Efficient clinical workflow for intraoperative electron radiation therapy with a mobile electron accelerator

Abstract

Purpose

Intraoperative radiotherapy (IORT) is designed to deliver a precise dose of ionizing radiation to a tumor or tumor bed after surgical resection. The IORT procedure can deliver targeted radiation to the tumor with increased normal tissue sparing, hence improving the efficacy of the procedure when nearby organs at risk (OAR) can be mobilized out of the field thus maximizing the therapeutic ratio. This report presents the institutional experience of establishing the IORT program, highlighting our workflow improvement, quality assurance (QA), and dosimetric practices.

Methods

The IntraOp® Mobetron® 2000 can be used to irradiate a target using megavoltage (MV) electron beams of energies 6 MeV, 9 MeV, and 12 MeV at a high dose rate of 1000 cGy/minute. Mayo Clinic Florida (MCF) has a dedicated and well-shielded QA suite adjacent to the operating room (OR) for both treatment preparation and daily QA purposes. The Mobetron is docked over the target assembly with the help of the autodocking software. The IORT workflow in the OR is assisted by the Mobility (IntraOp Medical Corp., Sunnyvale, CA) system which ensures safe and efficient maneuvering of the Mobetron. The treatment planning and beam control using the treatment console is done from inside the shielded QA room.

Results

Since its commissioning in 2016, 138 patients have been treated with IORT using the Mobetron ® 2000, at MCF. The efficacy of IORT has been utilized in the treatment of locally advanced and recurrent sites like pancreas, rectum, retroperitoneal sarcoma and some gynecologic cases.

Conclusion

A well-planned housing of the Mobetron, adjacent to the OR, and its timely upgrades has resulted in an efficient and safe IORT workflow.

Introduction

Intraoperative radiation therapy (IORT) delivers highly targeted radiotherapy where a collimated beam of ionizing radiation can be directed onto an unresected tumor or tumor bed [1], especially in cases where there are close or positive resection margins [2, 3]. The treatment area is carefully determined by the surgeon and radiation oncologist, and a collimated beam of radiation is delivered to the target in a single fraction. IORT can be delivered with a few different techniques: (a) kilo-voltage (kV) X-ray IORT [4]; (b) electronic brachytherapy [5]; (c) radioactive sources using Harrison-Anderson-Mick (HAM) applicator [6]; and (d) IORT using megavoltage (MeV) electrons [7, 8]. The purpose of using shallow penetrating radiation, such as kV photons or MeV electrons, is to ensure treatment of the tumors up to the depth of a few centimeters while sparing tissues and organs distal to the tumor. All IORT procedures, excluding those involving a radioactive source, are performed by using either an X-ray source [4] or a mobile or stationary linear accelerator (LINAC) [3]. Historically, IORT has been performed either by transporting anesthetized patients from the OR to the department of radiation oncology or by using a stationary LINAC placed in the OR [3, 9]. The use of mobile LINACs has the advantage that the patient could be treated in a sterile environment in the OR under anesthesia, without having to move the patient and the surgical team. At the same time, a mobile LINAC can facilitate a safer and more efficient IORT workflow in an OR with reduced shielding requirements [3].

The Mobetron® 2000, which stands for MOBile ElecTRON beam linear accelerator, is a moveable electron beam radiation therapy system. It is designed to be used in the OR with minimal amount of shielding due to reduced radiation leakage and the presence of a beam stopper. It can generate high energy electron beams of energy: 6, 9, and 12 MeV at a dose rate of 1000 cGy/minute which ensures fast dose delivery during an IORT procedure. A tissue equivalent bolus can be used for cases where more superficial treatment is desired.

IORT is often used as a boost to pre-operative external beam radiation therapy (EBRT) for locally advanced gastrointestinal malignancies, primarily pancreatic and colorectal cancers as well as soft tissue sarcomas [10–15]. Additionally, it has shown efficacy in managing gynecological cancer and hepatocellular carcinoma [16–18]. When used as an intra-operative tumor-bed boost, IORT has the potential to reduce local recurrence rates with acceptable toxicity [2, 10, 19]. Randomized trials have indicated that IORT can serve as a viable alternative to whole breast irradiation (WBI), demonstrating comparable local control in selected early stage disease [20]. In the case of high-risk soft tissue sarcomas of the extremities, combining IORT with EBRT after limb-preserving surgery has resulted in high local control rates [11, 12]. Overall, IORT boosts have demonstrated excellent local control rates in both early breast cancer and soft tissue sarcomas [2, 21]. In terms of efficiency, the IORT procedure with the Mobetron adds only about fifteen minutes of OR time, as detailed within this manuscript.

