Read the Full Video Transcript

Andrea Miyahira: Hi, everyone. I'm Andrea Miyahira at the Prostate Cancer Foundation. With me are doctors Karen Sfanos of Johns Hopkins University and Dr. Jason Ridlon of University of Illinois, Urbana-Champaign, to discuss their new nature microbiology paper in expanded metabolic pathway for androgen production by commensal bacteria. Thank you both for joining.

Karen Sandell Sfanos: Thank you, Andrea. Thank you so much to Andrea for having us here today to discuss our new study just published in Nature Microbiology. And I just want to mention that this has been a wonderful collaborative study from many institutions. And I'm here with Dr. Jason Ridlon today to tell you about this work. But I just want to point out that he was the real champion of this project and the very key microbiology findings that came out of this study.

So I just to start. So from the prostate cancer perspective of this study, what we were looking at was this hypothesis regarding androgen production by gut bacteria, and how that can influence prostate cancer treatment resistance. And for the prostate cancer data that we'll be showing you, the patient samples, I just want to point out that this is in the setting of individuals with advanced metastatic prostate cancer that are being treated with standard androgen deprivation therapy, which prevents testosterone production by the testes.

And then also with abiraterone acetate, which prevents adrenal androgen synthesis. So these are drugs that are used to shut down all human androgen production in the body. But what we've been studying is the fact that there are bacteria in the gut that are part of the normal gut microbiome, that can produce androgens on their own. And our hypothesis is that these androgens that are produced in the gut can be absorbed and can circulate, and can interfere with these anti-androgen therapies that are given in the setting of metastatic prostate cancer.

And Jason's previous work had shown, or had discovered this bacterial gene that we call desmolase, which is part of this operon. This des operon. The gene is desAB. And he had previously shown that precursors such as cortisol can be converted by desmolase into an androgen called 11-beta-hydroxyandrostenedione that in itself can signal through androgen receptor, but also be further metabolized into 11 keto derivatives that are potent androgen receptor agonists.

But then what Jason also showed in previous work is that prednisone can also be converted by this bacterial desmolase into another androgen that I'll call delta 1-ATE. Both of these metabolites have been shown to lead to prostate cancer cell proliferation. And importantly, in our setting of studying this with abiraterone acetate, prednisone is given as the replacement glucocorticoid along with abiraterone acetate. So we're studying how this is being actively converted into an androgen by gut bacteria.

So the key findings of this study, or two new bacterial genes that are part of this des operon. And the two genes are desF and desG. The desF was discovered in gut bacteria. And for example, Clostridium scindens is one of the main strains that contains this bacterial desF gene. And what this does is it can convert-- well, so first off, if you have a precursor like deoxycortisol, like I already said, the desAD or that desmolase gene can convert it to androstenedione.

And what Jason discovered was that this other strain of Clostridium scindens that contains this desF gene can convert androstenedione into epitestosterone. And this happens in the gut. He also showed, in our study, that there are strains of urinary bacteria. For example, one called propionate microbial lithophyllum. That there's a pathway where cortisol, again, can be converted by this desmolase into our 11-beta-hydroxyandrostenedione.

And then this new gene, desG, can convert that precursor into 11-beta-hydroxyandrostenedione. So here's a testosterone derivative that can be produced by urinary tract bacteria. So these androgens that I've been talking about, these bacterial androgens, are not what you standardly think about in terms of androgen receptor signaling testosterone dihydrotestosterone. These are alternate androgens that seem to also be androgen receptor agonists.

And there was prior-- well, there's prior dogma, that epitestosterone that I told you could be produced by this desF and gut bacteria. That that's actually an androgen receptor antagonist. But there was a recent study showing that perhaps epitestosterone has some agonist activity towards androgen receptor. And in this study, we showed that as well. So in cell proliferation assays, we showed that epitestosterone actually induces stronger cell proliferation at 96 hours after exposure than testosterone.

And we know that this is acting through androgen receptor. Because if you add enzalutamide, which is an androgen receptor antagonist, you can block the proliferative effect of this epitestosterone. So this is very much against dogma in the prostate cancer field. That epitestosterone could be an androgen receptor agonist, and similarly they also showed that AR-target gene expression of KLK3 or PSA and also TMPRSS2 is very much upregulated when LNCaP cells are exposed to epitestosterone.

