Enzalutamide in biochemically recurrent prostate cancer: key findings from the EMBARK study

Hello everyone, I’m Dr. Neal Shore. I’m the medical director of Carolina Urologic Research Center in Myrtle Beach, South Carolina, and it’s a great pleasure to be joined by my good friend and colleague from Australia, Dr. Henry Woo. Maybe, Henry, you’d like to introduce yourself, please?

Sure, it’s great to be joining you on this podcast, Neal. I’m an academic urological surgeon, based in Sydney, Australia

So, it’s really a great pleasure to have everybody come back. I hope you listened to our first podcast in this series examining the phase III EMBARK study. Today’s podcast is entitled “Enzalutamide in biochemically recurrent prostate cancer: key findings from the EMBARK study.” And I know Dr. Woo and I are very proud to have been coauthors on this, along with so many other wonderful people, and wonderful sites, patients, and families. The podcast is funded by Pfizer and Astellas, and they assisted with some medical writing and editorial assistance. Our disclosures are within the podcast article.

Today, we’re going to discuss the key efficacy and safety findings from EMBARK, we’re going to consider some of the patient-reported outcomes [PROs] and their implications for quality of life [QOL], and review some of the implications for clinical practice. I’d like to start by providing just a brief reminder of what the EMBARK design looked like. This will be familiar to those who’ve already listened to our first podcast, but it’s helpful to get an overview for putting the findings into that context. So, EMBARK is the first study designed to look at whether early treatment with enzalutamide plus leuprolide—we call that enzalutamide combination therapy—or enzalutamide monotherapy in patients with high-risk, nonmetastatic castration-sensitive prostate cancer—the acronym nmCSPC—improves outcomes versus placebo plus leuprolide, which was at the time of our trial design sort of the standard of care [1]. EMBARK is an international, randomized, phase III clinical trial. It was sponsored by Pfizer and Astellas, the developers of enzalutamide. So, the full details relating to the design of the study have been previously published in BMJ [British Journal of Medicine] Open, and they’re again very nicely discussed in our first podcast [1].

So, briefly, patients were randomized in three arms, three groups in a 1:1:1 ratio to receive enzalutamide plus leuprolide, the combination group, versus a placebo plus leuprolide, or leuprolide alone group. Now, these two groups were blinded, so that was double blind. Or in the third group or arm, they could receive enzalutamide monotherapy and that was open label [1].

So, treatment—and this was kind of a really interesting aspect when we designed the study—was suspended at week 37 if the PSA [prostate-specific antigen] level was <0.2 ng/mL at week 36 and was restarted when the PSA level was at least 5.0 ng/mL if the patient had not had previous radical prostatectomy [RP] or at least 2.0 ng/mL if the patient had previous RP [1]. And that, those were sort of based upon, at the time, the feeling that that would be more likely to show the potential for metastatic disease. These thresholds were designed to allow patients that 37-week interruption, so they can have off time from therapy, but low enough to prevent metastasis [2]. So, that was our sort of our clinical construct.

The primary endpoint was MFS, or metastasis-free survival, in the combination group compared with the leuprolide alone group; that was the primary endpoint [1]. MFS was utilized as it has been demonstrated to be a very strong surrogate for overall survival [OS] in localized prostate cancer [3]. Important key secondary endpoints included MFS in the monotherapy enzalutamide group compared with the leuprolide alone group; time to PSA progression, time to the use of a new antineoplastic therapy, and OS, and this would be for all three groups [1]. Safety and PROs were also important secondary endpoints [1]. So, let me pass it over to you, Henry, and if you could talk about some of the key findings from the study.

Sure, Neal, very happy to walk our audience through this. Now, we had a total of 1068 patients who were enrolled and randomized into EMBARK at 244 sites in 17 countries—so, this is a truly global study—with enrollment completed between January 2015 and August 2018. Now, here, we discuss the results after 5 years of follow-up, with full details which have been published in the New England Journal of Medicine [NEJM], as well as in NEJM Evidence [4, 5].

Now, the baseline characteristics were balanced across the three groups. Overall, the median age was 69 years; and from an ethnic racial perspective, most patients were White, that was 83%; and the median PSA was 5.2 ng/mL; and the median PSA doubling time was 4.9 months [4]. And just as a reminder, the entry criterion was a PSA doubling time of ≤9 months [1].

Neal, do you have any other comments you’d like to make at this juncture before I move on to the key efficacy results?