The Mobetron ® 2000 was commissioned at MCF in 2016 and started treating patients in 2017. After a few years of use, the Mobetron® 2000 was upgraded with a new beam control board, new beam stopper with two ¾ inch lead plates, a lead shell to shield the collimator, and a camera-based auto-docking system in 2020. Then in 2021, the clinical use was transitioned from a standard OR, to a specially shielded OR with a steel plate in the floor, covering the area directly below the Mobetron® treatment beam. This new OR is equipped with an adjacent dedicated closet to store the Mobetron which has shielded walls that allow for routine Quality Assurance (QA) measurements during business hours. In the current manuscript, we present the institutional experience of commissioning and workflow improvement measures taken towards establishing an efficient and successful IORT program at MCF.

Materials and methods

Mobetron® 2000: The portable electron accelerator

The Mobetron® 2000 is an X-band LINAC designed to deliver electron beam radiation to a surgically exposed target in the operating room (Fig. 1). It operates at 9 GHz which is 3 to 4 times higher than S-Band (2–4 GHz) accelerators. This higher frequency provides greater accelerating gradients (100 MV/m) allowing for shorter wave guides, reducing the system size and shielding which results in less weight. The system’s high dose rate ensures that irradiation times remain acceptable within the time constraints of the surgical workflow. The removal of the bending magnet helps limit the leakage radiation from the LINAC head while the beam stopper tracks the rotation of the gantry in all directions hence intercepting the primary beam and the small quantity of patient and table-generated bremsstrahlung scatter radiation [3, 22]. The Mobetron® 2000 gantry head has a rotational range of ± 45° while it can tilt (+ 10°/-30°) in the forward and backward direction. In addition, the gantry can travel 30 cm vertically, 10 cm laterally, and 10 cm in the longitudinal direction.

Fig. 1.

Schematic diagram of the Mobetron® 2000 LINAC

A total of 45 cylindrical anodized aluminum applicators that come in variable diameters from 3 cm to 10 cm with a 0.5 cm increment are available. In case the tumor bed lies at an angle to the gantry head, each applicator is further available in 3 bevel angles of 0°, 15° and 30°. A set of target and assembly clamps are used to hold the desired applicator over the treatment area. The gantry head is then aligned to the target using the auto soft docking system assisted by three built-in cameras. At the nominal gantry position, the nominal treatment SSD (source to surface distance) is 50 cm. Additionally, for the treatment of sarcoma of the extremities and retroperitoneum, rectangular sarcoma applicators are available in three sizes: 7 cm x 12 cm, 8 cm x 15 cm and 8 cm x 20 cm with each of them available in 0° and 20° bevel angles. Due to their larger field aperture, each sarcoma applicator is fitted with a beam spreader that creates a uniform field across the applicator aperture.

Commissioning of Mobetron® 2000

The acceptance and commissioning of Mobetron® 2000 at MCF was performed in 2016 for the three available energies of 6, 9 and 12 MeV. Commissioning data acquisition included acquiring percent depth dose (PDD) and output factors for all available energy, applicator size and angle combination in a water tank. Profile widths at various depths were acquired for all applicators of various sizes and angles for each energy. This resulted in the determination of the width of the 90% isodose line, along the bevel direction, at different depths in water using the IC-10 ion chamber. Our Mobetron® commissioning data were compared to those from one of our associate institutions, for additional validation.

Shielding and radiation safety

The exposure limit of 1 mSv/year of total effective dose equivalent to individual members of the public, according to Florida Administrative Code 64E-5.312 1a, was used for all workload limit determinations for spaces surrounding the Mobetron treatment and QA operation. An additional limitation of 2 mrem in any hour was followed for all locations. To avoid excessive construction costs, the shielding design was primarily constrained by floor load capacities and door weight limitations.

Operating room

Figure 2 shows the layout of the OR with a dedicated, shielded room for QA and equipment storage. The two entrance doors to the OR are interlocked to the beam and have “Beam On” lights linked to the machine operation. The beam is also interlocked to the two “last man out” buttons and it cannot turn on until both buttons are pressed.

Fig. 2.

