Read the Full Video Transcript

Neeraj Agarwal: Hi, my name is Dr. Neeraj Agarwal. I'm here today to welcome Dr. Wilbert Zwart from the Netherlands Cancer Institute in Amsterdam. We will ask Dr. Zwart to talk about his exciting talk he delivered at the ESMO 2024 meeting on epigenetic targeting in prostate cancer. Welcome, Dr. Zwart, to UroToday.

Wilbert Zwart: Thank you very much. Good day, everyone. So indeed, my talk for today is on targeting epigenetic pathways for improved treatment of mCRPC. And I would specifically like to focus today's presentation on a study where we looked for novel biomarkers for response prediction for AR-targeted therapy. And this is work which was spearheaded by these three individuals, and it was all embedded within a single-arm trial, which we performed within our institute, within the Netherlands Cancer Institute in Amsterdam, the Netherlands. This is a single-arm study where prostate cancer patients with mCRPC were enrolled. There was an imaging-guided biopsy taken prior to treatment. Patients were then treated with Enza or Abi combined with ADT. And for a subset of these patients, we also had an imaging-guided biopsy taken afterwards. This entire study was spearheaded by my close collaborator Andre, who's also a medical oncologist within our institute.

So in this study, we specifically focused on a specific epigenetic mark called Histone 3 Lysine 27 acetylation, which is a mark indicative of active promoters and enhancers, and it's found at sites where the androgen receptor binds the genome and is indicative of active transcriptional programs. When we compare patients who responded to treatment and patients who did not respond to treatment, we could find this very clear, distinct epigenetic profile which would demarcate patients who were resistant to the therapy, as indicated here in the green box.

This is also something we can very nicely quantify, and we can actually see that at these particular regions—that's a concise list of over 600 sites—there was a lot stronger signal for this active histone mark in patients who were later on not responding to the Enzalutamide treatment. We could further confirm this also in patient-derived xenograft models where tumors were implanted into mice. And on these tumors, again, the very same histone mark was profiled. We could confirm the expression or the level of that specific histone mark also in the mouse tumors, which also nicely associated here with differential signals. So basically, we could lift over the observation from patients to animals, and we could again also confirm that in mice where the signal was weak—so indicative also of response to castration in patients.

We can also see those mice also very nicely respond to castration, while when we do the very same analysis in mice where that specific histone mark was strong at this particular region, indicative of therapy resistance in patients, we could directly lift over all the predictability of resistance to castration in mice. So it's a full translation of the observation from humans to patient-derived xenograft models.

Next question was which proteins are now driving this phenotype? And for this, we made use of an in silico computational model where we analyzed over 14,000 individual experiments looking for overlap of specific proteins with our epigenetic regions we found to be associated with the therapy resistance, trying to find out which proteins bind there. And you can see this entire list here of proteins which were predicted to bind to these regions. And when we then do a knockdown in castration-resistant cell lines, we can actually see that when we knock down these factors, we can also diminish proliferation capacity of these cells.

There was one specific hit which was most interesting to us called HDAC3. Because we can target this specific protein, we can actually have small molecule inhibitors, and we can use these inhibitors to nicely, hopefully diminish the activity of HDAC3. And when we did this, we could find a very strong synergy, as indicated here in red, between Enzalutamide, our AR-targeted therapy, and the HDAC inhibitor Vorinostat.

And when we combine those two, we can actually see there's a nice drug-drug synergy, as indicated here, like I said, in red, not only in the parental cells but also in the castration-resistant cells. And even also in cells which we took from our patient xenografts—I saved those from the animals—again tried the drug synergy, and nicely we could show again synergy there.

And finally, when we tested the experiments also directly in mice in vivo, we can also see that we get a very strong effect of the HDAC inhibition in normal castration-resistant models as shown here in decrease of the proliferation capacity. So in a nutshell, I've shown that the epigenetic profiling in mCRPC reveals not only epigenetic-based biomarkers for therapy response prediction but also new synergistic drug combinations, which we've now shown that works in vitro, in mouse models. And we're very keen to translate this further to patients. And that's my presentation. This is my acknowledgment slide and everyone involved in this work. And yeah, I'm looking forward to this question. Thank you very much.

Neeraj Agarwal: Wow, Wilbert, this seems like a next generation of personalized clinical trial, if you will. So just to summarize for our audience, which are across the world actually for UroToday, you made it a very personalized approach. You grafted the tumor from the patient in the mice, which is patient-derived xenograft. Those experiments showed that patients who were not responding to Enzalutamide, the xenograft also did not respond to Enzalutamide and had increased histone markers. I'll keep it simple. HDAC3, for example, in this case. And if you treat those tumors with Enzalutamide, with Vorinostat, which is obviously an epigenetic modulating drug, you see better responses. Is it a fair summary?

Wilbert Zwart: Yes, there's a little nuance there because actually we use the patient-derived xenograft as an independent validation cohort. For the simple reason that such epigenetic datasets are not available for comparable types of clinical trials. So for this, we reached out to Eva Corey and Pete Nelson, and they already generated these patient-derived xenografts from an independent cohort of mCRPC patients. So we do the clinical observation in our phase two clinical trial. We identify the epigenetic profiles, we confirmed those epigenetic profiles in patient-derived xenografts in mice from where castrate-resistant tumors are implanted from patients. And we could directly confirm the original observation. And indeed, as you're pointing out, could show that we now find a new drug combination where we now target the HDAC, use the HDAC inhibitor together with Enzalutamide, then showing there's clear effects.

Neeraj Agarwal: So it looks like we may have stumbled upon a next generation of personalized medicine for patients with mCRPC who either have disease progression on Enzalutamide or are inherently resistant or short-term responders to Enzalutamide.