No, that was a really good, concise review, and I think our audience really wants to understand about the results and that was what was really exciting after it took us close to 9 years to get these results. So, thanks for doing this, Henry.

All right, in that case, we ought to cut to the chase and get into the key efficacy results. Now, for the primary endpoint, after a median follow-up across the three groups of 60.7 months, enzalutamide combination therapy was shown to significantly improve MFS compared with leuprolide alone [4]. Now, the 5-year MFS was 87%, and 95% confidence intervals with this was 83.0% to 90.6% in the combination group, and 71% with confidence intervals at 65.7% to 76.3% in the leuprolide alone group [4]. The risk of metastasis or death was 58% lower in the combination group compared with the leuprolide alone group and this had a hazard ratio of 0.42, which is highly statistically significant [4].

The enzalutamide monotherapy group also had a significantly improved MFS compared with leuprolide alone and this was a secondary endpoint of the study [4]. The 5-year MFS was 80%, and the 95% confidence intervals for this were 75.0% to 84.1% in the monotherapy group, which represented a 37% lower risk of metastasis or death versus leuprolide alone, and the hazard ratio here was 0.63, and again, this was highly statistically significant [4]. Now, compared with leuprolide alone, both enzalutamide combination as well as enzalutamide monotherapy significantly lowered the risk of PSA progression; significantly increased the time to first use of new antineoplastic therapy; and resulted in improvements to other secondary outcomes, such as longer time to metastasis, symptomatic progression, and first skeletal event [4].

Now, OS was improved for enzalutamide combination compared with leuprolide alone; although, this did not meet the predefined threshold for statistical significance. The hazard ratio for death, nonetheless, was 0.59 [4].

So, once again, we might just stop at this juncture. Neal, if you had any further comments that you’d like to make with regard to the efficacy data or if you think we’ve said it all?

Well, I think you did say it all, but it’s kind of fun to realize that this was almost two studies in one, right? We had the combination—LHRH [luteinizing hormone-releasing hormone] and enzalutamide—versus LHRH alone with a placebo; and then, we also had enzalutamide monotherapy—no ADT [androgen deprivation therapy], no LHRH—versus LHRH alone in our statistical analysis plan [1]. And really great to see that the combination as well as the monotherapy beat leuprolide alone with statistically significant findings; hazard ratios, as you point out, of 0.42 and 0.63, respectively [4]. So, I think this was why it was a plenary at the AUA [American Urological Association] in 2023; this is why it had an NEJM publication [4]; and this is why it’s been now approved by the FDA [US Food and Drug Administration] and EMA [European Medicines Agency] [6, 7], and I think most international guidelines [8–10]. So, great overview, Henry. I think we clearly hit the efficacy MFS endpoint, but, maybe you can tell us now about some safety and the PROs.

Yeah, that’s also going to be very important to consider safety and PROs. And I think you’re absolutely right with the excitement about the monotherapy side of this study, because there hadn’t been any major randomized controlled trial evaluating an ARPI [androgen receptor pathway inhibitor] as a monotherapy. And to some extent, I think that we might be a little bit distracted from considering that because there were so many other profound findings that came out of this study. But anyway, we might move on to the key results.

There were no new safety signals observed during the EMBARK study [4]. So don’t forget, we’re using a drug that has already been through numerous randomized controlled trials, and so, therefore, we already actually know a lot about this particular drug. And so, the findings with this study were consistent with the known side effects in patients with advanced prostate cancer. Over 97% of the patients in all treatment groups reported an adverse event [AE] of any grade [4]. Although the frequency was similar amongst all three groups, there were some differences in the nature of those AEs observed [4]. The most common AEs, which occurred in >30% of patients, in the combination and leuprolide alone groups were hot flashes and fatigue [4]. The most common AEs in the monotherapy group were fatigue and gynecomastia [4]. Although still observed in over one in five patients in the monotherapy group, patients who received enzalutamide monotherapy had fewer hot flashes compared to those in the combination and leuprolide alone groups [4]. However, gynecomastia, nipple pain, and breast tenderness were more commonly reported in the monotherapy group compared to the other groups [4]. Most AEs were mild to moderate in severity [4].

