Read the Full Video Transcript

Andrea Miyahira: Hi everyone. I'm Andrea Miyahira at the Prostate Cancer Foundation. Today, Dr. Andrew Goldstein, an associate professor at UCLA, is joining me to discuss his group's latest paper, "MYC is a regulator of androgen receptor inhibition induced metabolic requirements in prostate cancer," recently published in Cell Reports. Dr. Goldstein, thank you for joining us today.

Andrew Goldstein: Thank you for having me. I'm very proud to present this work from my group that, as Andrea said, was recently published in Cell Reports. The project was led by two outstanding graduate students, Preston Crowell and Jenna Giafaglione, and they received a lot of support from members of my group as well as a lot of collaborators with a range of expertise including mitochondria, metabolomics, epigenetics, computation, clinical trials, pathology, patient derived xenograft models, showing science is truly a team sport and I'm personally very grateful for everyone's contributions. We know that the androgen receptor is a central target in advanced prostate cancer because the AR promotes prostate cancer growth and survival. But we also know that most advanced patients will eventually recur, which means that some prostate cancer cells can survive treatment, and these cells are responsible for ultimately driving disease progression. So, we wondered a few things.

First, how do tumor cells initially respond to treatment prior to the outgrowth of recurrence? We wondered if there are opportunities to exploit that initial response to AR inhibition, and then what are the regulators of the treatment induced phenotypes? So we started looking in clinical datasets where tumor biopsies were profiled before and after hormone therapy. And when we looked at the pathways representing the genes that change with treatment, a third of them were related to metabolism. So we thought about employing a range of techniques to understand how metabolism is reprogrammed after AR inhibition. We used first metabolic profiling and nutrient tracing in prostate cancer cells treated with enzalutamide, apalutamide, or the PROTAC AR degrader, ARCC-4. And this led to very similar results in terms of metabolite abundance and utilization of glucose, utilization of glutamine, suggesting that there's really a conserved metabolic response to AR blockade.

Using Seahorse we found a number of interesting phenotypes, but the one I want to highlight is that after AR inhibition the cells consistently generate a higher proportion of ATP from mitochondrial oxidation, and we think that this is likely something that would make them particularly susceptible to targeting that process. And so we used a potent inhibitor of complex 1 of the electron transport chain and found this had far greater effects on the growth of enzalutamide treated cells compared to control. And this was consistent with all of the AR inhibitors. In vivo, we found that complex 1 inhibition could cooperate with AR inhibition to suppress proliferation in patient derived xenograft models. So all of this is telling us that there's a clear metabolic shift in the cells that survive AR inhibition, and this is characterized in part by increased reliance on mitochondrial oxidation as well as increased sensitivity to complex 1 inhibition.

And so we wanted to understand what are the factors that are regulating this treatment induced response? We looked closely at the mitochondria using confocal microscopy, and we noticed that after AR inhibition the mitochondria appeared to be more elongated. We could quantify this using aspect ratio, which is the length of the major axis over the minor axis. Now, a major regulator of mitochondrial morphology is this protein called DRP1. It gets phosphorylated at serine 616, which then oligomerizes around and constrains mitochondria leading to its fission. And there were previous reports in the literature that DRP1 expression might be regulated by AR. We didn't really see evidence of that consistently across models, but we did find phosphorylation of DRP1 was strongly reduced using every AR inhibitor we tried and in every model we looked at. And so we thought about rescuing DRP1 phosphorylation.

We used a phosphomimetic and this was sufficient to rescue the elongation phenotype. And notably, we could actually reduce sensitivity to complex 1 inhibition, which I was pretty surprised at. So taking this back to our model, we can add one more phenotype to the list, which is increased mitochondrial elongation in the cells that survive AR inhibition. We found that phosphorylated DRP1 seems to play a role in some of these metabolic phenotypes, but we wanted to know what was upstream of this and regulating this entire treatment induced metabolic shift. And that turned out to be MYC, so expression of MYC and some of its well-known metabolic targets are reduced consistently after treatment, both in experimental models and clinical samples. And so we looked at what would be the response of rescuing MYC. First looking at glucose utilization, enzalutamide causes a significant shift and that's essentially completely reversed by MYC expression.

Enzalutamide increases the percentage of ATP generated from mitochondria oxidation, and this is reversed by MYC. Enzalutamide decreases phosphorylated DRP1 and this is rescued by MYC. And finally, enzalutamide increases sensitivity to complex 1 inhibition, and this is completely rescued by MYC. So bringing it around to the beginning, since advanced prostate cancers will eventually recur, it's important to understand these prostate cancer cells that survive AR inhibition. So how do they respond? We see clear evidence of a metabolic shift characterized by this increased reliance on mitochondrial oxidation. Can we find vulnerabilities? These cells appear to be more sensitive to inhibitors of oxidative phosphorylation. In terms of what regulates these phenotypes, MYC appears to be a major regulator of the metabolic shift, and DRP1 phosphorylation also plays a role. Now taking this to the clinical setting, if we look at pre versus post AR inhibition, RNA expression, and focus on how MYC signatures change with treatment, we can see in terms of the clinical response that the good responders seem to have a strong reduction in MYC, and these are probably the ones most likely to exhibit that metabolic shift that we found.

