Journal Club: Oral Relugolix for Androgen-Deprivation Therapy in Advanced Prostate Cancer - Zachary Klaassen
June 15, 2020
Christopher J.D. Wallis, MD, Ph.D., Instructor in Urology, Vanderbilt University Medical Center, Nashville, Tennessee
Zachary Klaassen, MD, MSc, Urologic Oncologist, Assistant Professor Surgery/Urology at the Medical College of Georgia at Augusta University, Georgia Cancer Center
ASCO 2020: HERO Phase III Trial: Results Comparing Relugolix, an Oral GnRH Receptor Antagonist, versus Leuprolide Acetate for Advanced Prostate Cancer
HERO Phase III Trial: Results Comparing Relugolix, an Oral GnRH Receptor Antagonist Vs. Leuprolide Acetate for Advanced Prostate Cancer - Neal Shore
Christopher Wallis: Hello and thank you for joining in this UroToday Journal Club. Today, we're discussing the recently published and presented HERO trial looking at oral relugolix for androgen-deprivation therapy in advanced prostate cancer. I'm Christopher Wallace, I'm a urologic oncology fellow at Vanderbilt. Joining me is Zachary Klaassen, an Assistant Professor in the Division of Urology at the Medical College of Georgia. We're discussing this paper published by the HERO study investigators in the New England Journal of Medicine and presented concurrently at ASCO Virtual Annual Meeting this past week.
Now, most urologists and urologic oncologists will intuitively know the importance of androgen-deprivation therapy in the management of prostate cancer, but just to circle back, this is 1941 in the seminal work from doctors Huggins and Hodges at the University of Chicago demonstrating the effect of castration and of estrogen and of androgens on serum markers in metastatic prostate cancer. Since that time, we've obviously evolved a long way, moving from surgical castration and the use of estrogens. An understanding of the hypothalamic-pituitary and testicular access led to the identification of GnRH and the introduction of GnRH agonist as a method to moderate expression along this axis. So, that was a Nobel Prize-winning work. To date, we now have four relatively commonly used GnRH agonists on the market, as well as a single GnRH receptor blocker or an antagonist in degarelix. All five of these agents are injectable agents. Comparing the approaches, LHRH agonists or GnRH agonists act through electronic stimulation leading to eventual suppression of LH and testosterone. FSH suppression, while initially induced, is not maintained longterm. In contrast, GnRH antagonists, which act more directly to block the activity of this hormone cascade, result in a more immediate suppression of FSH, LH, and testosterone, as well as more prolonged suppression of all three of these hormones.
As we mentioned before, there's a single GnRH agonist on the market, and it is injectable as are all the LHRH agonists. There's some data comparing these approaches, in terms of clinical outcomes. This is potentially the best data from Dr. Klotz, looking at a randomized comparison of degarelix and leuprolide. What we see is that the cumulative probability of testosterone suppression is high in all regimes. These are monthly injections, that's why you'll see the leuprolide dosing is somewhat different than typically used in clinical practice, then there's two different doses of degarelix, both with a leading initiation dose followed by a monthly maintenance. In all cases here, testosterone suppression from 28 to 364 days was high with grade range in between 96 and 98%. However, when you break it down to the first month or so, we see that there's a substantially faster decline in testosterone levels when using GnRH antagonists, and there is a small flare phenomenon, peaking around three days after injection for patients receiving leuprolide. These serum testosterone changes also translate into differences in PSA kinetics. We see a more rapid suppression of PSA in patients who receive the GnRH antagonists as compared to those who received leuprolide.
Today, degarelix has been the only available option, and this is an injectable GnRH antagonist, however, for a number of Phase I and Phase II trials, this novel oral GnRH antagonist has been developed, it was originally known as a TAK-385. So, we moved first from dose findings Phase I studies in healthy men to use in hormone-naïve, non-metastatic prostate cancer through to use as a neoadjuvant or adjuvant androgen-deprivation therapy in men undergoing external beam radiotherapy for localized disease. Finally, this is a presentation, and not yet a publication from Dr. Saad, but an interim analysis of Phase II in a more general population. It's worth noting that all of these clinical data have been published in the last year and a half or so. This is a very rapidly evolving aspect of prostate cancer care.
Now, that brings us to the Phase III trial that was presented and recently published from the HERO study investigators. This is the study overview. This is a multinational Phase III open-label randomized trial across 155 different sites. In men with advanced prostate cancer, and we'll talk about the details of that definition later, there's a two to one randomization to oral GnRH antagonist or intramuscular or subcutaneous LHRH agonist. And primary endpoint was suppression of testosterone, and again, we'll look at the details of both the inclusion criteria, treatment, and outcomes in the coming slides.
