While proton therapy is becoming a standard treatment option in radiation oncology – there are currently 92 operational proton facilities worldwide and a further 45 under construction – many challenges remain in terms of the fundamental physics, radiobiology and clinical use of protons for the treatment of cancer. Those challenges, and plenty more besides, were front-and-centre at the UK’s Fifth Annual Proton Therapy Physics Workshop held early in February at the National Physical Laboratory (NPL) in Teddington.
The timing of this year’s event was apposite. In 2018, the UK’s National Health Service (NHS) opened its first high-energy proton therapy centre at The Christie Hospital in Manchester, with a second facility at University College London Hospital (UCLH) scheduled to come online for patient treatment in 2020. Proton Partners International, a private health provider, is also rolling out a network of four proton-therapy facilities across the UK, with the first of its Rutherford Cancer Centres now treating patients in south Wales.
That upsurge in UK activity – spanning construction, commissioning and clinical go-live of new proton facilities – is being supported by the Proton Physics Research and Implementation Group (PPRIG), a consortium of “interested organizations” that includes NPL, The Christie, Clatterbridge Cancer Centre, University Hospitals Birmingham, UCL and UCLH.
“PPRIG was set up by NPL in 2012 to progress UK deployment of high-energy proton therapy,” explained Russell Thomas, senior research and clinical scientist at NPL and chair of PPRIG. “We aim to help coordinate research activities, encourage multicentre collaboration and minimize duplication of effort. Our annual proton-therapy physics workshop is a logical extension of PPRIG’s remit, bringing the UK proton-physics community together with leading researchers from overseas.”
Another objective of PPRIG is to promote the work of early-career researchers, helping them to build networks, collaborate on specific research problems, and support their grant applications. “The workshop fosters open, robust, but always good-humoured debate around the hot topics in proton therapy,” Thomas added.
Ana Lourenço, a postdoctoral research scientist at UCL and NPL, agrees that the PPRIG meeting provides a welcoming platform for younger scientists. “The PPRIG workshop was a great opportunity for early-career scientists to present their work and have feedback from world-leading medical physicists and clinical scientists,” she said. “The reduced registration fee for students allowed many to participate, with plenty of time in the programme dedicated to more open, informal discussion to facilitate the interaction between students and senior researchers.”
Standardize and verify
Given NPL’s role as the UK’s national measurement institute, much of the discussion at last week’s meeting eddied around issues of dosimetry and quality assurance (QA). In other words, how to maximize clinical outcomes by ensuring that patients receive standardized and rigorously audited proton therapy – irrespective of where they’re being treated.
With those outcomes in mind, Thomas reported on NPL’s work to develop a code of practice for proton-beam dosimetry in collaboration with the UK Institute of Physics and Engineering in Medicine (IPEM). Underpinning that code is absolute dosimetry for calibration of the proton beam, something that NPL is striving to improve through the development of a primary standard for protons based on a portable graphite calorimeter, in which the temperature rise due to a typical patient dose is measured to quantify the amount of “dose” deposited.
Previously, the only option for proton-beam reference dosimetry was a 60Co-based calibration, which has an uncertainty in terms of reference dosimetry that’s often quoted at 4.6% – but which anecdotally may be somewhat higher. With the increase in patient numbers for proton therapy, it is desirable to bring this uncertainty down to a similar level achieved with the reference dosimetry of conventional photon radiotherapy, which would be closer to 2%.
Thomas says that the NPL calorimeter has been transported to proton-therapy centres in Liverpool (Clatterbridge), Manchester (The Christie), Newport (Rutherford Cancer Centre), Sicily, Prague and Japan, where it has been operated successfully in clinical settings.
“We need to improve the uncertainty of the dose delivered to patients to ensure the best possible consistency across the patient population and to fully understand and interpret the patient outcomes,” he explained. “By bringing the uncertainty on reference dosimetry down to a similar level as that currently achievable in conventional photon radiotherapy, the primary standard will aid in the comparison of the results from high-quality, multicentre clinical trials featuring different treatment techniques.”
