this scenario of a 1 cm port

MA

displacement in the same location for all 25 fractions is contained within the expander itself. The CTV D99% (99% of the CTV volume receives this dose or more) remained a clinically acceptable 46.5 Gy [RBE], compared with 48.0 Gy [RBE] for the base plan. Finally, to assess

ED

the impact of interfraction volume deformation (along with setup, respiratory, and rotational

PT

uncertainty) we evaluated the resulting dosimetry when proton treatment plans were cast on verification CT scans acquired during the course of treatment. For the 12 patients, a total of 30

CE

verification CT scans were acquired. The resulting CTV coverage and normal tissue doses (supplementary table 1) again compared favorably with the worst case scenarios from routine

AC

setup and range uncertainty analyses performed as part of our routine treatment plan evaluation process. Just one patient underwent a re-plan. This patient had developed a 25% hot spot in the CTV near the chest wall that was felt to have resulted from reproducible change in tissue expander volume deformation. The hot spot was caudal of the expander port and was not felt to have been related to the port.

Acute toxicity The median follow-up is 10 months. The maximum physician-assessed acute radiation dermatitis was three in one patient, two in five patients and one in six patients. Two patients developed mild symptomatic (i.e. CTCAE grade 2) esophagitis that did not alter their eating or

ACCEPTED MANUSCRIPT swallowing. One patient developed a 1 cm wound over the mastectomy incision 3 months following the completion of radiotherapy in the setting of active tobacco use that was managed

T

conservatively with prophylactic antibiotics and hyperbaric oxygen. An additional patient

RI P

developed cellulitis of the reconstructed breast one month following implant exchange and 8 months after the completion of radiotherapy that resolved following IV antibiotics. There has

SC

been no rib fractures, pericarditis, pneumonitis or reports of chest wall pain. Baseline, end of treatment and short-interval follow-up digital photographs are displayed in two representative

MA

NU

patients in figure 3. There have been no locoregional recurrences.

Discussion

We demonstrated that IMPT is feasible in the setting of immediate subpectoral or

ED

subcutaneous tissue expander reconstruction, and associated with favorable CTV coverage and

PT

normal tissue sparing, even when taking into account setup and range uncertainties. These findings are significant, as there has been a marked increase in the number of women electing

CE

immediate two-stage reconstruction with expanders over time in the United States.[3] In addition, there has been a rise in the use of preoperative chemotherapy, which limits the time

AC

available for tissue expansion and implant exchange prior to the administration of PMRT, resulting in an increased number of patients undergoing irradiation of expanders. Based on our findings, this growing population of patients with expanders, and indications for PMRT, should be eligible to participate in future breast cancer clinical trials studying the potential role of proton therapy. We caution, however, that each institution considering IMPT in patients with expanders should carry out similar analyses of the physical properties of the particular expanders used in their practice as well as the day to day setup uncertainty of the position of the metallic port in order to confidently assess the potential dosimetric impact of expander reconstruction on proton PMRT.

ACCEPTED MANUSCRIPT The analyses reported here were performed using two-field IMPT which in our experience provides more flexibility to reduce hot-spots on the skin if desired and is slightly less

T

sensitive to setup and beam range uncertainty, when accounting for the metallic port in

RI P

expanders, compared with single-field optimized plans.

Macdonald and colleagues previously reported early outcomes of a prospective clinical

SC

trial of passively scattered proton PMRT.[11] Five of the 12 patients treated on that study had undergone immediate reconstruction and had permanent implants at the time of radiotherapy.

