0:00
I think our next speaker is Dan.
0:03
He's the BD person at Biognosys and he's going to give a quick presentation on the unbiased targeted proteomics and biomarker discovery.
0:14
Dan, the floor's yours, buddy.
0:23
So yeah, good afternoon everyone, and thank you for sticking around for the CRO talk at the end of the session.
0:29
That's very much appreciated.
0:32
So I'm, my name is Dan.
0:33
I'm one of the BDMs for diagnosis.
0:35
So working here in the UK, but also across the Nordics and across Europe as well.
0:40
And I suppose the focus of the talk today is to give you an overview of what we do.
0:46
I'm going a bit into some of the applications, also a little bit of a case study as well.
0:51
But yeah, just to give you an idea of what we can do with mass spec based proteomics as a CRO of diagnosis.
1:00
But not to bore you too much, just to start with a little bit of a history lesson, I suppose as to where we came from and how long we've been around.
1:07
So the company was founded almost 16 years ago now as a spin off out of ETH Zurich.
1:13
So still a university very that is very prestigious and very much involved in proteomics.
1:19
So we specifically spun out of the lab of Ruedi Aebersold, who some of you may know as one of the Pioneers of what's called DIA proteomics.
1:28
So the idea of unbiased mass spectrometry.
1:31
So taking all of your ions forward and looking at deep coverage of the proteome.
1:35
And that's very much where we started from sort of doing running DIA mass spectrometry.
1:39
But we've expanded a lot since then try to move a bit more away from the technical side of things as we cater more to the drug discovery arena and focus on specific biological applications.
1:52
But we're a very high throughput facility.
1:53
So I think one of the myths of sort of mass spec proteomics and mass spec in general is that it's expensive and it's very low throughput and very slow as well.
2:02
But I would say over the last 10 years or so, we're very much at a high throughput rate in terms of mass spec proteomics as well as without compromising on the deep coverage that we get.
2:13
So yeah, we process sort of thousands of samples a year, most of which are happening currently in Schlieren.
2:19
So that's the district of Zurich in Switzerland.
2:22
So that's our HQ.
2:23
But actually last week I came back from Boston where we launched our high throughput proteomics lab in Boston.
2:29
So that is mainly to cater to the clinical market in the US to save shipping samples over to the EU, but a very exciting development for us. In terms of who we work with.
2:40
So as I mentioned, it's very much the drug discovery community, also academia as well.
2:45
But we have just an excess of 800 clients, some of which are in the large pharma, but a lot in biotech as well.
2:51
We have, it's actually almost 4000 publications now using the various proteomic platforms and we can process, as I said, in a very high throughput manner up to it's almost 2000 samples per day across the sites now. And some fairly big news.
3:06
So we were a private company, we functioned on our own and we're still our own entity.
3:10
But we recently, well last year now joins the Bruker family.
3:14
So Bruker very much a leader in mass spectrometry amongst other things.
3:18
And essentially what that gives us is access to the latest technology and the ability to keep improving our workflows.
3:27
But just to touch a bit more on the sort of Bruker family and how that all fits specifically related to proteomics.
3:33
So you can see the diagnosis side of things on the bottom there.
3:35
So the Contract Research Services also the mass spec software, which any of you, if any of you are mass spec users, you may know the software's the Spectonaut and Spectromine and it's something we use in the CRO as well.
3:48
But as you can see, it's very much a proteomics ecosystem.
3:51
So we have the sample prep side of things again for mass spec uses from a company called PreOmics based in Germany.
3:58
We then of course have the Bruker mass spec technology which we use ourselves and then coming into the services and the software at the bottom.
4:05
And what I want to focus on today is very much the CRO services I think is most applicable to the audience here today.
4:13
So yeah, that's what I will be focusing on in the subsequent slides.
4:19
But just first of all, why mass spectrometry?
4:21
So I'm sure a lot of you are aware of other proteomics technologies.
4:25
I'm sure some of you know Soma, you know Olink, so very much in the biofluid space, so antibody and aptamer based technologies.
4:32
But why do we use mass spectrometry?
4:35
So it's a very reliable technology.
4:37
So we detect multiple epitopes, multiple peptides per protein, allowing for a very accurate quantification and also identification as well.
4:46
We can detect isoforms as well, subtle changes in the amino acid sequence of peptides as well, something that's not possible in other techniques.
4:57
Being mass spec based.
4:58
We are very much both matrix and species agnostic.
5:02
So we of course work in the biofluid space, but we do work in cells, tissue and that is fresh frozen and FFPE as well.
5:09
So we're very agnostic there and have the ability to transfer different assays across those species and across those matrices as well.
5:19
And I suppose, you know just touching again on the being species agnostic, of course that's very translatable.
5:24
We do develop a lot of targeted assays in the preclinical space in mouse models in rat PDX, CDX and then move them into to sort of other matrices, so FFPE and oncology and fresh frozen tissue and biofluids too.
5:40
So in terms of how we sort of position these services that we have, it can, you know, it can come across as quite a black box if you like the way we market things.
