Designing a Multileaf Collimator for Proton Therapy
Reviewer: Christine Hill, MD
Abramson Cancer Center of the University of Pennsylvania
Last Modified: May 28, 2008
Presenter: C Ainsley, S Avery, R Maughan, J McDonough, J Metz, R Scheurermann, Z Tochner Presenter's Affiliation: Department of Radiation Oncology, University of Pennsylvania Type of Session: Poster
Multileaf collimator (MLC) systems are frequently used to deliver photon-based radiation, and allow conformal shaping of treatment beams. These devices are made up of individual leaves, composed of material of high atomic number, which move independently in and out of the path of the photon beam.
Multileaf collimators have largely replaced blocks which may be used during photon radiation delivery, and have become widespread throughout radiotherapy centers, allowing efficient and cost-effective delivery of conformal treatments.
Multileaf collimators are essential to the delivery of intensity modulation radiotherapy (IMRT), allowing modulation of photon fluence to increase dose sculpting and precision.
As particle beam-producing machines have been developed, interest has gained in development of MLC devices for use during proton and other particle beam radiation.
Many proton beam centers currently make use of aperture and snout systems, which involve use of a snout to shape and focus the proton beam, a brass aperture to modify field shape, and an acrylic compensator to modulate depth.
Benefits to use of an MLC system versus a snout/ aperture system during proton beam delivery are expected to be many:
Elimination of need for snout changes, which may be quite time consuming.
Reduction of required therapist labor associated with lifting heavy apertures.
Reduction of time and expense required for aperture creation/ machining.
Allowance of multiple fields per patient/ treatment session to be easily treated.
Allowance of field shape changes to provide more conformal target coverage.
This study was carried out with the intent of designing an MLC system as an alternative to a snout/ aperture system.
Materials and Methods
The intention of this study was to utilize MLC systems that are currently used during photon-based radiation as a model for MLCs compatible with protons.
The research group attempted to adapt current experience with conventional radiotherapy MLCs for use with protons.
To this end, they partnered with Varian Medical Systems and IBA in order to modify the current Varian MLCs and IBA nozzle to allow MLC use with a proton beam.
Clinical specifications were used to guide changes to these systems to increase compatibility; several iterations were performed to accommodate mechanical constraints on size and weight.
Additionally, activation and dose due to primary proton beam leakage and the presence of secondary neutrons were taken into account during these iterations.
Leakage dose was studied using the Monte Carlo code GEANT4 version-9.0, which was used to simulate the IBA nozzle.
Analytical calculations were used to study the effects of leaf material on activation.
After several iterations, the prototype designed consisted of 50 leaves per bank, each projecting to 0.5 cm at isocenter from a nominal nozzle position.
Leaves were made of a tungsten alloy.
Leaves were each 8 cm high and 11 cm long.
The central axis could be over-travelled by 1.5 cm.
In double scattering mode, the maximum field size was 16 cm by 16 cm.
In uniform scanning mode, when limitations of the scattering system were absent, the maximum field size was 25 cm by 18 cm.
The system was designed so that the MLC system could be retracted into the nozzle during raster scanning, allowing maximum field size of 29 cm by 39 cm.
The MLC system was able to move remotely, and retained the ability to contain apertures for cases when aperture use is deemed more appropriate.
The MLC system was able to rotate about the proton beam +/- 95 degrees.
Two compensators could be placed and held within the MLC system.
Neutron leakage was calculated to be less than or equal to that resulting from the standard IBA snout/ aperture system.
Activation was felt to be insignificant; this was the case because individual MLC leaves are largely self shielding.
Additional shielding structures were added based on the GEANT4 simulations to minimize primary and scatter proton leakage.
In this study, an MLC system was designed without necessitating significant change to the IBA nozzle design or the Varian mechanical or electronic MLC design.
This system may serve as an alternative to the standard IBA snout/ aperture system.
Prototype component testing is planned for the upcoming months.
As proton beam therapy becomes more widespread, focus on increasing patient accessibility to this resource is extremely important.
Through past experience with conventional radiotherapy, the benefits of MLC systems have been witnessed by radiation oncologists currently practicing; these include increased efficiency, increased cost-effectiveness, and increased dose conformality (Brewster, 1995).
Each of these factors offers important benefits to patients, allowing more patients per day to receive radiotherapy, and, in many cases, improving dose delivery to minimize toxicity.
Most current proton-based systems make use of snout and aperture systems, which can be quite cumbersome. These systems require manual snout changes, which can occupy large amounts of time, and require significant manual labor on the part of radiation therapists.
Additionally, snout/ aperture systems limit the number of fields per patient that can be treated each session, and do not allow the field shape to change during treatment.
Use of an MLC system for proton radiotherapy delivery would be expected to greatly improve efficiency of proton beam delivery. This is essential to providing maximal accessibility and cost-effectiveness.
Additionally, use of MLCs in place of snout/ aperture systems will likely greatly increase the flexibility of the proton radiotherapy system, allowing intensity and fluence modulation. These, in turn, allow field and dose changes to take place, providing more conformal target coverage.
The design, presented here, of an MLC system compatible with systems already in use is particularly appealing, as clinical implementation would be expected to be expedited greatly by this.
Concerns regarding proton and neutron leakage around the MLCs are valid, and have been addressed within this study. As the authors note, neutron leakage does not appear to be increased with use of MLCs versus the current snout/ aperture system. Use of necessary shielding structures was feasible to reduce proton leakage in this study, as indicated by Monte Carlo calculations.
Certainly, clinical testing of prototype designs presented here is essential, as accurate assessment of leakage and scatter will be paramount to safe patient care; however, the system presented here is promising, and will hopefully be utilized to allow increased patient accessibility, cost-effectiveness, and flexibility with regards to dose conformality.