Interpreting EMBARK Trial Results: Intensified Treatment and Hormone Therapy in High-Risk Patients - Christian Gratzke

June 14, 2023

Zach Klaassen hosts a high-level discussion with Christian Gratzke about the potential practice-changing outcomes of the EMBARK trial, specifically the impressive results of intensified treatment in high-risk patients. Dr. Gratzke suggests that intermittent hormone treatment could also be effective. The conversation also touches on enzalutamide monotherapy as a standalone treatment and how its success raises new questions about the necessity of ADT. Furthermore, the doctors discuss the impact of the EMBARK trial on side effects, as well as the controversial role of extended versus limited lymphadenectomy in prostate and bladder cancer patients. Finally, the conversation turns towards ASCO's upcoming presentation on AI clinical and histopathological biomarkers, which have the potential to significantly change approaches to the duration of ADT in treatment.


Christian Gratzke, MD, Urologic Surgeon and Chair, University of Freiburg, Freiburg im Breisgau, Germany

Zachary Klaassen, MD, MSc, Urologic Oncologist, Assistant Professor Surgery/Urology at the Medical College of Georgia at Augusta University, Georgia Cancer Center

Read the Full Video Transcript

Zach Klaassen: Hello, my name is Zach Klaassen. I'm a Urologic Oncologist at the Georgia Cancer Center in Augusta, Georgia. We are here live at ASCO 2023 in Chicago. I'm pleased to be joined by Dr. Christian Gratzke, who is the Chair at the University of Freiburg in Germany. Welcome, Dr. Gratzke.

Christian Gratzke: Thanks very much for having me. Pleasure.

Zach Klaassen: We're going to have a very high-level discussion today, looking at some of the highlights from the AUA we just had a few weeks ago, and then looking forward to some of the data that we are hearing about today at ASCO as well. Let's start with AUA. For urologic oncologists, no bigger trial than the EMBARK trial.

This was a big trial to have presented at AUA, potentially practice changing. Just to sort of go over that trial, this was patients with high-risk non-metastatic castrate-sensitive prostate cancer. These patients were treated with three arms. There was a luteinizing hormone arm, as well as the Enza plus LHRH, as well as the Enza alone. Basically, as we expected, with some impressive hazard ratios, we saw a decrease in metastases-free survival in both the doublet versus LHRH as well as the Enza alone. How do you interpret that data? We've had a few weeks to think about it now. How does it resonate with you?

Christian Gratzke: Well, I think this goes very well in line with our ideas of intensified treatment in those high-risk patients. When you look at the hazard ratios of I think 0.42 and 0.63, that's really impressive, and I think we need to reconsider our approach to our high risk patients very well. Looking at these results, I think in a clinical scenario with a biochemical recurrence, intensified treatment needs to be offered to our patients.

One of the remarkable aspects also was the fact that it's, I think, 36 weeks, if the PSA had dropped to below 0.2, they could have a pause of their medication, recovering from side effects. And if then PSA went up again above 2 for the radiation patients and about 5 for the surgery patients, they restarted medication. So the idea of intermittent hormone treatment comes into play as well. So I think, first of all, intensified treatment is key in that patient cohort, and maybe pausing treatment and starting it at a certain level is something we need to reconsider.

Zach Klaassen: Absolutely. I think what was interesting is these were truly high-risk patients, PSA doubling time less than 9. And also, when we look at those patients in the clinic until this trial came out, we had nothing for them. It was basically waiting. A couple other things I want to get your thoughts on is, do we need overall survival data to pan out for this to be practice changing?

Christian Gratzke: Well, I don't think so. I think with data that we have and those remarkable significant differences that we have, this is all we need in order to show our patients that, in that scenario, intensified treatment is definitely needed. Do we need to show overall survival? We don't need to. I think if we have data, it is nice to show, but it is not needed to indicate that sort of intensified treatment in that cohort.

Zach Klaassen: And MFS is a very meaningful endpoint for these patients, right? I think that's important too. Let's talk a little bit about the Enza monotherapy, which is unique. This is basically not including ADT whatsoever, just going straight to the enzalutamide, and these patients also did very well. What's interesting too is the testosterone levels don't go down. So, how do we reconcile what happens to these patients when they progress? They haven't seen ADT yet.

Christian Gratzke: Well, this is an area, I think, where we truly don't know the answer yet.

Zach Klaassen: Yeah.

Christian Gratzke: Because we don't have any data and no studies yet. But the notion, if we think about it, has been around for a long time that, why do we really need to add ADT to our novel hormonal agents? And this study has very nicely showing that it works without the ADT arm.

Zach Klaassen: Right.

Christian Gratzke: Now, the exact differences of which patients would actually then still profit from doublet treatment and which one doesn't, I think nobody's able to tell right now. But the observation that mono-treatment with Enza was achieving a remarkable result is a standalone result showing that there can be mono-treatment. But I think we need further studies to really trim down which patient would need doublet treatment, which patient just is probably well approached with monotherapy.

Zach Klaassen: The last point from EMBARK that I'd like to hit on is, when you look at the adverse events, it was actually quite similar with Enza and Enza plus ADT. How do we explain that and does this maybe affect how we approach patients and who we're going to treat with what?

Christian Gratzke: That's a very interesting question.

Zach Klaassen: Could be a discussion on all on its own.

Christian Gratzke: I know, I know. Well, I think that those side effect rates were not truly surprising given the abundant evidence that we have with Enza in the various clinical scenarios. How does those two drugs combine with regard to side effects in contrast to other combinations is still to be found out in further studies. But the fact that the side effect rate was very well tolerated by our patients, you look at the number of patients stopping the trial because of side effects, which was really low. I think that was a very good result.