5:50
But the way I would sort of describe our services is something that really does translate all the way from discovery through to preclinical and into clinical trials as well.
6:00
And we're involved in a number of phase one trials running targeted panels as we speak at the moment.
6:07
But there's a lot of different platforms here.
6:09
And maybe to move forward a slide just to deconvolute that slightly.
6:13
So there's three main platforms that we have at Biognosys.
6:16
So the first one which we market is TrueDiscovery is very much unbiased proteomics.
6:21
It's using what's called DIA mass spectrometry.
6:24
So what we're looking at here is things like biomarker discovery, looking at MOA studies, the deep coverage of the proteome, looking at significantly regulated proteins in an unbiased way in specific molecular pathways as well.
6:36
So the mass spec itself, the data acquisition purely unbiased, but then we can really add some biology to the data and interpret that in a much more relevant context.
6:47
But in terms of discovery, there's a lot of spin offs that I'll go into in a minute.
6:51
But we can enrich for certain parts of the proteomes.
6:54
So PTMs are often something that's of interest to people and something we can do with mass spec phosphoproteomics in particular.
7:02
But yeah, so that's the first platform we have.
7:04
Very much unbiased deep coverage.
7:07
We then have our TrueSignature platform, which is almost the opposite.
7:10
So this is targeted proteomics and in this workflow, we're looking at more like more sort of 10s of proteins, looking at PK/PD biomarker panels and these are entirely customizable panels.
7:21
So unlike other companies, we don't have a set panel of proteins, and you will see those proteins every time.
7:26
We can very much develop exactly what you want often out of these initial discovery studies that we do on the left there and maybe just another sort of difference dimension.
7:36
So discovery, of course, you identify the proteins in the sample.
7:39
You also get a relative quantification as well in that sample set targeted proteomics, the TrueSignature there is an absolute quantification.
7:47
So we spike in a reference peptide.
7:49
And again, I can go into that in a bit more detail with a bit of with a bit of context later on.
7:55
And then third and finally, it's we have a third platform that I'm not going to dwell on too much.
7:59
This revolves more around target ID of small molecules.
8:03
So we have a proprietary platform known as LIP-MS where we identify targets of small molecules from phenotypic screens or target-led approaches looking more at selectivity and we're able to identify targets with no form of labelling in a cell lysate.
8:18
But if that's of interest to anyone, I'm happy to speak about that afterwards separately.
8:24
So yeah, as I said, I will focus on these two platforms on the left here, which is I would say, most relevant to this audience.
8:30
So to start with TrueDiscovery, just to go into a bit more detail.
8:34
So as I mentioned, we use what's called DIA mass spectrometry where traditionally in other acquisition methods in mass spec, we pick specific master charge windows and we select certain ions. In DIA, very simply, we take everything and that's what allows us to quantify up to 14,000 proteins in tissue, 11,000 in cells and up to almost 7000 these days in plasma and other biofluids as well.
8:59
So it's a truly unbiased method when it comes to acquiring the data.
9:04
And I think a question that comes up from a lot of people is that's a lot of proteins.
9:08
That's very interesting as a technical feat, that's very cool.
9:11
But which proteins you can see and what they do is what's key.
9:15
And a question that is asked a lot is will you give me a very long Excel and a very long list of proteins?
9:21
And then I have to work out what to do with it.
9:23
And the answer is no.
9:24
So of course we do give the raw data and the proteins, but what we also look to do is more of a functional analysis.
9:31
And we do things like look into molecular pathways, GO enrichment, but of course we can cater that and tailor it to exactly the question we want to answer.
9:40
But it's not a case of a scary list of thousands of proteins.
9:44
It's very much a functional analysis, if you like, as you can see on the right there.
9:50
So I mentioned this a little bit earlier, but we do look at specific parts of the proteome as well.
9:55
So phosphoproteomics is notoriously a difficult thing to do with mass spec.
9:59
But what we do is we enrich for phosphopeptides, see up to 80,000 phosphor peptides and normalised the protein level and allows us to identify true PTM events and significant changes.
10:11
So we do a lot of work with PTM biomarkers, for example, for customers both in this unbiased space, but also in the more targeted absolute quantification as well.
10:21
Maybe just a final comment on this slide.
10:23
So we also look at the immunopeptidome as well, not specific so much to the to this agenda, I know, but we do a lot of HLA 1 and 2 presented peptide profiling and it's very much a growing area for us.
10:38
And again, something happy that I'm happy to talk about later on if you guys want to stop by the booth and that is relevant, but now moving more into the targeted proteomic space.
10:50
And I also want to go into add a bit of biology later on with a case study as well.
10:54
So this is the absolutely quantitative method that we use where we look at specific customizable panels from customers.
11:00
And it's quite often a platform we employ poster discovery study.
11:04
So we look at the significantly regulated proteins in the discovery sample set, we add the biology, we look into it, we develop a panel for customers and we can then use that in a new sample set and be much more accurate and precise in what we quantify.