Schematic diagram to indicate the treatment setup in the OR

Last person out 1 button is to be pressed after the OR is cleared of all staff and both straps are pulled across the doors (Fig. 3). Then last person out 2 button is pressed, and the physicist, radiation oncologist, and engineer (as needed) prepare for treatment at the console area. Cameras in the OR are viewable from the computer console at the console workstation, to allow the treatment staff to always observe the anesthetized patient. Anesthesia monitors are viewable from a workstation in the hall outside the OR while staff are waiting for treatment completion. The floor and ceiling decks of the OR are concrete and steel composite of thickness 7.5 inches. A 2-inch steel plate is placed in the OR floor, right below the gantry, for additional shielding during treatment. The floor tiles covering the steel plate are colored green to indicate its exact location.

Fig. 3.

(a) Last person out button (left); (b) Door strap pulled during Mobetron use in OR (right)

Dedicated quality assurance room

During normal hours, the Mobetron is stored in a dedicated QA room next to the OR as shown in Fig. 4. The QA room has leaded walls ranging from ¼ inch lead to 2 ¼ inch, with all lead walls extending to a 7ft height. All QA procedures for the Mobetron® 2000 are performed inside the well-shielded QA suite with the LINAC placed inside the suite pointed towards the back wall and the operating console outside a partially closed shielded door as shown in Figs. 4 and 5. The LINAC orientation was selected to accommodate a maximum of ¼ inch leaded doors to minimize construction costs.

Fig. 4.

Schematic diagram of physics suite during a QA procedure

Fig. 5.

Mobetron® gantry and console positions during QA

Optimized workflow for quality assurance

Traditionally, Mobetron® daily QA is performed early in the morning, before the OR staff arrive for room preparation. This workflow ensures that QA is completed without interfering with OR setup. More recently, we have begun adopting a parallel workflow in which daily QA is performed while the OR staff prepare the room. This new approach leverages the shielding of the dedicated QA room, allowing QA to be conducted safely in parallel with OR activities without exposing staff to radiation. This change optimizes efficiency by allowing the OR and radiation oncology teams to proceed independently, while reducing the need for the RadOnc team to arrive very early in the morning. Daily QA is performed by the engineering team, who are trained by the physicists, and includes output and energy constancy checks for all three energies using an Exradin A12 ionization chamber (Standard Imaging, Middleton, WI), with output tolerances set at ±3% in accordance with AAPM TG-72 recommendations. Energy constancy is verified within the range corresponding to a ±2 mm shift in the percentage depth-dose (PDD). Additional tests include auto-docking verification, beam interruption checks, door interlock functionality, and applicator inventory confirmation to ensure readiness for clinical use. Figure 6(a) shows the typical PDD curves for the 3 available electron energies while figure 6(b) shows the effect on the target coverage due to a ±2 mm shift in the percent depth dose (PDD) curve. It demonstrates that with sufficient margin around the target, there is much less impact on target coverage at 90% isodose line (IDL) even with a seemingly large change in the energy constancy ratio.

Fig. 6.

(a) PDD curves for the three electron energies (left); (b) and the range of energy constancy relative to shifts in the PDD showing the effect on target coverage (right) (both images are provided by IntraOp®)

Optimized workflow for treatment delivery

IORT with Mobetron® 2000

Since its commissioning in 2016, 138 patients have been treated with IORT at MCF by the time of this manuscript preparation. Pancreatic and rectal cancers are the two most common IORT indications, followed by gynecologic malignancies, extremity sarcomas, and hepatocellular carcinoma cases. The commonly used surgical tables for the IORT procedures are Allegro Mizuho (Mizuho OSI, Union City, CA), Skytron UltraSlide 3603, and the Skytron EZ Slide 3501 C (Skytron, Grand Rapids, MI). For the colorectal cases, the patient’s head is oriented closer to anesthesia, while for the pancreatic resections the patient is rotated 180 degrees to have their feet towards anesthesia. This is due to the placement of the steel plate in the floor, and the planned shielding evaluation location. The Mobetron® is steered from the side of the table and docked over the target assembly. For treatment sites involving the pancreas or regions superior to it, the patient’s feet must be positioned over the table base to allow gantry and beamstopper clearance for beam delivery over the extended portion of the table. For rectal cases, an additional inserted table piece is used to allow for greater extension of the patient beyond the base of the table stand. The table may also need to slide down to its limit of extension and be tilted so that the Mobetron® can access the target area. Additional measures may be required to ensure patient safety and stability in this setup [23]. At MCF, we employ supplemental counterweights to prevent the table from tilting in case of accidental or unexpected unlocking of the table base from the floor.