NU

Of note, patients with expanders were specifically excluded. The maximum reported skin toxicity

MA

in the entire population was grade 2 in 9 patients (75%) and there was no reported radiation pneumonitis or cardiac toxicity. They concluded that proton PMRT was feasible and suggested that along with selected patients with unfavorable cardiac anatomy, immediate reconstruction

ED

may be a setting in which proton therapy may be warranted. Cuaron and colleagues reported

PT

dosimetry and acute toxicity in a series of breast cancer patients treated with uniform scanning proton therapy. Fourteen patients (47%) treated had undergone mastectomy and implant

CE

reconstruction but none were treated with expanders. The incidence of acute grade 2 dermatitis in the entire population was 71%, 28% of whom had moist desquamation. Bradley and

AC

colleagues recently reported the dosimetric and clinical feasibility of proton therapy in 18 patients with breast cancer undergoing regional nodal irradiation.[20] Two patients in that study had expanders and underwent proton therapy to the regional lymph nodes while the chest wall and expander was treated with matched photon tangents. Grade 3 dermatitis was reported in 4 of 18 (22%) patients. Although the same level of skin constraining likely would not be possible, further investigation is needed to determine the feasibility of comprehensive passively scattered proton PMRT in patients with expanders. Although our sample number is small, a limitation of this study, the skin toxicity observed in patients with expanders treated with post-mastectomy IMPT appears promising. The low skin toxicity is likely due to application of a skin constraint to limit hot spots at the skin surface while

ACCEPTED MANUSCRIPT attempting to maintain the chest wall skin D90 ≥ 90% using pencil beam scanning technology delivered with two oblique beams, a skin dose expected to control microsocopic disease. In

T

addition, six patients applied Safetac-based Mepitel Film prophylactically as part of a pilot study

RI P

at our institution, which has previously been shown to be effective at reducing moist desquamation in breast cancer patients after lumpectomy or mastectomy without

SC

reconstruction.[21] The other 6 patients received standard skin care including moisturizing cream twice daily. Long-term follow-up of a larger number of patients will be required to

NU

determine the impact of IMPT on cosmetic outcomes, rates of implant loss and other

MA

complications. Of note, patients with implant-based reconstruction are a population known to be at high risk of radiation-induced toxicity with conventional photon PMRT techniques.[22] Two of twelve patients early in our experience developed mild symptomatic esophagitis

ED

that resolved shortly after the completion of radiotherapy. Others have reported esophagitis

PT

following proton and photon therapy for breast cancer.[12, 23] In patients with indications for regional nodal irradiation treated with either photon and proton therapy, careful attention to the

CE

medial border of the supraclavicular CTV is important. Physicians must balance adequate coverage of the areas at highest risk of harboring microscopic disease with the potential risks of

AC

acute esophagitis and late secondary esophageal malignancy.[24] In patients treated with photon therapy, low rates of regional failure have been reported when the medial supraclavicular field edge border was placed at the anterior edge of the sternocleidomastoid muscle.[25] In addition, supraclavicular lymph node mapping studies performed to provide guidance for CTV delineation in the more modern era of volume-based planning have demonstrated that disease medial of the internal jugular vein is unusual.[15, 17] Therefore, the para-tracheal, anterior jugular, pre-laryngeal and pre-tracheal lymph nodes (i.e. level VI of the neck, as defined by Grégoire and colleagues) are not targets in women with breast cancer undergoing elective nodal irradiation and the CTV should not abut the esophagus and trachea.[26] Based on this data, in the absence of gross disease near the internal jugular vein

ACCEPTED MANUSCRIPT and common carotid artery, at a minimum the supraclavicular CTV should not extend medial to the common carotid artery. Furthermore, in women without gross supraclavicular disease, we

T

consider limiting the medial border of the supraclavicular CTV to the lateral edge of the internal

RI P

carotid artery or medial edge of the internal jugular vein, as has been suggested by others, in order to minimize unnecessary exposure to the esophagus and other central structures. When

SC

possible, we also attempt to limit the esophagus D0.01cc to less than 36 Gy [RBE].[27, 28] In conclusion, post-mastectomy IMPT is feasible in breast cancer patients who undergo

NU

immediate reconstruction and have expanders at the time of radiotherapy planning. IMPT may

MA

offer advantages in terms of target volume coverage and normal tissue sparing and should be the subject of further clinical study. Prior to treating patients we caution that a similarly rigorous dosimetric analysis should be performed by the treating institution on any expander that will be

AC

CE

PT

ED

treated.

ACCEPTED MANUSCRIPT

1.

9.