Now, with regards to PROs, the baseline assessments suggested that most patients in the trial were asymptomatic at the baseline [5]. Overall, the PRO scores were generally stable across all treatment groups throughout the trial, with no clinically meaningful differences amongst the groups [5]. Notably, there was no difference observed in the functional status as measured by the Functional Assessment of Cancer Therapy – Prostate, that’s the FACT-P total score [5]. And there were no differences in the time to first and time to confirmed clinically meaningful deterioration, and worse pain in the past 24 hours determined by the BPI-SF [Brief Pain Inventory – Short Form] item 3 [5]. A high level of health-related QOL was preserved with both enzalutamide combination and monotherapy whilst improving MFS compared with leuprolide alone [4, 5]. So basically, we’re seeing that patients were deriving a great benefit from enzalutamide without paying a price, a significant price, in terms of their health-related QOL.

Yeah, now, that was fantastic, Henry, a lot to unpack there. I encourage our colleagues to read the NEJM Evidence paper, which goes into a lot of good detail on the PROs [5].

I think it’s important to note that in the enzalutamide monotherapy group, we did not offer or suggest any prophylactic strategies to mitigate any breast-related symptoms. Our audience is very savvy, folks who have been treating prostate cancer for a long time really understand that ADT alone is—we say is no free lunch, right? It’s remarkably hormone sensitive as a solid tumor, probably the most—prostate cancer compared to others is hormone sensitive. But we do see when you lower T [testosterone] levels and estradiol levels, patients can develop pretty significant fatigue and clearly hot flash, the thermal regulatory disequilibrium is affected. And then there’s other things too that our colleagues are familiar with, probably less so on the enzalutamide or the ARPI monotherapy; a lot of our colleagues over the years have maybe given bicalutamide monotherapy. And I think more and more we’re going to see that because we have this approval now of enzalutamide monotherapy and the ability to have the shared decision‑making with patients, versus combination ADT and enzalutamide versus enzalutamide alone. It’s really important to keep up to date with the ongoing literature and there are papers coming out on some strategies not only to mitigate the sequelae of T suppression and ADT that you and I both practice, but even now also there are interesting things coming out on enzalutamide monotherapy.

So, Henry, that was really great taking us through these key findings, but before we wrap up, I want to briefly examine some clinical discussion points that have come up from the EMBARK study. And here are some of the things we regularly hear. I think we’ve already addressed some of it, but maybe we can talk about it before we finish the podcast.

During the EMBARK trial, there were three separate phase III trials in patients with nonmetastatic castration-resistant prostate cancer, nmCRPC, and that had a PSA doubling time of ≤10 months that reported improved clinical outcomes, and that was the SPARTAN trial with apalutamide [11], the PROSPER trial using enzalutamide [12], and the ARAMIS trial using darolutamide [13]. Because of those three important nmCRPC trials, treatment intensification with an ARPI became the standard of care for patients with nmCRPC. From February 2019, EMBARK investigators were, therefore, informed by the central laboratory if a patient had PSA progression with a doubling time ≤10 months as those patients could become eligible for these treatments from those three different nmCRPC studies. As EMBARK is a global study, the availability of treatment varied across our sites and the choice for intensifying treatment was naturally a shared decision-making moment between the patient and their healthcare provider [14].

EMBARK was designed with a single treatment suspension period. Patients who reinitiated treatment after an initial suspension did not then suspend treatment a second time if they reached an appropriate PSA threshold. We get that question a lot, and I think that’s an area of ongoing investigation and further discussion [14]. You know, what are the real risks and benefits of true serial intermittent therapy with enzalutamide in combination or enzalutamide monotherapy? And we really need to better understand that. We had one separate defined treatment interruption at 37 weeks.

There have been considerable developments in imaging technology during the course of EMBARK and notably—and we mentioned this in podcast 1—is the development of PSMA PET, or prostate-specific membrane antigen positron emission tomography—a lot easier to just say PSMA PET. And our colleagues—and Henry, I know you’ve been integral on this in Australia—have really been key in leading the way, the world’s understanding of PSMA PET and the whole field of theragnostics. In EMBARK, we used conventional imaging, but PSMA PET imaging has the ability to detect now tumors in patients with high-risk BCR [biochemical recurrence] that may not have been detected by the conventional or standard imaging that we used in EMBARK, because that was what was available at the time [14]. So, I think it’s important to note that further research on how to apply PSMA PET results to these treatment decisions for our nmCSPC high-risk BCR patients is required.

