Podcast 3: Molecular testing techniques for patients with mCRPC
Dr Alessia Cimadamore (Italy), Dr Niven Mehra (the Netherlands), Dr Pasquale Rescigno (UK) and Dr Michelle Shiller (USA) share their experiences from across the globe on the evolving diagnostic methods for patients with mCRPC.
[00:00:00] Dr Cimadamore: Welcome, everyone. I am Alessia Cimadamore, a uropathologist based in Italy at the University of Udine, and your host for today. So target therapies based on molecular biomarkers are increasing in use and transforming prostate cancer care. [00:00:15] Across this podcast series, we will be diving into the world of diagnostic and prostate cancer, raising awareness on the importance of considering molecular testing for all patients with metastatic castration-resistant prostate cancer, and [00:00:30] addressing key knowledge gaps, providing a variety of perspectives from different specialties and regions of the world.
In the previous podcast sessions, a review of the genetic and genomic testing guidelines has been covered, and it has been discussed [00:00:45] how the identification of homology recombinational repair and BRCA mutational status may impact patient outcomes. So, for a comprehensive dive into these topics, please go back and listen to them. Today, in this podcast, we are going to [00:01:00] discuss the testing techniques available for the metastatic castration‑resistant prostate cancer molecular testing paradigm in line with the updated guidelines. Today, I'm joined by Michelle Schiller, a pathologist specialising in molecular genetic [00:01:15] pathology from Texas, US, and then two medical oncologists, Niven Mehra from Radboud University from the Netherlands, and Pasquale Rescigno from Newcastle University, UK. So welcome, everyone. [00:01:30] And let's start introducing each testing method, starting from the gold standard. So, starting from Michelle with the tissue testing. So, what type of samples can be utilised for molecular testing in metastatic castration‑resistant prostate cancer, [00:01:45] and what type of information can we gather from them?
[00:01:48] Dr Shiller: That's such a great question, Alessia, and something that comes up a lot. There are several different sample types to consider for testing. First, a diagnostic biopsy, [00:02:00] and that's a common one to consider. The convenient aspect about testing the biopsy is that it offers the opportunity to perform testing earlier in the patient's course of disease.
For some institutions, they may want to test the [00:02:15] tissue at the time of diagnosis. At that time, if a mutation is detected in the homologous recombination repair pathway, for example, it will likely be stable. Such detection may also assist with routing the patient to [00:02:30] genetic counselling early in their disease course to determine if the mutation that's detected is also present in the germline and, therefore, causing them to be genetically predisposed for their cancer, as well as if cascade testing and surveillance is necessary for the [00:02:45] patient and their germline mutation positive relatives.
Points to consider with biopsy testing is timing, and if testing is postponed until disease progression, this may be many years after the initial diagnosis. Data [00:03:00] shows that the likelihood for success decreases with each year and that tissue that is older than 10 years has approximately a 40% or less chance of successful testing due to DNA degradation. Additionally, there may not be sufficient tumour volume to support [00:03:15] testing. One way to circumvent the possible low tumour volume consideration is to pool tumour from multiple cores if there is more than one core with tumour from the same biopsy interval. It's important that such a [00:03:30] practice be performed only on tissue that is acquired at that same time.
[00:03:35] Dr Cimadamore: Great. Thank you, Michelle. So, Niven, do we have alternative to tissue testing?
[00:03:40] Dr Mehra: Yes, we do. We can look at circulating tumour DNA, [00:03:45] so liquid biopsies. This is still something which is not standard practice in many countries. It's adopted particularly, I think, in the US and a bit less in Europe, but for about 60% of patients who are [00:04:00] progressing on ADT or combinations, they are amenable to liquid biopsy testing.
But there are some difficulties and some pitfalls, but we can discuss those later. But I think liquid biopsies are a good alternative.
[00:04:12] Dr Cimadamore: And from both tissue testing and [00:04:15] ctDNA testing, we can identify both inherited and acquired mutation, right?
[00:04:21] Dr Mehra: Right. So from ctDNA testing, you can, if you only look at ctDNA, you will also not be 100% sure if it's germline [00:04:30] or somatic. Sometimes it's quite clear, but sometimes not clear. So, I think every patient with a BRCA mutation or a DNA damage repair mutation, such as ATM, with a family history, should be referred for a specific germline test.
[00:04:44] Dr Cimadamore: And this is one [00:04:45] of the main differences with whole blood testing, am I right Pasquale?
[00:04:49] Dr Rescigno: Yeah, no, definitely. I mean, as Niven and Michelle were saying, you can capture a somatic alteration on tissue or ctDNA, but for confirmation of [00:05:00] a germline origin of this alteration, you need to do a sequencing of whole blood. So, usually you will look at the DNA coming from healthy cells, not tumoral cells.
