AIM High: Epigenetic Modulation and Immune Stimulation in Bladder Cancer - Elizabeth Koehne

March 15, 2024

Sam Chang engages in an conversation with Elizabeth Koehne who delves into her research, AIM HIGH: DNA Methyltransferase Inhibition and Immune Stimulation in Bladder Cancer. The project explores the potential of making bladder cancer more susceptible to immunotherapy by reversing epigenetic mechanisms, like DNA methylation and chromatin remodeling, that the cancer uses to evade the immune system. With an eye on future clinical applications and the potential for phase one clinical trials, Dr. Koehne's work could significantly impact the treatment landscape for bladder cancer, striving to convert "immune-cold" tumors into "immune-hot" ones that are more amenable to immunotherapy.


Elizabeth Koehne, MD, Urologist, University of Washington, Seattle, WA

Sam S. Chang, MD, MBA, Urologist, Vanderbilt University Medical Center, Nashville, TN

Read the Full Video Transcript

Sam Chang: Hi, I'm Sam Chang. I'm a urologist in Nashville, Tennessee, at Vanderbilt University Medical Center, and we are very fortunate to have a rising star in urologic oncology join us today, Betsy Koehne. I've known actually since she was a resident and has recently finished her urologic oncology fellowship at the University of Washington.

She was one of the most sought-after fellowship candidates, and it showed as she won a Urology Care Foundation Research Scholarship Award, and was actually just recently named as the outstanding graduate of that scholarship award. So she really is a star, and the University of Wisconsin is very, very fortunate to have her, and she'll be joining them sometime, I guess this summer, after she graduates.

So Betsy, thank you so much for spending some time with us. We've asked you to discuss some of the research behind this scholarship award, this research scholar award that you got from the Urology Care Foundation, which I guess is looking at a variety of things, looking at the immune environment, its impact on bladder cancer or rather what bladder cancer does to maybe hide from immunotherapy and what we can do to adjust the immune environment so that we can actually perhaps maybe get more successful therapies.

But I'll let you talk a little bit about your research. We'll ask some questions and then we'll look forward to hearing what kind of future research you have on hand. So first of all, thanks so much for joining us.

Elizabeth Koehne: Thank you. Thank you so much for having me. I'm really delighted to be with you today. Okay, so the title of my project was AIM HIGH: DNA Methyltransferase Inhibition and Immune Stimulation in Bladder Cancer. And I'll talk a little bit more about where the name AIM HIGH comes from in a minute.

Essentially, our motivation for this project was that, well, first, immunotherapy represents a critical therapeutic field for patients with all stages of bladder cancer. However, some patients still don't respond, and those response rates certainly vary with the disease stage and the agent used. But we always need more options for our patients who may not have any other options.

And so one of the ways that potentially some patients don't respond to immunotherapy or their cancer doesn't respond, is that bladder cancer uses epigenetic tactics to hide from the immune system. Two of those mechanisms are DNA methylation and chromatin remodeling, and those are both ways that bladder cancer modifies its DNA to repress transcription of tumor suppressor genes and other things that would alert the immune system that it's there.

And so we wanted to see if we could make bladder cancer more susceptible to immunotherapy by overcoming mechanisms of resistance such as these two ways. And so first, we started looking at DNA methyltransferase inhibitors, and one of those drugs is 5-azacytidine, which is an old drug that used to be used for myeloid malignancies and isn't used very commonly now in the clinical setting, but when it's used in very low doses, it becomes incorporated into DNA and then prevents methylation at that site.

And so, over successive rounds of replication, this hypermethylation that potentially cancer is doing could be reversed, thereby maybe making or reactivating transcription of tumor suppressor genes. And then histone deacetylase inhibitors or HDAC inhibitors are kind of similar but work in a different way in which they also loosen chromatin to make genes more amenable to transcription.

And so we were inspired by another study in which a group of researchers looked at breast, colorectal, and ovarian cancer cell lines, human cancer cell lines, and they found that when they treated those cells with 5-azacytidine or AZA, they could increase expression of a panel of 317 immune-related genes, which they termed the AZA immune gene panel or AIM. And so we hypothesized that bladder cancer with a low AIM expression at baseline could be therapeutically converted to have high expression of these genes for a future combination with immune checkpoint inhibitors or other immunotherapy options.

