Charles Ryan: Hello from PCF 2019. I'm delighted to be joined by my colleague, Alex Wyatt, who is an Assistant Professor who focuses on prostate cancer genomics, and he's a Senior Research Scientist at the Vancouver Prostate Cancer Center in Vancouver, British Columbia. Thank you for joining me.

You are really one of the top minds, I think, in our field on the area of cell-free DNA and its limitations and its promise. So considering that we may be educating clinicians, and clinicians are watching this, what's the current state of what a clinician can use in prostate cancer from cell-free DNA for the current testing and where they are?

Alexander Wyatt: No, that's a good question. I think this field has emerged so rapidly and expanded so quickly that it's been difficult even for someone in the heart of the field to understand all of the changes.

At the moment, there are several commercial companies around the world, mostly in the US, that are offering cell-free DNA testing for advanced patients. Not all of them have immediate clinical implications, but there's an ever-evolving landscape of clinical tests.

I think there's a couple of key things to be aware of. One is that there are multiple potential clinical indications for circulating tumor DNA in the blood. One, which is more relevant for early disease settings, is about the detection of recurrence. So if you can detect circulating tumor DNA at all, it means that you have cancer somewhere in the body. And so that's really promising for looking at patients post-surgery or post-treatment in the early setting, and looking for that concept of minimal residual disease.

Charles Ryan: Before we go on, is that a bit of an expensive way, in a way, to measure for recurrence in a patient, because these tests run panels of several genes, and really what you're looking for is presence of anything or nothing, right?

Alexander Wyatt: Yep, that's correct. I think in prostate cancer it's going to be very difficult to beat PSA as a detection of recurrence.

Charles Ryan: But in cancer overall?

Alexander Wyatt: In cancer overall, I think there are some settings, bladder cancer, for example, post-cystectomy or post neoadjuvant treatment even, where there's really an unmet need for biomarker. And so it's going to depend on the precise situation. You're right.

Charles Ryan: But now the cell-free DNA test that we get, as I said, they run a panel, they can tell us about mutations, they can tell us about amplifications, but not so much about deletions, right? So if I have the choice of looking at tissue or looking at cell-free, should I do both? If I had to do one, what would I do?

Alexander Wyatt: Yeah, so a patient that has metastatic disease normally has some level of circulating tumor DNA in their blood, but there's this big spectrum in terms of how much DNA there is.

And so in a patient that has high amounts of circulating tumor DNA, it normally associates with high burden of disease, so sometimes visceral metastases or a lot of bone metastases. Those patients, you'll get the same amount of tumor DNA that you might get from a tissue biopsy. So in those patients, you can get all that information, copy-number changes, so, say, a deletion of BRCA2 or an amplification of AR.

But in the patients with a lower burden of disease, maybe just one or two progressing metastases, sometimes the ctDNA amount is much lower, and you may be able to get mutations from that information, but certainly looking for copy-number changes. Those are the patients where you might need to consider a tissue biopsy or going back to look at archival tissue.

Charles Ryan: So right now, in our current state, most clinicians who are doing tissue or DNA analysis in a man with CRPC for example, they want to know, does this patient have a DNA repair alteration? Does this patient have MSI-high status? Because those will directly lead to a therapeutic recommendation. Are the cell-free DNA tests that we may order, the commercial ones, sufficient for ruling in or out those two entities currently?

Alexander Wyatt: So I think the current crop of commercial assays are not sufficient to rule out all DNA repair defects. They can certainly detect a great number of them.

Charles Ryan: So a negative test does not necessarily mean the alteration is not there.

Alexander Wyatt: That's an important concept to understand, yes. So a negative test does not mean the patient's tumor is well type for a given gene. There are alterations that are very complex and cryptic to detect, particularly for mismatched repair genes. There are structural rearrangements that affect introns, for example. There are even epigenetic alterations. And the same is true for BRCA genes and others. So you may be able to identify a somatic mutation in BRCA2 but the more complex deletion or biallelic deletion, it's going to be more challenging.

Charles Ryan: Right, and a biallelic deletion is precisely the most important defect you would want to discover in a patient, for example, for the use of a PARP inhibitor, right?

