Deep inspiratory breath-hold radiotherapy on a Helical Tomotherapy unit: Workflow and early outcomes in patients with left-sided breast cancer

Highlights

• •DIBH treatments in Helical Tomotherapy units are challenging.
• •A workflow incorporating a frame providing tactile feedback and staggered junctions is feasible.
• •Treatment delivery is monitored and gated manually.

Abstract

Introduction

The clinical implementation of deep inspiratory breath-hold (DIBH) radiotherapy to reduce cardiac exposure in patients with left-sided breast cancer is challenging with helical tomotherapy(HT) and has received little attention. We describe our novel approach to DIBH irradiation in HT using a specially designed frame and manual gating, and compare cardiac substructure doses with the free-breathing (FB) technique.

Material and methods

The workflow incorporates staggered junctions and a frame that provides tactile feedback to the patient and monitoring for manual cut-off. The treatment parameters and clinical outcome of 20 patients with left-sided breast cancer who have undergone DIBH radiotherapy as a part of an ongoing prospective registry are reported. All patients underwent CT scans in Free Breathing (FB) and DIBH using the in-house Respiframe, which incorporates a tactile feedback-based system with an indicator pencil. Plans compared target coverage, cardiac doses, synchronizing treatment with breath-hold and avoiding junction repetition. MVCT scans are used for patient alignment.

Results

The mean dose (Dmean) to the heart was reduced by an average of 34 % in DIBH-HT compared to FB-HT plans (3.8 Gy vs 5.7 Gy). Similarly, 32 % and 67.8 % dose reduction were noted in the maximum dose (D0.02 cc) of the left anterior descending artery, mean 12.3 Gy vs 18.1 Gy, and mean left ventricle V5Gy 13.2 % vs 41.1 %, respectively. The mean treatment duration was 451.5 sec with a median 8 breath-holds; 3 % junction locations between successive breath-holds were replicated. No locoregional or distant recurrences were observed in the 9-month median follow-up.

Conclusion

Our workflow for DIBH with Helical-Tomotherapy addresses patient safety, treatment precision and challenges specific to this treatment unit. The workflow prevents junction issues by varying daily breath-hold durations and avoiding junction locations, providing a practical solution for left-sided breast cancer treatment with HT.

Introduction

Cardiac exposure to radiation increases the risk of cardiovascular disease and cardiac mortality [1]. In a landmark paper by Darby et al., the risk of acute coronary events, comprising coronary revascularisation, myocardial infarction and death, was noted to increase by 7.4 % for every 1 Gy increase in the mean heart dose (MHD) in patients undergoing radiotherapy for breast cancer [2]. In a recent prospectively assessed cohort, the risk of any cardiac event was 17.5 % in left-sided breast cancer patients at a median follow-up of >10 years [3].

Radiotherapy in deep inspiratory breath-hold (DIBH) is considered the standard of care in the treatment of left-sided breast cancer. The manoeuvre leads to a separation between the breast/chest wall target and the heart, in addition to the latter becoming narrower and reduced in volume. This, in turn, reduces the radiation dose received by the heart and various cardiac substructures. This reduction ranges from 26 % to 67 % for the heart and 31 to 71 % for the left anterior territory [4], [5], [6].

Various methods have been described for DIBH - tracking of respiratory excursion and gating using infrared marker devices placed on the patient's chest/abdomen, tracked by a camera [7], [8] or using a valve-based breathing device with a visual guide that helps the patient to reach the index lung volume, i.e. the lung volume attained during the planning CT [9]. In addition to the systems mentioned above that vendors of radiotherapy equipment provide, there are other techniques based variously on imaging the surface of the patient, i.e. surface guidance [10] or a wearable belt or laser tracking of respiratory movement [11], [12].

A non-commercial, voluntary breath-hold system based on using in-room cameras zoomed in to confirm the matching of in-room lasers with skin marks, the beam being interrupted manually was used in the UK Heart Spare trial and has also been described by Conroy et al. [13].

DIBH treatments have not been described for Helical Tomotherapy. We have previously discussed a tactile feedback-based Respiframe (Fig. 1), which may be used to gate radiotherapy in DIBH manually [14]. We describe our in-house developed workflow for a DIBH solution, based on this frame, for use with the Radixact Helical Tomotherapy unit.

