© The Authors, published by EDP Sciences, 2024

Introduction

Cancer is one of China’s five high-priority disease areas, with patients experiencing a significant health burden and unmet needs, as emphasized in the Chinese government’s Healthy China Action Plan (2019–2030). Cancer treatment at Chinese hospitals is based on an integrative multidisciplinary approach that includes radiotherapy, chemotherapy, intervention, and surgery. Compared to conventional photon radiotherapy, proton, and heavy ion therapy offer superior relative biological effectiveness (RBE) in the treatment of deep-seated tumors due to their Bragg peak feature of energy deposition in organs. Currently, only five particle therapy institutes are treating patients in China, with dozens more under construction or conducting clinical trials. This demonstrates the potential demand for particle therapy in light of China’s aging population. Many cancer patients have benefited from particle therapy, which offers unique advantages over surgery and chemotherapy. By the end of 2020, nearly 8000 patients had been treated with particle therapy in China [1].

The latest generation of particle therapy is named intensity-modulated particle therapy (IMPT), which improves upon the pencil beam scanning technique for a conformal dose distribution. This technique sculpts doses to the tumor region while minimizing doses to the surrounding normal tissues. For this type of beam scanning technique, accelerators are required to produce ion beams with energies up to around 230 MeV (protons) and 430 MeV/nucleon. Additionally, fast beam energy and intensity modulation are critical for treatment accuracy and efficiency. The particle therapy system is a complex system, consisting of many subsystems such as the beam production system, beam transport system, and beam delivery system. Every subsystem is critical for the precision and safety of the treatment, with the accelerator (i.e. beam production system) being the core component.

Two types of circular accelerators have been widely used in particle therapy: the cyclotron and the synchrotron. Meanwhile, cyclotrons can be classified into two types: isochronous cyclotrons and synchrocyclotrons. These accelerators use oscillating radio frequency (RF) electric fields to accelerate electrically charged particles. Figure 1(a) indicates a growing interest and investment in cyclotrons and synchrotrons in particle therapy over the last 30 years, with occasional fluctuations in the numbers of each type of accelerator. Especially, there seems to be a high increase in the number of cyclotrons built or operational in recent years. This could reflect advancements in accelerator technology, such as the superconducting techniques that further reduce the cost, size, and weight of the facility.

Figure 1 (a) Number of particle therapy facilities in operation by accelerator type (cyclotron, synchrocyclotron, and synchrotron) worldwide during the past decades. (b) Number of particle therapy facilities in operation per country sorted by accelerator type. Data was extracted from the PTCOG website: https://www.ptcog.site/index.php/facilities-in-operation-public.

The United States focuses exclusively on proton therapy, whereas Japan provides both proton and carbon ion therapies, as shown in Figure 1(b). Both countries use cyclotrons and synchrotrons, but Japan has a higher prevalence of synchrotron facilities, particularly for carbon ion therapy. This indicates a broader adoption of advanced particle therapy technologies in Japan. Meanwhile, Europe has a diverse and well-established particle therapy infrastructure with a significant number of facilities across various countries.

Pompos et al. [2] estimated that the cost of developing a center with a capacity of 1000 patients per year is roughly twice as expensive as a proton center of the same size. Much of this cost comes from the complexity of the system, such as the need for a synchrotron and additional shielding. Vanderstraeten et al. [3] summarized the investment in building and equipment for proton and carbon ion facilities including all investment costs, personnel requirements, and other operational costs.

Peeters et al. [4] listed the cost of external beam radiotherapy with carbon ions, protons, and photons, as summarised in Table 1. The information suggests that both proton and carbon ion therapy facilities entail significant upfront capital costs, with carbon ion facilities being particularly expensive due to the complexity of the system. It should also be highlighted that the design and configuration of the particle therapy facility, particularly regarding the gantry system and number of treatment rooms, play a crucial role in determining the cost implications of the accelerator.

In the following sections, we discuss the differences between a cyclotron and a synchrotron-based particle therapy system in terms of ion source injection, particle acceleration, and beam extraction. We further explain the attributes of both accelerators, focusing on aspects such as beam quality, reproducibility, stability, and treatment efficiency.