And so that's where our title AIM High: DNA Methyltransferase Inhibition and Immune Stimulation came from. And this was sponsored by the AUA Urology Care Foundation. And so just again to review the principle of our project: we wanted to see if we could turn immune-cold bladder cancer, that is hiding from the immune system and therefore probably not very susceptible to immunotherapy, if we could convert it to an immune-hot bladder cancer that would potentially be more responsive to immunotherapy.

And so this was a project that I mostly did in the lab during my research year at the University of Washington. And the first part of the project was just finding very low doses of these drugs, AZA, and then actually used an HDAC inhibitor and explored multiple HDAC inhibitors but ended up using one, and then even looking at those in combination.

And we wanted to find a low dose that would leave more than 70% of the human bladder cancer cells alive because we weren't really trying to kill all of the cells, more so priming them for future treatments and increasing expression of immune-related genes.

And so, once I determined the doses of these drugs that we could use, I treated human bladder cancer cell lines in the lab, and they were each treated for 72 hours, and then I collected these cells and harvested RNA for RNA sequencing, and then used that RNA to look at things such as consensus molecular subtype class and potentially if that switched, which is a way to categorize bladder cancer by its clinical and molecular features.

And then also looked at clustering of immune-related genes and immune pathways. And finally, I'll discuss briefly at the end, we used our repertoire of clinical bladder cancer specimens to see if this AIM gene signature or immune gene signature correlated to pieces of bladder cancer that came from patients.

So first, as background, we looked at the Cancer Genome Atlas to see if this AIM gene signature was relevant or expressed in human bladder cancer. And so this represents muscle-invasive bladder cancer, and we found that about 20% of muscle-invasive bladder cancers in the TCGA are AIM high at baseline or intrinsically have a high expression of immune-related genes, and then maybe about 30% are low or immune-cold bladder cancers. And the middle is somewhat mixed.

Comparing that to the consensus molecular subtype classification system to see if that correlated, we found that the immune-high or AIM high group was enriched with the basal squamous molecular class. And then, similarly, actually, on the flip side, the AIM-low group was enriched with LumP, luminal papillary.

And then I used 11 different human bladder cancer cell lines, and so far have sequencing from four of those. We looked at the Cancer Cell Line Encyclopedia to see if any of these human bladder cancer cell lines were already immune-high or expressed a high amount of this AIM gene set.

We only found one that was at baseline, and that was SCaBER, which is noted here. And then the other cell lines were not. Most of the cell lines were basal squamous, and then two were luminal papillary that I used. And so now moving on to our RNA sequencing results, we did find, and so these are after the cells were treated, and then we did the RNA sequencing to look at the expression of this AIM gene set. We did find an increased expression of a subset of the AIM genes. And so about 124 of the original 317 genes were upregulated overall in our cohort of these four different cell lines.

And so those genes represent genes involved with inflammation, interferon, cytokine, and chemokine signaling, and cancer testis antigens or CTA, which is a family of immunogenic proteins. And so first, the two cell lines that had the most uptake were both low at baseline, and one was SW780, and the other was T24. And so one of those was a LumP line, and the other one was basal squamous. The third line that had the third most uptake or second least uptake of AIM genes was SCaBER, which is the one that we thought was already high at baseline. And then lastly, this line RT4, which is a LumP line, had the least change with AZA treatment.

And so this is highlighting that some of these genes that were upregulated do have clinical relevance as they were shown in the PURE-01 study when patients were treated with neoadjuvant pembrolizumab. These are genes that we were able to upregulate that were shown in this study to correlate with response to pembrolizumab in those patients.

And then in the secondary analysis of the PURE-01 cohort, with another couple of cohorts included, these are genes that they found correlated with response to immunotherapy that we were able to upregulate with AZA treatment.

The next part of our study was that we also looked at co-treatment with low-dose HDAC inhibitors, and ultimately we were able to find some additional upregulation of AIM genes that wasn't seen with just AZA alone. So potentially, this combination could also be interesting in the future. Lastly, we did also look at consensus molecular subtype classification, and that did not change with treatment.

Essentially, these cell lines retained their original molecular phenotype, and then we did an unbiased analysis to look at pathways of immune-related pathways or other pathways that may not be represented with just the AIM gene set.

We were able to find some additional activation of T-cell pathways, which are shown here, that wasn't necessarily represented in the AIM gene set but were upregulated by AZA treatment.

