Nima Sharifi: Thank you, Andrea. Thanks so much for the invitation and thanks to UroToday as well. Thank you so much for the invitation to talk about our work on androgen biosynthesis. I'll talk about how androgen biosynthesis may occur independent of the previously known pathway that requires CYP17A1, which is a target for abiraterone and also perhaps independent of CYP11A1. So I'm Nima Sharifi at the University of Miami Miller School of Medicine, the Desai Sethi Urology Institute and the Sylvester Comprehensive Cancer Center. And I'll start with takeaway points from this very short presentation. The points are the following. First is that all endogenous vertebrate androgens are thought to require CYP17 for biosynthesis. Blockade of gonadal and adrenal androgens extends survival, these are many of our standard-of-care therapies in the clinic for prostate cancer. Inhibition of all known androgen sources are incomplete. So this comes from a combination of clinical studies that I'll share with you as well as preclinical studies from the laboratory. Number four is that we've identified a CYP17-independent pathway of androgen biosynthesis, which is also parenthetically relevant to estrogen biosynthesis because you need to make androgens in order to make estrogens. The next point is that oxysterols are substrates for androgen biosynthesis.
This is also a surprising new development. And finally, CYP51A1 is an enzyme that's necessary for the CYP17-independent pathway of biosynthesis. So the canonical pathway of steroid biosynthesis or androgen biosynthesis, all of this really starts from cholesterol, which has 27 carbons. The canonical pathway requires CYP11A1, which takes off the six carbons off the cholesterol side chain. And the second step in the process requires CYP17A1, which is a target for abiraterone, and that takes off another two carbons. And this results in this product of 19-carbon DHEA, but all androgens have 19 carbons. There's a different pathway that we're describing here, but this time starting from cholesterol instead of requiring these two enzymes, it actually requires a different enzyme, and that's CYP51A1, which is capable of taking off these eight carbons off the cholesterol side chain at once. Instead of stepwise taking off six carbons and two carbons. So the clinical problem that we're tackling here is that CYP17A1 inhibition with abiraterone leaves androgens behind. And this comes from in part from data from clinical trials, including a neoadjuvant clinical trial from Mary-Ellen Taplin and colleagues at the Dana-Farber Cancer Institute where they treated men with prostate cancer prior to prostatectomy with either medical castration or medical castration plus abiraterone. And here they're looking at the androgen levels, either testosterone or DHT, within the prostatic tissues.
Here, looking at the androgen levels or the testosterone levels that are left with medical castration versus medical castration plus abiraterone, you can see there's a decline in the testosterone in the tissues, but there's still testosterone left behind. And instead, if we're looking at DHT, which is a more potent androgen, you can see that the decline in DHT with abiraterone on top of castration is actually a greater proportionate decline, but there's still DHT that's left. So together it's not hard to see how the testosterone and the residual DHT could be responsible for continuous androgen stimulation, and more importantly, prostate cancer hormone therapy resistance in the context of castration plus abiraterone acetate. In the laboratory, we see that there is also CYP17-independent androgen biosynthesis. Here, what we're doing is we're looking at the C4-2 prostate cancer cell line model in tissue culture on plates. And you can see that when we take a look at the androgens that are left in the cells by mass spectrometry, there's T and DHT that's there in the absence of serum. And when we also incubate these cells with pharmacologic inhibitors of CYP17, similar to abiraterone, it really doesn't put a dent in these intracellular androgens.
So this appears to suggest that this is all independent of the target for abiraterone, which may be relevant to this clinical process as well. The next point here is that there's a sterol derivative or cholesterol derivative called 17,20-dihydroxycholesterol that can be converted in these prostate cancer cells to androgens. So here we're treating prostate cancer cells with this dihydroxycholesterol, and you can get the generation of all kinds of androgens, DHEA, testosterone, and DHT. And when we do this in a time-course-dependent fashion, you can see that you get the generation of all of these androgens as well. And then we wanted to take a look at this a bit deeper to figure out, what is the enzyme that's required to make androgens from dihydroxycholesterol? And of all the P450 enzymes that are present in humans, there's only a single one that's capable of taking dihydroxycholesterol all the way to DHEA and other androgens. And nothing else is capable of doing that in our experiments. CYP17 doesn't do it. CYP11 doesn't do it as well. So this is really the takeaway that here you have the canonical pathways. It really simplified canonical pathways going from cholesterol, CYP17 all the way to various androgens, but there's a completely different pathway that goes around CYP17, requires an oxysterol and utilizes CYP51A1 instead to get to androgens, which is really unanticipated.