Fig. 1.

Respiframe. Respiframe with a pointer to provide tactile feedback to replicate DIBH, (A) Without Breath Hold and movable pencil in expiration, (B) With Breath Hold and movable pencil in DIBH position with a florescent strip. DIBH-Deep Inspiratory Breath-Hold.

Using this technique, we also describe the treatment and dosimetric parameters of DIBH radiotherapy in 20 consecutive patients enrolled in the prospective DIBRAD Registry (IEC App No.APH-C-S-002-10/21).

Material and methods

Twenty consecutive patients with left-sided breast cancer treated with deep inspiratory breath hold technique were analyzed. All patients were treated utilizing the helical tomotherapy technique. Only patients who achieved a breath-holding duration exceeding 30s were included. The patient characteristics including age, comorbidities, location in quadrant, surgery and neo-adjuvant chemotherapy details are shown in Table 1.

Table 1.

Patient Characteristics. BCS-Breast Conservation Surgery, MRM-Modified Radical Mastectomy, AC-Adriamycin Cyclophosphamide, TCHP-Docetaxel, Carboplatin, Trastuzumab, Pertuzumab, Doce-Docetaxel.

S No	Patient characteristic	Number
1	Total Number	20
2	Median Age (range)	55 (24–75)
3	Comorbidities
	Prior Cardiac ailment	1
	Diabetes	7
	Hypertension	9
	Obesity	2
	Multiple	4
4	Left-sided	20
5	Quadrant
	Upper outer	8
	Lower Outer	2
	Lower inner	2
	Upper Inner	5
	Central	3
6	Surgery
	BCS	8
	MRM	12
7	Reconstruction using autologous flaps	4
	Silicon implant	1
8	Any Nodal Irradiation	17
	3 nodal stations treated (internal mammary chain + supraclavicular fossa + axilla)	6
9	Neo-adjuvant Chemotherapy Adjuvant Chemotherapy Hormonal Therapy alone	10 7 2
10	Neo-adjuvant chemotherapy	10
	AC → paclitaxel + carboplatin	1
	AC → paclitaxel	4
	TCHP	5
	Adjuvant Chemotherapy
	AC → paclitaxel	7
	Docetaxel + cyclophosphamide	2
	Docetaxel	3
	AC → Paclitaxel + Carboplatin	1

BCS-Breast Conservation Surgery, MRM-Modified Radical Mastectomy, AC-Adriamycin Cyclophosphamide,

TCHP-Docetaxel, Carboplatin, Trastuzumab, Pertuzumab, Doce-Docetaxel.

Workflow

Breathhold training

Once a patient has been scheduled for radiation treatment, she undergoes breath-hold training in the mould room in the supine position. The patient is familiarised with the Respiframe(C-shaped rectangular frame with pointer) as shown in Fig. 1, and instructed to hold her breath till she encounters the tip of the pointer. This training involves a series of inhalation, breath-holding and exhalation exercises. The patient is instructed to practice these breath-holding exercises twice daily for two days, with an emphasis on achieving a minimum breath-hold duration of 30s. On the third day, she is requested to come to the hospital for an assessment of her breath-holding ability. Radiation planning is initiated if she successfully holds her breath for more than 30s. However, if the patient is unable to meet the 30-second threshold, she is advised to continue practising for an additional two days. In case the patient is still unable to achieve the required breath-hold duration, she is treated in the free breathing mode. In certain patients, breath-hold training has been initiated on the day on which the patient visited the hospital for her final chemotherapy cycle, thus allowing a 2.5 – 3 week period for practising breath-hold. Patients practising breath-hold as a part of their daily yoga / meditation routine have been able to hold breath for longer than 1 min, even at the onset of training.

Simulation

Previously described Respiframe (Fig. 1), a C-shaped rectangular frame, is designed to fit into slots of the appropriate base plate, with a horizontal arm extending parallel to the patient's body. This horizontal bar holds a pointer with a tapered end perpendicular to the bar. Within the pointer is a movable pencil that touches the patient's skin. The far end of this pencil carries a fluorescent/coloured strip that becomes visible during DIBH. In a modification from the previous design, we have also used a frame wherein the movable pencil makes contact with a superiorly placed metallic connector that activates a battery-operated lighting system on the horizontal bar.