Enhancing efficiency with Mobetron mobility

In 2023, the existing Mobetron® 2000 was equipped with the Mobility system, replacing the previous transport jack. The Mobility system provides significantly improved user experience, allowing the Mobetron to be maneuvered with a hand control pendant rather than the manual jack. The control pendant offers two speed settings: a slow mode, in which the Mobetron travels 1 m in 17.65 s, and a fast mode, in which it travels the same distance in just 4.2 s. These modes can be switched seamlessly depending on the proximity of the Mobetron to the patient table. Operation requires pressing and holding the green motion enable button while directing the joystick in the desired direction (Fig. 7a); an ESTOP knob immediately halts any motion if needed. The system must maintain at least 24 V (Fig. 7b) and should be plugged in to recharge if the voltage drops below this threshold.

Fig. 7.

(a) Mobility hand control pendant showing the two speed settings (left); (b) Emergency ESTOP knob (red) and the voltage panel showing the Mobility operating voltage (right)

The integration of the Mobility system has not only improved maneuverability around the treatment table but has also enabled a more efficient workflow: the Mobetron is now brought into the OR only immediately before IORT delivery and removed after treatment is complete. This approach frees valuable OR space during surgical preparation and eliminates the need for engineering staff to wait until the end of the day to remove the unit. Importantly, this modified process adds less than 10 min to the total OR time while providing greater flexibility and confidence in positioning the Mobetron safely and efficiently.

Patient selection

All MCF patients are reviewed in a multidisciplinary tumor board. IORT recommendations are given by the tumor board and largely based on preoperative imaging indicating extensive vascular involvement by malignancy. Nearly all patients selected for IORT qualify as “borderline resectable” or “locally advanced” by NCCN guidelines.

Staff training

Physician training for Mobetron® treatments includes a formal credentialing process for radiation oncologists, involving the observation of three cases followed by proctoring of three additional cases prior to independent practice. Surgeon training is minimal and typically guided in the OR by the radiation oncologist and physicist as needed, with most surgeons achieving comfort within five cases.

Proper operation of the surgical table is an important focus during onboarding of the surgical teams. The table base must be aligned with the green floor tiles marking the embedded steel plate and securely locked in place. Only the tabletop should be moved for alignment, as the base must stay locked to the floor to prevent instability while extending the patient beyond it. To prevent tipping, especially if the base is accidentally unlocked, adequate counterweights are placed at the base as a safety measure. Education of surgical and instrument processing teams regarding appropriate handling of the Mobetron® accessories is also crucial to ensure safe handling.

The physics team undergoes a more extensive training process, which also includes observing three cases and being proctored for three more. They receive hands-on equipment training to become familiar with key components such as the target assembly, ball clamp, and applicators, which improves confidence and reduces handling errors. Physicists must have complete understanding of the daily QA to review and approve prior to treatment. Additionally, they must be proficient in performing MU calculations and operating the Mobetron and its auto-docking system.

Results

Radiation survey results

All survey readings were measured at 1 foot beyond the barriers and the measurements around the QA room were performed with the Mobetron® in place facing the back wall of the closet (Fig. 5). The lateral x-ray scatter to be measured is expected to primarily originate from the treatment head of the machine and from the beam stopper. The highest exposure on the floor below occurs in areas laterally adjacent to the beam stopper. Based on the survey results, a workload limit of up to 1972 Gy annually and 40 Gy in any one hour was implemented for treatment operation. A workload limit of up to 3809 Gy annually was implemented to the QA operation in the storage room. Based on the survey results at the hallway below the QA room, a workload limit of 40 Gy in any one hour was implemented for QA operation. Therefore, the storage room use is limited to daily QA in one hour. Additional monthly QA is performed during a different hour of the day. Additional service beams that must be run are performed after hours and on weekends when surrounding rooms are unoccupied. Service staff avoid occupying high exposure zones and are considered radiation workers with a 50 mSv annual limit.

Some neutron production is possible due to high energy photon scatter particularly from the 12 MeV beam. A neutron survey 1 foot beyond the barrier, in front of the gantry in the QA position, measured a maximum of 0.5 mR/hr, confirming safe neutron exposure levels. Other areas surrounding the room were at background ensuring that no additional neutron specific shielding was required.

Mobetron output

Output calibration

Absolute dosimetry for the Mobetron® initially presented some challenges with the high dose per pulse electron beams of 0.56 cGy/pulse which are approximately ten times higher than most stationary LINACs. This difficulty was primarily due to the utilization of the Exradin A12 (Standard Imaging, Middleton, WI) ionization chamber for absolute calibration. AAPM Task Group 51 [24] advises against using ionization chambers for absolute dosimetry if the P_ion value is greater than 1.05 and traditional Farmer type chambers may exceed this for the Mobetron. Although published programs or fits exist that may be implemented to address this [25, 26], we have implemented absolute calibration with an Advanced Markus parallel-plate chamber. The Cross-calibration is performed on a Varian TrueBeam linear accelerator against a NIST traceable farmer-type chamber to determine the to be applied in the Mobetron® electron beam calibration. Between the time frame of 2017 to 2025, the reported IROC (Imaging and Radiation Oncology Core) Houston results consistently stayed within 3% accuracy for all electron energies to support our clinical reference dosimetry practice.

