Dr. Shah: Welcome to PracticeUpdate. I’m your host, Dr. Aman Shah, and I’m delighted to be here with Dr. James Hsieh from Washington University. Welcome, Dr. Hsieh.
Dr. Hsieh: It’s a pleasure to be here.
Dr. Shah: Dr. Hsieh, you’ve been one of the pioneers in investigating genomics and precision medicine in renal cell carcinoma, and it looks like we are now on the cusp of actually using this in clinical practice. How would you describe where we are and what developments are likely to be most important for the practitioner?
Dr. Hsieh: Yeah, so in the past, the kidney cancer is basically resistant to chemotherapy and radiotherapy, and at the very beginning there were only two immunotherapeutic drugs, but not working very well, and later on with the advent of more targeted therapy, and the targets have become effective, is because they are targeting a specific pathway that occur, the mutated in clear cell kidney cancer. So that’s why all the VEGF inhibitors work very well. And that’s how we changed our practice initially.
But now because the immunotherapeutic era has come and can help generate very exciting data that we can put the patient into more stable remission or even a sort of a cured state. So, it becomes very important to figure out how to select all the right patients to do the right treatment.
And the most…at this moment, we all still depend on clinical criteria. We are treating all these patients almost exactly the same except the blood, performance status, all those kind of things, but I think in the future we really need to understand this is a very heterogenic disease. It can break into…simply, it’s two, three camps, based on genomic profiling. One is PBRM1-mutated patients, [which] accounts for about 50% of kidney cancer, clear cell kidney cancer patients...
Dr. Shah: Sorry to interrupt you there, but so PDL-1 has, in many cancers, not been a very definitive marker for efficacy for immunotherapy. Where does that stand in term for RCC?
Dr. Hsieh: Yeah, it’s still very unclear.
Dr. Shah: Okay.
Dr. Hsieh: Right. At this moment it’s like…
Dr. Shah: Unclear in clear cell.
Dr. Hsieh: Yes. Exactly. Unclear in clear cell. So that’s exactly right. So basically, if you would have had PD-L1 expression, they are likely to respond. If they don't, they can respond.
So, the idea is that you just cannot use these to exclude patients. Some patients. If you look at…because...highly heterogeneous ... the genes. If you just look at one area, you may miss the whole picture, and looking at a primary metastasis are a different issue.
Dr. Shah: Right. Right. And…sorry, go ahead. Beyond PD-L1, you were talking about other potential…
Dr. Hsieh: No, no, I was saying the…based on genomics, they are three different…I think there are two molecules, the genetic markers that people use now, and we use...kind of pioneer all those. One is called PBRM1 mutation. One is called PEP1 mutation.
And one is...have none of these. So, that’s very simple. So, the PBRM1 mutation tends to be…it kind of can predict the patient has the disease for 20, 30 years, and a PEP1 it can predict they have maybe like 5 to 10 years. And the rest is kind of a mixture. So, it kind of tells you how aggressive the disease is, how to manage that. So, we’ve got one PBRM1 patient that responds to almost all, any treatment I give them, VEGF treatment, mTOR treatment and PD-1 treatment work very well. And PEP-1 is not responding to many of the treatments. So, there are a few things that, when you start treating a patient, you can not only to get their Heng criteria and IMDC criteria. You need to look into their genomics. So that’s why I cannot profile every single patient on my...
Dr. Shah: Right. Okay.
Dr. Hsieh: ...in my practice.
Dr. Shah: And are you discovering any mutations that were previously not known to be important? Because we’re finding a whole host of actionable mutations in unexpected cancers that suddenly clear open an avenue of treatment.
Dr. Hsieh: Yes. Yeah.
Dr. Shah: Do we see that in RCC?
Dr. Hsieh: Yeah, so I think we are seeing that, but I don't...I see there are multiple opportunities beyond genomics because you know, transcriptomics is very important, at least when we can actually decode the convoluted transcriptomics data and try to tease out what it is, lot of VEGF signaling or whether there [is] lot of immune signature, and that way we can actually target those patients, design a specific treatment on top of that.