Functionality of cyclotron and synchrotron

Both cyclotrons and synchrotrons utilize magnetic and electric fields to accelerate charged particles. Table 2 provides a comparison of the most crucial technical parameters for clinical cyclotron and synchrotron-based particle therapy systems. Cyclotrons, including isochronous and synchrocyclotrons, are the most commonly used accelerators in proton therapy. As shown in Videos 1 and 2, the IBA Proteus 235 and Varian ProBeam systems use a cyclotron and a superconducting cyclotron, respectively. The superconducting cyclotron is significantly smaller in size and weighs 60% less, with a weight of 90 tons compared to 220 tons. However, they have not been utilized in carbon therapy due to technical challenges. Cyclotrons designed to accelerate heavier ions like carbon are extremely large and heavy. For instance, the proposed designs weigh between 400 and 550 tons [5], presenting significant challenges in construction, installation, and maintenance. Meanwhile, cyclotrons require specialized ion sources that can produce ions with the appropriate charge to accelerate different ions. This adds to the complexity and cost of the system. Achieving sufficient turn separation at the extraction point is crucial and demands considerable design effort. Ensuring precise control over the ion beams further adds to the engineering challenges. Recent proposals for coupled cyclotrons, which include a larger accelerating stage weighing around 100 tons, suggest ongoing efforts to find more practical and manageable solutions. However, these designs are still in the development or proposal stage and may not be widely implemented yet [6].

For institutes offering both proton and carbon ion therapy, a synchrotron is currently the only viable option. The isochronous cyclotron maintains a fixed radio frequency (RF) during acceleration and employs an azimuthally varying field (AVF) to enable isochronism and vertical focusing. Conversely, the synchrocyclotron operates with a timing-varying RF frequency and a decreasing magnetic field, resulting in a pulsed model with vertical focusing. A synchrotron, on the other hand, features a varying bending field, varying accelerating RF frequency, and varying energy in circulation. All three variables are synchronized to ensure stable beam extraction.

Indeed, the timing structure of the beam for these accelerators is quite distinct. Unlike cyclotrons that provide beams in a continuous wave (CW) model at a fixed energy, synchrotrons produce pulsed beams (typically at 1 kHz, with a duty cycle of around 1%) at varying energies throughout the acceleration cycle. A CW beam has an output that lacks macrostructure but includes microstructure on the nanosecond scale. This results in a nominally constant output over intervals of seconds or longer. Even though there may be tiny fluctuations (microstructure) happening in nanoseconds, the overall effect is a smooth and constant flow of particles over a longer period, like several seconds, which ensures a stable and predictable dose of particles to the target area.

In contrast, a synchrotron generates beams in a spilled structure (typically at 0.5–0.1 Hz) with a dead time exceeding 10%. In clinical practice, the continuous beam from the cyclotron can be utilized for fast scanning to minimize the interplay effect. This is a challenge in synchrotron-based particle therapy institutes due to their pulsed beam structure and extended dead time. Synchrotron performance strives for increased current output and control, as well as faster energy-switching time. The synchrotron-based system’s beam may face challenges in delivering the beam effectively or precisely, especially under respiratory-gated modes.

The beam intensity can be adjusted on a millisecond scale for cyclotrons, and the intensity stability is within 5%, ensuring reproducibility and efficiency for treatment. However, synchrotrons exhibit a relatively large variation in beam intensity of 10–20% and limited intensity modulation in magnitude. Regarding energy modulation, the extracted beam from a cyclotron has a fixed maximum energy of 250 MeV, which can be varied via a degrader combined with an energy selection system to the clinical energy range in IMPT. On the other hand, the synchrotron adjusts the energy during circulation. Consequently, a cyclotron-based particle therapy system requires additional shielding near the degrader and radiation protection instructions for radiation workers.

Table A1 in Appendix 1 provides a comprehensive overview of the parameters for existing PT systems in China. These include systems from overseas vendors, such as IBA, Varian, MeVion, Siemens, and Hitachi. The table also includes some domestically developed synchrotron-based PT systems that are currently in use. One of these was developed in Lanzhou by the Institute of Modern Physics (IMP) and is used exclusively for carbon ions. The other system, developed in Shanghai Rujin Hospital by the Shanghai Institute of Applied Physics (SINAP), is used solely for protons. The synchrotrons developed by Siemens and Hitachi are capable of providing both high-energy protons and carbon ions. In contrast, other cyclotron-based PT systems can only provide protons.