10.

11.

12.

13.

14.

15. 16.

SC

NU

MA

7. 8.

ED

6.

PT

5.

CE

4.

AC

3.

RI P

T

2.

References McGale, P., et al., Effect of radiotherapy after mastectomy and axillary surgery on 10-year recurrence and 20-year breast cancer mortality: meta-analysis of individual patient data for 8135 women in 22 randomised trials. Lancet, 2014. 383(9935): p. 2127-35. Katipamula, R., et al., Trends in mastectomy rates at the Mayo Clinic Rochester: effect of surgical year and preoperative magnetic resonance imaging. Journal of clinical oncology : official journal of the American Society of Clinical Oncology, 2009. 27(25): p. 4082-8. Jagsi, R., et al., Trends and variation in use of breast reconstruction in patients with breast cancer undergoing mastectomy in the United States. Journal of clinical oncology : official journal of the American Society of Clinical Oncology, 2014. 32(9): p. 919-26. Atisha, D., et al., Prospective analysis of long-term psychosocial outcomes in breast reconstruction: two-year postoperative results from the Michigan Breast Reconstruction Outcomes Study. Annals of surgery, 2008. 247(6): p. 1019-28. Rowland, J.H., et al., Role of breast reconstructive surgery in physical and emotional outcomes among breast cancer survivors. Journal of the National Cancer Institute, 2000. 92(17): p. 1422-9. Al-Ghazal, S.K., L. Fallowfield, and R.W. Blamey, Comparison of psychological aspects and patient satisfaction following breast conserving surgery, simple mastectomy and breast reconstruction. European journal of cancer, 2000. 36(15): p. 1938-43. Thiruchelvam, P.T., et al., Post-mastectomy breast reconstruction. BMJ, 2013. 347: p. f5903. Pusic, A.L. and P.G. Cordeiro, Breast reconstruction with tissue expanders and implants: a practical guide to immediate and delayed reconstruction. Seminars in plastic surgery, 2004. 18(2): p. 71-7. Motwani, S.B., et al., The impact of immediate breast reconstruction on the technical delivery of postmastectomy radiotherapy. International journal of radiation oncology, biology, physics, 2006. 66(1): p. 76-82. Ohri, N., et al., Quantifying the impact of immediate reconstruction in postmastectomy radiation: a large, dose-volume histogram-based analysis. International journal of radiation oncology, biology, physics, 2012. 84(2): p. e153-9. Macdonald, S.M., et al., Proton therapy for breast cancer after mastectomy: early outcomes of a prospective clinical trial. International journal of radiation oncology, biology, physics, 2013. 86(3): p. 484-90. Cuaron, J.J., et al., Early toxicity in patients treated with postoperative proton therapy for locally advanced breast cancer. International journal of radiation oncology, biology, physics, 2015. 92(2): p. 284-91. Jimenez, R.B., et al., Intensity modulated proton therapy for postmastectomy radiation of bilateral implant reconstructed breasts: a treatment planning study. Radiotherapy and oncology : journal of the European Society for Therapeutic Radiology and Oncology, 2013. 107(2): p. 2137. Dijkema, I.M., et al., Loco-regional conformal radiotherapy of the breast: delineation of the regional lymph node clinical target volumes in treatment position. Radiotherapy and oncology : journal of the European Society for Therapeutic Radiology and Oncology, 2004. 71(3): p. 287-95. Brown, L.C., et al., Delineation of Supraclavicular Target Volumes in Breast Cancer Radiation Therapy. International journal of radiation oncology, biology, physics, 2015. 92(3): p. 642-9. Brown, L.C., et al., Delineation of Supraclavicular Target Volumes in Breast Cancer Radiation Therapy. In Reply to Yang and Guo. International journal of radiation oncology, biology, physics, 2015. 93(3): p. 723-4.

ACCEPTED MANUSCRIPT

23. 24.

25.

26.

27.

28.

T

RI P

SC

NU

22.

MA

21.

ED

20.

PT

19.

CE

18.