Most of the time you do sequencing of what is called PBMCs, so the [00:05:15] peripheral blood mononuclear cells, and that's usually a really easy technique. So, you don't have to look at tissue, you don't have to look at ctDNA, you can just collect a sample on an EDTA tube. That is the normal tube that you will use for a full [00:05:30] bloodcountackout, basically.
And then isolate cells like PBMCs and do sequencing from there. As Niven was saying, for ctDNA, that's not a routine or a standard for determination, especially for BRCA1/BRCA2 mutation. The germline testing is usually done either to [00:05:45] confirm if the somatic alteration found on tissue testing or on ctDNA is a germline mutation or not, or as done when a patient is referred for genetic counselling in case the patient has a really [00:06:00] strong family history. So these usually are the recommendation to perform germline testing.
[00:06:05] Dr Cimadamore: Thank you very much. So getting into the technique of somatic testing, how these perform. So as a pathologist, I am required to [00:06:15] select the tumour area with the highest cellularity, avoiding areas of necrosis, inflammation, and this is because the DNA yield is the main issue, especially from biopsies, and the limiting factor to proceed further with [00:06:30] the test. So what are the next steps, Michelle, for tissue testing?
[00:06:34] Dr Shiller: Yeah, so for tissue testing specifically, after you identify the tumour and have those considerations, as you mentioned, the next thing, it could vary based [00:06:45] on labs, but what we do in our laboratory is we micro-dissect, going specifically where the tumour is. Again, as you say, Alessia, to remove the background nuclei because every nucleus has DNA in it and so we [00:07:00] want to try to really optimise for isolation of tumour DNA for these results.
And then after we micro-dissect, we isolate the DNA from the tumour and then we will quantitate the amount of that we have at [00:07:15] that step. And that's a pivotal step because, at that point, we know whether or not we have enough DNA to do the test. Once we have adequate DNA, then it next goes on to the sequencer, and there's also internal quality control for that [00:07:30] on the sequencer.
Sometimes you may have enough DNA, but the DNA may not be of sufficient quality for sequencing. Then once the data comes off, depending on your method, there may be a bioinformatics pipeline for the results to go through, [00:07:45] and then ultimately it gets analysed by the pathologist to report.
[00:07:49] Dr Cimadamore: Okay, let's say in case of a positive result, let's say a point mutation in the BRCA2 gene from tissue testing, how should we report it and what are the [00:08:00] next steps?
[00:08:01] Dr Shiller: Yeah, so if there's a point mutation from BRCA1 or [BRCA]2 testing on tissue, it's important to be very clear as to whether or not it is one that changes protein function. So, the first consideration is whether or not [00:08:15] it is. If the testing is done on next-generation sequencing, then we get an appreciation for how much of that mutation is present, and we can then say, oh, you know, we're seeing this at around 50%, and we do think this [00:08:30] might segregate out as one that's germline, so please make sure to follow the patient with germline testing.
And we'll just kind of keep it simple and leave it basically at that, and the molecular pathologist should let you know whether or not it is one that's been detected in the germline, [00:08:45] as another measure to see whether or not additional germline testing is necessary. But either way, it does provide some informative components for our oncology colleagues with respect to treatment consideration.
[00:08:58] Dr Cimadamore: Thank you. And [00:09:00] Pasquale, in the specific case of a germline variant discovered in a metastatic prostate cancer patient, what are the implication for his family, specifically for sons and daughters of the patient.
[00:09:11] Dr Rescigno: That's a great question, actually, Alessia. So, [00:09:15] usually, when the BRCA mutation is of germline origin, it behaves as an autosomal dominant alteration in terms of transmission to the offspring. So, if a patient has a BRCA2 mutation detected, he has the chance to have [00:09:30] passed this to 50% of his offspring.
What is really important to understand and to clarify is that when a BRCA2 alteration is germline, that means that there is a predisposition for cancer in the family, but not [00:09:45] necessarily that the person that has inherited the BRCA2 mutation will develop cancer 100%. It's important for females or the daughters of our patient because there are follow-up pathways for female patients that inherited the BRCA2 [00:10:00] alteration, because of course in female this will, especially for BRCA2, means they're likely to develop breast cancer, or for BRCA1, to develop ovarian cancer.
It is less clear right now what a man that has inherited the BRCA2 mutation from his [00:10:15] father has to do in terms of follow-up. First of all, BRCA2 mutation in men could mean the possibility to develop male breast cancer or again, prostate cancer. However, the guidelines are not really clear what the sons of the patient have to do.
The [00:10:30] NCCN guidelines suggest PSA testing at least 10 years earlier than when the father has developed prostate cancer. However, I would say that could be quite suboptimal in terms of the chance of detecting cancer [00:10:45] in a BRCA2 son of a patient that had a prostate cancer.