Study inclusion, as posted on clinicaltrials.gov, included all men aged 18 years and above who had a histologic diagnosis of prostate cancer and were considered by their treating physician to be candidates for at least one year of androgen-deprivation therapy. At the time of screening, they had to have a testosterone level above 150 nanograms per deciliter and a PSA above two if there was no local therapy or evidence of biochemical recurrence in patients who had a local therapy that's using a PSA threshold 0.2 in the surgery patients and the Phoenix criteria in the radiotherapy patients. So, we talked about the advanced prostate cancer definition a little bit on the flow diagram, and this is the three scenarios that study investigators included to consider patients with advanced prostate cancer. The first is biochemical or clinical recurrence following curative intent to local therapies, so that's surgery or radiotherapy. They also included men with newly diagnosed metastatic castrate-sensitive prostate cancer. And patients who had locally advanced disease, who are unlikely to be cured by local therapy and were treated with ADT monotherapy.
Patients were not eligible for the trial if they were felt by their treating physicians to require chemotherapy or surgery due to symptomatic burden of disease within two months of starting androgen-deprivation. But, if they had received prior ADT for more than 18 months, and for patients who had had ADT less than 18 months, they needed to be off treatment for at least three months prior to randomization. Patients who received cytotoxic chemotherapy for prostate cancer, prior to randomization, were also excluded, as were those with brain metastases scheduled for a recent surgery or had prior surgical castration.
Now, again, the flow diagram gave a slight overview of this, but the study used a two to one randomization strategy and randomization was stratified according to region, grouping North and South America, Europe, and then Asia and Pacific regions. It was also stratified according to the presence of metastatic disease as a binary variable, and according to patient age at randomization, again, as a binary variable, less than 75 or greater than 75 years of age. The primary trial endpoint was sustained castration rate. This was defined as the cumulative probability of testosterone suppressed below 50 nanograms per deciliter from day 29 through week 48 following initiation of therapy. There was a variety of secondary outcomes. The two most prominent were considered in a hierarchical testing strategy. The first was to assess the non-inferiority of the new agent to the standard of care for sustained castration with a non-inferiority margin of 10% considered. And finally, to consider that as a superiority analysis, as opposed to inferiority.
Other secondary outcomes included the probability of testosterone suppression of days four and 15, looking at the speed of castration, as well as the percentage of patients with a significant PSA response at day 15. The authors further defined a profound castration rate as testosterone less than 20 on day 15, and additionally looked at FSH levels at the end of week 24, as we know patients who have LHRH agonist, as opposed to a GnRH antagonist, may have loss of suppression of FSH as time goes on. Finally, they considered castration resistance-free survival. Although, this outcome has not yet been fully assessed, as followup is ongoing and thus the data for this are not yet reported. In terms of statistical analysis, the authors base their sample size on an assumed castration rate of 94% and 96% respectively and presumed a dropout rate of 15%, which actually turned out to be quite conservative. For their non-inferiority analysis, these assumptions led to a sample size calculated at 915 total patients required to give 99% power. Now, I'll hand it over to Dr. Klaassen to talk about the results.
Zachary Klaassen: Super. Thanks, Chris. For this trial, from April 2017 to October 2018, 1,327 patients were screened and subsequently, 934 were randomized two to one. As Chris mentioned, the target randomization number was 915, so they had adequate power for this study. That ended up being 622 patients receiving relugolix and 308 patients receiving leuprolide. There was more than 99% treatment adherence for this group, which we'll discuss in further slides. The median follow-up for both groups was 52 weeks.
Looking at the table one baseline characteristics, you can see that the median age was around 71, 72 years of age. There was a good balance between the age ranges in terms of about three-quarters of these patients being less than 75 years of age. Looking at the clinical disease presentation, the majority of these patients had evidence of biochemical or clinical relapse after local primary intervention with curative intent. You can see that looking at the Gleason scores, the majority of these patients had high-risk disease, either Gleason 7 or Gleason 8 to 10. Most of these patients had great performance status, ECOG zero in the majority of these patients. Roughly 11 to 12% of these patients had prior androgen-deprivation therapy. And about one-third of them have had prior radiotherapy. You can see that the median PSA level was 11.7 for relugolix and 9.4 for leuprolide. And all of these men had adequate testosterone at baseline. Finally, you can see that, of importance for this study, the cardiovascular risk factors, 67% of patients had lifestyle risk factors, nearly 80% had cardiovascular or cerebral vascular risk factors, and about 14% of patients had a history of MACE.
In terms of the primary endpoint, which was sustained castration rate, the HERO study hit both of the superiority threshold, as well as the non-inferiority threshold. You can see here on the left, that 96.7% of patients had sustained castration rate, and we can see that with the lower confidence interval being 94.9, this was a success for criteria of the lower boundary being greater than 90%. So, this was a positive study based on this primary endpoint. Subsequently, you can see that, in the subgroup analysis for sustained castration rate, the majority of these subgroups did favor relugolix, and you can see here that based on geography, based on the presence of metastatic disease, in age, all of these treated the GnRH antagonist.
