Around half of newly diagnosed cancer cases would benefit from treatment with radiation therapy. And hadron therapy, which uses proton or ion beams to deliver precision tumour targeting with zero exit dose, could improve therapeutic outcomes in 15–20% of these cases. But the number of potential patients — some 1.5 million per year at present — far outweighs the capacity of currently installed hadron therapy systems, which treat some 20,000 patients per year.

So why not just build more hadron therapy facilities? One obstacle is the vast size, weight and cost of current instruments. To maximize conformal irradiation of the tumour target, treatments require beam delivery from different directions. And as it’s neither practical nor comfortable to move the patient any great amount, this necessitates the use of either multiple beam lines or, more commonly, a rotating gantry.

Such gantries are complex structures that represent a considerable fraction of an installation’s cost and size. And for carbon ion therapy, there are only two gantries in the world. The one at the Heidelberg Ion-Beam Therapy Center in Germany is 25 m long and weighs over 600 tonnes. The other, in Chiba, Japan, is a superconducting gantry that is smaller and weighs 250 tonnes, but comes with the added challenge of a rotating cryogenic system.

But what if a new type of gantry could be created, one that doesn’t need to rotate and is small enough to fit on one room? That’s the goal of CERN scientist and magnet expert Luca Bottura. Along with PhD student Enrico Felcini and colleagues at CERN and EPFL, Bottura is working to create a lightweight and compact superconducting gantry for hadron therapy — called GaToroid.

“Current hadron therapy gantries are massive because they use large and heavy magnets of relatively low field, and require a very robust structure to rotate them precisely,” explains Bottura, speaking at a recent CERN Knowledge Transfer seminar. One way to reduce the size is to use superconducting magnets to increase the bending field in large bore magnets, an approach that’s being studied by several research teams.

A more radical idea is to devise a magnetic configuration that does not need to rotate in order to bend the treatment beam onto the patient. This will reduce the stability requirements on the gantry, and hence lower its mass and footprint. Such a configuration can be achieved using a toroidal magnet, and this is the idea underlying GaToroid.

“A toroidal field has axial symmetry and has the property of bending particles from any angular direction towards its centre,” Bottura explains. “This is exploited in GaToroid to direct beams from different angles towards the patient, without the need to rotate the magnet.”

System designs

In the GaToroid gantry, the hadron beam from the accelerator first passes through a vector magnet, which acts as an X-Y kicker to direct the beam at an appropriate angle into the toroidal magnet. The magnet comprises a series of superconducting toroidal coils and the beam is directed into one of the inter-coil spaces. The coils are designed with a graded winding, which shapes the field profile to help control the beam path.

“We shaped the coils to have a very large ‘acceptance’. This means that beams of different energy are directed towards the same point by making use of a large field volume, without the need to change the magnetic field,” says Bottura. “And because the field lines close inside the torus, it does not need a heavy iron yoke, and the space for the patient, in the bore of the torus, is field-free.”

As the gantry does not physically move nor change its field, it only needs the kicker magnet to direct the beam, and the magnetic field configuration does the rest. This enables rapid changes in the delivery direction (which can take several seconds for a classical gantry), as the speed of beam delivery is only governed by the speed of the kicker magnet and the accelerator itself.

To investigate the design’s potential, the researchers simulated tracking of single particles of 70–250 MeV directed onto the patient by the graded coil toroidal magnet. They observed excellent acceptance and isocentric properties. Repeating the simulation for clouds of particles with slightly different beam parameters (initial position, direction and energy) showed well superimposed input and output beam clouds at 70, 150 and 250 MeV.

The team also demonstrated a proof-of-principle of beam painting using the GaToroid. They achieved this by using different kick angles at the vector magnet to rapidly change the beam path in the transverse (±50mm) and sagittal (±60mm) directions.

Proposed specifications

At the seminar, Bottura shared some potential design specifications for the GaToroid gantries. A proton gantry, for example, would include 16 coils with a peak field on the coil of 8 T and an operating current of 1800 A. This gantry has a radius of 1.5 m, a length of about 6 m and an estimated mass of 12 tonnes. A carbon ion GaToroid would also have 16 coils, with a peak field of 13.8 T and an operating current of 6 kA. This gantry would be larger, with a radius of 2.5 m, a length of about 10 m and an estimated mass of 50 tonnes.

“The idea is on the path of being demonstrated rather than proven, but if you compare the proton solution to other systems, the size and mass reduction is remarkable. For the carbon gantry, the comparison is more striking in terms of mass,” says Bottura. “GaToroid will be at least two times smaller and ten times lighter. I think it has big potential and we need to explore it.”

The team is now half way through the three-year project. Last year they started work on the basic GaToroid design, addressing issues such as magnet design and geometry, beam tracking and the ability to paint a target, and then tackled the mechanical design.

“We are still focusing on technical challenges — the choice of material, mechanics and fabrication, beam optics and beam dynamics, quenches and thermal engineering — but we also started to think about a staged prototyping plan, and integrating beam and dose monitoring instrumentation,” says Bottura. “This will require careful planning, and collaboration with other institutes and companies that are expert in specific areas.”

Ultimately, a compact non-rotating gantry could prove a game-changer for ion therapy — where the size and complexity of the few existing ion gantries restrict their use as turnkey commercial solutions for ion treatment centres.

Bottura thinks that GaToroid could also prove of interest for proton therapy: “Because fully-rotating proton gantries represent a large portion of the cost and footprint of multi-room proton centres, and also because the high speed at which the beam direction can be changed will open up interesting possibilities for treatment,” he tells Physics World.

“This idea has a touch of insanity, but if it works, it could be a quantum step towards compact gantries and ease widespread application of hadron therapy,” says Bottura.