Development of a Proton Therapy Research Beamline with FLASH and Minibeam Capabilities at the 18 MeV Bern Medical Cyclotron

This paper reports the successful adaptation of the 18 MeV Bern Medical Cyclotron's beamline into a flexible research platform capable of delivering both conventional and FLASH proton beams with spatially fractionated minibeam capabilities, thereby enabling systematic pre-clinical radiobiology studies to optimize emerging radiotherapy modalities.

The Gist

Imagine you have a very powerful, high-speed water hose (a particle accelerator) that usually shoots water so hard it can only be used for heavy industrial cleaning. But scientists want to use this hose to water delicate flowers (living cells) in a very specific, gentle way to study how plants react to different watering schedules.

This paper describes how a team at the University of Bern took their existing "water hose" (a medical cyclotron) and built a special attachment system to turn it into a precise gardening tool. They wanted to test two new, cutting-edge ways of "watering" (treating) cells:

1. The "Flash" Method: Instead of a slow drip, they wanted to blast the cells with a massive amount of water in a split second.
2. The "Grid" Method: Instead of a solid sheet of water, they wanted to shoot the water through a sieve, creating a pattern of tiny, separate streams (minibeams) with gaps in between.

Here is how they did it and what they found, explained simply:

1. Taming the Beast (The Setup)

The machine they used is a cyclotron, which usually shoots protons (tiny particles) at 18 million electron volts. This is like a bullet. To make it safe for delicate cell experiments, they had to slow it down and shape it.

• The Scatterer (The Fan): They put a thin sheet of aluminum in the path of the beam. Think of this like putting a fan in front of a firehose. It spreads the tight, powerful stream out into a wide, soft mist. This made the beam cover a larger area and made it much more even, like a gentle rain rather than a jagged spray.
• The Chopper Wheel (The Dimmer Switch): To get the "Flash" effect, they couldn't just turn the hose on full blast. They built a spinning wheel with a slit in it. As the wheel spins, the slit lets the beam through for a tiny fraction of a second, then blocks it. By changing how fast the wheel spins or how wide the slit is, they could control the dose from a slow drip (conventional therapy) to a massive, instant blast (FLASH).

2. Measuring the Water (Dosimetry)

You can't just guess how much water hit the flower; you need a ruler. In this experiment, the "ruler" was a special film (like a high-tech photographic paper) that changes color when hit by radiation.

• The Problem: This film is tricky. When hit by slow-moving protons (which are heavy and stop quickly), the film gets "confused" and doesn't change color as much as it should. It's like a sponge that gets so full of water in one spot that it can't absorb any more, even if you keep pouring.
• The Fix: The team did a lot of math and extra tests to figure out exactly how much to "correct" the film's reading. They realized that because the protons lose energy as they pass through the plastic walls of the cell flask, they hit the film with different "oomph" than expected. They created a formula to fix this, allowing them to know the exact dose the cells received.

3. The Grid Test (Minibeams)

For the "Grid" method, they used a metal plate with tiny holes cut into it (like a stencil). They wanted to see if they could keep the pattern sharp even if the cells weren't touching the stencil.

• The Result: They found that if you move the cells even a tiny bit away from the stencil (like holding a stencil a few millimeters away from a wall), the sharp lines of the water start to blur together. The "valleys" (the dry spots) start getting wet because the water sprays sideways in the air.
• The Lesson: To keep the grid pattern perfect, the stencil must be held very close to the target, and the distance must be exact. If the distance varies, the pattern changes, which could change the biological results.

4. What They Achieved

The team successfully built a system that can:

• Shoot protons at speeds ranging from a slow drip to a super-fast flash.
• Create a wide, even field of radiation (about 20mm wide) that is very consistent.
• Create sharp, grid-like patterns of radiation (minibeams) for studying spatially fractionated therapy.

They proved that this setup works for testing how cells react to these new, experimental radiation styles. They also highlighted that measuring the dose accurately is hard because the protons are moving slowly, but they figured out a way to do it correctly for their specific setup.

In short: They took a heavy-duty industrial machine, added a fan, a spinning shutter, and a stencil, and turned it into a precise scientific tool. They showed it can deliver radiation in both "slow and steady" and "super-fast flash" modes, and they figured out how to measure exactly how much radiation the cells got, paving the way for future biological studies.

Technical Summary

Technical Summary: Development of a Proton Therapy Research Beamline with FLASH and Minibeam Capabilities at the 18 MeV Bern Medical Cyclotron

Problem Statement
Emerging radiotherapy modalities, specifically ultra-high dose rate (FLASH) irradiation and Spatially Fractionated Radiation Therapy (SFRT), show promise for improving therapeutic ratios, yet their biological mechanisms and optimal delivery parameters remain uncertain. Progress in this field requires accessible research platforms capable of flexible temporal and spatial dose delivery. While clinical proton facilities are often optimized for therapy rather than systematic radiobiology, radiopharmaceutical production cyclotrons offer high-current proton sources suitable for FLASH studies. However, adapting such facilities for precise, low-energy proton radiobiology research—particularly for in-vitro and in-vivo studies requiring specific energy ranges and spatial fractionation—presents challenges in beam shaping, dosimetry, and dose-rate control.