So we were very intrigued by this finding. And as I already mentioned, we're interested in understanding, could these desF gut bacteria be detected in individuals undergoing treatment with abiraterone acetate? And so we had a clinical cohort that was established here at Johns Hopkins, where we had individuals who had prostate cancer, hormone sensitive prostate cancer, but who were not undergoing any sort of active treatment, were not undergoing any sort of ADT.

And then a cohort of patients that were undergoing treatment with abiraterone acetate plus prednisone, who either were responding to treatment, what I'm calling stable here, they had stable PSA trends or decreasing PSA trends, versus individuals whose PSA was progressing on abiraterone acetate. And then what we did was we collected fecal samples from these patients. And we used a quantitative PCR assay for this desF gene.

And what we showed was that in comparison to the control patients that had hormone sensitive prostate cancer, that were undergoing treatment, people that were being treated with abiraterone acetate had higher levels of fecal desF. And this is just two different ways of looking at it. Either, if you normalize to the total amount of bacteria in the fecal sample, versus if you're just looking at copies of the desF gene.

It's increased in people that are taking abiraterone acetate. And interestingly, those progressing on abiraterone acetate had a significantly higher amount of fecal desF levels versus the controls. And if you look over here in these last two charts, these are people where we had samples taken when they were stable on abiraterone versus a matched sample when they were progressing on abiraterone.

And what you can see is that a subset of the patients, not everyone, but a subset of the patients had very significant increases in the amount of fecal desF that we could detect when they were progressing on their drug. So then on the last slide here, I want to show you a little bit about this very cool technique that was developed by Dr. I at University of Illinois, Urbana-Champaign, where they were able to micro encapsulate bacterial species in an anaerobic environment and then co-culture those with LNCaP prostate cancer cells.

So this is a really cool part of the study where we had isolated a bacterium, propionate microbial lithophyllum, 17 years ago from a prostatectomy specimen and isolated this strain of bacteria that is what we initially-- that we found contained the desG gene. Then 17 years later, we collected a urine sample from the same individual. And we're also able to isolate that very similar strain of propionate microbial lithophyllum. And this is how we help to discover this desG gene.

So those cells were encapsulated and co-cultured with LNCaP cells in the presence of cortisol. So the enzyme, the desG enzyme, produced by the bacterium could convert cortisol to 11-hydroxytestosterone. And what we showed was that, in cell proliferation assays, is that when you have both the cortisol and this, what we call API-2 strain of bacteria, you get cell proliferation versus the cortisol or the bacteria alone.

So again, this is a urinary tract bacterium. But also showing that androgen production from a urinary bacterium can cause prostate cancer cell proliferation. So the take home messages of this study are that-- are several. That bacteria in the gut and the urinary tract can produce androgens that support prostate cancer cell proliferation. And I told you about some of these alternate androgens that might be surprising androgen receptor agonists, such as epitestosterone.

But I also want to point out that a big part of this study was also showing that prednisone itself can be metabolized by these enzymes into androgens that can cause prostate cancer cell proliferation. Epitestosterone might be an unrecognized androgen receptor agonist. And we showed in our cohort of prostate cancer patients that desF levels or the enzyme that can produce epitestosterone is enriched in advanced prostate cancer patients with disease progression on abiraterone acetate and prednisone.

And then finally, although we don't know the significance of it yet, androgen production by bacteria in the urinary tract could potentially contribute to disease progression, maybe in the setting of primary prostate cancer. And that is work, certainly, to be determined. So with that, I just want to end this part by acknowledging all the many, many people that were involved in this study. I already highlighted Dr. I from University of Illinois, Urbana-Champaign, but then I also want to just a highlight. Taojun Wang and Saeed Ahmad, a postdoc and a graduate student that were very instrumental in this work.

And then all of the funding, certainly funding by the Prostate Cancer Foundation that made allof the work, and the biobanking that we've done in individuals with advanced prostate cancer possible, and all of the other funding sources listed here. Thank you.

Andrea Miyahira: OK, well, thank you so much, Dr. Sfanos for sharing that. So your study noted the prevalence of desF expressing microbes rose in some patients when progressing on abiraterone prednisone. Do you think there are any feedback loops between the microbe and the cancer?

Jason Ridlon: Yeah, that's a great question. I mean, what we have right now, is as an association. I think we need larger studies to verify this. Having some in vitro data showing that these metabolites cause proliferation, I think, is quite important. But we really need to move on to more preclinical sort of animal models and more advanced cell culture studies to really answer questions relating to feedback loops.

