MIRAGE Trial Phase 3: CT / MRI SBRT Which is Better? [2022]
Which is Better: CT or MRI SBRT in Prostate Cancer?
So traditionally, radiation therapy, when it’s delivered in an external beam format, has relied on x-ray or CT-based imaging. MRI imaging provides very good space, full contrast, and resolution for certain targets, for example, targets in the pelvis such as the prostate.
But historically, it has been technically difficult to integrate an MRI with a linear accelerator, which is what delivers the dose. High-energy x-rays and external beam radiotherapy There are two commercially available MRI-guided linear accelerators on the market now, and they both present technical solutions to this complicated physics problem.
This is important for us because, when doing external beam radiotherapy, we are often reliant on the imaging that we have at the time of treatment to guide our treatment and to plan our treatment. So having the best imaging available is obviously critical to delivering radiation. And so that’s the premise behind MRI-guided radiotherapy: the ability to finally use MRI guidance to direct our radiation rather than an x-ray or a CT-based image.
Again, this will be helpful for some diseases because an MRI has better spatial contrast and resolution. MRI-guided radiotherapy is technically doable for any disease state. However, the areas in which it probably poses the best advantages are those in which the target may be better visualized using an MRI-based platform.
And/or there’s some type of significant target motion, which the modern MRI-guided linear accelerators are able to account for. The prostate is both better seen on MRI than on CT and a highly mobile target. And the ability of the MRI-guided machine to track the motion of the prostate is integral to why the outcomes of the MIRAGE trial, which I’ll discuss in a second, may suggest a superiority to MRI-guided radiotherapy.
MRI-guided radiotherapy can also be used in patients with abdominal tumors. For instance, pancreatic cancer and cancers in the liver, which may be primary liver cancer or metastatic cancers to the liver, Because again, in those areas, there may be complex and mobile anatomy. The same also applies for upper abdominal tumors like renal cell carcinomas when treated with radiation.
What Is The Current Standard Of Care For Men With Prostate Cancer. Why Use MRI-guided Technology In Radiation Oncology?
The trial was chosen to help men with prostate cancer, and this is statistically significant. The vast majority of men diagnosed with prostate cancer are diagnosed with localized prostate cancer and are amenable to definitive and curative local treatment.
Typically, that’s either a radical prostatectomy, which is surgery, or radiation therapy. Radiation therapy can take multiple forms. It could be either implanted or external beam radiotherapy. And within external beam radiotherapy, there has been a move towards shorter and shorter courses of radiation where higher and higher doses per day are delivered.
This is because prostate cancer appears to respond better to a higher dose of Radiesse delivered per. But historically, we didn’t have the technology to deliver such high doses of radiation safely, and therefore we instead delivered longer courses of radiation with a low dose per day.
As we have developed better imaging and treatment delivery technologies, we have moved towards these shorter and shorter courses. Stereotactic body radiotherapy, or SBRT, sometimes also called SABR for stereotactic ablative radiotherapy, is a type of radiation treatment wherein five or fewer doses of radiation are delivered towards a target for prostate cancer.
The most common regimen used is indeed a 5-fraction regimen. This has been supported by randomized trials conducted outside of the United States as well as numerous phase 2 trials conducted within the United States, and it is considered a standard-of-care radiation modality for patients with localized prostate cancer.
Now, as I mentioned earlier, the prostate is a highly mobile target. When you’re delivering external beam radiation, you need to account for this motion. You need to put a margin around the prostate in case it moves during the radiation treatment. You don’t want to miss the target. By adding this margin, it does mean that certain adjacent organs are going to be exposed to radiation.
In this context, that would mainly be the bladder and the rectum. This will in turn lead to some side effects due to irritation of the bladder and rectum. With respect to localized prostate cancer, again, most men are diagnosed with a localized disease, and we’re treating them with curative intent, and thankfully, many men are cured by radiation.
Therefore, it is important to minimize the post-treatment burden on these patients by limiting post-treatment toxicity, whether that’s urinary toxicity, bowel toxicity, or even sexual toxicity from the radiation. The premise of the Miras trial was that we could use the advanced imaging technologies of the MRI-guided lunar accelerator, in this case, the Meridian VRA device, to improve the side effect profile of SBRT for patients getting this type of treatment for their localized prostate cancer.
