ASCO 2016

Sequencing analysis with circulating DNA

Dr Oliver Zill - Guardant Health Inc, Brentwood City, USA

I am excited to be here because we’re presenting results from over 15,000 advanced stage cancer patients who we have analysed using our circulating DNA sequencing platform. So that abstract was accepted for oral presentation on Tuesday and my co-authors are really excited about that.

Circulating DNA was part of a press release that came out on Friday; it seems to be attracting quite a lot of attention in the field as a way of being prognostic, diagnostic indicators with less of the intrusive use that, for example, surgical samples may include. So can you tell us more about how this is being set apart from other samples and the impacts it could have for clinicians, patients and everyone involved?

Sure, yes. It is increasingly used for basically genotyping these cancers, so to help patients get targeted therapies. The traditional way of doing that has been direct tumour biopsy, so actually taking a sample from a person’s tumour and then isolating the DNA from there. Now, one of the advantages of using a blood-based method, in addition to just being non-invasive, it’s less harmful and as general as that. Sometimes the biopsy method fails for various reasons – the tissue gets exhausted and you don’t have a chance to completely genotype all of the actionable biomarkers or just there’s not enough material there, especially in lung cancer that’s often a frequent complication. So with blood we actually have the ability to help patients and we show in this abstract that there are a number of patients where we find actionable biomarkers where patients can get a targeted therapy where their tissue is exhausted or just what’s called QNS – quantity not sufficient. So that’s one of the major applications for this technology is in helping out those patients.

Has it been useful with ongoing patient surveillance as well?

Yes, some patients are using it for monitoring which is, of course, another application. Since it’s non-invasive it’s quite difficult and usually not recommended to serially biopsy a patient’s tumour. But we do see for some of the patients who are doing serial testing or being serially tested that you can see emergence of firstly a primary driver or driver mutations. And then, if that patient is put on targeted therapy, emergence of resistance over time of those drugs and that could give them a heads up in saying this therapy may last for a little bit longer but you might want to start thinking about what the next option would be.

One advantage of solid tumour samples would be genotyping being able to determine if there is any heterogeneity within the tumour. Do circulating DNA samples offer the same kind of variability with content that way?

Yes, actually you could argue it both ways. We like to think of circulating tumour DNA as providing a global summary of tumour heterogeneity and that’s because as long as a sufficient number of tumour cells are shedding their contents into the blood we can isolate that and find all these subclonality, this heterogeneity as you put it, from multiple sites within the same solid tumour and potentially multiple metastases within a patient. In fact, tumour tissue biopsy sometimes has a disadvantage where if you’re taking one slice or a needle poke through a tissue you may miss a lot of the heterogeneity. In a metastatic case these are non-resectable cancers, you’re not going to capture the entire solid tumour when you go in there. So we think right now that circulating tumour DNA does have the ability to capture all of that heterogeneity.

Are we going to see this being brought forward to a wider clinical use?

It is already used quite widely, we have over 2,000 ordinary oncologists just within the US, for example. We’re aware that we’d like to continue to improve our sensitivity and we have a lot of efforts to do that. We recently announced an effort to try to do earlier stage detection. Right now tests are just indicated for stage 3 and stage 4, so very late stage disease, but then again you have the sensitivity problem where the less tumour material substances within a patient potentially the less DNA that would be shed into the blood.

Now we actually have had some very promising results in a couple of different studies, some of which is being presented at this meeting, showing that we can detect circulating tumour DNA in stage 2 patients, at least in some subsets of them. But it’s very early days so we’re working very hard to try and potentially expand that role.