Brian Helfand, MD, PhD, and Steven Finkelstein, MD, FACRO, review results from the phase 2/3 EMPIRE-1 trial examining the use of ¹⁸F-fluciclovine-PET/CT imaging to guide postprostatectomy salvage radiotherapy for prostate cancer.
Brian Helfand, MD, PhD: Welcome to this Cancer Network® program. Today’s featured article is on 18- fluorinated fluciclovine-PET/CT imaging versus conventional imaging alone to guide postprostatectomy salvage radiotherapy for prostate cancer for the EMPIRE-1, a single-center, open-label, phase 2/3 randomized controlled trial. I’m Brian Helfand from NorthShore University in Chicago, and I’m here with Steven Finkelstein from the Florida Cancer Affiliates, US Oncology Network in Panama City, Florida.
This is one of the most important articles that people should be aware of, both for community and academic-based practices. Most people who are watching this understand that prostate cancer is a very complex disease, and many decisions that we make aren’t perfect. This is one technology that can help us with some decision-making. Nonmetastatic or localized prostate cancer can be treated with definitive therapies, including surgery—or radical prostatectomy—or radiation. For those men who undergo either of these therapies, unfortunately, it’s still cancer, and the cancer can come back. Many of these procedures, prostatectomy and radiation, have inadequate treatment responses. Disease can return in 20% to 40% of patients. Some of this is based on the aggressiveness or Gleason score of the surgical specimen or tumor type, and is influenced by many other adverse pathologic features. But if you underwent surgery, certainly adjuvant, meaning immediate, or salvage, waiting until your PSA [prostate-specific antigen] value starts to rise from zero, can be used to treat disease recurrence.
Molecular imaging is one way to help guide us in these treatment decisions because historically, we used conventional imaging in the form of CT scans and bone scans to try to locate where that disease recurrence is coming from. Unfortunately, many of these were inadequate, and even though it’s still considered standard, CT and bone scans aren’t that sensitive for picking up disease recurrence. Often, we have to wait until PSA values go as high as greater than 10 ng/mL or greater than 20 ng/mL before we can identify a target. Within recent years, fluciclovine, or a fluorinated PET [positron emission tomography]/CT imaging agent has been used that offers increased sensitivity and specificity for identifying disease recurrence after treating prostate cancer. Certainly, the PET/CT imaging has been FDA approved in the space of any type of failed therapy. This could be surgery, radiation, or it could be for men who have been on hormone therapy and whose disease continues to progress.
Specifically, the article that we’re going to talk about today focuses on men who underwent surgery and experienced a disease recurrence. This technology, fluciclovine-PET/CT imaging, has been used to help us identify or locate targets that can theoretically make subsequent radiation therapy more accurate and more improved. We’ve all accepted this as a paradigm to say that if we can better localize it and we can treat it, then that should give better oncologic outcomes. However, which is true in many aspects of all urology and oncology these days, is that even though we hypothesize something, we have to prove it. The EMPIRE-1 study is the first to attempt to answer that question, if using PET/CT imaging can influence oncologic outcomes.
Steven Finkelstein, MD, FACRO: EMPIRE-1 is a single-center, open-label, phase 2/3 randomized control trial. When the study was conceived and designed in 2011, molecular imaging was not routinely being done with PET radiotracers for prostate cancer. When we think about radiotracers prior to 2011, there were no trials that were randomized in the context of postprostatectomy that had tested if an agent could improve upon cancer outcomes. So, their eligibility was patients with prostate cancer. They needed to have a detectable PSA after prostatectomy and they had to have negative conventional imaging, which meant no extrapelvic or bone findings. Of those, there were 165 patients who were randomized in a 1:1 fashion. As you can see by the orange bar, they were randomized as patients to get radiotherapy directed by conventional imaging alone, which is the way that we had been doing radiotherapy, versus what we’re seeing here in the gray bar, which is radiotherapy directed by conventional imaging plus fluciclovine-F18 PET/CT scan. The primary end point was a 3-year event-free survival, and events included either a biochemical or clinical progression, or initiation of systemic therapy.
Brian, what are your thoughts about this particular EMPIRE-1 trial schema?
Brian Helfand, MD, PhD: Based on the actual design of the trial and when it was engineered, Emory University was and continues to be the site where the fluciclovine PET/CT imaging was born. It’s a single center so we acknowledge that, and I’m not sure that we could get beyond that. But we have to understand that Emory itself may be superior in its interpretation of all of the images because these are radiologists who have been working with this imaging from the start. Certainly, we have seen in looking at the trials that have all been based for fluciclovine, an improvement over time. Even at baseline when radiologists are first trained, they do a great job, but there is a learning curve. Emory has been well through its learning curve. Looking at the study design, as they randomized patients and patients had a recurrence, many of them were actually on hormone therapy for various amounts of time and some of them had various types of hormone therapy prior to enrollment in this study. Do you think that’s common, since back then it wasn’t standardized how we were giving the hormones?
Steven Finkelstein, MD, FACRO: That’s a really good point about androgen deprivation. Androgen deprivation is being used more commonly than previously in the setting of postprostatectomy radiation therapy as a result of randomized trial data. Additionally, RTOG-0534, which is a randomized trial showing significant incremental improvements going from prostate bed radiotherapy only to prostate bed radiotherapy plus short-term androgen deprivation, to prostate bed plus pelvic lymph node radiotherapy plus short-term androgen deprivation, could be used to justify both androgen deprivation and pelvic nodal radiotherapy for a higher proportion of patients, particularly patients with PSA concentrations that were higher than 0.34 ng/mL or greater. When we think about the clinical trials, the failure-free survival rates in RTOG-0534 were higher than those reported herein, with a 3-year failure-free survival of approximately 80% in the study group. You’ll see what the actual Kaplan-Meier data were with respect to EMPIRE-1.