Most proton therapy systems offer a similar energy range of approximately 70–230 MeV for protons, corresponding to a water-equivalent depth of around 4 to 30–35 cm. The maximum carbon ion energy can be around 480 MeV/u for a synchrotron developed by Hitachi, meeting most of the clinical requirements with a maximum energy of 400 MeV/u. The clinical dose rate for most PT systems is around 1–2 Gy/min/Liter, corresponding to around 2 nA at the isocenter. The definition of dose rate in photon and particle therapy is slightly different. In PT, the scanning mode delivers precise doses (spots) directly to the target, eliminating the need for a collimation system to scatter the beam, as required in photon and electron therapy. Therefore, the dose rate of 2 Gy/min/Liter represents an average estimation of the energy deposited on a physical target during spot scanning. In the case of a synchrotron-based system, the maximum extraction intensity per spill is on the magnitude of 10¹¹ for protons and 10⁹ for carbon ions, such as the Hitachi system as listed in Table A1. However, for a cyclotron-based system, the maximum beam intensity can be a few hundred nA (e.g., 300 nA for IBA C235 and 800 nA for Varian ProBeam). Even though most of the beam intensity is significantly lost when the beam passes through the energy selection system, the average beam intensity at the isocenter is only a few nA.

Indeed, a synchrotron generally has a much larger size than a cyclotron, which can increase the construction cost. As indicated in Table A1, the circumference of the synchrotron at the Shanghai Proton and Heavy Ion Centre is around 64.8 m, while the diameter for a superconducting (SC) cyclotron is only 1.8 m for the MeVion Hyperscan system. The world’s most compact synchrotron, developed by Hitachi, has a circumference of 18 m [7]. Most of the cyclotrons in Proton Therapy (PT) applications have an internal cold cathode Penning ion source (PIG), while synchrotrons require an additional accelerator (normally a LINAC) to pre-accelerate the protons to a few MeV to overcome the residual magnetism on the synchrotron. This makes the cyclotron more compact and integrated compared with the synchrotron. The Radio Frequency (RF) in a synchrotron increases simultaneously (i.e., from 1 to 8 MHz for protons) with the beam energy/speed, as does the magnetic field. On the other hand, in an isochronous cyclotron, the RF is fixed with an azimuthally varying magnetic field of 2.4 and 4 T at the central and extraction positions in the Varian ProBeam system.

In terms of beam extraction, most synchrotron systems listed in Table A1 utilize RF-knockout slow extraction, while the isochronous cyclotron employs electrostatic deflectors to peel off the beam on the last orbit (as seen in the Varian ProBeam system and IBA C235 system). Fast beam ON/OFF technology (less than 0.15 s) plays a crucial role in discrete spot scanning. The beam energy switching time in a cyclotron is ten times faster compared with that in a synchrotron in the case of the Single Energy Extraction (SEE) technique. However, the recently developed Multi-Energy Extraction (MEE) technique improves that to the same magnitude as a cyclotron (i.e., around 0.2 s). This is one of the most important factors regarding treatment efficiency. In the following section, we will delve into ion source injection, beam acceleration, extraction, and energy/intensity modulation for both types of accelerators, discussing the technical complexity in a clinical particle therapy institute.

Ion source and beam injection

Figure 2 shows a schematic drawing of a cyclotron with the magnetic and oscillating electric fields (left side) and some key components of the cyclotron (right side). The ion source of a cyclotron is located in the central region, where particles are injected into the electric field for acceleration. At the same time, the particles are bent by the magnetic field on the Dee, reaching their maximum energy (250 MeV) at the largest radius of the cyclotron for beam extraction.

Figure 2 Acceleration of protons in a cyclotron. Both a magnetic field and an oscillating electric field are applied in the cyclotron to bend and accelerate the proton bunches. The key components of the cyclotron are demonstrated on the right side of the figure [14].

