HRR Mutation Testing in mPC
Transcript: Liquid vs tissue biopsy and HRR testing
Colin Pritchard, MD, PhD, and Alicia Morgans, MD, MPH
Interview recorded October 2025. All transcripts are created from interview footage and directly reflect the content of the interview at the time. The content is that of the speaker and is not adjusted by Medthority.
- [Presenter] Hi, and welcome to another episode of our Expert exchanges in HRR testing in metastatic prostate cancer. I'm so excited to have, here with me today, Dr. Colin Pritchard, who's going to talk with us about liquid versus tissue biopsies, and how we make the right call in HRR testing. Thank you so much for being here with me today, Colin. - [Colin] Hi, I'm pleased to be here. Thanks for the invitation. - [Presenter] Wonderful. So you're a molecular pathologist, and have a huge amount of expertise, and certainly a unique perspective, as we clinicians who see these patients, in our practises, will so appreciate, I think in these conversations. Can you just lay out the very basics for us? What are the key differences between tissue and liquid biopsy approaches for the HRR testing that we need to do in practise?
- [Colin] Sure, yeah. So for tissue, we're talking about tumour tissues, you know, biopsy, resection, what have you. For liquid biopsy, we're talking about the DNA that's shed from tumour tissue into the plasma. The blood plasma, right? So that's the basic differences. There are, you know, a lot of times, in prostate cancer, where you have enough tumour shed into the plasma to do an adequate liquid biopsy test, and there's a lot of times where you don't. So I think the one really big take home is that liquid biopsy isn't always gonna be an adequate test for every metastatic prostate cancer patient, just because there isn't gonna be enough prostate cancer in the specimen. - [Presenter] That makes a lot of sense. And I think, at this point in time, our tissue testing is still considered the gold standard. Is that right? - [Colin] It is. And I think, largely for the reason I just outlined, and just from greater experience using tissue, we have a lot more, in the molecular pathology world that I'm in, we have guidelines that our professional organisations outline in terms of how you qualify a tissue specimen for testing. Like a pathologist has to look at a slide, and decide there's enough cancer there, and then there's actually guidelines of, like, you even circle the cancer area, and then you just scrape only the cancer tissue into the test tube that you're using to do the testing, all that kind of stuff. So there's all these guidelines to really make sure that the quality of tissue testing is adequate. It's the longer lived modality. Whereas on the liquid biopsy side, those guidelines are emerging, but they aren't quite there yet. So, you know, most labs will just test whatever specimen you send them without any sort of qualification step. It's getting better. A lot of the labs are doing things to look, to make sure there's enough cancer in the blood plasma, for example. But that's one of the reasons why tissue testing is still the gold standard. But not to say that liquid biopsy isn't useful, it certainly is, and, in some ways, it can be more useful, in certain situations, and we can talk about that.
- [Presenter] Well, let's do talk about that. I think, you know, one of the things that I think about liquid biopsies is that they give us the opportunity, perhaps, to understand some of the molecular changes, mutations, that may be present, if we have a patient who has really heterogeneous mutations that may be happening in distinct areas of disease spread. That's just one thought that comes to mind. But I wonder are there other situations, and can you comment on that too, if you'd like, when a liquid biopsy may be helpful, and a tissue biopsy may be either inadequate, or just too difficult to obtain?
