MRI-Guided SBRT Could Represent Safer, More Precise Radiotherapy Option for Prostate Cancer

Through slimmer margins and automatic shutdown when a target shifts out of range, MRI-guided stereotactic body radiotherapy (SBRT) could serve a potential alternative to the standard of care of computed tomography (CT)–guided SBRT for patients with prostate cancer, according to Amar U. Kishan, MD.

Findings from the phase 3 MIRAGE trial (NCT04384770) published in JAMA Oncology showed that in a prespecified interim futility analysis of 100 patients who were randomly assigned to MRI-guided SBRT (n = 49) or CT-guided SBRT (n = 51), MRI-guided SBRT with 2-mm margins led to an absolute reduction of 19.0% for acute grade 2 or greater genitourinary (GU) AEs compared with CT-guided SBRT with 4-mm margins. Grade 2 or higher GU toxicities rates were 24.4% (95% CI, 15.4%-35.4%) for the MRI arm vs 43.4% (95% CI, 32.1%-55.3%; P = .01) for the CT arm.

No patients in the MRI arm experienced grade 2 or greater gastrointestinal toxic effects (95% CI, 0.0%-4.6%) arm vs 10.5% (95% CI, 4.7%-19.7%; P = .003) of patients in the CT arm.

CT-guided SBRT “is a standard that’s thought to be highly effective. What we're looking at is the potential for reduced toxicity while maintaining the efficacy of the treatment,” Kishan, the study’s lead author, said in an interview with OncLive^®.

“With prostate cancer, we know that cure rates are very high. One of the primary focuses of research interests globally is to try to reduce the adverse effects [(AEs) associated] with treatment,” he added. “People are living for a long time after their diagnosis and treatment, which is great, but we want to make sure they're living without toxicity. Steps like this—reducing the margins, more accurate treatment, and trying to drop toxicity—are steps in the right direction.”

Kishan discussed the safety results of MRI-guided SBRT compared with the current standard of care, highlighted the benefits of using an MRI to guide radiation for patients with prostate cancer, and explained other areas where MRI-guided SBRT could explored. Kishan is the vice-chair of Clinical and Translational Research and chief of the Genitourinary Oncology Service for the Department of Radiation Oncology at the David Geffen School of Medicine at the University of California, Los Angeles (UCLA), and the UCLA Jonsson Comprehensive Cancer Center.

OncLive^®: What are the potential benefits for MRI-guided SBRT compared with CT-guided SBRT?

Kishan: The MIRAGE trial was a single-center, randomized trial where we compared the standard of care, CT-guided SBRT for prostate cancer, vs MRI-guided SBRT for prostate cancer.

When you’re treating prostate cancer with any form of external radiation, you need to account for the fact that there are some uncertainties in radiation delivery. We designed radiation to precisely target the prostate, but we need to account for the fact that the prostate is a moving target. It sometimes isn't easily imaged on CT or X-ray, and we need to bring in other imaging modalities. Traditionally what we have done is put markers into the prostate that show up on x-ray, because the prostate itself doesn’t show up as well on X-ray or CT.

We rely on those markers as an indirect ability to track the prostate, both as it moves from day- to-day and between radiation treatments. SBRT is a high-dose per day treatment, where we are able to condense a radiation course for prostate cancer from, [for example], 39 to 45 treatments, all the way down to 5 treatments.

In that situation, we’re delivering 5 high-energy, high dose fractions of radiation. You need to make sure you’re accurately hitting the prostate. Instead of just targeting the prostate, we target the prostate with a margin around it. That way, if the prostate is moving or if there are any uncertainties, such as a 1 mm change in set up [of the prostate] on any given day, we make sure we're hitting the target. That's the goal. You also want that margin to be as tight as it can be while safely hitting the target so that you reduce [radiation] doses to the bladder and the rectum, which are nearby, and other structures such as blood vessels and nerves that may be involved with erectile function.

When we’re doing standard-of-care, CT-based SBRT, we put metallic markers into the prostate. Those metallic markers are easily seen on an X-ray, which we have readily built into standard radiation delivery devices. We target the prostate with this 4-mm margin around it, and we deliver our treatment.