Enzalutamide is now recommended in treatment guidelines, pretty much globally, with or without leuprolide for high-risk patients with nmCSPC. And that is nonmetastatic by conventional imaging; a doubling time of the PSA of ≤9 months; a PSA of >2 ng/mL above the nadir if they have had radiation therapy [RT] or >1 ng/mL after RP with or without postoperative RT; and they are not necessarily considered a candidate for pelvic-directed salvage RT. Treatment intensification with enzalutamide monotherapy or enzalutamide in combination with ADT, as we have said, should be a shared decision between the patient and the healthcare provider [14]. As Dr. Woo pointed out, there are differences in the AE profile [4]. Both of these treatment options met their statistically significant endpoint of MFS with statistically significant hazard ratios [4]. But, it’s a really nice opportunity now to individualize patient care. Both of these treatments delay disease progression while maintaining QOL, that’s what EMBARK taught us [4]. But there are differences in treatment duration; there are some differences in the AE profile and the QOL that you really want to have that nice shared decision-making conversation with patients to see what they prefer and what is most appropriate [4, 5, 14].

You know, with all clinical trials, there are always some limitations to the study that we should briefly consider. Let me sort of just outline these. Right now, our OS data are immature [4]; it is trending positively, but it is immature. We are following all these patients and we are excited to report on the results of that; we will keep everybody posted. The non-White patient group was definitely underrepresented [4, 5]. The long-term consequences of prolonged enzalutamide exposure and tolerance to subsequent therapies are being investigated, but not fully established [4]. That said, enzalutamide has been used now for over a decade in more advanced disease, but we will want to continue to follow and see the long-term implications in the high-risk BCR group.

The unblinded monotherapy group confounds somewhat the interpretation of the PRO analysis [5]. Furthermore, some of the PRO analyses were limited in their nature; for example, in the instruments that we had, the sexual function score was captured as a functional domain of QLQ-PR25 [Quality of Life Questionnaire – Prostate 25], but dedicated sexual function questionnaires were not performed [5]. And I think this is a very interesting and important area of ongoing active investigation, and enhancing or improving better questionnaire instruments and tools. So, the meaning of these detailed analyses of some of these PRO endpoints was not necessarily possible. I think that’s where folks are going to have some empiric considerations to using the combination versus monotherapy for the patients where libido and sexual function is something that you would want to discuss. All of our QOL analyses also need to be interpreted within the context of a trial. In other words, the setting of the combination group with a one-time treatment suspension because there are potentially differences that happen in the real-world setting [5].

I really want to thank you all for listening. Before I do any key takeaways and summarize really the outstanding synopsis you did, Henry, any other final comments?

No, I think that you have wonderfully summarized the limitations and I think also have articulated where there’s a lot of food for thought on what more we can learn. It’s interesting how we have learned a lot from this study, but, in a way, it makes us hungry to learn a lot more.

Yeah, extremely well said. That’s why we keep doing these clinical studies; they are invigorating, and their intellectual vigor is there. I always think to help colleagues that there’s a lot of challenges in healthcare with regulatory overburden and things of that nature, but you can avoid a lot of the burnout by getting excited about clinical trials and clinical research.

So, key takeaways, I will just briefly summarize that EMBARK illustrated, teaches us that enzalutamide has clinical benefits in patients with high-risk BCR, or the BCR population, who had definitive treatment for their prostate cancer, classically an RP, RT, or both. The combination group of enzalutamide and leuprolide as well as enzalutamide monotherapy, both resulted in statistically significantly longer MFS and a longer time to PSA progression and receiving the next antineoplastic therapy versus leuprolide alone, while maintaining overall QOL [4].

Enzalutamide was approved by the US FDA for nmCSPC with BCR at high risk for metastasis based on the results of the EMBARK study [6]. And the Committee for Medical Products for Human Use, CHMP, of the EMA also approved the use of enzalutamide as monotherapy or in combination with ADT for the treatment of patients with high-risk BCR nmCSPC who are unsuitable for salvage RT [7].

I really thank the audience for their attention, and Henry Woo, thank you so much for doing this together and for all the great work you are doing in research.

Author contributions

Neal D. Shore (NDS) and Henry H. Woo (HHW) critically reviewed and edited the discussion guide for this podcast before submission and during revision, and agreed on the journal where the podcast was submitted. NDS and HHW agree to take responsibility and be accountable for the contents of the podcast as recorded, and to share responsibility to resolve any questions raised about the accuracy or integrity of the published work.