- [Colin] Yeah, I mean, so you hit the nail on the head. I think one of the key issues is that you can have heterogeneity between different metastatic sites. And so one of the advantages of a liquid biopsy is that it's sampling all sites. I mean, every site of metastasis is shedding cell-free DNA into the blood. So if you have enough, you should be, in theory, able to sample those sites, which is potentially a big advantage. One thing I just go on, before I get to some of the other advantages and disadvantages, I would like to emphasise though that just because you see it like an actionable HRR mutation in a liquid biopsy, it may be only from one metastatic site, right? So because of that heterogeneity, it's both a strength and a limitation, right? So, you know, as a medical oncologist, you always have to think to yourself, okay, so I have it an HRR mutation, or another action mutation. We're talking about HRR here. If I drug that, am I only drugging one of the patient's cancer sites or all of them? So that's a question you should be asking yourself. And, again, it's a potentially a strength, but also limitation. Another strength I think of liquid biopsy is if you're looking for resistance mechanisms over time, we do have well-established resistance mechanisms, to say PARP inhibitors that are used to treat HRR. They're not really actionable yet in the same way that like, for example, lung cancer and EGFR inhibitor resistance mutations are, where you might use like a different PARP. We haven't quite got there yet, but we're close. So, that's more of a theoretical advantage, but it's certainly something you can detect, is resistance mechanisms in the plasma. So that's a big advantage, if and when those become medically actionable. And then the other thing, I think the other big advantage that you sort of alluded to is, you know, it's not always easy to get a metastatic biopsy, right? So if you have a patient who, let's say, you know, 5, 10 years ago they had their primary prostate cancer diagnosis, you know, and they had maybe a prostatectomy years ago, or even just a biopsy years ago, and now they have metastatic disease, and, you know, typically it's bone, and maybe bone only metastasis would be the typical. It may be not that straightforward to get a tissue biopsy. I'd still consider it a gold standard to do that, but it's technically challenging. Not all institutions are set up to do bone biopsies. Even if you are set up to do it, there are limitations in terms of, you know, again, that tissue qualification piece for a bone biopsy, in particular, is really important, because you have to make sure that the specimen's treated right, that you don't use like acid solutions, that they typically use in bone and pathology labs, that destroy DNA. So there's sort of technical aspects around that. So, a lot of times a liquid biopsy, again, if there's an adequate amount of cancer in the plasma, is a really attractive option that more represents the cancer as it exists now, than like, say, going back 5, 10 years, to a primary prostate cancer tissue biopsy, or resection, that might be in the bank. That said, the key HRR mutations, we've done work on this, other people have looked on this, looking at sort of the correlation between primary and metastasis. And the good news is that the correlation is very good. It's not perfect, but it's very good, maybe 80, 90% correlated. So it's not that going back to that old primary specimen isn't gonna tell you anything. It'll probably give you the right answer, but it's not a hundred percent gonna give you the right answer. So the liquid biopsy will be the up to date information on what the cancer is doing.
- [Presenter] Yeah, no, that makes a lot of sense. And I so appreciate that you kind of walked us through some of the challenges with processing, and ensuring that the tissue that we collect is going to be interpretable, or usable, for these assays, and then interpretable when we get them. Now, these circulating tumour DNA, or these liquid biopsies, also have challenges. And I wonder if you can speak a little bit to the issues that we have when a patient's responding really well to treatment, is that a good time to do ctDNA testing? Or when a patient, maybe, is developing things like CHIP, over time as they age, and as their systems are being exposed to more and more treatments that may accelerate the development of CHIP, can you speak to that a little bit? And maybe even share what is CHIP?
- [Colin] Yes, very good. Yeah, so I'll define CHIP. So CHIP is clonal hematopoiesis of indeterminate potential, or you also see it in the literature as just CH for clonal hematopoiesis, because increasingly we kind of know what it's potential is. So that's it. But CHIP has kind of stuck. What that is is it's somatic mutations in the white blood cell component, specifically the myeloid white blood cell component that are common as people age. Like they're even common in younger people. It's not a matter of if you have clonal hematopoiesis, it's at what level. So even young people, even people in their thirties, if you look with sensitive enough methods, have some level of clonal hematopoiesis in their blood cells. And, inconveniently, these genes that you can get blood cell mutations in are the same genes, the same HRR genes even, that we're looking for in cancer. So that can really interfere with the liquid biopsy because the mutation that you're looking for is actually in the sample, but it's not the prostate cancer, it's this clonal hematopoiesis thing. You're not trying to drug the clonal hematopoiesis, you're trying to drug the cancer. So it'll give you a false positive result. You know, it's not that it's a false positive in the sense that the mutation wasn't actually in the sample. It really is. But it's a clinical false positive in that it's not telling you what you need to know about the cancer, so. So, that's a big issue. We've published on this for prostate cancer, several years back, and showed that, you know, even if you do a plasma only liquid biopsy, that is if you don't have a control for clonal hematopoiesis, like doing a paired white blood cell sample in order to, kind of, subtract out the clonal hematopoiesis, if you just do plasma only, maybe even up of the half the time that you have an actionable HRR mutation that is in fact clonal hematopoiesis and not the prostate cancer. So this isn't like, you know, an edge case scenario. This is like every single time you should be thinking about that.