The experimental arm of the study was to use the newer MRI technology, which has the ability to obtain an MRI four times per second and very precisely track the prostate as it’s moving. We know from many studies that the prostate is moving maybe 4 to 3 millimeters in each direction during the course of treatment, and that’s the amount of margin that we need to add.
Typically, a 4 millimeter, because of the high precision of the MRI-guided device, we hypothesized that we could shrink that margin into 2 millimeters, and that margin reduction, which is very aggressive, would lead to fewer side effects, specifically fewer urinary side effects after SBRT for localized prostate cancer.
What Is The MIRAGE Clinical Trial Design For Patients With Prostate Cancer?
It’s worth getting into the specifics of the design of the trial, as that’s crucial to the interpretation. When designing a clinical trial, you must consider things like power, statistical power to detect a difference, hypothesizing what the difference might be, and what is practical to do in a single-center setting, because the MIRAGE trial was, after all, a single-center randomized trial.
Because prostate radiation in any form is the most common acute toxicity in the context of prostate SBRT, acute urinary toxicity is the most important consideration. We decided to make that our primary endpoint, and acute physicians scored urinary toxicity specifically at grade 2 or greater. which generally means a moderate annoyance for patients and something that necessitates medication to control the side effects.
Now we know from well-designed prospective trials—as I mentioned, there have been some randomized trials outside the US that have been run using CT-guided SBRT. We knew that a good rate to assume in the controlled arm would be around 29. And, once again, this is the rate of acute grade GU toxicity with prostate SBRT.
That specific trial where he got that estimate from the pay speed trial used a lower dose of radiation, 36.25 grad, to the prostate PTV or planning target margin, which was in general five millimeters, maybe four millimeters in some cases, around the prostate target, which I mentioned is the typical margin that’s used to account for prostate motion among some other uncertainties that may exist.
Now that trial has focused on patients with more favorable intermediate-risk prostate cancers than the typical mix of patients that we see at my institution, At my institution, we hypothesized that we’d be treating a larger number of patients with a more aggressive disease, either unfavorable, intermediate-risk, or high-risk prostate cancer.
And based on some earlier studies we’ve done, we thought it would be more important to treat with a higher dose of radiation, closer to 40 Gy in 5 fractions, which is actually a substantial dose escalation. And we know that dose escalation is associated with an increase in acute toxicity. However, because no large-scale trial had been done using 40 Gy in 5 fractions, we did not know what to hypothesize as our control rate of acute grade 2 toxicity in the context of a closely monitored randomized trial. So again, we hypothesized a 29% incidence of acute greater toxicity in the controlled arm and a reduction to 14% in the experimental arm. However, because we were treating at this higher dose, we stipulated an interim analysis to be done after 100 patients were valuable for the primary outcome so that we could evaluate the actual rates of toxicity.and the fact that this would lead to a reduction in toxicity because there would be less dose going to the bladder and the rectum because of the much tighter margins—a 50% reduction in margins.
So go another way. This was primarily a test of whether aggressive margin reduction would be effective. Leveraging the MRI guidance technology would lead to less toxicity. We followed patients with prostate cancer essentially at the one-month timeframe and at the three-month time point. And any toxicity during those 90 days following SBRT would be considered acute toxicity.
We also gave the patients patient-reported outcome instruments like the EPIC-26 and IPSS, which are well-validated forms to track quality of life, and we hope to assess urinary bowel and sexual toxicity as scored by patients as well. The trial, however, was powered and designed to investigate acute physicochemical urinary toxicity.
Watch and Share the Video here: https://oncologytube.com/v/41393
The trial opened and began enrolling patients in May of 2020. And by September 20, 2021, 100 patients had proven to be useful for the primary endpoint. So we did the interim analysis that I had indicated, and at that time, the interim analysis actually showed that there was a higher rate of toxicity in the CT-guided group.
Then, in the original hypothesis, which we expected because we were treating at a significantly higher dose, it also showed that we would need far fewer patients to show a difference in toxicity and would actually need closer to 154 rather than the hypothesized 300 patients. Because of this, we were actually able to close the trial early because of the favorable results of the interim analysis.