Jing, H., et al., Mapping Patterns of Ipsilateral Supraclavicular Nodal Metastases in Breast Cancer: Rethinking the Clinical Target Volume for High-risk Patients. International journal of radiation oncology, biology, physics, 2015. 93(2): p. 268-76. Depauw, N., et al., A novel approach to postmastectomy radiation therapy using scanned proton beams. International journal of radiation oncology, biology, physics, 2015. 91(2): p. 427-34. Wan Chan Tseung, H., J. Ma, and C. Beltran, A fast GPU-based Monte Carlo simulation of proton transport with detailed modeling of nonelastic interactions. Medical physics, 2015. 42(6): p. 2967-78. Bradley, J.A., et al., Initial Report of a Prospective Dosimetric and Clinical Feasibility Trial Demonstrates the Potential of Protons to Increase the Therapeutic Ratio in Breast Cancer Compared With Photons. Int J Radiat Oncol Biol Phys, 2016. 95(1): p. 411-21. Herst, P.M., et al., Prophylactic use of Mepitel Film prevents radiation-induced moist desquamation in an intra-patient randomised controlled clinical trial of 78 breast cancer patients. Radiotherapy and oncology : journal of the European Society for Therapeutic Radiology and Oncology, 2014. 110(1): p. 137-43. Momoh, A.O., et al., A systematic review of complications of implant-based breast reconstruction with prereconstruction and postreconstruction radiotherapy. Annals of surgical oncology, 2014. 21(1): p. 118-24. Ma, J., et al., Post mastectomy linac IMRT irradiation of chest wall and regional nodes: dosimetry data and acute toxicities. Radiat Oncol, 2013. 8: p. 81. Clarke, M., et al., Effects of radiotherapy and of differences in the extent of surgery for early breast cancer on local recurrence and 15-year survival: an overview of the randomised trials. Lancet, 2005. 366(9503): p. 2087-106. Nielsen, H.M. and B.V. Offersen, Regional recurrence after adjuvant breast cancer radiotherapy is not due to insufficient target coverage. Radiotherapy and oncology : journal of the European Society for Therapeutic Radiology and Oncology, 2015. 114(1): p. 1-2. Gregoire, V., et al., Delineation of the neck node levels for head and neck tumors: a 2013 update. DAHANCA, EORTC, HKNPCSG, NCIC CTG, NCRI, RTOG, TROG consensus guidelines. Radiotherapy and oncology : journal of the European Society for Therapeutic Radiology and Oncology, 2014. 110(1): p. 172-81. Nielsen, M.H., et al., Delineation of target volumes and organs at risk in adjuvant radiotherapy of early breast cancer: national guidelines and contouring atlas by the Danish Breast Cancer Cooperative Group. Acta oncologica, 2013. 52(4): p. 703-10. Verhoeven, K., et al., Vessel based delineation guidelines for the elective lymph node regions in breast cancer radiation therapy - PROCAB guidelines. Radiotherapy and oncology : journal of the European Society for Therapeutic Radiology and Oncology, 2015. 114(1): p. 11-6.

AC

17.

ACCEPTED MANUSCRIPT

Figure Legends

T

Figure 1. Arm up (A) and arm down (B) patient immobilization for post-mastectomy IMPT.

SC

RI P

Figure 2. (A and B) Template of expander port constructed from measurements. (C) Axial image from a planning CT scan at the level of the tissue expander port with the expander port showing the overlay of the expander port template. (D) Planning data set showing overrides of relative stopping power (RSP).

PT

ED

MA

NU

Figure 3. (A) Axial image with color-wash from the planning CT scan at the level of the tissue expander port. Arrows indicate typical beam angles. (B) Colorwash at the same level with the port considered water-equivalent material (HU = 0) showing a potential worst case normal tissue impact with magnet motion. The overridden HU is present for the plan calculation but is not displayed in figure. (C) Dose volume histogram showing the clinical target volume (CTV), ipsilateral lung and heart doses as planned and with the port considered water-equivalent as shown in A and B, respectively, demonstrating minimal differences in dose to the normal tissues under a worst case scenario. (D) Axial image with colorwash from the planning CT scan at the level of the tissue expander port. (E) Colorwash at the same level with the port hypothetically displaced 1 cm in the superior direction. (F) Dose volume histogram showing the CTV, ipsilateral lung and heart doses as planned and with the port hypothetically displaced 1 cm in the superior direction demonstrating a small, clinically acceptable impact on CTV coverage under a worst case scenario.