Now the good news is that, couple things on the good news front for the clonal hematopoiesis interference, or whatever you wanna call it, in liquid biopsy, one, labs are getting better at subtracting this out. So some labs are now, including ours, but some of the commercial labs are doing this now too, are now doing paired testing, which I would consider really the gold standard in liquid biopsy, so that you can definitively say this is clonal hematopoiesis versus prostate cancer. Paired testing, meaning doing a white blood cell test. And a plasma test together. Very logistically convenient. You have the same. You only need one blood draw to do that. So it adds costs to the test, of course, but logistically it's just the same specimen. You can just separate the two components. Other labs, that are still doing plasma testing, are getting better at bioinformatic ways of understanding what is CHIP to subtract that out. So even the labs that are still doing plasma only testing, which I think is not quite as good, they're getting better at understanding that. So that's part of the good news there. But, you know, it remains an issue. The other thing I should clarify in the colonial hematopoiesis, or CHIP, is that it's genes like BRCA1 and BRCA2 that we care most about for HRR genes, those can be CHIP. But it's rare. Okay, so, that's the other good news. So for those genes in particular it's rare. It's not unheard of though. And we've seen, and we published on this, BRCA2, and we've also seen BRCA1 CHIP. So it happens. Genes like ATM, CHEK2, common, okay, so those are just genes.
Those two genes which are on label for some PARP inhibitors, you know, depending on their approval in your country, those genes in particular are very prone to CHIP. So if you see an ATM or CHEK2, those two genes in particular, and you're considering using a therapy, like a PARP inhibitor on the basis of that, from liquid biopsy, be very, very cautious. That's a pretty decent chance that that's in fact clonal hematopoiesis. And, you know, talk to your molecular pathologist, if have one on hand. I realise not everyone does. You know, there's a lot, you know, a lot of times someone like me can, you know, look at the test results. Ideally you have a relationship with the person, or the lab, that's doing your testing that actually reviewed the test, and signed their name on the paper, just like your anatomic pathologist. And you can talk to them about the details because they can usually tell you, "Yeah, I'm pretty sure that's clonal hematopoiesis." Or, "No, I don't think that's clonal hematopoiesis, for this reason, that reason, and that reason." A lot of times like the variant fraction, and other things that a molecular pathologist is looking at, so. So, a long winded answer to your question, but certainly something we need to be concerned about it is getting better. Particularly if you see ATM or CHEK2, and you're thinking about drugging that from a liquid biopsy, talk to your molecular pathologist or the lab, to get more information would be my recommendation.
- [Presenter] I think that makes a lot of sense and I appreciate you giving those kind of specific examples around the genes that really more commonly have CHIP, because that's, I think, pretty easy for all of us in clinic to remember those two, ATM, CHEK2. Really think twice. And those are not necessarily genes that we think are going to have really strong responses to these drugs anyway. So it is important to make sure that we know what we're looking for, and we certainly would not want to drug something that really is just CHIP. You know, the other thing that I alluded to, as I was asking that extremely long question for the last one, so sorry for that, is that we don't usually want to do a liquid biopsy when a patient is being treated, and is having an excellent response. Maybe PSA is going down to zero, maybe they don't have a large disease volume to begin with. What are your thoughts there? And would you recommend maybe waiting until the cancer starts to grow again? Or would you take that test when they're responding?