And then, our correlate from this project is that we used our rapid autopsy program and looked at RNA sequencing data from patients who had metastatic bladder cancer, and we wanted to see if this immune signature was consistent across all specimens from a patient, so their primary bladder tumor and their metastatic sites, or if it varied by tissue origin, such as it would just be high in lymph nodes and not in the primary, something like that.

Ultimately, we found that this enrichment of AIM genes displayed a patient-dependent pattern and was consistent across anatomical metastatic sites. And so, this group of some patients who had high expression of these AIM genes, it was consistent across all of their metastatic sites and primary if we had that too, and not just in one spot, which could be interesting for future clinical applications.

And that's highlighted here just by zooming in on this upper part of this heatmap. And so, our results thus far indicate that epigenetic modulation can increase the expression of immunomodulatory and T-cell-related genes in bladder cancer cell lines.

Now I'm working to elucidate a more bladder cancer-specific immune gene set because this original AIM gene set is not exactly applicable to bladder cancer, and there may be some genes that weren't represented in that group of genes. And then also hoping to better understand the heterogeneity of bladder cancer in response to treatment and why some types of bladder cancer are more amenable to becoming more susceptible to immunotherapy or increasing their expression of immune-related genes versus some that are more resistant and try to stay cold.

And then thinking about clinical correlations, perhaps ultimately could develop something that's a little more targeted for bladder cancer or for different types of bladder cancer than the AZA and HDAC inhibitor that I used in the lab. And then I'll leave this a little bit for later, but am hoping to one day bring this to a phase one clinical trial to prime bladder cancer for immunotherapy.

And thank you to my amazing team and collaborators, especially Dr. Ming Lam and Dr. Wright, who were my mentors for this work.

Sam Chang: So that's me applauding Betsy for all the work that you've done. The amazing thing is you've just scratched the surface, and there are so many questions now, to be honest. I have a general idea of what we've looked at in our immunotherapy attempts, and we have sometimes contradictory results.

And for some areas, we're already down the line, and we've hit with therapeutics, and we've got this, we're trying to figure out who responds to whom. I'm going to start off with just some general questions in those cancers, so I'm going to use that generally — cell lines, whatever things you've looked at, human metastatic sites, primary sites in general — are the tumors that are AIM low generally, have we found them to be more aggressive or less aggressive or more likely to be metastatic or less likely to be metastatic? That's a simple question.

Elizabeth Koehne: Yeah, that's a great question. And I think, not necessarily even thinking about localized or non-muscle invasive bladder cancer, some low-grade papillary tumors are very immune cold. And actually, the cell line, the human bladder cancer cell line, RT4, that stayed the coldest originated at some point a long time ago from a low-grade papillary tumor.

Sam Chang: Got it.

Elizabeth Koehne: And then in a different way, but in our metastatic setting, there were also some patients who had just very immune AIM scores.

Sam Chang: Very high AIM scores.

Elizabeth Koehne: Yeah, some high and then yeah, and some cold too. And so it doesn't necessarily mean that out of those people, we don't think they necessarily had more aggressive disease.

Sam Chang: Great. So then have you been able, or have you reached out to the different studies that have looked at various immunotherapies in advanced urothelial carcinoma and everybody's attempted to predict who's going to respond or who's not? By default, it's been either PD-1 or PD-L1 expression.

And honestly, half the time it seems to be predictive, and half the time it doesn't. And when you ask clinicians, the vast majority that I ask, they don't do any PD-1 testing or PD-L1 testing. And if they do, they don't know what it means because they'll still end up trying it many times.

So have you been able to, or have you yet reached out to these large companies saying, "Hey, can we look at your Atezo data, your Pembro data, your Nivo data," and just to look at responders versus not, and being able to look at AIM gene expression in those that responded versus those who haven't?

Elizabeth Koehne: So I haven't spoken with the companies specifically, but at least for the PURE-01 study, they published a subset of genes that they thought correlated with response to immunotherapy. And some of those were AIM genes. And as I mentioned in the study, some of those were genes that we were able to activate with either AZA or an HDAC inhibitor.

And then, when I'm thinking about developing more of a panel of genes that may be prognostic, I am looking at those genes too, which perhaps we weren't able to activate. And then, I think that's just another question, but yeah, I do think that looking at those genes too, to try to predict responses is also good.