The PIG ion source is the most commonly used in cyclotrons. Figure 3 shows a schematic view of the cold cathode PIG in the Varian ProBeam system, which is the core of the cyclotron. Two cathodes that placed at each end of a cylindrical anode. The cathodes are at negative potential relative to the anode (the chimney), emitting electrons that are needed to ionize the hydrogen gas and create plasma. The ion source is located in the center of the cyclotron, where hydrogen gas is introduced and dissociated into ions via a high-voltage (HV) plasma. As shown in the figure, there are two cathodes at negative HV on the top and bottom of the PIG source. A homogeneous magnetic field is also applied in the vertical direction. Hydrogen gas is introduced into the ion source and discharged by the electric field into plasma (hydrogen ions, protons) in the upper and lower halves of the source. The hydrogen ions (protons) move up and down the chimney in the magnetic field, while electrons in the PIG make small circles up and down in the ion source due to the magnetic and electric fields. In the middle of the chimney, there is a voltage difference between the puller nose in the PIG and Dees (110 kV and 80 kV for the puller and Dees, respectively). These voltages pull protons from the chimney and accelerate them to their final energy.

Figure 3 Schematic view of the cold cathode penning ion source in the Varian ProBeam system.

In contrast, a synchrotron is composed of a ring of small magnets, including dipole, quadrupole, and sextupole magnets, as shown in Figure 4. The Hitachi synchrotron has a circumference of 60 m and maximum beam energies of 480 MeV/u for carbon and 250 MeV for protons. The synchrotron consists of a beam injection system (bump magnets and injector), acceleration (RF cavity), and extraction (bump magnets, transverse RF, deflector, and septum magnets). Unlike the PIG source in a cyclotron, protons are injected into the ring and begin traveling around it about 10 million times per second. An RF cavity within the ring increases the energy of the protons each time they travel around the ring. By varying the magnetic and electric fields, synchrotrons can produce protons of various energies.

Figure 4 Schematic view of Hitachi synchrotron. DM: Dipole magnet; QF, QD: Quadrupole magnet; SF, SD: Sextupole magnet; STM: Steering magnet; IBump/EBump: Injection/Extraction bump magnets [15].

A linear accelerator (LINAC) is used to pre-accelerate proton or carbon beams before they are injected into the synchrotron using the multi-turn injection method. This accelerates proton beams generated by an ion source up to 7 MeV with about 10 mA. Hitachi has developed a compact microwave proton source for LINAC pre-acceleration. Figure 5(a) shows the ion source generator with a high-frequency microwave (2.45 GHz), which improves the manageability and maintainability of the ion source. A permanent magnet case is used to make the microwave ion source compact and a stable proton current of more than 20 mA can be generated for acceleration. Additionally, the commonly used Electron Cyclotron Resonance (ECR) ion source has a typical H⁺ output current of 25 mA, and no maintenance is necessary except for the replacement of the magnetron tube and the H² gas bottle.

Figure 5 (a) Compact microwave proton source of a Hitachi synchrotron. (b) A spill of the synchrotron operation pattern including the beam injection, acceleration, extraction, and deceleration [16].

Beam extraction

Extraction is the process of beam transfer from an internal orbit to a target that is placed outside of the magnetic field. Synchrocyclotron: regenerative extraction is used; the beam is also extracted during resonance build-up. The resonance is driven by a second-harmonic gradient bump of the field. The most commonly used beam extraction methods include stripping, resonant extraction using a deflector, regenerative extraction, and self-extraction. Beam extraction describes the final round of protons within the cyclotron. For example, in the Varian ProBeam system, there are two extraction deflectors, some permanent magnets, and some steering magnets at the exit of the cyclotron. The front of the deflector is a thin metal sheet called the septum, and the electric field only exists within the deflector. Protons passing the septum on the right side stay inside the cyclotron for another run, while those passing the septum’s front edge on the left side are extracted from the cyclotron. The deflectors are charged up to 55 kV. To maintain these high electric fields, the main magnet must be on, allowing arcs to travel vertically but not horizontally. Meanwhile, a small amount of oxygen is used in the deflectors to keep the extraction system stable.