11:20
So in terms of set of advantages and why to do targeted assays with mass spec, I mean, I've mentioned one before and that it's very translatable from discovery studies into more targeted assays.
11:31
So what's really key when developing a target targeted assay is we have a good flying peptide in the mass spec from the protein of interest.
11:38
So we can use that discovery data, and we synthesise a heavy labelled version of that peptide that we will spike into samples.
11:46
And we look at the ratio of the endogenous peptide to the heavy labelled and allows us to absolutely quantify the protein in that sample.
11:55
And with these assays, what we see very commonly as I mentioned earlier, as we start in the preclinical space often in animal models and we've picked peptide sequences that are shared amongst all species, particularly human as well.
12:07
And we can then very easily transfer these assays more into clinical samples as well.
12:14
So I've talked through this already a little bit, but just to go through the very general workflow.
12:19
So as I said, we select a peptide from the proteins of interest and that the acts as a surrogate, we developed the assay.
12:25
So we'll spike in the heavy labelled version of that peptide and the endogenous and just develop the assay from there.
12:31
And then we can either go straight to the sample measurement itself or we can also look at characterising these assays as well.
12:38
And the level of characterization we do here can vary depends on the exact specification, but I would see them say the most simple is looking at the lower limits of quantification and also detection as well.
12:53
So finally wanted to write a bit of biology and save you all from a lot more mass spec and talking through all that.
12:59
So I've got a few slides here outline outlining one of our, I suppose key collaborators based in Barcelona.
13:05
So Peptomyc.
13:07
So Peptomyc are a very exciting biotech who've actually just been published in Nature recently regarding and so detailing their various analysis of Omomyc.
13:17
So a new in class MYC inhibitor and what they've used over the years is a number of our platforms, but I wanted to focus specifically on the targeted proteomic side of things and what they did here in terms of the pharmacokinetic analysis.
13:32
So just a bit of background.
13:33
So they're currently to date is there is no direct MYC inhibitor approved for clinical use.
13:38
So Peptomyc, as far as I'm aware, are very much the first people to have some kind of inhibitor that can directly interact with MYC and inhibit the interaction with MYC and MAX leading to cell proliferation.
13:51
And what they developed was this Omomyc sort of mini protein that is based on MYC but has a four amino acid substitution.
13:59
And they use this, as I said, to inhibit this MYC protein.
14:04
But what did they do with us?
14:06
So what Peptomyc wanted to look at was the pharmacokinetic analysis.
14:10
They wanted to quantify Omomyc in various different sample types and look at the distribution of the drug within different matrices.
14:21
So what we did was develop a targeted assay.
14:23
So you can see there, we've got a few peptides highlighted.
14:26
So these were the peptides that we chose to develop an assay for this Omomyc protein so we could quantify in different samples.
14:34
And then after this, so once we developed this targeted assay for Omomyc, we then looked to do initial an initial pilot in mouse tissue samples.
14:43
So fresh frozen to start with.
14:45
So we took 180 samples, two time points, 2 doses.
14:49
And I don't have an exact figure of the quantification data here, but the trend we saw as expected was as you increase the dose, we see more Omomyc present in the fresh frozen tissue.
15:02
So after the success of that study, we looked into more of our sort of PK analysis.
15:05
So we've one dose of Omomyc and four different time points.
15:10
And what we also did was look at fresh frozen tissue.
15:13
Again, we also looked at serum as well to see how that would change over the time points.
15:17
But we also wanted to include FFPE as well because ultimately FFPE being the sample matrix they would have available to use in the clinic.
15:26
And we took a couple of 100 samples to look into this from each of these matrices and again, using that developed panel that I mentioned earlier.
15:36
And what we see is, I suppose a very clear result.
15:40
So what you can see here is on the four of these graphs is each of the peptides from the Omomyc mini protein.
15:47
And you can see that first of all, the amount of Omomyc in serum decreases over time, which was as expected.
15:55
We see that there is much more of this Omomyc protein in the tissue as you go across time and it very much stays there as well.
16:03
And what's interesting too is when you compare the FFPE and fresh frozen tissue samples, we see that they're very comparable, meaning that we can use this assay in the clinic as well.
16:13
And that's something that is moving forward as we speak.
16:18
And yeah, and just as a summary, so Omomyc is asymmetrically distributed between serum and tissues where it lasts longer and very much backed up their hypothesis.
16:28
And this data also features in the Nature publication that came out a few weeks ago.
16:35
So it's a very quick summary.
16:36
So MS based proteomics for drug discovery, both unbiased and targeted analysis very much flow into one another and yeah, work very well.
16:47
We are matrix and species agnostic.
16:49
So we do work in preclinical into clinical trials as well.
16:54
And we can quantify with minimal volumes, so, 100 microlitres of plasma, serum, CSF and very low inputs for tissue as well with a couple of milligrammes too.
17:05
But with that, thank you very much for listening to the CRO talk and happy to take any questions.
17:09
Thank you.