Zach Klaassen: And I think to your point, if they get a good result at 36 or 37 weeks, they go off treatment for a while too, which is nice. Anything else from AUA that piqued your interest?

Christian Gratzke: Well, personally, I'm very interested in the fact and the question whether or not we should do extended versus limited lymphadenectomy in both prostate and bladder cancer patients. I think it was Dr. Karim Touijer from MSKCC who did present, again, his trial that was published 2 years ago, where he showed that extended lymph node dissection did not result in a better biochemical-free survival in those patients with prostate cancer, which was confirmed by this previous Brazilian study, which was very nicely done. The one notion that came up was always, well, maybe we are missing the one arm without any lymph node dissection. This trial is now up and running here in the US and it's very interesting to see what the outcome will be.

Now, the more important question may be the question whether or not modern imaging using a PET scan, PSMA PET scan, will be able to replace lymphadenectomy, which I think, at this time, we don't know and we can say it is already able to replace it, but there are trials ongoing and we may be able to stratify our patients before treatment into those who need dissection of lymph nodes and the ones who don't.

And so this year at ASCO, we will see in the area of bladder cancer and Dr. Seth Lerner presenting his SWOG trial showing very similar question that extended lymphadenectomy in bladder cancer patients did not result in a better overall survival and disease-free survival versus limited. And that study very nicely confirmed the previously shown results of LEA trial by the group from Juergen Gschwend in Munich that was presented at the EAU in Milan this year confirming very similar results. The SWOG trial, also with the extended lymph node dissection, there was a higher comorbidity. So we think we can say at this time of day that we don't need extended lymph dissection. The question is, which patient will eventually profit and which one won't? But I think there's abundant data that extended lymph dissection is not necessary in every patient, that's for sure.

Zach Klaassen: Yeah, it's very interesting. You hit on some great points. Going from prostate and bladder, similar questions for both disease spaces and I totally agree. As surgeons, both of us do these operations. When you're looking at that patient with a negative PET scan, in 5 years, are we going to be doing lymph node dissection? We'll find out. I think that's very interesting to see how this is going to play out over the next several years.

You kind of focused a little bit on ASCO, which is perfect. So let's talk about ASCO. One abstract that's going to be presented this week is very interesting and it kind of incorporates what we're seeing around us in general with AI. We're going to see data from an AI clinical and histopathological biomarker, basically taking high risk localized patients. And these patients are going to either get long-term or short-term ADT. Basically, what this biomarker showed was that 34% of patients that previously would've been given long-term ADT plus radiation are getting short-term with the same outcomes. So treatment de-intensification. Whereas 43% of patients in intermediate risk would've gotten short-term, which should benefit from long-term. Those are big numbers. How do you interpret that study? This is going to change how we think about the duration of ADT.

Christian Gratzke: This is terrific, because, in other words, that will mean that we reduce over and under treatment in those patients. And if you think about it, in real life, we talk with our patients and most patients do not want to have a hormonal treatment while undergoing radiation. So it is really important in a clinically relevant question whether or not you can actually reduce the amount of time that you expose them to hormonal treatment.

Zach Klaassen: Yeah.

Christian Gratzke: Having this AI-based biomarker will and may eventually be able to enable you to stratify those patients. Whether or not it is a prognostic marker, I think we don't know. It is a little too early to tell, but predictive for sure. And that's in line with many other studies trying to use AI in a sensible way. We are hearing this term all over the place.

Zach Klaassen: Not even just in medicine, just in real life.

Christian Gratzke: Right, right. Machine learning, AI. I think AI is as good as the data it is based on, so we need a very good database that is really solid to base our algorithms on. And if we have that, we may be able to use AI in a sensible way. That's my feeling.

Zach Klaassen: No, that's great. I think just from a broader sense, from urology standpoint, if we look at AI to grade biopsy slides for reading MRIs for minutes, it's here. It's going to continue to infiltrate where we're at. Is there any specific AI avenue of interest to you at this point?

Christian Gratzke: We've been looking at AI not only in uro-oncology but also in functional urology, where maybe if you think about a bread and butter field like BPH, you treat those patients with a drug for such a long time, can you actually find out at the very beginning which patient will profit from which drug so you don't lose any time by offering the wrong drug, in a way? So being predictive in treatment response to various drugs, I think that's the most interesting area, because we can avoid unnecessary exposure to a certain drug and avoid side effects.

Zach Klaassen: Absolutely. And maybe we can figure out how, when we do lymph node dissections, maybe that'll be the next AI battle. We've had a great talk. A little bit about AUA, a little bit about ASCO. Any take home messages for our audience today from either meeting?

Christian Gratzke: Well, ASCO, looking forward to all those posters and presentations. What was sort of the predominant field to my opinion at ASCO GU was the PARP inhibition. PROpel. There is TALAPRO-2 data, again, showing here, more data extended cohorts. There is very interesting combinations. Radioligand treatment in combination with PARP inhibition by Dr. Shahneen Sandhu from Melbourne. There is more interesting imaging data, for example, looking at data that showed that the positive predictive value of bone scans is really low. So maybe in a previously recorded oligometastatic setting, these were not truly positive, but negative, and you were actually talking about a localized disease.

Zach Klaassen: That's right.

Christian Gratzke: And those very simple observations maybe able to change our daily practice. So I think, yes, imaging is the one field where we see so many big changes, and that might impact our daily business very soon.

Zach Klaassen: Yeah, absolutely. It just shows, all these big meetings, there's big stuff at every single ASCO GU, ASCO, ESMO. It's incredible how much every single meeting we're learning a little bit more. Thank you so much for your time. It was a fantastic discussion. We really appreciate you joining us.

Christian Gratzke: Thanks for having me.

Zach Klaassen: Thanks.