[00:10:49] Dr Cimadamore: Thank you. So, focussing now on somatic mutation, so on the tissue test and ctDNA test, what are the benefits and the [00:11:00] limitation of different methods. Pasquale, can you answer this question?
[00:11:04] Dr Rescigno: Well, I think that both have great potential, a great limitation. The main limitation, for example, for tissue testing is the availability of the tissue. [00:11:15] Most of the time the tissue is old, especially if we are talking about the archival, so there is a need to repeat a biopsy, and this is a really important issue, especially here in UK, because the UK is one of the countries where you don't need the [00:11:30] histological confirmation to start treatment for prostate cancer.
So, you can start treatment if the PSA is high and the patient has, for example, a lot of bone massmet[astase]s. So, we are in the situation now where we have to determine if the patient has a BRCA alteration, but we [00:11:45] don't have any tissue for that patient. So that would be one of the main pitfalls for tissue testing, for example.
[00:11:52] Dr Cimadamore: And we know that tissue test fails in about 30% of the case. So what are the main reasons why the [00:12:00] tissue test fails, Michelle? And what we can do to try to increase the success of the tissue test?
[00:12:08] Dr Shiller: So, there are a number of reasons that tissue may fail, Alessia, and a common one for prostate to think about [00:12:15] is the fact that prostate cancer has a proclivity to metastasise to the bone, and as a result, in order for us to perform testing on it, we have to decalcify it. Now the way that tissue is decalcified is through a very [00:12:30] acidic process, which breaks DNA down into very small fragments and makes it extremely hard to sequence.
And this is why most laboratories will not test decalcified tissue. However, the decalcification products are in [00:12:45] evolution and there are some newer ones that are less toxic to the DNA, and in our hands, there are a couple that are providing us around a 40 to 60% success rate, so that might be one thing to consider with respect [00:13:00] to management and increasing your success rate on tissue testing.
Other concerns are how the specimen was stored, especially if it's archival tissue. We may not know all of the environmental variables that that [00:13:15] archived specimen saw, which can also impact the DNA quality. Other things are having a sufficient amount of DNA for the testing. Some things that we can do on the front end is to, as a pathologist, when we sign a [00:13:30] case out, put a comment that says block A, block C, whatever, is a good candidate for downstream testing.
[00:13:39] Dr Cimadamore: And Niven, in case of tissue test failure, what are the pros and cons of ctDNA [00:13:45] tests?
[00:13:45] Dr Mehra: So, especially if the failure is on old archival tissue, then I think ctDNA progression is a good option because, of course, you will capture disease heterogeneity and the resistance mechanism. So, on the whole broad spectrum, you may [00:14:00] find more actionable alterations. But as we said earlier, especially in homologous repair and BRCA1/BRCA2, we think that at least these genes are truncal and clonal from the beginning.
So, these you should also [00:14:15] capture early as late, but for other genes and other alteration, possibly also in homologous combination repair genes other than BRCA, you might find new mutations. So, those are one of the benefits that you may capture more. But [00:14:30] there are some limitations, I think, to ctDNA, which we should state is that in ctDNA, as in tissue, you need to have sufficient circulating tumour DNA fraction in the total cell-free DNA what's available.
Otherwise, you will, may be, [00:14:45] picking up some mutations, but it's quite hard to find copy number alterations. And it's good to state that, as for tissue, as for ctDNA, that loss of both alleles is quite common for BRCA2, so about 30% of all the [00:15:00] patients with the BRCA2 alteration is a deletion, and those are quite hard to pick up in ctDNA because the ctDNA fraction is commonly lower than 30%.
And then it's from 20% to lower, it's quite hard to [00:15:15] pick up this kind of deletion. So, that's a kind of a pitfall for ctDNA.
[00:15:19] Dr Cimadamore: So Pasquale, the big question for you, what is the concordance between tissue and ctDNA testing?
[00:15:27] Dr Rescigno: I mean, all the registrative studies for PARP inhibitor [00:15:30] that have compared the results from tissue testing versus what I call liquid biopsies, highlight that usually the concordance is quite high. It says between 80-85% of the time there is concordance, especially for those alteration.
[00:15:44] Dr Cimadamore: [00:15:45] Great. So, Niven, what are the potential causes of discordances between ctDNA and tissue testing?
[00:15:52] Dr Mehra: Yeah, that's a good question. So, I think we need to be quite careful with ctDNA testing as sometimes you can pick up what we [00:16:00] call clonal haematopoiesis mutations or ‘CHIP’. So, as patients age, their bone marrow ages and you get certain mutations in stem cells, which leads to certain clones which have TP53 [00:16:15] but also like BRCA mutations or other HRR mutations.
So, sometimes we pick these up at a very low frequency compared to mutations which we know are prostate cancer. But sometimes it's quite difficult, because if the variant allele [00:16:30] frequency of picking up these mutations is in the same range, it's quite difficult to find a difference between a real somatic mutation from the tumour or a bone marrow derived CHIP mutation.
