January 2017

Real-Time Radiotherapy
By Beth W. Orenstein
Radiology Today
Vol. 18 No. 1 P. 16

MRI has the potential to improve the efficacy of radiation oncology.

What if radiation oncologists could see tumors and surrounding normal tissue as they were treating them in real time? "It may result in a revolutionary change in our field," says John Christodouleas, MD, MPH, a radiation oncologist and vice president of medical affairs and clinical research at Elekta, one of two companies that are introducing MRI-guided radiation therapy systems to the market.

Elekta, based in Stockholm, Sweden, plans to take its first commercial orders for its high-field MR-linac, an MRI-guided linear accelerator, in 2017 with deliveries in 2018, assuming it receives the applicable regulatory clearances. Cleveland-based ViewRay, which nearly three years ago introduced the world's first clinical MRI-guided radiation therapy system using cobalt sources, is now introducing a linear accelerator version of its system, the MRIdian Linac. The MRIdian Linac is for sale and clinical use in Europe and for nonclinical research in the United States, while the company pursues 510(k) approval from the FDA. Some cancer centers in the United States hope to have ViewRay's MRIdian Linac operational later this year.

Long Time Coming
Twenty years ago, radiation oncologists used a combination of skin marks and plain film X-rays to guide radiation treatments in cancer patients. Since the early 2000s, radiation oncologists have been using cone beam CT to align treatment to tumors.
"Cone beam CT transformed the clinician's ability to see where they are putting the radiation," says Kevin Brown, Elekta's global vice president of scientific research. But it doesn't go far enough. "There still are a lot of targets you cannot see on CT because there isn't sufficient contrast," Brown says. MR is much better at seeing soft tissues and enables physicians to "actually put radiation where they want to put it, not where they think they should put it."

The combination MR scanner and linear accelerator has been at least 16 years in the making, Brown says. "That's when the concept for the machine was conceived at our research partner, UMC Utrecht."

The challenge for physicists was to find a way to deliver radiation treatment in the presence of a strong magnetic field. Radiotherapy treatment is typically done with a linear accelerator, which generates high-energy X-rays. Physicists had to find a way to stop the electrons that the radiation beams' photons generate from being diverted by the MRI's magnetic field. Their solution was to modify the MRI in a special way so the electrons wouldn't be affected by its magnetic field. The linear accelerator is mounted on a rotating gantry so that the treatment beam can pass directly through the body of the MRI, enabling the patient to be imaged and treated simultaneously.

Another issue was stopping leaks that create noise and interfere with the MR images. To solve this problem, ViewRay researchers took a cue from the military, which has learned how to hide airplanes from radar, says Jim Dempsey, PhD, ViewRay's founder and chief scientific officer. Stealth aircraft are coated in carbon, which absorbs microwaves. Copper reflects microwaves. In its patented technology, "we use layers of copper and carbon shielding to meet up and absorb any leaking radiofrequency interference," Dempsey says.

Personalized Therapy
The MRI-guided linear accelerator can be used to treat all types of cancer at almost all locations, says Ben J. Slotman, MD, PhD, a professor and chairman of the department of radiation oncology at Vrije University Medical Center in Amsterdam. "We have the MRIdian system in use for six months now and have focused on patients with tumors in the pancreas, liver, adrenal gland, kidney, prostate, rectum, lung, and breast. In all these cases, we use stereotactic or hypofractionated treatment. However, if we have free time slots available, we also use it for some other indications."

The MR-linear accelerator combination could solve several challenges that radiation oncologists currently face. One of the huge challenges is not being able to predict when a healthy organ exhibits severe toxicities from radiation, Christodouleas says. "One reason is that human beings are different in the way they are sensitive to radiation," he says. Some patients are just more sensitive than others. "Another reason is that we really don't know what dose normal tissues have gotten over the course of treatment. We don't have good ways of tracking that." One of the major benefits of having superior soft tissue imaging during treatment "is that it can help us target and track the cumulative dose to both tumor and healthy tissues over the course of therapy," he says.

The combination system also allows radiation oncology to be tailored to the patient's tumor. "The way most radiation treatment works is you create a plan in advance, and you deliver that exact same plan without changing it," says Michael F. Bassetti, MD, PhD, a radiation oncologist at the University of Wisconsin's Carbone Cancer Center in Madison, one of the first centers to treat with ViewRay's low-field system, beginning in 2014.

"[The MR-linear accelerator] allows the potential to adjust the treatment as you go along based on response. It really offers potential useful information about the patient's individual cancer biology and the ability to personalize the treatment" vs one-size-fits-all treatments, he says. "With good quality imaging on a daily basis, we can actually watch the response as it is happening during their treatment course and make adjustments, if we want," Bassetti says.

In the future, MR-linear accelerator systems may allow oncologists to decide whether a patient benefits from further treatment based on the in-treatment imaging response. "For example, some patients may have such a good response that they may not need surgery, while poor responders might need more aggressive surgery or chemotherapy as a follow-up," he says.

Better Dose Control
While the MR-linear accelerators can be used for most tumors, those using it find it is most helpful for tumors in locations that can change from day to day because of movement of nearby organs. Good examples are tumors of the lung, which can move with each breath, and tumors of the abdomen that can be affected by how full the patient's bowel or bladder are. Today, Slotman says, physicians are able to control around 90% of early-stage lung cancers with stereotactic ablative radiotherapy. With the MRIdian Linac, the radiation oncologists could push their success rate even higher because they can see the lung tumor as it moves up and down with each breath. Knowing where the tumor is during treatment means they can deliver the radiation only when it is in the target field, he says.