- [Colin] Yeah, thanks. I hadn't answered that part of it. That's a great point. You know, if you don't have much cancer burden, you're not likely to have much cancer in your plasma, which is great for you if you're the patient, but not great if you're trying to do a liquid biopsy. So I wouldn't personally recommend doing liquid biopsies, for the purpose of treatment, if your patient's, you know, PSA is zero, and you're having a great response clinically. Some exceptions, like, you know, if you have evidence of still significant burden of disease, based on imaging, and maybe you have reason to think that it's like a PSA negative cancer, maybe it's neuroendocrine, or go on its way to neuroendocrine, in that scenario, you might still get a good liquid biopsy, you know, even with like a low PSA. But, again, if all of your clinical indicators are that this patient basically has low burden disease, and like the PSA is low, or undetectable, that's not the time to do a liquid biopsy. It's gonna be garbage in, garbage out, right? And the problem, again, getting back to CHIP, is that the lower the probability that you actually have prostate cancer in your sample, the higher the probability that whatever positive you get are just these kind of clinical false positives, right? So, sort of, it's like a lose-lose.
So absolutely be cautious about doing testing in that scenario. I would recommend against it if you really think that the patient's responding well, and the PSA is is low or undetectable. We published on this, gosh, five, six years ago now, and showed that in the setting of AR positive disease, PSA, this is a really a rough thing. It's not a hundred percent, but PSA about 10. If you have greater than 10, you tend to have really adequate amount of shed DNA into your plasma for liquid biopsy. If you have PSA less than 10, it starts to get sketchy. Now again, that's in the setting of AR positive disease, not necessarily neuroendocrine. That can be PSA low expressing. - [Presenter] That's a great rule of thumb, though. So, thank you for that. You know, one of the things, especially as these assays become more sensitive, that I find frustrating, and sometimes challenging, to interpret is when you do a liquid biopsy, and you find a bunch of things, and you're really excited. Wow, look at all these different findings. But the allele fraction is extremely low in all of them.
And then you wonder, are any of these real, or are these common mutations in the majority of the cancer cells in my patient? How do you think through that, and what should someone think about multiple mutations if they're 0.05% of the DNA that's found? - [Colin] Yeah, it's a great question, and, you know, liquid biopsy, even in the best case scenarios, tends to have low-ish amounts of cancer. So the fractions can be low, even in the setting of an adequate test. Bottom line is the lower the fraction, like certainly anything lower than 1% variant allele fraction, start to be sceptical of, but you can interpret it in context, right? And, again, this is where a molecular pathologist can come in, if you have access to someone, or someone equivalent, to discuss results, like a molecular tumour board or something. But it's contextually dependent. So if you have clear prostate cancer mutations, let's say you see like an androgen receptor resistance mutation, or a TMPRSS2-ERG fusion, or something that's very highly characteristic, that's definitely prostate cancer. And let's say you see that mutation at about 0.5%, or 1% fraction. Well, that kind of anchors you now on, well that's about the amount of cancer in my sample, right? So then other mutations at that same fraction are therefore more likely to be prostate cancer related. Whereas, let's take the opposite scenario. Let's say you have a prostate cancer mutation up at, let's say, 40% fraction. And then you see another mutation, let's say in BRCA1 or 2, that's down at 0.05, or something, percent fraction. Well, now, that's at best a tiny subclone, right? So what's going through my mind in that scenario would be like either you have a tiny subclone, you know, maybe one metastatic site is just developing a tiny little bit of BRCA2-ness, in which case drugging that, what's that gonna do? It's gonna kill that little tiny subclone. It's probably not gonna be clinically that beneficial. Or maybe it's clonal hematopoiesis, right? Depending on how the test is being done. So either one of those two scenarios isn't really clinically something you want to drug. So it is kind of contextually dependent, kind of, interpreting that variant fraction in context of other mutations. And it can get kind of complicated.