Looking at secondary endpoints, all of these did favor relugolix as well. You can see that the non-inferiority analysis for sustained castration rate, which we previously mentioned was 96.7, for relugolix and 88.8% for leuprolide. The cumulative probability of testosterone suppression to less than 50 on day four was 56% for relugolix and 0% for leuprolide. Cumulative probability of testosterone suppression, less than 50 in day 15, significantly favored relugolix without almost a hundred percent of the patients being less than 50 nanograms per deciliter at day 15. Certainly, PSA response at day 15, with confirmation at day 29, was 79% for the treatment group and 19.8% for the control arm. Subsequently, this profound testosterone suppression to less than 20 nanograms per deciliter was 78.4 for relugolix and 1% for leuprolide. Certainly, as we mentioned before, the FSH level at the end of week 24 was much lower for the treatment arm compared to the control arm.
When you look at the mean testosterone levels among all patients, you can see this rapid decrease in testosterone with relugolix. At day four, the median testosterone level was 38 nanograms per deciliter, and not unexpectedly, as Chris mentioned in the introductory slides, we see this surge in testosterone with the leuprolide injection. You can see this profound decrease and sustained decrease with relugolix at day four. They also followed a subset of patients for testosterone recovery, and you can see here that the percent of patients with testosterone greater than 280 nanograms per deciliter at 90 days was 54% for relugolix and 3% for leuprolide. This is chartered nicely in this graph on the right. Looking at suppression of FSH and LH, you can see that relugolix suppressed FSH and LH very quickly and had sustained durable suppression all the way through week 49.
In terms of adverse events, any great adverse event was 92.9% in the treatment arm and 93.5 in the control arm. You can see that there are some very interesting data looking at the major adverse cardiovascular events. The overall number for relugolix was 2.9% compared to 6.2% for leuprolide. In the patients without a history of an adverse cardiac event, the percentage was 2.8% for the treatment arm and 4.2% for the control arm. What's really interesting is the patients that had a history of a major adverse cardiovascular event, the incidence of a subsequent event was 3.6% for relugolix and 17.8% for leuprolide. The most common adverse event occurring in more than 10% of patients was hot flashes at about 50% for both groups.
In terms of the cumulative incidence of major adverse cardiovascular events, you can see that the cumulative incidence of MACE at the end of week 48 was 5.6% for leuprolide and 2.8% for relugolix. You can see an early separation of these curves, and this led to a reduction of 54% event rate for MACE with relugolix with a hazard ratio of 0.46.
In terms of the discussion points for this trial, this was a very positive trial, with primary endpoint significant for the new oral GnRH antagonist. This was a beautiful figure from Dr. Mostaghel, who was an ASCO 2020 discussant, just basically highlighting the potential advantages in the groups that would likely benefit from relugolix, and you can see here, the rapid response with a lack of flare. This is important probably for patients with metastatic castration sensitive prostate cancer. It also avoids the anti-androgens that are often necessary for those receiving leuprolide. Certainly, we've discussed the fewer major adverse cardiovascular events, and this is certainly pertinent for men with a history of MI or stroke. We see very deep responses to testosterone with rapid recovery. This is particularly important from a clinical standpoint on those that may be considered for intermittent ADT or ADT in conjunction with radiation therapy. Certainly, in the current climate of the COVID pandemic, an oral administration may be more desirable for men to continue to take this therapy at home versus having to come to the clinic every three months or so for an injection of a GnRH or LHRH agonist.
Certainly, there was a possible concern for treatment adherence. When we see these men in the clinic and they get their three-month injection, we don't worry about them until the next three-month injection. We know that they're getting adequate treatment. To have a patient rely on oral therapy on a daily basis could be concerning. But, we see here in the HERO trial, 99% of patients were adherent to relugolix, and you may say that this is a clinical trial population, they're being closely followed, but when we look at data in the real world for oral therapies for men with CRPC, it's also translated to 92 to 96% adherence to oral medication. So, I think, that moving forward, these patients in our clinics will continue to be followed closely, but we can see from this real-world experience, as well as data from the HERO trial, that men are used to taking medications, especially at their age, and this is incorporated into their daily regimen of the medications they take.
Certainly, one of the major implications of this trial is the reduction in major adverse cardiovascular events. Certainly, the prostate cancer population is one that's at risk given their age and comorbidities. This is an excellent meta-analysis of six Phase III prospective randomized studies published in 2014 by Dr. Albertsen and colleagues, where they looked at 2,328 men recruited from trials from 2005 to 2012. They found that, in men with pre-existing cardiovascular disease, risk of cardiac events within one year of their ADT initiation was lower for the GnRH antagonists versus the LHRH agonist with a hazard ratio of 0.44 at a confidence interval of 0.26 to 0.74. When we combine this with the data from the HERO trial in men with a history of major adverse cardiovascular events, the HERO trial showed that relugolix had an incidence rate of 3.6% versus 17.8% for leuprolide, which is a 4.8 higher rate of these events in patients receiving leuprolide.
Several conclusions from this study, relugolix achieved castration as early as day four and demonstrated superiority over leuprolide in sustained testosterone suppression through 48 weeks of the trial. Second, relugolix had faster testosterone recovery after discontinuation, and perhaps one of the more important and exciting outcomes is that relugolix resulted in a 54% reduction in major adverse cardiovascular events. Taken together, relugolix has the potential to become a new standard of testosterone suppression in men with advanced prostate cancer. Thank you very much for your attention for this Journal Club discussing the HERO trial.