By the time 100 patients with prostate cancer were valuable for the primary endpoint, we had actually already enrolled and randomized 156 patients. So we had slightly over accrued, and we were able to stop the trial in September 2021. And what I presented just a few weeks ago at our national median were actually the final results of the trial, looking at acute toxicity in the whole cohort of 156 patients.
What Were The Inclusions and Exclusions Of The Mirage A Phase III Trial?
Just to remind you, this trial included any patient at my institution who could get SBRT. A patient would basically be evaluated through different consultations. Different options for treatment would be discussed, including SBRT but not just SBRT.
Again, many patients have many different available options. Once a patient agreed to being treated with SBRT, we did a simple one-to-one randomization of patients either receiving CT-guided or MRI-guided SBRT. The trial was not blinded. There are numerous reasons for this. The first is that it would be impossible to blind patients to this because patients getting CT guided SPRT need to have fiducial markers placed into the prostate, and we didn’t feel it was ethical to do a SAM procedure on patients that are not going to need fiducial markers placed.
The machines are very different from each other; in one case, the patient is being treated inside of an MRI. In the other case, they’re not, so it wouldn’t have been possible to be blind. And third, the physicians were not blinded in terms of assessing outcomes. Because the first trial was run during the height of the COVID-19 pandemic, as I mentioned, it opened in May 2020, and it was felt that it would be inappropriate to bring in a separate person to adjudicate toxicity because that would potentially increase exposure. And second, many times patients reference the type of treatment they receive. For example, a patient may say, “Oh, the worst part of the procedure was getting the fiducials placed, or at least the treatments weren’t as bad as I thought,” even though, as in the tube of a bore, many patients, yeah, tube inside the bore of an MRI.
Many patients will say things like that unprompted. That would also make blinding the physician adjudicator of toxicity impossible. So the trial was not blinded, but we feel that patient-reported outcomes, which we did collect, are relatively free of bias, at least in terms of the physician assessing toxicity.
So in terms of the final results of the Mirage trial, at least with respect to the primary endpoint, we found a statistically significant reduction in acute grade 2 or greater toxicity with CT guidance, with a toxicity rate of about 43.4%. And MRI guidance has a toxicity rate of 24.4%. So that was a statistically significant difference, met our RP value criteria, and showed that MRI guidance. Allowing this aggressive margin reduction led to significantly reduced acute grade 2 or greater GI toxicity in these patients. That was driven largely by frequency and retention side effects. There also was a significant reduction in acute grade 2 or greater GI toxicity, which we did not power the trial to show.
We actually found a 10.5% frequency of grade 2 bowel toxicity in patients getting CT-guided SBRT versus 0%, no events, in patients getting MRI-guided SBRT. And this was driven by proctitis and irritation-type diarrhea type symptoms. Now, as mentioned, we also looked at patient-reported outcomes because that is arguably more meaningful to patients and is perhaps less prone to, per se, the bias of someone adjudicating toxicity.
And what we found was that there was significantly less decrement in patient-reported urinary domains. For example, the EPIC-26 urinary incontinence score, which is a metric of leaking or dribbling with urination, showed significantly worsened in patients receiving CT-guided SBRT versus MRI-guided SBRT.
We also looked at the IPSS score, which is a metric. Prostatism symptoms that patients experience these are lower urinary tract symptoms and when we look at the proportion of patients who that have a 15 point or more increase in the symptom score, which is an indication of a huge amount of symptoms that were significantly reduced with MRI guidance as well.
It was about 19.4% of patients with that large magnitude increase in the CT-guided arm versus 6.8% of patients in the MRI-vetted arm. As a result, patient-reported urinary outcomes have improved significantly. We saw a similar pattern with bowel outcomes as well. So when looking at the EPIC-26 bowel domain score, there was a significantly greater decrement in patients getting CT-guided SBRT versus MRI-guided SBRT.
And when we look at the proportion of patients who have a clinically significant decline in their bowel quality of life, there are about 50% with CT-guided SBRT and 25% with MRI-guided SBRT. So all of those were statistically significant differences in patient-reported outcome metrics for both bowel and bladder toxicity.
Now, these differences have largely resolved in three months, and part of this is because we are intervening. If a patient is assessed after one month and they’re complaining of urinary or bowel symptoms, we would treat them. And so those differences, we would assume, go away, and the second is the tincture of time.