AC

CE

Figure 4. Digital photographs of two representative patients reconstructed with expanders and treated with IMPT. (A, E) Baseline. (B, F) Last day of RT with mepitel film in place. (C, G) 1 week after RT. (D, H) 2 weeks after RT. Supplementary Figure 1. Axial image with color-wash from the planning CT scan of a patient treated in the arms down position demonstrating exposure of a small volume of the ipsilateral arm. Supplementary Figure 2. Sagittal images with colorwash demonstrating field specific doses for the right anterior oblique (A), left anterior oblique (B), and total dose in a multi-field optimized plan for a patient treated for left breast cancer.

AC

Fig. 1

CE

PT

ED

MA

NU

SC

RI P

T

ACCEPTED MANUSCRIPT

Fig. 2

AC

CE

PT

ED

MA

NU

SC

RI P

T

ACCEPTED MANUSCRIPT

MA

NU

SC

RI P

T

ACCEPTED MANUSCRIPT

AC

CE

PT

ED

Fig. 3

MA

NU

SC

RI P

T

ACCEPTED MANUSCRIPT

AC

CE

PT

ED

Fig. 4

ACCEPTED MANUSCRIPT

Priority 3

MA

NU

SC

RI P

T

Table 1. Target and normal tissue dose volume objectives Target/Organ at Risk Priority 1 Priority 2 CTV D90% ≥ 45 Gy (RBE) D95% ≥ 47.5 Gy (RBE) D0.01cc ≤ 57.5 Gy D0.01cc ≤ 55 Gy (RBE) (RBE) Chest wall skin* D1cc ≤ 52.5 Gy (RBE) D90% ≥ 90% Supraclavicular skin* D1cc ≤ 45 Gy (RBE) Heart mean ≤ 1 Gy (RBE) LAD D0.01cc ≤ 15 Gy (RBE) RCA D0.01cc ≤ 15 Gy (RBE) Ipsilateral lung V20 Gy (RBE) ≤ 20 % Esophagus D0.01cc ≤ 45 Gy (RBE) Brachial Plexus D0.01cc ≤ 54 Gy (RBE) D1cc ≤ 50 Gy (RBE)

Thyroid

AC

CE

PT

ED

Abbreviations: CTV = clinical target volume; D90%, and D95% = 90%, and 95% of the volume receives this dose or more; D0.01cc, and D1cc = 0.01cc, and 1cc of the volume received this dose or more; V20 Gy = the volume receiving 20 Gy or more; LAD = left anterior descending artery (for left-sided cases only); RCA = right coronary artery (for right sided cases only). *Skin is defined as the first 3 mm of tissue under the body surface.

D1cc ≤ 48 Gy (RBE) D1cc ≤ 40 Gy (RBE) mean ≤ 0.5 Gy (RBE) D0.01cc ≤ 3 Gy (RBE) D0.01cc ≤ 3 Gy (RBE) V20 Gy (RBE) ≤ 15 % D0.01cc ≤ 36 Gy (RBE) D0.01cc ≤ 51 Gy (RBE) D1cc ≤ 45 Gy (RBE)