The system is also ideal for treating tumors such as those found on the pancreas that are close to critical sensitive structures, Bassetti says. The pancreas lies in very close proximity to the small intestine, stomach, kidneys, and spinal cord. Many patients' tumors are adjacent to organs, such as the intestine or stomach, that cannot tolerate high doses of radiation, Bassetti says. "So the doses of radiation we would like to deliver are limited by our ability to avoid that dose to critical or adjacent organs. If we can see what we're doing, while we're doing it, we understand what we're delivering to that organ."

The combination system also could lead to higher radiation dosing, if it is needed, for a number of reasons. One reason is physicians don't have to be concerned about delivering such high doses to healthy tissue because they can clearly see when it is and isn't in the way, Brown says. "I'm much more comfortable giving a very high dose of radiation to a tumor if I can see the intestine right next to it and know it isn't moving into the field," Bassetti agrees.

Treatment Response and Planning
Also, if physicians can image the patient during treatment, as opposed to before and after, they can see how well the treatment is working as they are delivering it. They might need to step up the dose if the tumor isn't responding, or they could perhaps lower the dose or number of treatments if they see it's responding well after only a few of the planned courses. In most cases, radiation oncologists have to wait three months or more after treatment to reimage their patients and see how well they responded.

In the future, Brown says, physicians will be able to use MRI diffusion weighted imaging to determine the effectiveness of the treatment as they are delivering it. "There are many things MR sequences can show you," he explains, "and one will tell you about the diffusion of water in tissue. That can tell you how densely packed the tissues are at the cellular level. If water is able to diffuse more readily between cells, that's a sign that there's tissue death. That's something we wouldn't be able to pick up before."

Researchers at UCLA reported in the March 2016 issue of Medical Physics on their pilot study using ViewRay's MRI-guided radiotherapy system to perform a longitudinal diffusion MRI strategy for assessing patient response to radiotherapy. They concluded that such an approach may enable response-guided adaptive radiotherapy, but more research is needed.

Eventually, Christodouleas says, researchers hope to automate the best sequences for different types of tumors and treatments based on experience with the combination machine. MR imaging should help, for example, to identify organs and their parts. With that information, "we may find that you need one sequence for a pancreatic tumor that's up against the duodenum and another for a tumor that's midpancreas and not constantly up against the bowel," he says. "It's not a question of whether it can be done with the MR images. It's really just a matter of putting these sequences together. In a CT world, you're still just getting the one parameter, and you simply can't solve these problems."

Another potential advantage is shortening time to treatment, especially when radiation therapy is being used for palliative care, Bassetti says. Now, patients have to come to radiology for a CT scan to plan their treatment. "We send them away or back to the hospital floor while we plan the radiation," he says. "With this system, you can plan on the fly entirely, so we can bring them down for one single visit, potentially." It not only is more convenient for the patient but also allows them to get the pain relief they need sooner, he says.

Costs vs Benefits
While combining MR and linear accelerators has many advantages, it also has some disadvantages. One disadvantage to the MRI-linear accelerator is that it is likely more expensive than regular radiotherapy. Radiotherapy is incredibly low-cost compared with chemotherapy or surgery or other techniques used for cancer treatment, Brown says. Radiotherapy has benefitted from efficiencies that continue to lower its cost, he says. "Using the MR-linac will likely be more expensive," Brown says. Reimbursement could also be an issue. "One of the things we will be looking at is enhanced reimbursement for the MR-linac. We believe, if it brings value to the health care system, our users should be reimbursed accordingly," he says.

Time also could be a disadvantage. A regular session using the MR-linac shouldn't take any longer than a session with a conventional linear accelerator, Brown says. However, using diffusion weighted imaging to determine tumor response could add some time to treatment sessions. Typically, radiation sessions last 15 to 20 minutes. Diffusion weighted imaging may add five to 10 minutes. "So you're not really talking about a significant amount of time, especially compared to the potential benefit," Brown says.

In addition, some patients have metal implants and cannot get MRIs, Bassetti says. Also, there's this: While MR is nonionizing and there's no safety concern for exposing patients to any unnecessary radiation, a strong magnet requires that proper safety procedures be put in place. Elekta uses a 1.5T MR imaging system from Philips similar to that used in its diagnostic systems.

Those familiar with the MRI-linear accelerator machines don't believe they will ever be the standard treatment for all cancers, but Slotman believes that if cost weren't an issue, more patients would be treated with them. "I believe that 15% to 25% of patients who are treated with radiotherapy will have a definite benefit of MR-guided radiotherapy," he says. "If costs were not an issue, this could easily be 80% to 90%."

Because it's more expensive, it won't be used for cases that don't require it, Brown agrees. "It would make sense to treat those cases on the less expensive technology that is still of high quality." However, Brown says, it's all about "seeing," and "seeing" what the radiation oncologists are doing as the treatment happens can make all the difference.

"This has the potential to improve the efficacy of radiation therapy," Dempsey agrees, and that can be life-changing for some patients.

— Beth W. Orenstein, of Northampton, Pennsylvania, is a freelance medical writer and regular contributor to Radiology Today.