Methodology
The authors adapted the Beam Transfer Line (BTL) of the Bern Medical Cyclotron (BMC), an 18 MeV IBA Cyclone 18/18 HC cyclotron, to create a flexible research platform. The experimental setup involves:

• Beam Shaping: A passive scattering system was implemented to reduce the extracted proton flux from >200 Gy/s to conventional levels (0.01–0.05 Gy/s) while improving field uniformity. This system utilizes a 1 mm to 35 mm pinhole collimator, a 350 µm aluminum scatterer, and a 137 cm extended drift space to generate a broad, uniform Gaussian distribution.
• FLASH Delivery: For ultra-high dose rates, a remotely controlled beam chopper wheel (CW) was developed. The CW, driven by a stepper motor, modulates the beam by rotating a slit aperture through the beam path. It operates at 0.2–4 Hz, enabling pulse durations from 0.5 s down to ~1 ms, covering dose rates from 0.01 to 2200 Gy/s.
• SFRT Implementation: The low proton energy (<20 MeV) allows for the use of thin metallic collimators (e.g., 0.5 mm tungsten) to create minibeam arrays. Wire electrical discharge machining (WEDM) was used to create apertures as small as 200 µm.
• Dosimetry Framework: A comprehensive dosimetric system was established using an in-beam ionization chamber (IC) for real-time current monitoring and Gafchromic EBT-3 films for spatial dose mapping.
  • Calibration: The IC was calibrated against film measurements and a Faraday Cup (FC) to link beam current to absorbed dose.
  • LET Corrections: Recognizing that EBT-3 films exhibit LET-dependent quenching at low energies (near the Bragg peak), the authors implemented corrections for both the "lamination layer effect" (energy loss in film packaging) and "relative efficiency" (underresponse to high-LET protons compared to photons). These corrections were derived using LISE++ simulations and experimental film measurements.
• Targeting: A two-axis positioning stage with 100 µm increments aligns targets (e.g., cell culture flasks) to the beam center.

Key Results

• Beam Uniformity: Passive scattering significantly improved beam uniformity compared to magnetic defocusing alone. Within a 20 mm radius of the beam center, the standard deviation of the delivered dose was 7.96% (scattered) versus 8.84% (unscattered). The scattered beam profile was also found to be more stable against short-term drifts caused by cyclotron temperature fluctuations.
• Dose Rate Tunability: The system demonstrated continuous dose-rate tunability over five orders of magnitude, from conventional levels (0.01 Gy/s) well beyond the FLASH threshold (40 Gy/s) up to ~2200 Gy/s. A linear relationship was established between the Faraday Cup current and the dose rate at the target for various collimator apertures (1, 4, 10, and 35 mm).
• Dosimetric Accuracy: The study confirmed the necessity of LET-dependent corrections for quantitative dosimetry at these energies. The combined correction factor for the film response in the specific in-vitro configuration (proton energy ~8.14 MeV at the cell layer) was determined to be 1.08(6). The overall uncertainty in the dose delivered to cells in a flask was estimated at 9.46%.
• SFRT Performance: Minibeam structures were successfully generated using tungsten collimators. Measurements showed that while peak-to-valley dose ratios (PVDR) remained distinguishable at target distances of 0, 2, and 4 mm, the PVDR decreased with increasing distance due to lateral proton scattering in air. Finer grid spacings (e.g., 50/200 µm) were more sensitive to these geometric variations than broader spacings (e.g., 500/1000 µm).

Significance and Claims
The paper claims to present a "flexible and accessible platform" for systematic pre-clinical proton radiobiology studies. The primary significance lies in the successful adaptation of a medical cyclotron, typically used for radionuclide production, for complex radiobiological research requiring both FLASH and SFRT capabilities.

Key claims include:

1. Accessibility: The facility bridges accelerator technology and radiobiology by providing a platform where dose rates can be precisely controlled from conventional to FLASH regimes.
2. Feasibility of Low-Energy SFRT: The low energy of the extracted beam (15.54 MeV in air, ~8.14 MeV at the target) facilitates compact SFRT implementation using thin collimators, allowing for well-resolved minibeams.
3. Dosimetric Framework: The authors established a preliminary but validated dosimetric framework that accounts for the specific challenges of low-energy proton dosimetry (LET quenching and energy loss), enabling quantitative dose assessment for in-vitro setups.
4. Stability: The passive scattering approach offers improved stability and uniformity compared to magnetic defocusing, which is critical for reproducible biological experiments.

The authors conclude that this adapted beamline supports the optimization of emerging modalities like proton FLASH and SFRT. They note that while the current setup is a proof-of-concept with specific limitations (e.g., geometry-dependent correction factors), it provides a controlled framework for future investigations into the interplay between LET, dose rate, and spatial fractionation. Future work will focus on improving beam shaping, reducing dosimetric uncertainties through systematic characterization, and expanding to in-vivo irradiation setups.