Brian Helfand, MD, PhD: This is really the crux of the paper, and this was how men did in each arm of the group. This is a Kaplan-Meier curve that demonstrates the probability of patients in each group experiencing event-free survival. The bars would lower if patients experienced a disease progression or recurrence, or PSA increase throughout the study period. And as you can see here, the red bar is fluciclovine, the study group, which again got radiotherapy planned based on conventional imaging plus fluciclovine findings versus conventional imaging alone, shown in the blue bar. And as you can see here, at over 3 years follow-up, the conventional imaging had more events, so more men in those groups experienced disease progression throughout the study period. This was statistically significant and associated with a 1.84 hazard ratio, meaning that men in the conventional imaging group had a 1.84-fold higher likelihood of experiencing a result throughout the study period compared to men in the fluciclovine group, suggesting that the fluciclovine group offered additional oncologic control that the conventional imaging group did not experience.
Steven Finkelstein, MD, FACRO: When we explore the univariate analysis, variables that were significant predictors of event-free survival were a PSA greater than or equal to 1 ng/mL, the absence of extracapsular extension, the absence of seminal vesicle invasion, a Gleason score of 8 or higher, and whole pelvis radiotherapy. Other variables assessed were not found to be significant predictors of event-free survival and they included age, race, margin positivity, nodal positivity, the use of androgen deprivation therapy, and study group.
Brian, we’ve just gone through the univariate analysis. What do you think about these variables?
Brian Helfand, MD, PhD: These are very stereotypical, meaning that most studies that have looked at adverse outcomes, whether recurrence or progression to cancer-specific mortality, these are generally the variables that we’re looking at. Once PSAs get elevated, when we have adverse pathologic features, including pathologic stage T3 or greater, or when we have high-grade, high-risk Gleason grade group 4 or higher, we are going to see a higher risk of having adverse features. Many of these are predictable. However, when we look at these in a multivariable way, it helps answer the question and helps us stratify and identify if the treatment group, fluciclovine or not, impacted outcome.
Steven Finkelstein, MD, FACRO: That’s key. We are interested in the multivariate analysis.
This is event-free survival. It’s either biochemical recurrence, a clinical recurrence, or progression, or if someone initiated systemic therapy.
Brian Helfand MD, PhD: These are the results of the multivariable analyses. When we took all of those significant factors or variables on the univariate and we put this all into a statistical model again that adjusts for all of these other variables, we do see that the event-free survival was significantly higher risk in the conventional imaging group compared with the fluciclovine group. This is not unexpected based on the other results, but it really emphasizes that the hazard ratio once we adjust for all of the variables and correct for this was 2.04. So that means that men in the conventional imaging group had an over 2-fold higher risk of having an event over the study period, which again, was only a 3-and-a-half-year follow-up. The median survival event-free was not reached in either group, which was not necessarily surprising given the fact of the short follow-up in terms of prostate cancer treatment. That’s not unexpected. Are you surprised by any of these results, or what do you think about that about 12% difference between groups?
Steven Finkelstein, MD, FACRO: It’s quite remarkable. We have to phrase this in the mindset of most imaging modalities are not approved for improvement in demonstrating cancer outcomes. This is the first time that we’re using a PET radiotracer for prostate cancer specifically in the setting of postprostatectomy salvage. There are more events in the conventional imaging group, and that’s actually with fewer patients who were in the intention to treat. When we look at 3-year event-free survival, there’s a 12% improvement, the primary study end point. Again, this is using a PET tracer to guide final radiotherapy treatment decisions and target volume design. I’m very excited seeing these data, and the next thing I’m starting to think about is, you’re improving cancer outcomes, so what about the toxicity issues?
When we look at failure-free survival and adverse events, again, the median follow-up for this particular trial, EMPIRE-1, was about 3 and a half years. As we see on the left side of the slide, the failure-free survival at 4 years was 51% in the conventional imaging group versus 75.5% in the fluciclovine PET/CT group. Again, from our previous slide, we know there’s a 12% improvement, as Brian said, in event-free survival at 3 years, which was the primary end point. Now, all this doesn’t really matter if there are tremendous differences in toxicity. When we think about adverse events, the most common adverse events were actually very limited, with late urinary frequency or urgency in 46% of patients in the conventional imaging group and 41% in the PET group. Again, these were any grade events. With respect to acute diarrhea, it was 14% in the conventional imaging group, 21% in the PET group. The key point is that when you look at significant adverse events, ie, greater than 3, grade 3 toxicity events were infrequent, and there were no grade 4 or 5 toxic effects. You have a study that is demonstrating that there’s a 12% improvement in event-free survival at 3 years with really no changes in toxicity by changing the radiation therapy fields based on these imaging findings.
Brian Helfand, MD, PhD: In conclusion, fluciclovine PET imaging really has the ability to change our management and improve overall oncologic outcomes over at least a 3-and-a-half-year period without adding additional toxicities. This should be considered because not only can it help us locate where disease may be, but certainly, not only that location can help and influence treatment decisions, which has previously been evaluated in other trials including the LOCATE trial. Now we know, according to this paper, that it actually can improve our overall outcomes. I’m very excited by this and look forward to continuing to talk about this. Thank you for joining us.
Transcript Edited for Clarity