Putting image-guided radiotherapy to the test

25 Sep 2018 Sponsored by Modus QA

Early adopters are exploiting a novel motion phantom to explore the possibilities of real-time magnetic-resonance guided radiotherapy for improved cancer treatment and neurosurgery

Upcoming techniques based on magnetic-resonance guided radiotherapy (MRgRT) could enable clinicians to compensate for patient movements to a much higher degree, thanks to the clarity offered by MR imaging. Real-time tracking methods that can pinpoint changes in the position of a tumour translate to improvements in dose conformality by keeping radiation on target and sparing healthy tissue.

As vendors, early adopters and clinicians bring new ideas to fruition, a key part of their success depends on having the right development tools, which includes motion (or 4D) phantoms. Accurate models give researchers the chance to safely explore solutions for overcoming hurdles that can be faced in the clinic as a result of tumour motion. Scenarios include when a patient breathes, causing organs and tumours to move, or when there’s peristaltic motion through the digestive system.

We’ve designed our system to be compatible and expandable, and even – to a certain degree – customizable

Enzo Barberi, director of MR product development at Modus QA
Tumour movement has always challenged cancer treatment and manufacturers have worked hard to mitigate the issue as much as possible.

“Image guidance for radiation therapy has been around for well over a decade and most linacs have some form of cone-beam CT or EPID imaging that allows to them to roughly see where the target is,” says Enzo Barberi, director of MR product development at Modus QA – a developer and manufacturer of quality assurance tools for advanced radiotherapy and medical imaging. “But those imaging techniques provide little information about soft tissue.”

In contrast, MR imaging can reveal soft tissue in exquisite detail, which – when linked to a radiotherapy system – shines a welcome light on where the cancer is at any moment in time.

Barberi, who’s been working in this field for almost three decades, confirms that it’s a very exciting time in terms of the technology and the clinical development of next-generation techniques exploiting MR linacs. “In both systems that are available today, you can image while you are applying radiation,” he points out.

Real-time imaging hits the target

On-board MR imaging offers numerous possibilities for advancing radiotherapy treatment. For example, if gas happens to pass through the intestinal tract of a patient during radiation treatment, real-time MR imaging can detect whether the tumour has moved. And, if the target is now positioned outside the safety margins, the beam can be turned off until the gas has passed through and the tumour moves back into position.

“It’s a dramatic example of how the combination of these two techniques in parallel and in real-time can make a big difference in terms of accuracy in hitting the target when it’s moving,” Barberi comments. Real-time imaging using MR could also see the end of so-called gating, where patients are required to hold their breath to keep their chest stationary – a development that could speed up treatment as well as reducing discomfort.

Bringing these new techniques into the clinic requires reliable tools for quality assurance (QA). MRI-compatible models make it possible to test the ability of novel imaging sequences to track a wide range of movements – such as those resulting from respiration. Verification is important too.

“Using phantoms like Modus’ programmable QUASAR MRI 4D motion product in combination with dosimetry inserts allows early adopters to calculate and measure the dose that is administered to a moving target and ensure that they are actually hitting this moving target and not the surrounding healthy tissue,” says Barberi.

These early adopters are important beta-testers for Barberi and his team, as they are at the frontier of MRgRT. Users require a phantom design that’s flexible, practical and easy to deploy, allowing them to gather as much data as possible for a range of possible patient scenarios.

“Modus focuses very heavily on workflow as we understand that time on the system is valuable,” Barberi comments. “If we can make our QA tools and QA procedures fast and efficient then sites are not only more likely to use them, but they will also appreciate the fact that we’re not taking up a lot of their magnet and linac time simply for setup or integration or when they have to switch over from one mode of measurement to another.”

Early adopters drive development

Features of the QUASAR motion phantom include a spherical target that can mimic numerous trajectories of a tumour in the body, including those seen during breathing. “We can add not only linear motion in and out of the phantom, but we can also add twist and offset that sphere so that it follows a complex 3D path as time plays out,” Barberi explains.

His team acknowledges that different investigators will have different demands, such as when it comes to dosimetry. “Users may wish to use ion chambers or film dosimetry or 3D gel dosimetry,” Barberi notes. “So we’ve designed our system to be compatible and expandable, and even – to a certain degree – customizable.”