In a typical clinical LINAC, a pulse forming network (PFN) is tuned to coordinate the timing of the gun and the RF system and to maximize the dose rate of the machine. This adjustment is commonly performed by engineers when underdose faults are occurring and physics does not need to evaluate dose afterwards. Adjustments for Mobetron underdose faults are directly correlated with the machine output and this is something to be mindful of and verify if dose rate adjustments are made. These underdose rate issues sometimes appear with flatness and symmetry faults but tuning the dose rate resolves both. Underdose rates may occur more often when the machine is not adequately warmed up. If adjustments are made while the machine is cold, it may not perform as well hours later during treatment, which is a critical concern for patient care. We have found that leaving Mobetron on overnight before an IORT case ensures stability of the beam delivery for QA and treatment. Prior work by Beddar et al. showed output constancy within 1% when left on and inactive overnight [30]. Lastly, when commissioned in 2016, the Mobetron beam delivery was affected by room temperature, but after a dose board upgrade in 2020, this has not been a concern.

Output factors

Mobetron output factor, for a given energy, applicator diameter and bevel angle combination, is defined as the measured output normalized to that for the reference applicator of 10 cm diameter and 0° bevel for that energy. The variation of Mobetron output factors with applicator diameter sizes are shown in Fig. 8 for all available energies and bevel angles. The two major contributors to the output at point on the central axis of an applicator are: the loss of electrons away from the field and the backscatter of electrons into the field, from the applicator surface. As a result, these factors generally increase with smaller applicator size due to the increased backscatter from the applicator until the applicator size becomes small enough when the loss of electrons becomes the predominant factor [23]. Figure 8 shows the output factors, measured with diode, for the 6 and 9 MeV energies peaking at about 4–5 cm applicator size. On the contrary, the increasing bevel angle only has a slight impact on the changes of the output factors for all electron energies.

Fig. 8.

Mobetron® 2000 output factors for all applicator-size, bevel angle, and energy combination

Percent depth dose curves

The PDD curves for the Mobetron® 2000 can vary significantly as a function of the bevel angles for a given beam energy, especially for larger bevel angles as shown in Fig. 9 for the 10 cm representative applicator. A larger bevel angle increases the obliquity of the incident electron beam and hence increases the surface dose and decreases the depth of maximum relative dose (d_max) and the depth of 50% relative dose (R₅₀). However, the change in the applicator diameter size has a smaller effect on the PDD curve characteristic of the Mobetron as shown in Fig. 10. The 6 MeV shows no significant differences, although higher energy use requires attention at smaller field sizes, particularly if higher bevel angle is also used. Mobetron commissioning data at MCF shows that the 90% depth of the 12 MeV beam for the 10 cm applicator with 0-degree bevel is 3.8 cm, while for the 3 cm applicator with 30 degree bevel it is at 2.4 cm.

Fig. 9.

PDDs for 10 cm applicator-size, bevel angle and energy combinations

Fig. 10.

PDDs for all energies and 3, 6, and 10 cm applicators with 0° bevel

Electron beam parameters

As part of the commissioning of our Mobetron® 2000, electron beam parameters were measured in a water tank and the data were successfully compared to those of our associate institution, which already had a well-established IORT program. Table 1 shows the surface dose relative to d_max and the values of depth of d_max, 90%, 80%, 50%, 30%, and 10% IDL for the 10 cm applicator. These values are in good agreement with some of the previously published data [23, 27]. Table 2 demonstrates the variation in the widths of the 90% IDL at the depth of d_max for three representative applicator sizes and all available energies. For larger bevel angles, the obliquity of the incident electron beam tends to provide slightly greater coverage at the 90% IDL. It is recommended that, some additional margin is considered to ensure adequate target coverage for the given applicator size, bevel angle and beam energy combination [23].

Table 1.

Electron beam relative dose values for the 10 cm applicator and 0° bevel

Energy (MeV)	Surface PDD (%)	Dmax (cm)	R 90 (cm)	R 80 (cm)	R 50 (cm)	R 30 (cm)	R 10 (cm)
6	92.7	1.2	1.9	2.1	2.6	2.9	3.3
9	94.1	1.7	2.9	3.2	3.8	4.2	4.7
12	96.4	2.1	3.8	4.2	5	5.4	6

Table 2.