ACCEPTED MANUSCRIPT

AC

CE

PT

ED

MA

NU

SC

RI P

Target/Organ at Risk Average (range) CTV D95% Gy (RBE) 49 (47-50) D0.01 Gy (RBE) 56 (54-58) Heart* Mean Gy (RBE) 0.9 (0.4-2.8) V25 Gy (RBE) (%) 0.8 (0.1-3.3) V5 Gy (%) 6 (2-15) LAD/RCA** Mean Gy (RBE) 5 (0.4-19) V25 Gy (RBE) (%) 3 (0-39) D0.01cc Gy (RBE) 17 (3.7-28) Ipsilateral lung Mean Gy (RBE) 7 (6-9) V20 Gy (RBE) (%) 13 (10-16) V5 Gy (RBE) (%) 38 (34-48) Chest wall skin*** D1cc Gy (RBE) 51 (48-52) Supraclavicular skin*** D1cc Gy (RBE) 45 (39-49) Esophagus D0.01cc Gy (RBE) 42 (35-48) Abbreviations: CTV = clinical target volume; D95% = 95% of the volume receives this dose or more; V25 Gy, V5 Gy, V20 Gy = the volume receiving 25 Gy, 5 Gy, and 20 Gy or more, respectively; D0.01 cc and D1cc = 0.01cc and 1cc of the volume received this dose or more; LAD = Left anterior descending coronary artery. RCA = Right coronary artery *Normal tissue doses include one patient with a contralateral internal mammary chain node who underwent bilateral internal mammary chain irradiation. **Dose to the LAD is reported for 10 left-sided patients and dose to the RCA is reported for two right-sided patients. ***Skin is defined as the first 3 mm of tissue under the body surface.

T

Table 2. Summary of selected dosimetry values obtained

Table 3. Setup uncertainty analyses of 5 mm perturbations along each translation axis and 3% beam range uncertainty on the CTV and organs at risk were performed for each patient. The average of the worst of the eight scenarios across all 12 individual patients for each dose volume parameter is reported.

AC

CE

PT

ED

MA

NU

SC

RI P

Target/Organ at Risk Worst Case Average (Range) CTV D95% Gy (RBE) 47 (44-48) D0.01cc Gy (RBE) 60 (58-65) Heart* Mean Gy (RBE) 1.2 (0.6-3.8) V25 Gy (RBE) (%) 1.5 (0.3-4.9) V5 Gy (RBE) (%) 11 (3-28) LAD/RCA** Mean Gy (RBE) 7 (3-12) V25 Gy (RBE) (%) 6 (0-10) D0.01cc Gy (RBE) 26 (12-50) Ipsilateral lung V20 Gy (RBE) (%) 19 (16-25) V5 Gy (RBE) (%) 43 (39-54) Chest wall skin*** D1cc Gy (RBE) 55 (52-58) Abbreviations: CTV = clinical target volume; D95% = 95% of the volume receives this dose or more; V25 Gy, V5 Gy, V20 Gy = the volume receiving 25 Gy, 5 Gy, and 20 Gy or more, respectively; D1cc = 1cc of the volume received this dose or more. *Normal tissue doses include one patient with a contralateral internal mammary chain node who underwent bilateral internal mammary chain irradiation. **Dose to the LAD is reported for 10 left-sided patients and dose to the RCA is reported for two right-sided patients. ***Skin is defined as the first 3 mm of tissue under the body surface.

T

ACCEPTED MANUSCRIPT

ACCEPTED MANUSCRIPT Supplementary table. Summary of selected dosimetry values obtained from CT verification scans Average (range)

RI P

T

48 (44-50) 59 (55-68)

NU

8 (5-12) 15 (7-24) 39 (27-52)

SC

1.3 (0.3-4.1) 1.3 (0-5.8) 7 (2-19)

MA

Target/Organ at Risk CTV D95% Gy (RBE) D0.01 Gy (RBE) Heart* Mean Gy (RBE) V25 Gy (RBE) (%) V5 Gy (%) Ipsilateral lung Mean Gy (RBE) V20 Gy (RBE) (%) V5 Gy (RBE) (%)

AC

CE

PT

ED

Abbreviations: CTV = clinical target volume; D95% = 95% of the volume receives this dose or more; V25 Gy, V5 Gy, V20 Gy = the volume receiving 25 Gy, 5 Gy, and 20 Gy or more, respectively; D0.01 cc and D1cc = 0.01cc and 1cc of the volume received this dose or more;*Normal tissue doses include one patient with a contralateral internal mammary chain node who underwent bilateral internal mammary chain irradiation.