RF-driven slow extraction technology is used to extract beams from a synchrotron. It involves the excitation of a 3rd-order resonance, forming a triangular stable phase space area called a separatrix. The optical elements remain constant during extraction. For spot scanning in particle therapy, researchers from Chiba, Japan [8] have developed a dynamic beam intensity control system intended for use in three-dimensional pencil-beam scanning. This control system, named RF-knockout slow-extraction, controls the spill structure (see Fig. 5(b)) and intensities of the beams extracted from the synchrotron. The amplitude modulation (AM) function of the transverse RF field for extraction is optimized to control the spill structure and beam intensity. In this case, the flat spill (constant current) and intensity control allow precise dose modulation over a large dynamic range.

Energy and intensity modulation

For cancer treatment, the spot scanning technique is the most advanced technique for both types of accelerators. Fast beam energy and intensity modulation are essential in clinical practice since the dose is accumulated layer by layer and the intensity for each spot can be varied for intensity-modulated proton therapy (IMPT). The design of these two types of accelerators for IMPT is completely different due to their intrinsic differences. In the case of a cyclotron-based particle therapy system, such as the Varian ProBeam system, beam intensity modulation is controlled by a component called the vertical deflector in the inner region of the cyclotron when protons are just pulled out from the chimney. As shown in Figure 6(a), the vertical deflector consists of two parts: the vertical deflection plates (with two electrodes) and the vertical slit collimator on the opposite side of the two plates. The electric field bends the protons, which are stopped by the vertical collimator for intensity variation. This means that the higher the voltage on the vertical deflector (VD), the lower the extracted beam intensity will be. The advantage of this design is that the beam intensity can be adjusted on a millisecond scale for fast scanning to minimize interplay effects in clinical practice (i.e., treating tumor sites with motion).

Figure 6 (a) The beam intensity from the PSI cyclotron as a function of the voltage on a vertically deflecting electrode pair [17]. (b) The multi-wedge energy degrader for energy modulation in the Varian ProBeam system.

When the beam is extracted by the extraction system, its energy reaches its maximum (250 MeV) and is then modulated to the clinical energy range (around 70–230 MeV) before being transported to the treatment room. A multi-wedge degrader (carbon) with fast variation in thickness is used for energy degradation. As the beam passes through the energy degrader, its emittance increases significantly and is then collimated by the size aperture, halo aperture, and divergence aperture, as shown in Figure 6(b). The beam emittance is nearly within clinical acceptance after passing through the divergence aperture. The intensity of the proton beams is significantly decreased in the energy selection system (losing almost 99%), which means that the transmission rate of the proton beam in a cyclotron-based system is very low and causes activation of the energy selection system. Extra shielding and radiation protection measures are necessary for the safety of proton engineers.

One of the most obvious differences between a cyclotron and a synchrotron is that an energy degrader is not needed for a synchrotron, meaning that energy modulation is achieved inside the circulation. As previously discussed, particles are filled into the synchrotron by the LINAC with a minimum energy of typically 7 MeV. Around 10⁸–10¹¹ protons are filled into the ring and accelerated to the desired energy using an RF cavity. Once the particles have been accelerated to the desired energy, they are stored in the ring for slow extraction by an RF kicker.

When the desired number of particles has been extracted, the remaining particles in the ring are deposited on a beam dump. One acceleration cycle or spill, which includes beam injection, acceleration, slow extraction, and deceleration/dump, lasts for around 0.5–5 s. Normally, the beam energy is the same within one spill and can only be varied until the next spill. Figure 7(a) shows the Single Energy Extraction (SEE) technique, where only one proton energy can be extracted for each of the four accelerator cycles/spills. Recently, Hitachi has developed a novel technique for energy modulation called Multiple Energy Extraction (MEE) (see Fig. 7(b)), during which protons of four different energies can be extracted in the same spill [9]. The MEE energy switching time is much shorter than that of SEE, improving delivery efficiency for cancer treatment. Thus, compared to a cyclotron-based system, an energy degrader is not needed in a synchrotron-based system, reducing activation. The beam transmission rate of a synchrotron is much higher than that of a cyclotron since fewer beams are lost in the synchrotron.