Width of the 90% IDL at D_max for three representative applicator sizes and bevel angles

Energy (MeV)	Width of 90% IDL (cm)			Width of 90% IDL (cm)			Width of 90% IDL (cm)
	for 3 cm applicator			for 6 cm applicator			for 10 cm applicator
	0°	15°	30°	0°	15°	30°	0°	15°	30°
6	2.2	2.5	2.7	5.5	5.8	6.5	9.6	10	11.1
9	2.3	2.4	2.9	5.5	5.8	6.4	9.5	9.8	10.9
12	2.4	2.6	3.1	5.5	5.8	6.4	9.1	9.5	10.4

Air gap factor

There is sometimes an air gap between the end of the applicator and the tissue target. To account for the dosimetric effect of the air gap, an air gap factor is included in the monitor unit (MU) calculation as shown in Eq. 1 below:

1

Only two patients treated at MCF needed an air gap correction, and the maximum air gap was 1.5 cm. 1 cm and 2 cm air gap factors were measured for each energy with a 10 cm 0-degree bevel applicator. Compared to Eq. 1, a maximum difference of 1.5% was seen for the 12 MeV beam with 2 cm gap. The rare occurrence of the use of gap factors did not necessitate the extensive commissioning required to measure specific gap factors.

Patient treatment with Mobetron® 2000

Mobetron® 2000 clinical setup

After tumor resection and subsequent pathology results, the surgeon and radiation oncologist decide if IORT treatment is beneficial for the patient. If treatment is warranted, the Mobetron is steered into the OR from the QA room in preparation for treatment. A sterile cap is fitted on the end of the gantry and acts as an electrical barrier between the machine and patient while preventing any contamination to the patient from the gantry head. The cap is designed specially such that when placed correctly, does not block the view of the three autodocking cameras that along with the Mobetron autodocking software guide the gantry head to dock over the target assembly. Figure 11 shows a typical treatment site for IORT with Mobetron at MCF with circle indicating the applicator position to encompass a resected pancreatic tumor bed. Once the applicator and target assembly are in place, the Mobetron is steered into close alignment with the target assembly and the autodocking software is activated. The software will be functional as long as two of the three cameras are functional (Fig. 12).

Fig. 11.

IORT with Mobetron after pancreatic resection at MCF

Fig. 12.

Mobetron autodocked over the target assembly with all 5 axes of motion in the “Green Zone” as shown in the display panel

Challenging cases

Colorectal cases are relatively challenging for IORT since their anatomical location means the Mobetron is often operated close to its rotational and translational limits. Additionally, in such cases, the patient table is extended far beyond the pedestal to make sure the Mobetron can be docked over the patient without its legs colliding with the table. Significant instability in patient positioning may occur in such cases and extra weights are utilized to counterbalance the patient’s weight and stabilize the table. OR staff often use straps on patient’s chest or thighs to prevent them from potentially sliding off the table. Figure 13 shows one such case where the patient bed needed to be extended and inclined to allow for Mobetron access, hence creating greater instability. A set of cameras placed strategically inside the OR can be streamed on the computer screen at the treatment console hence allowing for continuous monitoring of the patient and the anesthesia screen during the beam delivery. OR staff can also monitor the patient from door or hallway windows and quickly enter the room if needed, with the radiation beam immediately turning off due to the door interlock.

Fig. 13.

A challenging setup of rectal IORT with Mobetron® 2000 at MCF

IORT dosimetry

The dosimetry for IORT is done on a spreadsheet where the monitor unit (MU) calculation is performed based on the electron energy, prescribed dose, applicator size, bevel angle and air gap according to Eq. 2. The decision on the size of the applicator and beam energy is made by the radiation oncologist with the help of the OR surgeon based on the extent and depth of the disease. Using 90% IDL width and PDD tables available in the dosimetry spreadsheet, the applicator size is selected to provide adequate lateral coverage to the target while the electron energy is sufficient to cover the depth of the target along the central axis.

2

IORT treatment statistics

As previously mentioned in Section “IORT with Mobetron® 2000”, the majority of the IORT treatment at MCF involves gastrointestinal malignancies (mainly pancreas and colorectal cases) (see Table 3; Fig. 14 (a)). Some of the other sites that are treated with IORT include pelvis for gynecologic malignancies, axilla for sarcoma, liver, and sarcomas of the retroperitoneum and extremities. Table 3 also shows the distribution of electron beam energies, applicator sizes, and bevel angles used for IORT at MCF.