An MRI allows you to see the prostate better. When we get an MRI to help map out and stage the prostate cancer, [or when] we get an MRI to help plan surgery for prostate cancer, this gives better anatomical detail of the prostate itself. However, historically, we haven't been able to combine a linear accelerator that delivers radiation with an MRI because of magnetic interference. That's where the MRI-guided radiation piece comes in; these are newer entities and devices that allow us to have an onboard MRI that’s operational during the radiation.

Instead of a patient getting treated with guidance based off x-rays or a cone-beam CT, we are delivering the treatment inside the bore of an MRI, and the MRI is on during the treatment. With that, you’re able to set up the treatment a little bit better because the MRI gives you better resolution when you're trying to align things at the time of the treatment. Most importantly, [this process] has what is called a cine MRI: it’s getting an MRI image 4 times per second and stitching that together to make a real-time monitoring of the prostate. If the prostate moves too much, the beam will automatically turn off. That allows us to be confident that we can hit the prostate and not need a 4-mm margin, but [rather] a 2-mm margin.

How were these 2 SBRT methods compared in MIRAGE?

This trial looked at this newer device where we can do 2-mm planning margins, and then we have the standard-of-care—which is still advanced, [precise] technology—with a 4-mm margin. We’re reducing these margins by 50%. It sounds like not that much, but this is high-value real estate. As I explain to patients, you have the prostate right there, then the bladder and the rectum. These millimeters matter. The question is: could we reduce AEs?

Specifically, we designed the trial to look at urinary AEs as the primary outcome. We also looked at physician-reported bowel toxicity and patient-reported outcomes.

A patient would come to UCLA who wanted to get this 5-day course of radiation. We [randomly assigned] patients to receive the standard-of-care, CT-based approach with the markers or they would get treated on the new MRI-guided device.

Did the MRI deliver better safety and toxicity results?

What we saw was a significant reduction in acute grade 2 or greater—which is moderate or greater—urinary toxicity. That was the primary end point of the trial. It was reduced from 43.4% [with CT-guided SBRT] to 24.4% with the MRI guidance. Acute grade 2 or greater bowel toxicity was reduced from 10.5% [with CT-guided SBRT] to 0% with MRI guidance.

We also saw significantly less symptom burden as interpreted by patients. We gave patients questionnaires to fill out to give us an idea of how much urinary or bowel symptoms were bothering them. We gave people an International Prostate Symptom Score [IPSS] to look at the burden of urinary symptoms. That scale is 0 to 35 points, and 0 to 7 points is mild, followed then by moderate, and severe symptoms. We said if a patient gets a 15-point increase on the IPSS scale, that's a significant burden of symptoms. That proportion of patients getting that type of significant symptom burden was reduced from 19.4% [with CT-guided radiation] to 6.8% [with MRI-guided radiation].

Similarly, with bowel [symptoms], we gave patients the EPIC-26 Questionnaire, which asked some questions about bowel bother and symptoms. There are well-established thresholds for what is clinically detectable and a clinically significant decrease in outcomes. When we looked at that, there was a 50% rate of clinically significant bowel function bother with CT-guided SBRT and that dropped to 25% with MRI-guided SBRT.

The key takeaway was that these much tighter planning margins that were allowed and facilitated by the new device led to less physician- and patient-reported bowel and bladder AEs.

Was the reduction in AEs a function of more precise targeting with MRI-guided SBRT or the ability for the MRI-guided radiation to turn off if the prostate shifted?

That is something that we are looking into; looking back at all the motion traces and seeing in what patients there was excess motion or how frequently [the radiation] was turning off.

Clearly, [the ability for the radiation to turn off] is playing a role [in reducing off-target radiation]. When we look at just the planning differences and compared a CT-based plan to an MRI-base plan, there were some differences in bladder dose. [The differences in dose] weren’t extremely dramatic, but the difference we saw in toxicity was very robust. Likely, it’s not just the fact that it's a tighter margin, but it may also [be related to] the fact that we’re able to turn off the beam, limiting off-target radiation.

Although MIRAGE did not look at survival, did you notice any differences there?

This is the first and, to my knowledge, only randomized trial that is looking at whether this machine improves outcomes. It's too early to look at survival data from this study.

