Amar Kishan, MD, outlines the rationale for the MIRAGE trial

In this video, Amar U. Kishan, MD, highlights the rationale for the commencement and design of the MIRAGE trial (NCT04384770), which compared the use of aggressive margin reduction with MRI-guided SBRT to standard CT-guided SBRT. Kishan is a professor and executive vice chair of radiation oncology at the University of California, Los Angeles.

Video Transcript:

When you're treating prostate cancer with definitive radiation, it's a highly effective treatment, but of course, one of the major concerns that patients and providers have is post-treatment toxicity. This arises from the organs that are close to the prostate when we're doing the radiation—that includes the bladder and the rectum and also the urethra, which is a tube that carries urine, running through the center of the prostate. All of these normal structures can get affected by the radiation leading to post-radiation toxicity. It's always an area of interest, from the radiation oncology standpoint to try to reduce the toxicity that's involved with radiotherapy.

Now, one important aspect that is intuitive but maybe not obviously apparent when you're treating the prostate with external radiation is you need to account for the fact that the prostate is a moving target. It moves side to side, up and down, front and back, about an average of 2 to 3 millimeters over a several minute time frame. Again, because of the close proximity to these critical structures, we want to make sure that we have a margin around the prostate to encompass that motion, but we don't want it to be so big that it's irradiating large portions of these nearby structures. It's very typical when we're delivering radiation to aim at the prostate with this margin around it but then use sophisticated imaging technologies to limit as best we can how big that margin needs to be. This is called motion management.

In the MIRAGE trial, what we looked at specifically was in the context of doing stereotactic body radiotherapy, which is a 5-day course of radiation for prostate cancer that already uses advanced technology to begin with. One group of patients had what we called CT-guided SBRT, which is our standard of care option that we had been delivering for over a decade at UCLA. Basically, these golden markers are placed into the prostate, and we use that as a proxy for the position of the prostate, and track those markers using an x-ray and a CT to help guide the radiation. We compared that approach with an MRI-guided radiation approach, where we used a newer MRI guided radiation device, linear accelerator, where patients were actually treated inside of an MRI tube. They're getting a continuous MRI, 4 frames per second. That's basically creating a movie that monitors the prostate itself and can start and pause radiation if the prostate is moving too much.

That was a 1:1 randomization, men getting CT-guided radiation and MRI-guided radiation. All the SBRT was 40 Gy in 5 treatments, 8 Gy per fraction, so high dose radiation. In the CT-guided arm, because of the type of monitoring that we were using for motion management, we'd used a 4 mm margin. So, the target plus a 4 mm margin. That's standard of care. If you look at other SBRT reports and ongoing SBRT trials, it's typically 5 mm margins, maybe reduced to 3 mm posteriorly in certain cases. But, 3-5 mm posteriorly; I think our 4 mm margin is well within that range. [In] the MRI-guided arm, that's where we leveraged the abilities of the machine to reduce that margin to 2 mm. So, 4 mm vs 2 mm.

The trial, then, was specifically looking at whether that aggressive margin reduction from 4 mm to 2 mm would lead to a reduction in physician-reported urinary toxicity, specifically in the first 90 days after radiation. A lot of people have asked me why was that the end point, as opposed to long-term urinary toxicity or long-term bowel toxicity, patient reported outcomes? Honestly, it was an issue of practicality. This was a single center randomized trial using a novel device that many other institutions, frankly speaking, had and were using, not on a trial. So, there was a small time window within which we could complete a trial like this in a single institution while we were still able to randomize between the 2 machines. We chose a toxicity that's common. Physician scored urinary toxicity within the first 90 days after radiation is fairly common, so we would be able to detect a difference.

Thankfully, radiation is pretty good when it comes to long-term toxicity. Though, I'm sure there'll be some discussion about that statement, but many men do quite well with radiation in the long-term. To be able to show a difference in long-term toxicity would have required a large multicenter trial with many fold more patients. That's just not practical. I think, to date, the MIRAGE trial, which has now finished accrual over 3.5 years ago, is the only randomized trial that looked at these 2 machines. So, I don't think it would have been practical to have a longer-term toxicity end point. It's also not something you can blind patients to, which is why it wasn't a blinded trial. And it was run in the height of the COVID pandemic, so there were some limitations in how frequently we could ask patients questionnaires and give them other things. We were able to eventually figure that out and do remote visits and electronic patient reported outcomes. But anyone who was practicing during that time knows that was a big shift to move to that format. So, we stuck with a hard end point—for us, physician scored toxicity—which was easy to do remotely if needed to.

This transcript was AI generated and edited by human editors for clarity.