Table 3.

Statistical distribution of treatment sites, electron energy, applicator size and bevel angles for IORT at MCF

	Energy (MeV)			Applicator size (cm)			Bevel angle (Degree)			Treatment site
	6	9	12	3-3.5	4–6	6.5–10	0	15	30	Pancreas	Rectum	Other
Patients treated	53	55	30	2	64	72	86	7	45	88	23	27

Fig. 14.

Distribution of treatment sites (left) and annual number of IORT cases (right) at MCF

Discussion

To establish an efficient clinical workflow for IORT, a clinical team must first determine room shielding design and limitations of use for safe operation that fit the desired scope of practice. Although the small amount of leakage and scatter photons and the low energy scatter electrons, with limited range, allow for IORT treatment in an OR with little or no shielding [28], additional shielding has been utilized to enhance the program at MCF. Constructing a steel plate into the floor, under the Mobetron gantry, ensures radiation safety of all staff on the floor below the OR during an IORT procedure. Due to weight limitations of construction, the plate needed to be smaller in width and length than originally desired. The final placement was based on prior experience with lower GI treatment sites, but the program then expanded vastly into pancreatic treatment sites which require opposite patient positioning on the table. Although it is less than ideal to have the patient’s head oriented away from anesthesia, the MCF program has continued to orient the patient this way to position the beam over the steel plate for ALARA concerns.

The dedicated, shielded QA room allows all daily QA measurements to be performed safely even during business hours. All MCF Mobetron QA procedures are performed in compliance with the radiation dose limits regulated for our established hourly use limit of 40 Gy. This capability has enabled a recent workflow change in which daily QA is conducted in parallel with OR setup, allowing the OR and radiation oncology teams to proceed independently without delaying surgical preparation. When additional maintenance is required, beam operation is scheduled during periods of reduced occupancy in adjacent areas, further ensuring staff safety. A key challenge in the shielding design was routing the cables between the LINAC and the console. Future programs are advised to consider integrating cable ports into door frames to support similar workflow flexibility and avoid passing cables through a partially open doorway.

The next major component of an IORT program is achieving sufficient coverage of the target with appropriate margins, electron beam energy, applicator size, and bevel angle which are selected based on measured data, a subset of which is presented in Tables 1 and 2. The effect of bevel angle on PDD is most pronounced for the 30° bevel, resulting in a shallower curve (Fig. 9), whereas variation in applicator size has a comparatively smaller effect (Figs. 10). In most cases, the applicator can be placed in direct contact with the target; when this is not possible, a small air gap (< 2 cm) may be present. Any change in Mobetron® output due to this gap can be corrected using the air gap factor from Eq. 2, with minimal difference expected from measured gap factors. Because surface dose increases with electron energy, treatment of superficial targets may benefit from the use of higher-energy electrons and/or tissue-equivalent bolus. Acrylic bolus of 5 mm or 1 cm thickness may be applied to shift the isodose line toward the surface for all energies.

Due to the constriction of the higher isodose lines with increasing electron energy, an extra margin on applicator size surrounding the target should be considered to ensure adequate lateral coverage. For larger treatment volumes, field overlap (for lower energies) or field abutment (for higher energies) may be required, as reported previously [29]. At MCF, the largest circular applicator (10 cm diameter) has been sufficient to cover most IORT targets, with rectangular applicators reserved for large sarcoma cases. A simple lookup table based on the full dataset from Tables 1 and 2 may assist the physician in making real-time dosimetric decisions.

Another important component of the clinical workflow for IORT is having the most optimal software and hardware capabilities to allow for an efficient delivery of treatment. In 2020, the Mobetron at MCF underwent major upgrades, including additional LINAC head shielding, an upgraded beam stopper, a new beam control board, and commissioning of the autodocking system with a new target assembly that replaced the previous mirror-and-laser docking system. The enhanced head shielding and beamstopper allows for less limitation of treatment workload and safer exposure limits. The new autodocking system was the biggest impact on efficiency of treatment, as what was previously a long manual alignment process, is now achieved within a couple of minutes in majority of cases. Beam stability greatly improved with the new beam control board, preventing excessive faults from occurring during beam delivery. Short term output and energy stability of the Mobetron 1000 has been shown to be within 1% for an overnight inactivity [30]. At MCF, we have found good equilibrium performance of the Mobetron 2000, when it was left in a powered-on state overnight before the IORT treatment. These improvements coincided with the launch of the PACER trial (Pancreatic Adenocarcinoma with Electron intraoperative Radiation therapy) for pancreatic cancer with vascular involvement and associated with a marked increase in IORT treatments beginning in 2021 (Fig. 14b).