Alexander Wyatt: Yeah.

Charles Ryan: And that's the one where this test could fail you. So I think that what I'm getting at is there's a place for tissue and there's a place for cell-free. Ideally, if cost weren't an issue and availability of tissue weren't to an issue, we may do both, because this cell-free will give us information on evolution of the tumor over time in the individual patient.

Alexander Wyatt: Mm-hmm, yeah. I think about it in the sense that there are patients with high enough levels of circulating tumor DNA that you can be pretty confident when you see a negative result that you're not missing something. But it's that gray zone, where there's lower amounts of ctDNA, and you may not be confident that a deletion would be picked up. And so there I think we need tissue either as tested in parallel or as a fullback.

Charles Ryan: And so for the clinicians who are listening or watching, there is always a proportion of circulating cell-free DNA in us, even those of us who don't have cancer.

Alexander Wyatt: Yeah.

Charles Ryan: Of that proportion, if there is a cancer in the body, as you say, a proportion will be circulating tumor DNA. When you say a lot, what do you mean in terms of the proportion of circulating tumor DNA that is of the full cell-free DNA fraction?

Alexander Wyatt: Yeah. So yeah, so cell-free DNA is shared by all cells in our body when they apoptose. Most of the normal cell-free DNA in our blood is from blood cells, from white blood cell lineage. And so, in a person with cancer, some of that cell-free DNA is contributed by cancer cells. But we have to also remember there's other somatic expansions in the body, even in a perfectly healthy individual. And so those are also contributing somatic mutations to that cell-free DNA compartment.

So when we're looking in a patient with advanced metastatic disease, I like to use thresholds of above at least 1% circulating tumor DNA.

Charles Ryan: So of the full body's cell-free DNA, 1% of it is circulating tumor DNA.

Alexander Wyatt: Yes.

Charles Ryan: And you interpret that as a substantial amount. That's a real amount.

Alexander Wyatt: Yeah.

Charles Ryan: Okay.

Alexander Wyatt: Yeah, that's a good amount. So a lot of those commercial tests can actually identify alleles at much lower amounts than that. At that level, it becomes questionable as to the exact origin of the alteration. Sometimes it is difficult to discriminate between something from a prostate lineage and another sematic expansion.

One thing that's interesting to note though is that we have seen patients, and they're not rare, where the amount of tumor DNA in the blood is over 90% of the total cell-free DNA. And so in those patients you'll get micrograms of DNA from a few mls of blood. And so-

Charles Ryan: I mean, are those patients with such advanced disease, or are those patients who have somehow are shedding more or-

Alexander Wyatt: So normally it does track with the extent of their disease. And there is certainly a link to patients that have explosive and neuroendocrine transdifferentiation for example. But there's also some biological factors that we don't understand yet.

Charles Ryan: I'm glad you brought up the neuroendocrine. And a final point on that is, can we use cell-free DNA to differentiate histological subtypes? And what would be the marker we would look for in cell-free DNA that would say this is the neuroendocrine process?

Alexander Wyatt: So with the current technology, which is mostly looking at somatic alterations within the genome, so deletions, mutations, we don't yet have a biomarker that could say presence, absence, you have this phenotype in your cancer cells. But what we can do is enrich for that probability. We can look for changes like RB1 alterations, P53 alterations, that you have that phenotype.

In the future, I think it is going to be more possible. We can look at the epigenome of the circulating tumor DNA to look for the markers of transcriptional activity, and so forth, that's imprinted on the genome.

Charles Ryan: We'll be able to look at methylation status of cell-free DNA.

Alexander Wyatt: Methylation status, look for... using nucleosome spacing in the genome to look at regions of open chromatin. And from that you can predict, amazingly, what the transcriptome is doing in their cells.

Charles Ryan: It is amazing.

Alexander Wyatt: And it's really the transcriptome where you need to understand, to be able to see neuroendocrines.

Charles Ryan: Well, I look forward to talking to you again in the future on updates, as all of this evolves and develops. And congratulations on your work. And always a pleasure to talk to you.

Alexander Wyatt: Yeah.

Charles Ryan: Take care.

Alexander Wyatt: Thanks.