Patients with left-sided breast cancer are assessed for the ability to hold their breath for at least 30s and coached for DIBH without arching the back. Good breath hold is defined as the patient’s capability to hold breath for a minimum of 30s. Patients are immobilised on a vacuum cushion, and an Orfit All-in-One base plate (Orfit Industries, Wijnegem, Belgium) with both arms elevated. The face is turned to the contralateral side if the supraclavicular fossa is to be irradiated. Axial CT scans, 2 mm thick, are acquired in free breathing (FB) and then in DIBH; for the latter, the position of the movable pencil is adjusted such that it makes contact with the patient in DIBH and its position is read off on a graduated scale and documented. A repeat scan is taken in DIBH to ensure reproducibility, this time with a wire on the scar in patients being treated following a mastectomy and around the breast in BCS patients.

Computerised treatment planning

The contours are delineated according to previously published departmental protocol in both FB and DIBH scans (15). Subsequently, treatment plans for both FB and DIBH scans are generated, fulfilling prescription parameters. The free-breathing plan is generated using a 2.5 cm jaw opening, while the DIBH plan is generated using a 5 cm jaw opening, the modulation factor being variable. The latter is done in an effort to restrict the treatment time to 400–450 MU. Target coverage and the mean dose received by the heart, V5Gy left ventricle, and the dose received by 0.02 cc of the LAD, MLD, and V17Gy lung in FB and DIBH plans are compared.

Quality assurance checks are performed prior to treatment as per institutional protocol.

Image guidance protocol

The patient is placed on the treatment couch in the treatment position, and an MVCT is acquired in the breath-hold position; the replicability of DIBH is confirmed using the Respiframe, as described. The scan duration is set as per the patient's breath-hold capability and initiated 5s after the beam-on button has been pressed. Since the scan starts 10s after the beam-on button is pressed, this allows approximately 5s for the patient to follow the command and hold her breath in optimal position. In patients with good breath-hold capacity, an adequate image for alignment can be obtained with only one scan. In case the breath-hold capacity is limited to 30–40s, two scans are performed over two breath-holds to align the patient; both scans are in regions that provide some anatomical information, i.e., the lower edge of the breast contour, carina, manubrium sterni, or surgical clips.

Treatment delivery

Following this, treatment is initiated after informed consent. The patient is asked to hold her breath 5s into the warm-up time. The pencil of the Respiframe is confirmed to be in the breath-hold position using the in-room camera and is then monitored. Beam interruptions are strategically incorporated in alignment with the breath-hold times during treatment administration. Treatment is continued in helical mode and terminated just before the previously assessed breath-hold capacity of the patient. This process is continued till completion of the session.

On subsequent days, the duration of the first breath-hold is varied, thus impacting the location of all subsequent interruptions. In addition, the temporal location of each interruption is noted in an Excel file by cumulatively adding all preceding breath-hold times. Individualising each breath-hold ensures that no match-line site is repeated more than 2 or 3 times during the entire course, i.e., the junctions are feathered.

Quality assurance

Gafchromic EBT3 film is employed as an in-vivo dosimetry tool, to monitor and assess the dose heterogeneity that arises from the interrupted treatment protocol. This film is carefully positioned over the target region during treatment. Subsequently, the acquired dose profiles from the Gafchromic film measurements are thoroughly evaluated and confirmed to fall within a range of ± 5 % of the normalised reference dose. The validation process for these dose profiles is carried out using RITG148 + software (version 6.6.32.000). This software assists in analysing and comparing the measured dose profiles with the expected or reference values to ensure that the treatment plan adheres to the prescribed dosimetry requirements.

Analysis

FB and DIBH plans have been compared for target coverage and OAR doses. Treatment parameters, including the number and duration of breath-holds, location of junctions and total treatment time, have been recorded. Clinical outcomes viz., regional relapse-free, progression-free, and overall survival are analysed. EQD2 is calculated for MHD, Mean and Max LAD doses for the DIBH plans using an alpha -beta of 2 Gy.