Barberi’s group is already working on a second wave of inserts for the MR-safe motion phantom, thanks to the early-adopter programme. The new inserts will focus not just on soft tissue sites, but also modelling deep organ areas and more complex types of motion.

There is no shortage of challenges coming down the pipeline, but Barberi has a great team and is confident in Modus’ approach – having seen its flagship products develop successfully along a similar path. “Working with many different clinicians, physicists and OEMs over the years, we have families of different inserts that we can draw upon,” he says.

Barberi describes MRgRT as a “game changer”, and companies such as Modus are part of a big global effort to support upcoming advances that serve to accelerate the adoption of MR-linac systems for clinical treatment. Initiatives include STARLIT (System Technologies for Adaptive Real-time MR image-guided Therapies), a consortium developing techniques for next-generation motion compensation that includes two large equipment vendors – Elekta and Philips – along with small- and medium-sized companies and academic centres. “We are also equally proud to be a partner with ViewRay, supporting the requirements of an equally respected vendor, and their early adopting customers,” he says.

For more information about the QUASAR phantom, visit https://modusqa.com/mri/motion

©

First UK radiation treatment using MR-guided linac

25 Sep 2018 Tami Freeman

The Royal Marsden and the Institute of Cancer Research (ICR) in London have performed the first treatment in the UK using an MR-linac – the Elekta Unity system.

The Elekta Unity, which received its CE mark in June 2018 and is being clinically implemented in European cancer centres, combines high-field (1.5 T) MR imaging, precision radiation therapy and intelligent software to deliver MR-guided radiotherapy.

“It’s hugely exciting to be able to trial this technology here at the ICR and The Royal Marsden,” says Uwe Oelfke, head of the Joint Department of Physics. “Together we’ve made world-leading advances in radiotherapy through our research and we expect Elekta Unity to allow us to make another step change in improving cancer treatment. This trial is for prostate cancer, but we anticipate Elekta Unity will help us improve radiotherapy for a wide range of cancers, including hard-to-treat forms such as lung and pancreatic cancer.”

The patient received treatment as part of the PRISM clinical trial, which will assess the feasibility of delivering radical radiotherapy for prostate cancer using the MR-linac. The patient had a localized prostate cancer and started hormone treatment in May 2018. His PSA (prostate-specific antigen) level indicated that he was ready to start radiotherapy and he was offered treatment on the Elekta Unity.

“Tumour shape and position relative to healthy tissue evolve over the course of treatment and can change during an individual treatment session,” explains Alison Tree, who is leading the PRISM trial. “The ability to detect those changes and adapt therapy in real time allows us to improve the precision of radiation therapy, more effectively treating the tumour while preserving healthy tissue.”

Tree notes that the Elekta Unity will also enable radiation treatment of patients who would not be candidates using more traditional radiation delivery systems.

“For decades, the radiation oncology community has dreamed of the day when we could see what we treat in real time just as our surgical colleagues do, and we are excited that this day has arrived,” says Oelfke. “Radiotherapy is important to the treatment of around 40% of the people who are cured of cancer. But if we want to fully unlock the potential of radiotherapy by making it even more precise, we need to be able to see a patient’s tumour while we deliver the radiation treatment. The MR-linac will make this possible.”

The Royal Marsden and the ICR are founding members of Elekta’s MR-linac Consortium, a collaborative industrial–academic partnership that Elekta founded with seven centres and technology partner, Philips, in 2012.

© ©

Machine learning: a game-changer for radiation therapy

19 Sep 2018

Deep-learning organ segmentation* will be featured in the upcoming release of RaySearch’s RayStation treatment-planning software. RayStation 8B* is set to ship in December.
Finance, automotive, agriculture, telecoms and professional sports: these are just a few of the diverse industries being fundamentally disrupted by the inexorable rise of machine-learning technologies (more generally known as artificial intelligence or AI).

Machine-learning algorithms can solve problems by learning from experience, and without being explicitly programmed. In so doing, they can already control self-driving vehicles, spot plagiarized academic papers and translate the spoken word from one language to another.

And that’s just for starters. The long-game looks even more compelling.

RaySearch Laboratories certainly thinks so. The Stockholm-based oncology-software company is currently making significant investments in machine-learning and “big-data” – a fusion of technologies and applications that promises to transform radiation therapy and other cancer-treatment modalities such as chemotherapy and surgical intervention.