Figure 7 Four energy layers are delivered in (a) single energy extraction (SEE) and (b) multiple energy extraction (MEE) [9].

Overall, the synchrotron provides adjustable energy, can produce high-energy ions (protons and carbon ions), and has a high beam transmission rate, resulting in low radioactivity. However, the beam intensity is noisy (unstable) and limited by ring filling, and the dead time between spills limits the average dose rate. The beam emittance is also larger compared to a cyclotron-produced beam. On the other hand, the cyclotron provides continuous beams and fast intensity modulation using the vertical deflector (VD). The beam intensity has great reliability and the spot size is small. An energy degrader on the beam transport line is needed for energy variation in a cyclotron-based system, causing activation near the degrader.

Discussions and conclusions

This report presents a comparative analysis of two types of accelerators, predominantly utilized in high-energy particle therapy. The cyclotron emerges as a cost-effective option in PT applications due to its compact size and ability to readily provide adequate beam energy and intensity. The advent of superconducting techniques has enabled the cyclotron to be highly compact, thereby reducing both construction costs and power consumption. In terms of daily setup, minor adjustments to a few variables can optimize the cyclotron’s status. These variables include the current of the main coil, the voltage of the RF system, and the arc current of the ion source. It is recommended that the Smith-Garren plot (a plot representing magnet current, phase, and extraction current) be conducted each morning to determine the optimal phase for extraction efficiency and beam stability. Furthermore, periodic maintenance is required for the cyclotron to maintain the ion source in an optimal state. This maintenance, which involves replacing certain components of the PIG ion source, is typically conducted biweekly.

In contrast, the synchrotron is significantly larger than the cyclotron, necessitating an additional accelerator for ion pre-acceleration and beam injection. The ECR ion source of a synchrotron requires periodic maintenance either annually or semiannually, depending on the components that need replacement. It is crucial to consider not only the construction and equipment costs but also the recurring maintenance costs for the widespread adoption of PT. In this context, single-room PT systems such as the IBA S2C2 and MeVion Hyperscan offer advantages in terms of space-saving design. The daily setup and periodic maintenance of the synchrocyclotron are relatively simple. Given China’s large and aging population, the demand for PT is high. Therefore, reducing the costs associated with PT systems can significantly benefit the nation’s citizens.

Treatment efficiency is indeed a crucial factor for clinical applications. The cyclotron, with its continuous beam, is well-suited for fast scanning of moving targets. Moreover, the energy switching time for a cyclotron-based PT system is approximately 0.2 s, which is ten times faster than the conventional synchrotron-based PT system with the Single Energy SEE technique. It’s worth noting that the MEE technique has not been widely adopted in clinical practice. Most PT systems support respiratory gating for motion management during beam irradiation. The continuous beam from a cyclotron can be easily synchronized with respiration to minimize the interplay effect during treatment. However, the dead time for a synchrotron-based beam model can significantly reduce treatment efficiency when gating is used. For scanning carbon ion therapy [10], have developed an innovative respiratory guidance method that synchronizes with the synchrotron’s flat-top phase to enhance overall treatment precision and efficiency.

Regarding the clinical implications, continuous beams provided by cyclotrons offer advantages in terms of beam stability and dose delivery efficiency. They are well-suited for treating static targets and more efficient for moving targets with rescanning technique. Pulsed beams from synchrotrons may require additional strategies to manage motion and ensure precise dose delivery, particularly for moving targets. However, advancements such as dynamic extraction control in synchrotrons help mitigate these challenges. Synchrotron’s RF-knockout slow extraction allows for more precise control over the beam, which is crucial for sparing organs at risk (OAR) and treating moving targets. The use of energy degraders in cyclotrons allows for the adjustment of beam energy, which also introduces beam spreading. However, advancements in energy selection systems help mitigate these effects, ensuring optimal treatment delivery and clinical outcomes.

Differences in beam modulation, extraction techniques, beam stability, and beam type (continuous vs. pulsed) between cyclotrons and synchrotrons have clinical implications for treatment effectiveness, patient comfort, and potential side effects in particle therapy. Understanding these differences is crucial for optimizing treatment planning and delivery to achieve the best possible outcomes for patients.