Stuart Green, director of medical physics at University Hospital Birmingham, told Physics World that the new IPEM code of practice for proton dosimetry relies heavily on calorimetry developments at NPL over the past 15 years. He says the new code will be ready for publication later this year and hopes that UK centres will transfer to the new approach soon after.
“What’s more,” Green added, “the NHS has a significant opportunity with the opening of the two new proton-therapy centres [at Christie and UCLH] to initiate definitive clinical trials. I am sure the rest of the world will watch with interest to see how well these trials are rolled out.”
Other speakers developed the QA theme in terms of reference and in-clinic proton dosimetry. NPL’s Lourenço, for example, reported findings from a team of UK and Danish scientists who compared the response of user ionization chambers at three clinical facilities against NPL reference ionization chambers.
Their study, which involved a low-energy passively scattered proton beam and two high-energy pencil-beam-scanning proton beams, showed “good agreement between the results acquired by NPL and the proton facilities”. Lourenço added that “reference dosimetry audits such as this are important to improve accuracy in radiotherapy treatments, both within and between treatment facilities, and to establish consistent standards that underpin the development of clinical trials.”
In the same session, Antonio Carlino of the MedAustron Ion Therapy Center, Austria, detailed a new approach for end-to-end auditing of the treatment workflow based on customized anthropomorphic phantoms featuring different types of detector. During the dosimetry audit, which was carried out at HollandPTC in Delft, the phantoms followed the patient pathway to simulate the entire clinical procedure, mimicking the human body as closely as possible in terms of material properties and movement. Carlino told delegates that human-mimicking phantoms deployed in end-to-end audits of this type “may serve as dosimetric credentialing for clinical trials in the future”.
Image and adapt
Standardization and QA notwithstanding, there was plenty of focus on new concepts and emerging technologies for the proton-therapy clinic, with imaging at the point of treatment delivery and online adaptive proton therapy very much to the fore.
“It is one thing to deliver the sophisticated treatment dose volumes that proton therapy is capable of, but it is another to be confident that the dose is being delivered to the right place,” explained Thomas. “During a course of treatment lasting up to six weeks, with daily fractions, the anatomy of a patient may change dramatically as a result of weight loss and/or tumour shrinkage. Imaging can ensure the dose is still being delivered correctly, while informing dynamic refinement of the treatment plan over the course of the treatment.”
Central to the success of adaptive proton therapy is a technique called deformable image registration (DIR), the use of powerful image-processing tools that maximize spatial correspondence between multiple sets of images (e.g. CT scans) collected over an extended treatment timeframe, and even across multiple imaging modalities.
Many developments will take place in the next few years, enabling dose rates to increase and treatment times to reduce
Tony Lomax, Paul Scherrer Institute
However, according to Jamie McClelland from UCL’s Centre for Medical Image Computing, a number of open questions remain around online deployment of DIR in the proton-therapy workflow: “What exactly do we want DIR to do? How do we know it’s doing it correctly? And how do the errors and uncertainties in DIR impact clinical applications?”
More broadly, what of the longer-term development roadmap for proton therapy? Harald Paganetti, director of physics research at Massachusetts General Hospital (MGH) and professor of radiation oncology at Harvard Medical School in the US, reckons proton therapy is currently at what he calls the “Adaptive 1.0” stage, with CT scans performed daily but treatment plans revised and adapted offline.
NPL gears up for UK proton therapy
MGH’s evolution to “Adaptive 2.0” will see that adaptation taking place online in the proton treatment workflow. Key enablers of the MGH approach include cone-beam CT-based imaging and online measurement of the prompt-gamma emissions from delivery of a “partial dose” to the centre of the target (enabling an initial assessment of range accuracy). This is then followed by a rapid adaptation before the remainder of the dose is delivered for a given fraction.
For protons, it seems, the future is bright, with no shortage of opportunities for progress across core physics, emerging technologies and clinical applications. “Pencil-beam-scanning proton-beam therapy is still in its infancy,” noted Tony Lomax, chief medical physicist at the Paul Scherrer Institute in Villigen, Switzerland. “Many developments will take place in the next few years, [enabling] dose rates to increase and treatment times to reduce.”