What we’re getting at is [evaluating whether] that 2-mm margin is too tight. We feel that is highly unlikely to be the case, because of the target. We are tracking the prostate itself as it is moving back and forth, and if the prostate was moving out of the field, the beam would turn off, and it would only turn back on either manually or if [the prostate] immediately came back into field.

We do know that the prostate is being accurately targeted. It was essentially almost impossible that there was a miss [of the target] in the MRI group. In the control group, because we don't have that real-time feedback there, it is more possible that something could have been outside the field when the radiation was delivered. However, that is a standard-of-care option that in other studies has been looked at aggressively and with longer-term follow up, so I don’t think there is likely to be a significant outcome difference in terms of survival and cure rates between the 2 groups.

This is the first study looking at MRI-guided radiation treatment in prostate cancer. Has this been explored in other tumors?

It has been explored in other tumors. The bang for your buck arguably might be higher in some other tumors where the standard of care is not as good. In prostate cancer, the standard of care with CT-based treatment is already very good. It is clearly a good option for a lot of patients, and we’re trying to improve the bar.

However, there are cancers like pancreatic cancer, where it is extremely difficult to get a high dose of radiation in there because the critical structures nearby move or change shape, and [there is a risk for more serious toxicities]. Instead of [an AE] of needing to go to the bathroom more frequently [that can occur with radiation for prostate cancer], a fatal bleed of an ulcer in the duodenum [can occur with radiation for pancreatic cancer]. In those contexts, having this tight radiation delivery may be more favorable, where we’re talking about maybe avoiding serious toxicities and being able to deliver radiation to some place that we may have never even tried to treat [with radiation].

The other area where [MRI-guided radiation] is being explored very aggressively is in the setting of gastrointestinal malignancies, particularly pancreatic cancer, or for liver metastases. [Like the prostate] the liver also highly mobile.

In the setting of GU cancers, [in] kidney cancer, there is an upswing in possibly treating patients who can't get surgery with high-dose radiation. The kidneys are mobile targets, and you want to try to maximize the dose you can get in. The MRI-guided technology poses an advantage there, as well.

Would brain cancer represent another potential target for MRI-guided SBRT?

In the context of brain cancer, such as glioblastoma multiforme, it is being studied [to understand if] having an MRI on board can help tailor the treatment fields.

[MRI-guided SBRT] is a little different [in the brain] because the target isn't moving and the anatomy itself isn't changing. However, you may be able to visualize tumor response in that context, and that could be valuable. Do you need to deliver a higher dose of radiation because [the patient] is not responding, or can you tailor things differently? That would also be an interesting application, but it’s a slightly different twist, because we're not looking at the motion and the targeting as much; rather, it would be some type of a functional readout during the course of radiation.

Are there patients who would not be eligible for MRI-guided SBRT?

There are some contraindications to MRI-guided radiotherapy, which are tied to the ability to get an MRI. If somebody has a pacemaker that cannot be turned off, they can't get MRI-guided treatment because we wouldn't be able to use the MRI part of it. For someone has shrapnel in their body, and they can't get an MRI for that reason, the same thing applies. That's very rare, I treat a lot of patients with prostate cancer, and there have maybe been 1 or 2 of the multiple hundreds I have treated that cannot get an MRI for those reasons.

Maybe a more common [contraindication] would be claustrophobia. An MRI can be uncomfortable, and sometimes patients can power through the test of getting 1 MRI for a diagnosis. However, the idea of going in multiple times for treatments when there's an option of doing it without it can be hard to swallow. I'd say that's slightly more common to be a reason that someone might decline this treatment. At the same time, we try our best to help patients get through it. There are medications we can prescribe in a clinic setting to try to get patients through it. There are some contraindications, but they are pretty limited.

For patients who are claustrophobic, would an open MRI be an option?

Unfortunately, not with this technology. Specifically, we looked at the ViewRay MRIdian linear accelerator. There are other linear accelerators that are MRI-guided as well, but none of them have an open MRI feature. So, unfortunately, no, [patients] do have to go inside of the bore [of the MRI to receive radiation].

Reference

Kishan AU, Ma TM, Lamb JM, et al. Magnetic resonance imaging–guided vs computed tomography–guided stereotactic body radiotherapy for prostate cancer: the MIRAGE randomized clinical trial. JAMA Oncol. Published online January 12, 2023. doi:10.1001/jamaoncol.2022.6558