Acute toxicity is thankfully self-limited, and so we do see improvements in toxicity even in non-treated patients over time. So predominantly what we were detecting was a significantly greater toxicity burden at the one month timeframe that was resolved. Whether that’s due to tincture of time or by intervention or both. We did see a resolution in both groups by the three-month time point. Another point I’d like to make is that in any heterogeneous patient population, there may be imbalances between trial arms in terms of nodal radiation hormone therapy, using the gland size IPSS score at baseline rectal space for delivery of a higher dose of the gross intratumoral volume, and we controlled for these in a couple of ways, the main one being stratification. So every patient, upon randomization, or, I should say, prior to randomization, patients were stratified by IPSS and GLAM-SZLAG. So that’s how we balanced those. And the use of spacers, nodal radiation, and gross tumor volume boost were equal between randomized trial arms, even though we didn’t stratify on that, but we did a multivariate analysis anyway, trying to adjust for those. And we still found that the trial arm was a significant predictor of toxicity, the odds of acute grade 2 or greater urinary toxicity were 60% lower in patients getting MRI-guided radiotherapy.
Final Thoughts On The MIRAGE Trial In Patients With Prostate Cancer
This randomized trial looked at this novel MRI-guided radiation technology and showed that the aggressive margin reduction afforded by this improved technology significantly lowers the physician-scored toxicity in both the bowel and urinary domains, as well as patient-reported symptom burden in both the bowel and urinary domains in the acute timeframe.
Of course, longer-term follow-up would be necessary to see if this translates into improved late toxicity outcomes, which arguably would be more relevant to patients. But it stands as a proof of principle that tighter precision is leading to less toxicity. And it’s a reasonable conclusion that there will likely be less late toxicity as well, because fewer tissues were exposed to radiation.
That’s what’s causing the acute toxicity from exposure to radiation. However, it would take a very large trial—a much larger trial—to show a significant difference in late toxicity. So it’s possible that we might not see a statistically significant difference with longer-term follow-up. We will be looking at this closely, following these patients, and plan on reporting another analysis after 2 years’ worth of data are accrued for patient-reported outcomes, because we should be able to see if there’s a hint of difference at that point as well. However, it’s important to note there are other advantages; for instance, patients getting MRI-guided radiotherapy do not need to have fiducial markers placed.
So there’s no need for an invasive procedure. There’s no radiation dose from any onboard imaging, so there’s less dose to patients. So one could also argue that, even if all we see is an improvement in acute toxicity or an equivalent late toxicity, you have one option now that has less invasive procedures required and less dose to patients, and it might be attractive just from that standpoint alone.
I would say that MRI-guided radiotherapy in general is a very exciting technological advance for all patients, whether they have prostate cancer or not, but particularly in prostate cancer, where our cure rates are There’s a lot of emphasis on reducing patient toxicity burden, and I think the MIRAGE trial shows that this enhanced novel technology is able to reduce toxicity because of the precision that it affords.
Amar Kishan, MD – About The Author, Credentials, and Affiliations
Associate Professor Amar U. Kishan is the Chief of the Genitourinary Oncology Service and Vice-Chair of Clinical and Translational Research at the UCLA David Geffen School of Medicine and the UCLA Jonsson Comprehensive Cancer Center. Dr. Kishan earned Bachelor of Arts degrees in both molecular and cell biology and public health from the University of California, Berkeley. He got his medical degree from Harvard Medical School, where he also got the highest grade in the Harvard-MIT Health Sciences and Technology Program. Upon completion of his internship at Scripps Mercy Hospital in San Diego, he was named the H.H. Jones Intern of the Year. He subsequently finished his radiation oncology fellowship at UCLA.
Dr. Kishan is an expert in the use of radiation to treat genitourinary cancers (particularly cancers of the prostate and bladder). He has a lot of scientific experience and has written over 170 manuscripts, including articles in JAMA, Lancet Oncology, JAMA Oncology, European Urology, Cancer Cell, and the International Journal of Radiation Oncology, Biology, and Physics, where he was the lead author. He has gotten a lot of grant money from the Department of Defense, the Prostate Cancer Foundation, the American Society of Radiation Oncology, the Radiological Society of North America, the Radiation Oncology Institute, and the Jonsson Comprehensive Cancer Center. He is in charge of a busy program of clinical and translational research and has started a lot of new radiation oncology clinical trials.