The recent integration of the Mobility system has further enhanced operational efficiency by improving maneuverability and control of the Mobetron during docking. This has enabled a just-in-time approach, where the Mobetron is moved into the OR immediately before treatment and removed promptly after completion. This approach keeps the OR space fully available for surgical preparation until the time of IORT and immediately after the procedure. Additionally, it eliminates the need for engineering staff to wait until the end of the surgical day to remove the unit. Importantly, this workflow modification adds only about 10 min to the overall OR time while providing greater flexibility, safety, and confidence in treatment delivery.

A successful IORT program requires precise and timely coordination among the surgical team, radiation oncologists, physicists, and engineers. At MCF, radiation oncology engineers are integral members of the IORT team, performing daily QA in the morning of treatment and providing technical support during delivery. They also complete monthly QA under the supervision of the Mobetron physicist. The surgical and radiation oncology staff are fully integrated into the program, with dedicated communication channels to ensure timely updates on case scheduling and equipment status.

Team training in handling the accessories and steering/docking the Mobetron are essential for a successful program. Thoughtful room design with more distance to viewpoints for reduced shielding considerations and evaluation for multiple treatment sites should be considered as well as cable placement during use and necessity of door interlocks. Adequate room accessories that aid in safety of delivery, such as a fixed pedestal table with six degrees of freedom motion and room mounted cameras for close monitoring of the patient should be considered. The biggest learning curve for our program has been steering the Mobetron into position for treatment. This is performed by physicists and requires close attention to clearance from the table base, drapes, and the patient. The Mobility upgrade significantly improved comfort and confidence steering the Mobetron, particularly with the fine speed control and ability to better see clearance around the patient while driving the machine forward.

Conclusion

A successful IORT program has been operational for the past 9 years at MCF treating 138 patients during this period. The IORT treatments are performed in a partially shielded OR with precise and controlled maneuvering of the Mobetron in the OR with the Mobility system and efficient autodocking over the patient. The dedicated shielded QA room allows for all routine maintenance, including daily quality assurance, to be conducted during business hours by staying within the exposure limits allowed by the state.

Abbreviations

IORT

Intraoperative radiotherapy

OAR

Organs at risk

MCF

Mayo Clinic Florida

QA

Quality Assurance

kV

Kilo-voltage

MV

Mega-voltage

HAM

Harrison-Anderson-Mick

LINAC

Linear Accelerator

EBRT

External beam radiation therapy

WBI

Whole breast irradiation

Mobetron

MOBile ElecTRON beam linear accelerator

PDD

Percent Depth Dose

SMA

Superior Mesenteric Artery

SMV

Superior Mesenteric Vein

IDL

Isodose line

Author contributions

Deepak Shrestha: Conceptualization of work, data acquisition, data analysis, data interpretation, draft manuscript preparation, draft manuscript revision, final approval of manuscript and corresponding author. Amanda Jackson: Conceptualization of work, data acquisition, data analysis, data interpretation, draft manuscript revision, final approval of manuscript. Steven M. Herchko: Program service, patient treatment, data acquisition, draft manuscript revision. Michael Lowe: Program service, patient treatment, draft manuscript revision. Christopher Serago: Program service, patient treatment, draft manuscript revision. Jingjing M. Dougherty: Program service, patient treatment, draft manuscript revision. Deanna Hasenauer: Program service, patient treatment, draft manuscript revision. Mi Huang: Program service, patient treatment, draft manuscript revision. Ralph Young: Program service, patient treatment, draft manuscript revision. Robert Miller: Program service, patient treatment, draft manuscript revision. Chris J. Beltran: Program service, patient treatment, draft manuscript revision. Amit Merchea: Program service, patient treatment, IORT surgeon, draft manuscript revision. John Stauffer: Program service, patient treatment, IORT surgeon, draft manuscript revision. Michael Rutenberg: Program service, patient treatment, draft manuscript revision. Byron May: Conceptualization of work, program service, patient treatment, supervising the PACER trial, draft manuscript revision, final approval of manuscript.

Funding

The authors did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

Data availability

All data displayed in the manuscript can be shared by the corresponding author upon reasonable request.

Declarations

Ethics approval

Not applicable. This study did not involve human participants, animals, or human data, and thus independent ethical review was not required.

Consent to publish

Not applicable. This study did not include any human participants.

Competing interests

The authors declare no competing interests.