Racial Differences in Genomic Profiling in Prostate Cancer - Brandon Mahal

December 7, 2020

Studies have historically underrepresented Black and non-White prostate cancer patients limiting our understanding of the disparities in the diagnosis and prognosis of prostate cancer among these populations. Brandon Mahal, MD, joins Charles Ryan, MD, to discuss what he reported in this New England Journal of Medicine paper titled "Racial Differences in Genomic Profiling of Prostate Cancer."

Biographies:

Brandon Mahal, MD, Assistant Professor of Radiation Oncology, Assistant Director of Community Outreach and Engagement, Department of Radiation Oncology, University of Miami Health System, Coral Gables, FL.

Charles J. Ryan, MD, the B.J. Kennedy Chair in Clinical Medical Oncology and Director of the Division of Hematology, Oncology and Transplantation at the University of Minnesota, Minneapolis, Minnesota, USA.


Read the Full Video Transcript

Charles Ryan:  Hello, I am joined today by Dr. Brandon Mahal who's an Assistant Professor of Radiation Oncology at the University of Miami, where he's also an Assistant Director of the cancer center there for community outreach and engagement. He published a really interesting letter in the New England Journal of Medicine recently that puts new context around the issue of racial differences in outcome in prostate cancer. So welcome, Dr. Mahal.

Brandon Mahal:  Thank you. Yeah, thank you very much for having me. I'm very excited to be joining and sharing some thoughts on my work.

Charles Ryan:  Great. So, before we go any further, just tell us a little bit about what you reported in this New England Journal of Medicine paper.

Brandon Mahal:  Yeah, so this study, first of all, was motivated by the fact that there have been long recorded and long known racial disparities in prostate cancer, in particular, Black men have a nearly two-fold increased risk of developing prostate cancer, and a more than the two-fold increased risk of dying from prostate cancer. And there is really good evidence about healthcare access and socioeconomic factors that may contribute to these disparities, and that definitely does.

But a new emerging concern is that we are now using precision medicine and precision omics. And my concern, and our group's concern, was how these precision medicine initiatives might have the potential to widen disparities or contribute to disparities, in particular because they are so Eurocentric and very few Black or other minority men are included in these studies. So we wanted to see, with a new set of data that was available, what differences might be, and alterations in genes that are important for prostate cancer prognosis and therapy.

And so, what we did is we evaluated 474 unique genes and tumors in both cancers that were non-metastatic and cancers that were aggressive and metastatic. And this was in over 2300 men, including just over 200 Black men. And this data was made available by Project GENIE, which is ran by the AACR.

And so, what we were able to do is basically look at, again, these 474 genes on the panel, and found racial differences in alterations in those important genes. And the most interesting and striking differences that we found were in the metastatic setting. So on primary tumors and patients who just had primary prostate cancer that did not metastasize beyond the prostate, there were not huge significant differences. There were some minor differences, it's hard to know exactly what those may mean, but in the metastatic setting, we found that Black men had alterations in the AR gene for androgen receptor at a rate of 18% compared to about 8% for White men. And that is very important because our therapies are targeted toward the androgen receptor and alterations in that gene are the most common for men who develop castrate-resistant or treatment-resistant, metastatic prostate cancer.

And then the other thing that we found was that alterations in DNA repair genes and what we call actionable genes, or genes where there are therapies designed for alterations in those genes, also occurred at a higher frequency in Black compared to White men. So, for DNA repair genes, we saw about a 23% rate where it was about 16% for White men. And then, actionable alterations, again, lumping them all together, that rate was about 27% in Black man and about 18% in White men.

And then the other finding that we found as well was BRAF alterations, which are targeted in many other different kinds of cancers, but we don't really target those for therapy in prostate cancer outside of a few really early clinical trials, occurred at a rate of about 7% for Black men, where it was about 1 to 1.5% for White men.

And so, these findings were very important because they showed, in this small sample of patients, that there could be population-level differences in alteration frequencies of important genes that either predict for outcomes, or that could be used to identify different therapies that may work better for different populations.

Charles Ryan:  So this is really fascinating because we are talking about the tumor genomes here. We are not talking about the germline of these subjects, which one might expect because of racial differences, there may be differences in the germline. Do you know to what extent the somatic alterations that you report may reflect germline differences? Or, I mean, I know a lot of them are not genes that have a high prevalence of germline alterations.

Brandon Mahal:  Yeah. Very good question. And unfortunately, with this specific data set, we don't have the germline alteration data for me to say that definitively, but yes, I agree with you that these are most likely alterations that were influenced in the prostate cancer tissue itself and may not have been inherited. And so, this is a critical observation because it says there are likely many factors at play that are contributing to these differences in alterations in genes in the prostate cancer that we are seeing, including treatment differences, differences in exposures and environment, differences in what folks are doing.

Charles Ryan:  Right. That's something we don't think about or we don't have enough answers in prostate cancer. We don't think about prostate cancers caused by whatever. Like one might do with smokers and lung cancer versus non-smokers et cetera, where we just don't have a good handle on the exposures and sort of the environmental backdrop for a lot of these patients, which may explain it. I mean, I was struck, for example, by the very high rate of TP53 alterations that you observed in Asians. And I can't imagine... That defies explanation to me, do you have a thought on why that was observed?