Results

Since the initiation of this technique, 30 patients were assessed for radiotherapy for left-sided breast cancer. Eight did not have adequate breath-hold, and 2 were not taken up for DIBH as no significant dosimetric advantage was noted in comparison with the FB plan. Twenty patients with left-sided breast cancer were treated using this technique for DIBH.

Dosimetric Parameters

The dose received by target volumes, lung and heart and cardiac substructures in DIBH and FB plans are shown in Fig. 2. The mean heart dose and the LAD D 0.02 cc were reduced 34 % and 32 %, respectively, in DIBH-HT plans compared to FB-HT plans [Fig. 2].

Fig. 2.

Dosimetric comparison between free breathing (FB) and Deep Inspiratory Breath-Hold (DIBH) plans.

EQD2 (for DIBH patients)

The EQD2 of the mean heart dose ranged from 1.15 to 3.47 GyE (median 2.1 Gy). The range for the mean LAD dose was 1.59–7.76 GyE (EQD2)(median 3.7 Gy), and the LAD D0.02 cc was 3.85–21.86 GyE (EQD2)(median 6.9 Gy).

Treatment parameters

The median number of breath-holds for treatment was 8 (range 5––15). The breath-hold duration ranged from 29.9 to 104.2 sec, median 64.4 sec, Table 2. The total number of junctions per patient through the entire course of a hypofractionated (15-fraction) schedule was 7 [range 4–14]. Eighty (3 %) of 2640 junctions were duplicated.

Table 2.

Breath-hold duration, frequency of breath-holds and treatment time for all patients.

Patient	Breath-Hold Duration (seconds)		Number of breath-holds		Total treatment time per fraction (seconds)
	Median (seconds)	Range (seconds)	Median	Range	Median (seconds)	Range (seconds)
1	78.6	36.1–102.3	6	5–8	434.5	350–526.2
2	45.95	30.2–57.7	13	12–15	429.1	370.4–526.2
3	59.8	30.8–81	9	9–12	443.9	355.8–517.8
4	65.2	32–74.1	8	6–9	435.25	337.8–524.4
5	67.3	44.9–76.8	9	6–10	442.5	356.2–523.9
6	67.2	30.6–82.9	7	7–10	443.1	355.2–524.1
7	58.4	37.4–76.4	9	8–10	425.3	352.6–523.9
8	51.8	33–65.1	11	10–12	429.55	371.8–526.3
9	71.3	43.4–84.8	9	8–9	435	352.3–516.2
10	81.6	37.4–104.2	6	5–7	433.9	355.9–524.5
11	55	31.5–84.1	8	6–11	429.9	356–524.5
12	64.5	42.5–80.9	7	7–9	430.25	331.8–523.9
13	52.5	33.4–74.6	11	9–12	430	313.9–523.7
14	65.5	31–94	8	7–9	435.15	368.5–526.4
15	69.95	39.4–92.1	8	7 – 8	430.75	352–526.4
16	70.95	32.9 – 76.5	6	5–8	443	432–463
17	64.35	39.3 – 76.5	8	7–9	531.3	454.1––542
18	49.5	29.9–66	9	8–11	432.8	377.9–453.5
19	56.3	42.4 – 75.8	10	9–12	601.2	538.4–613.5
20	63.2	34.9–80.8	8	8–9	515.2	515–535.5
Total	64.4	29.9–104.2	8	5–15	434.75	313.9–613.5

The average total treatment time from initiation of the first breath-hold to completion of treatment was 451.5 sec, median 434.7 sec (range 313.9–613.5). There were an average of 7 treatment interruptions.

Quality Assurance

The film placed longitudinally along the patient to detect hot or cold spots at junctions revealed no significant variation across the longitudinal extent of treatment in any patient, on dosimetric analysis [Fig. 3].

Fig. 3.

Film strip at the surface with dose remained within +/- 5% dose variation.

Outcome

Seventeen patients had grade 1 oesophagitis, and none had grade II/III esophagitis. Skin reactions, Grade I in 10, were uniform; no grade 2 reactions were noted during treatment.

No patient has recurred locoregionally or at a distant site at a median follow-up of 9 months.

Discussion

We discuss an in-house developed workflow to administer radiotherapy on a Helical Tomotherapy Unit using DIBH and manual gating. We also discuss the early outcomes of this treatment technique in 20 women.