I mean, probably just what I've described here, this is like, well, he's getting kind of technical, but it could get even more technical than what I described. So, again, if you have access to an expert, ideally you have a relationship with the lab is doing your testing, and you can talk to the actual person who reviewed the test, and you trust them, that would be the ideal in those kinds of scenarios. - [Presenter] Yeah, that's great. And I so appreciate that you're really emphasising the multidisciplinary collaboration that is very important for these patients. When you think about the multidisciplinary team, and the way that we all work together, and think about the way that it really is overlapping with emerging technologies, and of the evolving standards that we have, what implications does that have for the future of clinical practise? How are we all going to be able to continue to work together, as we're integrating newer technologies? And what technologies are you most excited about? - [Colin] Oh yeah. I mean, the exciting thing from my standpoint, as a molecular pathologist, is that we've got so many new, you know, it's like the kid in the candy store. We have so many new technologies, right? And it's getting way more complicated, which, you know, if you're a busy clinician, is not necessarily a good thing, it's a bad thing. But from my perspective, and I think from the perspective of precision oncology, writ large, is really a good thing. But it does, you know, to your point, I think it, as we get all these new technologies, like we're starting to do RNA testing, right? Like, you know, sequencing to look for expression patterns. Or epigenetics. We're starting to do methylation testing and increasing numbers of saturate. So we're getting all these omics, right? And, eventually, probably, we're gonna be doing whole genome sequencing on every tumour. You know, that's sort of the end game, right? Instead of just looking at gene by gene. And then we're gonna be doing, probably, whole transcriptomic RNA testing, and epigenomics, and probably proteomics at some point, as well.
So, as these technologies build on, and get more complex, I think it really highlights what you're pointing out, that the need for, kind of, a multidisciplinary team really bringing in, ideally, a molecular pathologist, or someone equivalently trained, at the very least into some kind of molecular tumour board format. It was being shown in many studies to improve care. And I think it's gonna get to the point, I've made this argument, you know, that sort of like with radiology, you know, like, we had all these like technologies, like X-ray, for actually maybe 30, 40 years, before the field of radiology kind of developed. And I think that this with genomic medicine, writ large, which includes precision oncology, but is broader than that, you know, not just limited to precision oncology, but genomic medicine, I view similar to radiology. I think we're gonna get to a tipping point where everyone just realises this is way too specialised and complicated, you know, for me as a busy oncologist, or other provider using these genomics for my care, to do on my own. I need a diagnostic physician specialist on the multidisciplinary team, sort of like a radiologist is understood to interpret these other tests. And I think that's where it's going. I think, you know, AI, and things like that, are gonna really help, and that's gonna disrupt all of our primary care, not just diagnostic medicine, but clinical as well. So we'll see how that fits in. But I do think that it's just getting more complicated, and this idea of, you know, kind of what I talked about, like throwing a specimen over a fence to like a reference lab, and then they throw back a result, and you never talk to each other, and you just look at a piece of paper. I don't think that that model is gonna last much longer. I think we're gonna need to really team up for optimal patient care.
- [Presenter] Wow, well, you have certainly given me a lot to think about. Every time I talk to you, I feel like I could say the same thing, but there are so many ways that we can work together as a team, I think at this point, to really interpret these results, work together to get the best specimens possible, use that information to help provide care to our patients. And we are looking forward to an incredibly bright, but also incredibly complicated, future. So, thank you so much for your time, and your expertise today. - [Colin] Well, I really appreciate it. I enjoyed the discussion, thanks.
Developed by EPG Health. This content has been developed independently of the sponsor, Pfizer, which has had no editorial input into the content. EPG Health received funding from the sponsor to help provide healthcare professional members with access to the highest quality medical and scientific information, education and associated relevant content. This content is intended for healthcare professionals only.