Looking ahead, Rossi et al. (2022) present a Eurpoean Collaboration to investigate superconducting magnets for next-generation heavy ion therapy). This collaborative effort aims to advance the design and technology of superconducting magnets for ion therapy (synchrotron and gantry), in the frame of the Eu-ropean H2020 HITRIplus programmes. Norman et al (2022) introduce the next Ion Medical Machine Study (NIMMS) for investigating the feasibility of a compact superconducting synchrotron for heavy ion therapy. In Japan, the quantum Scalpel Project Iwata et al. (2023) focuses on developing a compact superconducting accelerator to promote the widespread use of heavy-ion therapy and maximize its therapeutic potential. Normandy Hadrontherapy (NHa) is developing, in collaboration with Ion Beam Applications (IBA), a full hadrontherapy treatment solution based on a new multi-particle cyclotron [11] and [12]. The NHa C400 Cyclotron delivers high dose rates of alpha particles to carbon ions at 400 MeV/u and protons at 260 MeV.

There is a growing trend towards the use of ultra-high dose rate (>40 Gy/s) FLASH Proton Therapy (PT), which maintains the treatment efficacy on tumors while reducing side effects on organs at risk. Although the biological mechanism underlying the FLASH effects has not been fully elucidated, clinical observations from animal experiments have demonstrated the protection of normal tissue in pre-clinical studies Vozenin et al. (2019) and Sørensen et al. (2022). However, the primary challenge of FLASH PT lies in providing sufficient beam current at the isocenter. The cyclotron can readily extract a maximum beam current of 800 nA (as in Varian ProBeam) for a maximum beam energy of 250 MeV. For low-energy beams, energy degraders cause substantial beam loss. Maradia et al. [13] from PSI has improved the transmission rate of low-energy proton beams by modifying beam optics. The ultra-high dose rate scanning beam technique holds promise for future PT, significantly improving treatment efficiency, precision, and therapeutic effects. However, the application of a low-cost PT system and ultra-high dose rate still necessitates further development of accelerator and beam delivery techniques, which will determine the widespread adoption of PT.

Acknowledgments

The author would like to thank Zhijie Huang and Yanrong Li for providing information about the synchrotron system at the Shanghai Proton and Heavy Ion Center.

Funding

This research received no external funding.

Conflicts of interest

The authors declare no conflict of interest.

Data availability statement

Data availability is not applicable to this article as no new data were created or analysed in this study.

Author contribution statement

The co-author Bing Liu has helped for video filming and the other co-authors provided information about the Varian ProBeam system.

Ethics approval

Ethical approval is not applicable for this article.

References

Appendix

All Tables

All Figures

	Figure 1 (a) Number of particle therapy facilities in operation by accelerator type (cyclotron, synchrocyclotron, and synchrotron) worldwide during the past decades. (b) Number of particle therapy facilities in operation per country sorted by accelerator type. Data was extracted from the PTCOG website: https://www.ptcog.site/index.php/facilities-in-operation-public.
In the text

	Figure 2 Acceleration of protons in a cyclotron. Both a magnetic field and an oscillating electric field are applied in the cyclotron to bend and accelerate the proton bunches. The key components of the cyclotron are demonstrated on the right side of the figure [14].
In the text

	Figure 3 Schematic view of the cold cathode penning ion source in the Varian ProBeam system.
In the text

	Figure 4 Schematic view of Hitachi synchrotron. DM: Dipole magnet; QF, QD: Quadrupole magnet; SF, SD: Sextupole magnet; STM: Steering magnet; IBump/EBump: Injection/Extraction bump magnets [15].
In the text

	Figure 5 (a) Compact microwave proton source of a Hitachi synchrotron. (b) A spill of the synchrotron operation pattern including the beam injection, acceleration, extraction, and deceleration [16].
In the text

	Figure 6 (a) The beam intensity from the PSI cyclotron as a function of the voltage on a vertically deflecting electrode pair [17]. (b) The multi-wedge energy degrader for energy modulation in the Varian ProBeam system.
In the text

	Figure 7 Four energy layers are delivered in (a) single energy extraction (SEE) and (b) multiple energy extraction (MEE) [9].
In the text

All Movies