Radiotherapy using DIBH is the standard of care for left-sided breast cancer, with the aim of reducing the risk of radiation-induced heart disease (RIHD). The technique helps reduce the mean heart dose 26–67 % [16] and the mean LAD dose from 31 to 71 % in various studies [7], [8], [9]. While there is no lower threshold identified, the risk of RIHD is especially high in certain patients, i.e., those with underlying diseases such as diabetes mellitus, hypertension, previous IHD, obesity, etc [2]. In the population being served by our hospital, the high incidence of these risk factors, the propensity for cardiovascular disease [17] and the high proportion of patients diagnosed with palpable lumps rather than via screening [18] and thus requiring anthracycline-based chemotherapy warrants access to DIBH-based techniques.

We have previously used a tactile frame-based technique for DIBH [14]. This technique allows for a wider application of DIBH, facilitating manual gating. Closed-bore gantry systems are, at present, not equipped for DIBH gated radiotherapy, thus precluding the delivery of standard-of-care radiotherapy in left-sided breast cancer patients. Challenges include the inability to monitor breath-hold using established methods such as surface guidance as well as the absence of an automatic cutoff. The use of the frame with the pointer-pencil combination allowing the breath-hold to be monitored addresses the challenge of monitoring and replicability of breath-hold, as noted previously [12]. The breath-hold was confirmed by using the Respiframe with an indicator pencil, allowing the therapist to monitor the patients using the in-room camera. In addition, the frame provided a tactile reference to the patient to ensure that the breath-hold position was replicated. This was confirmed by doing a helical MVCT prior to each fraction; since the duration of the MVCT was dependent on the breath-hold capacity of the patient, its longitudinal extent varied from 6 cm to 12 cm.

The other challenge of closed-bore helical treatment is that unlike linear accelerators, which allow a maximum field size of 28–40 cm, the maximum jaw size is 5 cm when treating in helical mode. Resultant multiple junctions at the location of treatment interruptions and the relatively long treatment time pose additional challenges. Our workflow has addressed each of the above issues.

Since the Radixact unit does not have an automatic cutoff of the beam when the patient is not holding the breath in deep breath-hold, manual gating was used. There is precedence of the same in the UK Heart Spare trial [12]and by Convoy et al [13], as well as when using other O-ring LAs [19]. The initial treatment sessions required careful titration with previous information about the patient's breath-hold capability. Since the Helical Tomotherapy unit has a warm-up time of 10 s after every interruption, the breath-hold was carefully initiated approximately 5––6 sec after turning the key, thus allowing the patient approximately 4–5 s to reach the breath-hold position before the beam was initiated.

The issue of the long treatment time when treating with Helical Tomotherapy was resolved using the largest jaw size, 5 cm and by treating patients with a minimum breath-hold capability of > 30s. A breath-hold time of > 20 s has been considered adequate for DIBH by Simonetto et al. [20]. However, in view of the relatively long treatment time on a Helical Tomotherapy unit, a longer breath-hold time and largest possible jaw size allowed the treatment time and the number of breath-holds and by extension, the number of interruptions to be restricted. In view of these interventions and a conscious decision to limit the total treatment time to approximately 400 sec, the average treatment time was 485.88 sec and the median 433.95 sec (range 313.9 – 526.4). The breath-holds varied from 5 to 15, with a median of 8; the number of junctions, therefore, ranged from 4 to 14, with a median of 7.

Following every interruption, the radiation delivery in the HT Radixact unit resumes from the same point. In our earlier analysis, as well as in the daily pretreatment MVCT, we were satisfied with the reproducibility of the breath-hold. However, as an added precaution, we ensured that the location of the junctions was staggered. By using this approach, the location was repeated in only 80 (3 %) instances in 2640 junctions in 300 fractions. In addition, we assessed the surface dose in each patient by placing a strip of film across the patient along the longitudinal axis and noted that the dose at the surface remained within +/- 5 % dose variation.

We did note a temporal trend in 6 out of 20 patients who were able to hold their breath 34 % longer in the last week of the radiation schedule versus the first week of radiation, discounting the first fraction. This is a reflection of the patient being more at ease with the treatment process along with better breath-hold capacity due to practice. We also noted, though did not formally document, that patients who had previous experience with deep breathing practices such as yoga, meditation or swimming were more at ease during these procedures.