But we're starting to do work of this nature. We have some funding to start addressing the relationship between the bacteria and the tumor.

Andrea Miyahira: Thank you. And what biological and therapeutic insights are your biggest take home messages?

Jason Ridlon: Well, I think, one of the things that Karen mentioned is in a lot of her reviews has to do with the fact that even a few years ago, it was thought that urine was sterile. And so now, the urinary microbiome was not part of the Human Microbiome Project. And so, researchers like Karen, started to look at relative abundances of organisms, between different disease states, prostate cancer, things like this.

And I think what-- where I came at this was asking the question, a lot of steroids are excreted via the urine in hundreds of nanomolar level. And it only takes a fraction of an animal to activate these receptors. And so my question was, coming into this, do the bacteria inhabiting the urinary tract, do they metabolize steroids? Can they generate androgens? And so, this work has basically shown that bacteria can metabolize steroids.

So I think very new and opens up a new area of study in terms of steroid microbiology of the urinary tract. And I think the other aspect of this is that it wasn't known before. Our earlier study and then this particular study, where drugs like prednisone are actually being side chain cleaved and the whole class of steroid is changing. So it's acting-- it's supposed to be a replacement glucocorticoid. And these bacteria are generating androgens from this.

So I think we need to have further metabolomic sort of studies in these patients to look at how much of this is actually accumulating in circulation? How much of it is getting in tumor tissue? Things like this. But this is a new area as well, and suggests that prednisone might be problematic in some individuals. And what we've shown was that also not all individuals have these genes.

So basically, this could be a biomarker for seeing how effective might this treatment be in some individuals.

Karen Sandell Sfanos: Yeah, and can I just jump in and add that from my perspective, another very fundamental aspect of this study that's so important is that there have been prior studies, including our own, that have indicated that androgen production by gut bacteria could interfere with endocrine therapy. Very great studies that have shown this. But unfortunately, up until this point, the actual bacterial enzymes that contribute to this androgen production are not fully understood.

And Jason's lab and his work has really pushed the needle forward so far with discovering these enzymes that are involved in androgen production. And the reason why that's important to prostate cancer patients is that if you know what the enzymes are, you could perhaps develop a targeted therapeutic to block those enzymes. So the pie in the sky goal is that you're blocking human androgen production, but can you also give a drug that can block the bacterial androgen production in the gut to prolong therapeutic efficacy of some of these drugs?

Jason Ridlon: Well, another point too, building off of that, is that we tend to think of the endocrine system as being only host tissues and host enzymes. And a lot of these host enzymes end up being drug targets or some 17-hydroxysteroid dehydrogenase, or people have been developing drugs against these and thought to be potentially therapeutic. And it turns out, the bacteria are also contributing these sorts of enzymes.

And even like very exciting result in here that Karen had isolated bacteria from prostate tissue, expressing a bacterial form of this enzyme. So, I think, part of microbiome research today is working out these steroid pathways in microbiomes, host associated microbiomes, that are contributing to steroid endocrinology.

Andrea Miyahira: OK, thank you. And what are your next steps, including any plans for translational research?

Jason Ridlon: Karen, do you want to speak to the trial?

Karen Sandell Sfanos: Sure. Yeah, I mean, that's one of the most amazing things for our labs, is that both Jason and I are basic scientists, we're PhD researchers. But this work from this study has led directly to a clinical trial that opened up at Johns Hopkins just a couple of months ago and is actively enrolling participants. That it's based on looking at abiraterone acetate response and glucocorticoid switch from prednisone to dexamethasone.

And then also an arm where we're giving a very short course of antibiotics as a current strategy to try to eliminate some of these gut anaerobes that can make these androgen metabolites. So I would say, in terms of our translational next steps, we've actually-- we're doing a clinical trial with a very, very rich correlative studies that are being funded by the Prostate Cancer Foundation, and we'll be able to study some of these hypotheses that we have more fully in the trial participants.

Jason Ridlon: And also funded by Prostate Cancer Foundation, we're also looking to understand the structure mechanism of some of these enzymes and start to develop some inhibitors that we can test in preclinical models. Not to give too much away, but also, we're finding other pathways in urinary microbiota that we'll, hopefully, be presenting soon.

Andrea Miyahira: OK, well, thank you both so much for sharing this study with us today.

Jason Ridlon: Thank you.

Karen Sandell Sfanos: Thanks for the opportunity.