Last slide, I want to acknowledge the people who actually did the work, the folks in our group, including Ziqi Zhu, who's a research assistant professor, but other contributors to this science in our group, our collaborators at University of Michigan, Vanderbilt, and other institutions and other collaborators and leadership at the Desai Sethi Urology Institute, Sylvester Comprehensive Cancer Center, and our funding from the Prostate Cancer Foundation, the NCI, the DOD and other sources. Thank you so much.
Andrea Miyahira: Thank you so much, Dr. Sharifi for sharing this study with us. So what are the known roles for CYP51A1 and in what cell types is it expressed and is it ever mutated or exhibiting altered expression in prostate cancer?
Nima Sharifi: Yeah, those are really important questions. So the known canonical function of this enzyme, CYP51, is actually that it plays an essential role in making cholesterol. And so it's expressed in a bunch of organs and cell types that are known to make cholesterol. One of the big ones is the liver, which has a major role in cholesterol biosynthesis. But this function is totally separate for making androgens and estrogens, in other words, sex steroids instead of cholesterol. And prostate cancer with collaborative work that we've done with Alex Wyatt at the Vancouver Prostate Center, we've seen that the copy number of CYP51 appears to be increased in a subset of cases of men with advanced and hormone-therapy-resistant prostate cancer, about 17% of cases. In terms of mutations themselves, we don't find any evidence so far of activating mutations in CYP51, but it's possible that we've missed that as well.
Andrea Miyahira: Okay, thanks. And do you think CYP51A1 would be a good therapeutic target? And if so, what context or patient population and what side effects would you anticipate if it was therapeutically targeted?
Nima Sharifi: Yeah, that's a great question. So I do think that it could be a new therapeutic target in prostate cancer. I think it could be particularly relevant in men treated with CYP17 inhibitors in combination with gonadal testosterone suppression, because that's the context where you have these residual androgens in CYP51 activity is particularly important. I think the key is can we separate out the enzymatic activity of making cholesterol versus making androgens? And so this is part of our current activities is to see if we can separate these activities. It's possible if we completely block cholesterol biosynthesis and androgen biosynthesis, that may be a good thing, but we don't know what the adverse effects of blocking cholesterol biosynthesis is versus the cholesterol that we ingest, maybe that's completely sufficient for the other functions in the body. So these are some of the basic things that we need to figure out in terms of both the biochemistry as well as the physiologic relevance of these two different activities of CYP51.
Andrea Miyahira: Thanks. And in your figure, it looks like 3β-HSD1 is required in both of these pathways. So would targeting 3β-HSD1 eliminate all androgen production by both the canonical and this new pathway?
Nima Sharifi: Yeah, that's a great catch. And so we think that blocking 3β-HSD1 could be part of the solution to blocking CYP17-independent biosynthesis of androgens as well as CYP17-dependent biosynthesis. We don't really know until we try fully in vivo systems as well as clinically if that's going to work as well as we anticipate that it should.
Andrea Miyahira: Thanks. And what research remains to fully rule out that there are no other human pathways to produce androgens? And do we know if there are any mutations that can convert alternate enzymes into androgen producers?
Nima Sharifi: Yeah, so I think this is one of the great black-box mysteries that it's hard to solve with using approaches like genomics alone. You have to get into the weeds of the biochemistry, which I think makes it really fun. There's a lot of active ongoing work looking at, for example, the potential contribution of the microbiome for making androgens, which I think is quite intriguing. That's being investigated by a number of different groups that I won't get into here. I think that's ongoing work. So I think it's very well possible there are other pathways that we haven't yet discovered or we just need to do additional work to figure this out. So these are really fundamental questions that's going to require a lot of hard work and analysis.
Andrea Miyahira: Thanks. And last, what are your next steps and do you have any translational plans?
Nima Sharifi: Yes. So one of the big next steps is to take a look at the structural requirements for this activity and the differences in the structural requirements for this activity making sex steroids versus the structural requirements of the same enzyme for making cholesterol to canonical function, and to really get in the weeds of the biochemistry. The other thing that we really need to do is to figure out what are the various intermediates and the pathway to get to androgens through this other pathway. I don't think we fully understand that yet. So we need approaches in biochemistry, we need approaches in metabolism, we need approaches in chemistry, and we're doing our best to take all of those to eventually develop better therapeutics for men with prostate cancer.
Andrea Miyahira: Okay. Well, very exciting, and thank you for sharing this with us.
Nima Sharifi: Thank you. Thank you so much for the opportunity to talk about this work. I appreciate it.