And sometimes even the CHIP mutations can be quite [00:16:45] high in frequency and can point to a leukemia. These are things which can be very discordant, so you need to be quite careful to know what's real and what's a CHIP. So, that's why in many ctDNA assays we also try to [00:17:00] add PBMC from blood, sequence that, and if you do then the cell-free DNA and the PBMCs, then you can really discern what's a CHIP mutation and what's a real mutation.
But in many commercial assays, this is not done, and then [00:17:15] you're never completely sure. Maybe the second point I can point out is that we know that prostate cancer is very heterogeneous. So, if you have a primary prostate cancer, the dominant cancer doesn't always have to give rise to [00:17:30] the mutation, so, and sometimes you can see that there are different types of cancer within the prostatectomy.
And there are quite some nice studies showing that there is also a mutational aspect. There are different clonal kind of mutations, and also one can [00:17:45] have a BRCA mutation, the other can not have it. If you do the archival sequencing and you find a BRCA mutation, but later on when the patient progresses to CRPC, maybe the patient has lost this mutation, has a maybe a neuroendocrine phenotype, these things can arise.
So, [00:18:00] that's why ctDNA has lots of promise because it captures all this heterogeneity, but we need to still be quite careful and probably test tissue, and if tissue doesn't work, test the liquid, but also always enhance where you [00:18:15] understand results.
[00:18:16] Dr Cimadamore: Thank you. Michelle, would you like to add something to these issues?
[00:18:21] Dr Shiller: I was just going to say some of the other things in dealing with tissue versus liquid biopsy or circulating DNA is that [00:18:30] with the tissue you have all of that background, as we discussed, and how we try to eliminate that when we do circle the tumour, and that is one of the reasons why you see all these other mutations better in the circulation because that background is [00:18:45] removed.
The other thing to think about as well, will there ever be 100% concordance between the two? And I think probably not, because we don't know if the tumour's shedding, and it may not shed in a sufficient quantity [00:19:00] to be detected. And so, no matter how we finesse the technology, I don't anticipate there will ever be 100% concordance, but that doesn't mean that either technology is not adequate for the patient.
[00:19:15] I also think I just want to make sure to be clear that, when something is detected on circulating, it is actionable with respect to identifying targets for treatment, and the clonal haematopoiesis of indeterminate potential is one that we're all really trying [00:19:30] to get a grip on because we see a lot of them as well even in the actual tumour testing.
[00:19:36] Dr Cimadamore: Great. Well, firstly, I would just like to thank you, to thank my fellow speakers for their time and fantastic input throughout this podcast episode. So after this podcast session, I feel like we all walk and walk away with a greater understanding of the pros and cons of the different testing methods we have spoken about today.
So, hopefully, this will drive us to [00:20:00] implement testing as widely as possible in our centres and optimise our sample handling practice. As my final remark today, thank you all for listening to this podcast today. If you enjoyed listening to our discussion, be sure to tune [00:20:15] in next time. If you missed the previous two podcasts, I encourage you to go back and listen to them too. Thank you.
Meet the experts
Dr Alessia Cimadamore is a pathologist known for her work in prostate cancer. She is currently an assistant professor at the University of Udine, Italy, where she focuses on genitourinary tumours.
Disclosures: Financial support from Amgen, AstraZeneca, and MSD.
Dr Niven Mehra is a medical oncologist specialising in prostate cancer. He is an associate professor and clinician at the Radboud University Medical Center, Nijmegen, the Netherlands. His work primarily focuses on understanding the genetic landscape of metastatic prostate cancer to develop new therapeutic strategies.
Disclosures: Financial support from Astellas, AstraZeneca, Bayer, Bristol Myers Squibb Foundation, Janssen-Cilag, MSD Oncology, and Pfizer.
Dr Pasquale Rescigno is a medical oncologist with a focus on genitourinary tumours. He is a clinical senior lecturer and honorary consultant at Newcastle University, UK. His work involves the development of clinical trials, aiming to further understand prostate cancer and enhance patient care.
Disclosures : Financial support from AstraZeneca, Gilead, Ipsen, Janssen, MSD, Pfizer, and Prostate Cancer Foundation.
Dr Michelle Shiller is a pathologist specialising in molecular genetic pathology. She is a member of the Precision Medicine Institute’s advisory committee, working on molecular diagnostics and genetic testing at the Baylor University Medical Center, Dallas, Texas, USA.
Disclosures: Financial support from Amgen, Astellas, AstraZeneca, Lilly, Merck, Mirati Therapeutics, Pfizer, and Takeda.
Dr Alessia Cimadamore
Dr Niven Mehra
Dr Pasquale Rescigno
Dr Michelle Shiller