Wilbert Zwart: So we clearly found a biomarker which could stratify patients on outcome. And this biomarker is based on very distinct epigenetic profiles. Bear in mind we find over 40,000 to 50,000 individual regions on the genome where this specific histone mark is found, and only a bit over 600 very specific sites, which are programmatically changed over and over and over again between responders and non-responders, allow us to make a prediction which patients respond to treatment.

So this is something that is a clinical observation in a clinical trial, and based on this we can lift over to the mice. And I think just on the bigger picture, it's not only about response prediction, whether or not we can predict whether a patient will respond to Enzalutamide, but also the mindset that we can now start to identify new targeted interventions—not focusing on AR itself, but on proteins which are involved in AR function, what we now show here with the HDAC inhibitor. But also there's other work with, for example, the EZH2 inhibition, P300 inhibition, SWI/SNF proteins, all components which are involved critically in androgen receptor function. There's work from us and from others which could show that these factors actually are meaningful potentially in improving patient outcome in mCRPC.

Neeraj Agarwal: So HDAC inhibitors and EZH2 inhibitors are already in the clinic. They are available. They are being tested, especially EZH2 inhibitors are being tested in large phase three trials right now. So it looks like exciting news for our patients and also clinicians who are taking care of these patients.

Wilbert Zwart: Yes. Well, I think we are truly now in the era of the epigenome. There's a lot of clinical trial activity going on, as you're rightly saying, on using epigenetic drugs to target cancer cells in alternative ways. And I think there's a lot of excitement, rightfully so, on this type of mindset of, like I said, not targeting the androgen receptor, the clinical driver of prostate cancer development and progression, but targeting processes which make androgen receptor work.

Neeraj Agarwal: Yeah.

Wilbert Zwart: The androgen receptor is so much dependent on a favorable epigenome for it to perform its action that the moment we start to tinker with the epigenome, then you're targeting androgen receptor in an alternative way, which is bypassing many of the therapy-resistant profiles or therapy-resistant mechanisms.

Neeraj Agarwal: So moving forward, say three years or five years from now, how do you see we are treating our patients with these drugs?

Wilbert Zwart: So we're designing a new clinical trial now based on these observations. Again, a phase two study, where we now start to then combine AR-targeted therapy with the HDAC inhibitor in the mCRPC setting. And this is a trial which we're very excited about. We hope to be able to actually get everything in place to start such a trial soon, even though we're still dependent on a couple of things.

In parallel, I think also the exciting work on, as you were saying, with the EZH2 inhibitor, but also with BET bromodomain inhibitors, SWI/SNF inhibitors, which will give us a tremendous expansion of the entire spectrum of different therapeutic options we have.

And now it really comes down to what's then the best drug combination for the individual patient and what specific features would make sure that drug A would then be a better combination treatment together with AR-targeted therapy versus drug B. And I think here also that epigenetic-based biomarkers for therapy response, as I was just showing with our bit over 600 sites, which demarcate which patients will in the future not respond or will respond to Enzalutamide treatment, would then be a very interesting way to go, where we can really get an epigenetics-based companion diagnostics. We find not only patients who are requiring additional treatment, but we can also then identify which therapy would actually be going along with it.

Neeraj Agarwal: And how many patients do you think mCRPC patients who are either progressing on Enzalutamide or yet to start treatment with Enzalutamide may be candidates for these kinds of drugs? I will classify them together as epigenetic modulating drugs.

Wilbert Zwart: Yeah, that's a very good question, to be absolutely honest. So I'm a translational researcher, and consequently it's very hard for me to make a prediction on which patient population would benefit most per se. But it is very much the case that patients in mCRPC who are then receiving AR-targeted therapeutics, eventually they will all progress. And consequently, I would say that somewhere in the clinical spectrum between AR-targeted therapy by itself versus eventually more aggressive, more intense chemotherapeutics within that spectrum, epigenetic drugs would be very interestingly positioned, if you ask me as a translational researcher.

Neeraj Agarwal: Yeah. And I can easily imagine a situation where patients have disease progression—they're slowly progressing disease on Enzalutamide, for example. We take the tumor, analyze the tumor, there is upregulation of HDAC pathway or other epigenetic pathways, and we can add these medications in a personalized fashion. And based on the data you showed and everything pans out down the line, assuming everything goes well, these patients may respond just by adding epigenetic modulators to ongoing treatment with Enzalutamide. Is that a correct assumption?

Wilbert Zwart: Yeah, well, I think it's really the essence of personalized therapeutics, about giving the right drug at the right time to the right patient. And now also with the era where slowly biomarkers and also epigenetic biomarkers are now transitioning from tissue-informed decision-making now towards a liquid biopsy type of decision-making where also there we can actually assess now epigenetic profiles in plasma. Now we're getting into a very interesting phase that you can then look at a plasma sample and then try to make a decision on what will be the most appropriate treatment for that particular patient at that particular phase of the disease. And when you get certain epigenetic profiles altered, that could then potentially be the right time to give another therapy instead, again, targeting potentially the epigenome.

So of course this is all future development, but I think we have arrived in a very, very exciting time and there's a continuous new development of new therapeutic options along with new biomarkers to actually select those patients. So it's a very, very interesting time to do cyber studies.

Neeraj Agarwal: Indeed. Congratulations again to Dr. Wilbert Zwart from the Netherlands Cancer Institute for this fantastic data, and thank you for taking the time to share your results with us. With that, we'd like to thank the viewing audience for your time, and we look forward to this new era of personalized medicine with epigenetic modulators in our clinic in the near future.

Wilbert Zwart: Thank you.