And as I also mentioned, bladder cancer is heterogeneous, and so it's hard to say perhaps how the panel would really apply to different stages, but thinking about that too when putting it together will be important.

Sam Chang: So as you look to move from the Pacific Northwest to the hearty Midwest, and I know you've spent time obviously in the middle, I don't know where you grew up, but at least I know with residency in Chicago, and as you start your research there, what do you think will be your first steps, once you get settled, once they make you take call, all those kinds of good things of becoming a new attending, where are you going to go next with the research?

You've talked about perhaps more specialized, either HDAC inhibitors or other types of interventions that may be more specific, but where do you go next? Are you going to still stay with the cell lines? Are you going to look at animal models? What's next?

Elizabeth Koehne: It's something that I don't know, but I'm very excited about. And what my plan is, when I get there, is to start meeting with people, and I have some connections and some ideas of new connections that I'm looking forward to making.

And I think that it's going to be really wonderful to bring my connections and all of my mentorship and collaborators that I have here at the University of Washington to bring all of that with me or have that still as a support for me when I go to a new place and then meet with a lot of different researchers. They have an epigenetics working group at the University of Wisconsin. I'm looking forward to joining.

And then they have a close partnership with the vet school there. And there's a person who works with animal models there. And so looking forward to meeting with them and starting to work with them. And then I'm glad that I have this cell culture work in my toolbox because that's something that I can pretty easily continue doing. And perhaps as I want to fine-tune some of our findings or look at things in a little different way, I'll be able to do some of that cell culture work just in the UW Urology lab there.

And so I think making new connections will just make this work really stronger and more applicable to our patients who have bladder cancer.

Sam Chang: Cancer, the transition, you can just keep on saying UW each time, and it simplifies things. It makes it easier for me too. Oh, she's at the University of W.

So, in thinking about all the new intravesical treatments that are coming out for non-muscle-invasive bladder cancer, do you see a possible application of doing a combination of one or two of these types of things that the HDAC inhibitors you're looking at, or something just in general to place, and then following that with some of the immune treatments that we've done?

What do you think will be the holy grail of looking at these AIM genes and the manipulation or expression of them and then ultimately impacting other types of therapies? How do you see it coming about in the next 10 or 15 years?

Elizabeth Koehne: Yeah, so I think being a urologist, my initial thought was for non-muscle invasive bladder cancer and high-risk BCG unresponsive disease, that field is just exploding. And a lot of the new options, some of which are intravesical immunotherapy, things look awesome, which is really great.

And so I'm not positive that all of them really may need a kind of primer or a boost. I think perhaps things that will affect that will be the ongoing BCG shortage and then just with the new drugs, what's available for different places. And so perhaps there still will be a role. I'm interested in something like ALT-803 or the IL-15 super agonist. I think potentially that could be a good partner, particularly if we're trying to conserve BCG.

And then what I'm thinking more about now and that I'm hoping to explore going forward is perhaps as we use more trimodal therapy for bladder preservation in patients with muscle-invasive bladder cancer, maybe they need a boost in the bladder if they still have tumor there, or just a boost to help their response to, as we start using immunotherapy in those patients as well, or maybe perhaps in replacement of the bit of chemo they get or I don't know, something like that.

Sam Chang: No, I think all those options are great. Non-muscle invasive disease, some type of intervention for bladder preservation where you already see a combination of immunotherapies with radiation therapy to replace chemotherapy.

But having an intravesical component or something else to help again increase or potentiate the effect of the radiation, I think, is very promising. And to your point of looking at the heterogeneous nature of bladder cancer, if you could have a mechanism that, in general, could increase the likelihood of any immunotherapy in whatever form it may take, will be incredibly, I think, attractive for the individual patients and for the different disease states that they may find themselves in.

So Betsy, just so excited for everything that you've been able to achieve in a short period of time and really look forward to, I know the folks at UW in purple will be missing you for sure, but I think onward Wisconsin and those folks, they'll be very excited to have you join them in Madison.

So thanks for spending some time with us and look forward to seeing you in person soon. And we look forward to visiting with you in the next year or two as you get settled in and you make your next research innovations. Take care.

Elizabeth Koehne: Well, thank you so much for having me. Really honored to be able to spend some time with you.