Brandon Mahal:  Yeah, that, again, is something that we don't have a great explanation for. And I really do think that it's important to, as we build on this research, and my hope is that we are able to capture very diverse cohorts and not just the patients and their tumor samples and their cancer outcomes, but what these patients are exposed to, where they live, where they were born, what they're eating. Very deep level, sort of deep phenotype along with the genotyping. So we can understand those exposures that you are mentioning and what they may do to the genome.

Charles Ryan:  Mm-hmm (affirmative). And for those listening, it's important, obviously, to think about the fact that there may be a whole separate body of work that could be done on the germline factors, not necessarily cancer-associated genes, but other factors. In some of my own projects, we look at androgen metabolism, which has a very high rate of diversity based on race and backdrop. And so, these are tumor factors, not host factors. And then, the triangulation would be tumor factors, host factors, and environmental factors and how they come into play.

So, it's a really fascinating initial observation here. And I was struck of course, because we moved into this era where we now have approved drugs regarding DNA repair. And we are doing tumor sequencing in the clinic very commonly, we are looking for actionable mutations, and you did a really nice job of breaking it down into what is actionable versus not actionable. And I was surprised that African Americans had a higher rate of actionable mutations than the Asian cohorts or the White cohort that you looked at. And I think that goes a little bit against what I had previously heard. Do you think of your observations as going against what we had come to think of as sort of the dogma in this situation?

Brandon Mahal:  Yeah, I think it's a great question, again. I feel like for the AR alterations, that that was something that people predicted or would have expected. But I totally agree that the more actionable or targetable alterations are the DNA repair alterations where you have PARP inhibitors that could help with those alterations. I do think that went against what maybe people would have expected.

And there was another cohort study that was led by Dr. Franklin Wong out of UCSF. And he looked at a few different DNA repair genes and he didn't find any differences, but there are a few differences in the way we analyze things. First of all, our cohorts included two more genes that they didn't have within the DNA repair pathways. And those genes actually drove a good amount of our signal, and we looked at all of the alterations together because there are still so few patients.

And so, if you go back and look at their paper, you could see that Black men had a slightly higher rate of all of the different DNA repair alterations, but didn't reach significance in those little areas. But I was surprised, and I do think that folks who are interested in clinical trials or developing therapies or pharmaceutical companies are taking notice of some of this data, because a lot of times, Black men are left off of trials because they have comorbidities, or there are just issues along the lines of academic mistrust, because of bad things that we've done, Tuskegee, that we haven't overcome. And then a lot of times just structural barriers to the way we either choose to or not, recruit patients.

But now, this data might say to companies and folks that there are patients that might really benefit from these therapies. And so, yeah, I think it was surprising for us. We didn't have that in mind when we went to discover.

Charles Ryan:  Yeah, there have been papers, I know, that have shown that Black men are less likely to have their tumor sequenced, for example, which hopefully is changing. That was a few years ago, and hopefully, systems now are coming around to doing it because it is now a standard of care, really because it's attached to standard of care therapies.

You bring up the work by Franklin Wong at UCSF because he showed, as you said, different alterations. MYC amplification was more common in African Americans than in Europeans, and P10 loss was also less common in African-Americans than in Europeans. And these are both aspects that may have treatment implications down the road, because we see, for example, Aurora kinase inhibitors being developed, we see potentially the Akt inhibitors, which may have a role in P10 loss disease.

So it's going to be more important over time that patients from all backgrounds be included in these analyses and that we make distinct efforts in approaching them for all different types of backgrounds, as we think about developing actionable targets and actionable drugs moving forward.

So congratulations, it is really a great contribution, I think, to our literature. I plan on citing it a lot, and hopefully seeing some clinical trials come out of this. And so, my final question is, do you or others have clinical trials that might be sort of more directly targeted at these populations?

Brandon Mahal:  Yeah, absolutely. So, we are starting to think a lot about what kind of trials we could develop or that would come out of this. And so, I'm working closely with folks at my university, folks where I came from, Dana-Farber Cancer Institute, folks at The University of Michigan, the Prostate Cancer Foundation. And so, obviously, a question would be about PARP inhibition and looking at different populations, and maybe, just in general, even before that, doing bigger genomic sequencing studies on more diverse cohorts. And there are efforts underway in that sense as well. So, we are planning to follow that up with properly designed trials, and I think that we are going to learn a lot about disparities, and how to address them.

But also, getting back to your point about exposures. And we know that we really only have three well-established risk factors for prostate cancer; age, family history, and race. Race is a social construct that is loaded. So, if we uncover what's driving the racial disparity, we are going to learn about what exposures drive prostate cancer. And that, I think, we will learn so much about prostate cancer from having more diverse cohorts in addition to addressing the disparities.

Charles Ryan:  Right. Well, thank you so much. It's a great conversation. And I look forward to reading more of your work in the future.

Brandon Mahal:  Great. Thank you very much.