Using DIBH on the Helical Tomotherapy unit allowed a reduction in the mean dose received by the heart by 36 %, the mean dose of LAD by 43 %, the max dose of LAD by 37 % and the mean lung dose by 0.8 %, similar to studies documenting dosimetry when treating on linear accelerators [7], [16].

The preferred mode of breast cancer radiotherapy has been FiF-based forward modulation techniques. Volume-modulated radiation therapy techniques allow for ease in the administration of SIB and shaping of high doses to avoid the heart. On the other hand, the low dose bath of the lungs and the heart is significantly higher, which is a concern both for cardiac side effects and the risk of second cancer [21], [22]. While the technique described above allows for the option of DIBH in departments equipped only with closed-gantry systems, not unusual in low and middle-income group countries (LMIC), the issues associated with the rotational mode of therapy remain. We have chosen to use Tomodirect, akin to non-helical tangential treatments, for patients not requiring nodal irradiation, to reduce the possibility of a low-dose bath.

The EQD2 threshold for any cardiac event and major cardiac event has recently been described by Zuriek et al [23]. The authors studied 375 consecutive patients irradiated for left-sided breast cancer, and noted that 36 and 23 patients suffered any and a major cardiac event, respectively. The thresholds for any cardiac event were calculated as 0.8 Gy for the MHD, 2.8 Gy for the Mean LAD and 6.7 Gy for the Max LAD while the threshold of Mean LAD of 3 Gy, Max LAD − 7 Gy and MHD − 1.3 Gy respectively, correlated with a major cardiac event, all dose values being expressed as EQD2. We have recalculated this for a 15-fraction schedule using an alpha–beta of 2 for both heart and LAD. We noted that reducing the EQD2 below the threshold for major events was not achieved for any patient for the heart and was achieved for the Mean LAD dose in 20% of cases and for the Max LAD (D0.02 cc) in 45% of cases. Unlike this, in our previously reported cohort of patients treated with proton therapy, the Dmean of heart (EQD2) was achieved in 100 %, Dmean LAD (EQD2) in 90 % and LAD D0.02 cc (EQD2) in 80 % of cases [15]. This underlines that while DIBH allows the reduction of heart and substructure doses, further reduction with proton therapy remains a desirable and achievable goal.

The limitations of this technique include issues associated with manual gating, i.e. the requirement of coordination with the patient, calibration of the treatment duration with the patient’s ability to breath-hold and additional caution during the beam-on position. Specific to Helical Tomotherapy, the issue of possible cold/hot spots at junctions is addressed above. The need for an AIO board, and difficulty in fixation of the pointer are potential difficulties in using the Respiframe. An additional concern, common to all DIBH techniques is the patient arching her back to reach the desired threshold; this can be obviated by educating the patient.

Besides its application in Helical Tomotherapy units, this technique allows DIBH in manual mode with better certainty of replicability than breath-hold without feedback. This has potential for widespread application in open and closed gantry treatment systems, including telecobalt units not equipped with DIBH solutions.

Conclusion

The above mentioned workflow for DIBH with Helical Tomotherapy addresses patient safety, treatment precision and challenges specific to this treatment unit. The workflow prevents junction-related issues by adjusting daily breath-hold durations and ensuring locations of junctions are not repeated during treatment. This technique offers a practical solution for minimizing radiation exposure in left-sided breast cancer patients treated with HT.

Author Contributions

Sapna Nangia Concept & design, interpretation of data, drafting & revising the article.

Nagarjuna Burela Data collection, analysis and interpretation of data, literature research.

Mayur Sawant Acquisition of data.

Aishwarya G Acquisition of data.

Patrick Joshua Acquisition of data, literature research.

Vijay Thiyagarajan Acquisition of data.

Utpal Gaikwad Acquisition of data.

Dayananda S Sharma Acquisition of data, revising the article.

Declaration of Competing Interest

Dr Sapna Nangia holds a patent for the Respiframe described in this article.

Mayur Sawant was affiliated with the Apollo Proton Cancer Centre when the work reported in this study was initiated. He is now employed with Accuray Medical Equipment Pvt Ltd.

The other authors have no conflict of interest to report.