“Machine learning has the potential to support and augment radiation oncology teams while freeing up their time,” explains Fredrik Löfman, head of machine learning and algorithm at RaySearch. “The power of sharing knowledge through machine-learning models will have a huge impact. Any clinic could potentially generate the same tumour target volume and radiation treatment plan as the best clinics in the world do. Radiation oncologists and medical physicists will all learn from each other through machine-learning models.”

Knowledge transfer

With these opportunities in mind, Löfman is intent on scaling RaySearch’s in-house capability and collective domain knowledge in machine learning. Right now, he heads up a dedicated division of machine-learning engineers located in RaySearch’s Stockholm headquarters.

“Ours is a multidisciplinary programme,” Löfman explains. “We have mathematicians, computer scientists and physicists with backgrounds in different industries [such as automotive and finance] that are further along with machine-learning technologies than healthcare. Cross-fertilization with these sectors is crucial.”

This open, outward mindset is evidenced in several high-profile R&D collaborations that, RaySearch hopes, will fast-track its innovation in machine-learning technologies. On the academic side, RaySearch is funding joint research at the KTH Royal Institute of Technology in Stockholm (automation of radiation therapy), while clinical partners include the Princess Margaret Cancer Centre in Toronto (automated treatment planning and model training) and Massachusetts General Hospital in Boston (deep learning for target-volume delineation and analytics prototypes).

“We have a history of collaboration with The Princess Margaret and Mass Gen,” says Löfman, “so when we started the machine-learning division it was natural for us to partner with these institutions.”

Partnership = progress

The tie-up with The Princess Margaret, Canada’s largest radiation-therapy facility, has been in place for more than a decade and focuses on the use of machine learning to automate treatment planning in the radiation-therapy clinic. The goal is twofold: to deliver workflow efficiencies versus manual treatment planning (with plans delivered in minutes rather than hours or days) and to generate personalized treatment plans tailored to the unique needs of each patient.

It’s been a productive collaboration. Last year, for example, RaySearch licensed The Princess Margaret’s AutoPlanning technology, a custom AI and machine-learning system that harvests information from a database of proven high-quality radiation-therapy plans – effectively learning from and optimizing against thousands of prior clinical treatments.

AutoPlanning itself is the result of a six-year cross-disciplinary initiative involving researchers at The Princess Margaret and Toronto’s Techna Institute. The development work was led by Tom Purdie, a medical physicist at The Princess Margaret, and his colleague Chris McIntosh, a computer scientist at the hospital.

“Machine learning is a natural fit for automating the complex treatment-planning process,” explains Purdie. “It will enable us to generate highly personalized radiation treatment plans more efficiently, [thereby] allowing clinical resources or specialist technical staff to dedicate more time to patient care.”

Ultimately, Purdie reckons that machine learning will help to lower the cost of cancer treatment. “We’re going to reach a stage – in the not-too-distant future – where we will be relying on machine-learning technologies to deliver the highest-quality cancer care,” he adds.

Into the clinic

RaySearch, for its part, is pressing ahead with the roll-out of advanced machine-learning capabilities into the radiation-oncology clinic. In December, the vendor will unveil RayStation 8B*, the latest release of its treatment-planning software with machine-learning-automated organ segmentation (quantitative 3D visualization) and machine-learning-automated treatment planning.

“We are evaluating the automated treatment planning with several clinics just now, so we are getting more and more data on how it performs,” says Löfman. “As a vendor, we can point to the efficiencies of automation, we can point to the consistency inherent to the approach. But in terms of treatment quality and patient outcomes, it’s the clinics that will need to provide the real-world evaluation and validation.”

The collaboration with The Princess Margaret has broadened in scope too, with the two partners signing an agreement in July to jointly develop RayCare, the vendor’s flagship oncology information system (OIS).

Watch this space, says Löfman: “Ultimately, RayCare OIS will use machine learning to improve workflow efficiency, manage resource allocation and enhance quality assurance across different treatment modalities – medical oncology, radiation oncology and surgical oncology.”

* Subject to regulatory approval in some markets

Fredrik Löfman and Tom Purdie will present a live webinar on “Machine learning and automation in radiation oncology” on 26 September 2018. Register now to view the webinar.

©