However, there are a subset of tumors that retain their MYC signature, and this is associated with poor clinical response as well as the lineage plasticity to a double negative phenotype. And so I think it'd be really interesting to try to look at what's happening at the metabolism in these different clinical tumors. So I just want to finish by thanking the people who worked really hard on this project and for the many collaborators lending their time and expertise. I'm personally really excited about the new ideas and projects that develop from this work, and I also hope that some of the metabolic resources that we've now published can be useful for others in the field. So thank you very much.

Andrea Miyahira: Okay. Well, thank you, Andrew. That was a really excellent presentation. So MYC amplifications are common in CRPC. How do you think this might change the metabolism and the impact of AR targeted therapies?

Andrew Goldstein: Yeah, I think it's very clear from our study that MYC seems to be majorly important in overcoming the effects of AR inhibition. So if you imagine a MYC amplified tumor having a lot of these phenotypes, a lot of these metabolic signatures, then I would imagine that the MYC amplification would be a risk factor for overcoming the effects of AR inhibition. But I think MYC amplification alone is probably not sufficient to understand what's actually happening in the tumor. I think the MYC signature is probably more important and that can happen with or without amplification. MYC is kind of like a complicated story in prostate cancer because there's a subset where MYC and AR seem really tightly aligned, and it's clear evidence that AR can regulate MYC expression, and when you inhibit AR, MYC activity can be reduced.

But then there's also a subset of AR independent or kind of these lineage plastic tumors that seem to have very high MYC signatures, so I think there's probably more work to be done to kind of separate the role of MYC in an AR driven context and an AR independent context, but probably in both contexts associated with this resistant phenotype.

Andrea Miyahira: Thank you. So how much of therapy resistance do you think is based on restoring metabolism and is altered metabolism alone sufficient to promote resistance?

Andrew Goldstein: I think, well, in the field we know there are a number of ways to develop resistance, so I think this answer probably has to kind of cover different aspects. So clearly there's a fundamental metabolism requirement if a cell is going to divide and grow and you need to make nucleotides, you need to make lipids, you need to make amino acids, so these kinds of things are going to be critical to overcoming resistance. And so if you're AR driven and your proliferation goes down with treatment, you need to somehow restore those fundamental metabolic aspects in order to be an actively dividing cell and to promote that tumor growth. But then there are other elements of resistance, which aren't just about helping the cell grow, but there's of course the resistance where you get around the androgen receptor.

And so we actually have another study that was recently accepted and hopefully it'll be published soon so other people can read it, which looks at the aspect of metabolism controlling lineage identity, like luminal differentiation. And so we see evidence that you can manipulate certain elements of metabolism and change the phenotype, make the tumors less luminal, less AR driven, and more resistant. So in that case, simply manipulating metabolism can actually promote resistance, but it's not fundamentally in that case about the basic requirements of a proliferative cell. It's more about metabolic regulation of epigenetics altering the sort of self fate to promote resistance. So I think that there are different ways that this works, but a lot of reasons why I think metabolism is fascinating.

Andrea Miyahira: I look forward to reading that next paper and seeing more of the biology unraveling metabolism. So have you evaluated the role of the tumor microenvironment and different metabolic niches on prostate cancer tumor metabolism?

Andrew Goldstein: I think it's a really important idea, a really important issue. We haven't done a ton of work in that yet. There is a really cool paper this past year from Molecular Cancer Research, a group from MD Anderson, that showed that tumors in different regions, different locations in mice, seem to have slightly different metabolic phenotypes and in some cases different sensitivities to metabolic inhibitors. So I think there is evidence that in some instances the location, the tumor microenvironment does have an influence on the metabolism. There's a ton more work to be done. So for us we're addressing that in two ways. One is we're trying to understand more complex in vitro cultures how different aspects of the microenvironment influence metabolism. And then we just started up a collaboration, or we're about to start this collaboration, with Adam Sowalsky's group looking at tumors in different niches. So hopefully that can be a follow-up in a few years, but I'm excited about that.

Andrea Miyahira: Really interesting. So are there any other next steps in your studies that you haven't, I guess, shared?

Andrew Goldstein: Yeah, we're really thinking about the mitochondria. So in our study that we are talking about, we found that the cells kind of shift, shortly after AR inhibition the cells seem to have this window where they become much more reliant on the mitochondria. Now I think once the resistant tumors grow out, I don't think that reliance stays. So it's kind of this window of opportunity almost in the castration sensitive phase. And so targeting the mitochondria is not really a straightforward thing. There are studies with really, really potent drugs that block oxidative phosphorylation and they just show massive toxicity because mitochondria are important to a lot of different parts of the body. So I think wiping out the mitochondria just as a blunt approach is probably not the way to go. So we're trying to think about and study molecules that are regulating mitochondrial oxidation in a unique way in advanced prostate cancers.

And we have identified a few molecules that seem to be not really expressed in the vast majority of normal cells and even are poorly expressed in most localized prostate cancers, but they get activated in advanced disease. So that's one thing that we're thinking about. And yeah, I mean, it's a very unique organelle. I personally remember when I was taking biology classes I never learned that the mitochondria can fuse together and then divide up. And so these weird dynamics that the mitochondria undergo, they're cool to look at, but I think they open up opportunities for targeting if we can really understand the molecular control of it and the elements of that that are kind of specific or enriched in advanced disease.

Andrea Miyahira: Okay. Well, thank you again for sharing this really interesting story with us, Dr. Goldstein.

Andrew Goldstein: Thank you so much.