And on top of it the other thing that I think is very important, we are kind of pioneer on that field is basically try to look at the spatial resolution. You know, understanding the relationship between immune cells and tumor cells and vessel cells. I think those are the information that are kind of critically missed in the field because we just do bulk sequencing. And bulk sequencing basically does not give you the resolution that you need to understand whether that immune cell is getting through a tumor cell or immune cell is excluded from the tumor cells.
Dr. Shah: Right. So, can you integrate sort of the microenvironment information with the genomics information to come up with … make some kind of sense of...
Dr. Hsieh: Oh, absolutely. That’s exactly what I was so excited about, and that is, if you take a look at this moment, I pioneer most of the genomic studies in terms of therapeutics and a targeted therapy type, but later on with immunotherapy, everything becomes…it will be more complicated and need a more sophisticated way of studying it.
At this moment, we…in the old days, before genomics, everybody says transcriptomic area-based, and after genomics, everybody’s genomic based. And then later on, basically, they published multiomics with two platforms, but they don't integrate.
So, that becomes a very, very intriguing issue, and what we try to do now is beyond these simple two platforms, we want to integrate additional platforms, and really, based on a very, very clear signature there can be established using a big data platform to analyze. We collaborate with a certain company to kind of develop…you know, to pioneer all those who work. So, I think we’re getting closer and closer to understand what’s really going to happen.
Dr. Shah: Right. Right. So, this is a question that I’ve had. For an academic research oncologist such as yourself, you have a whole lot of tools that the community oncologist does not have, and for genomics, there are companies that community oncologists can use to get a profile. How does a community oncologist figure out the microenvironment?
Dr. Hsieh: Yeah. So, it’s almost exactly the same question. First of all, the genomics, because the idea is when we do a lot of the convolution.
Dr. Shah: Right.
Dr. Hsieh: It’s when we do the convolution, you can…by looking at DNA, you will understand the tumor content.
Dr. Shah: Right.
Dr. Hsieh: And by looking at the RNA, then you can start to understand how many cells are from immune cells and how many cells are from vessels and how many cells are from tumor. So, I think that’s why the RNA sequencing…if you can do a very, very good RNA sequencing analysis, it’s become very, very powerful.
Dr. Shah: So, this is something that could be done by commercial entities to help the...?
Dr. Hsieh: Oh, absolutely. Absolutely.
Dr. Shah: So, this is not something which will automatically involve a referral from the community to the academic...?
Dr. Hsieh: No, it’s very easy. For example, like, I work closely with Tempus, and they provide me with the service of DNA sequencing, RNA sequencing. And then I work with another big data company called BostonGene to analyze the genomics data.
So basically, that way…actually, we found out it’s very, very interesting by looking at all these different big data kinds of platform, we were able to kind of find a very, very interesting thing, and that…the plan is really applicable, applicability, and the idea is not really just for academia centers, it should be applicable for everybody.
So, the idea is that we try to kind of work with these companies, try to develop specific algorithms to help people understand how to treat it, because at this moment is still to me, I don't…it’s real like…very rudimentary in terms of how we interpret...you get. You can…you know, they help you to say, oh, there’s 10 different sites.
I think that one interesting, based on genomics, it’s very misleading is even the MATCH trial, right? So, you find out patient mutation in a specific, and say, oh, you have this mutation, we will treat you as this disease. But the problem is how do we call mutations because the mutation frequency is very important. If it just is mutating 5% of the tumor it’s called passive, and people will say it’s actionable. But you only deal with the 5% of the tumor. The other 95% [you] just don't care about it. So, that becomes very, very misleading and the very, very beginning for the genomics error.
So, once you start attacking the genomic heterogeneity and the immune heterogeneity and combine them together, so you’ll start to see a much more clear picture, and I think we are kind of at a forefront try to kind of bring the field that kind of integration into the clinical setting.
Dr. Shah: That would be wonderful.
Dr. Hsieh: Yes.
Dr. Shah: Thank you so much, Dr. Hsieh, for all the work that you do, and thank you for speaking with us.
Dr. Hsieh: Thank you. Thank you. It’s a pleasure.