0:29
Yeah, My name is Chris.
0:31
I'm here presenting on behalf of Miltenyi Biotech.
0:38
So today I'm going to be talking about Miltenyi Biotech's offerings in the multiomic space, what we can do specifically with our MACSima platform, our new RNA Sky technology, which recently came on to the market.
0:53
And yeah, this talk's going to be divided into two parts.
0:56
First, we'll talk kind of generally about the company, about this technology, the platform, and then we'll spend some time getting into a case study where this technology was used really for the exploration of a tumour microenvironment.
1:13
Yeah, so starting Miltenyi Biotech.
1:15
A little bit about us.
1:17
We are driven by this core mission to make cancer history.
1:21
So to do this we provide research and clinical workflows to our customers.
1:28
This spans multiple fields where we provide really end to end solutions from, you know, sample preparation all the way through data analysis and visualisation.
1:37
Whatever is really needed to make the workflow come to life and get all the insights that you need from it.
1:43
Yeah, across multiple different areas, including in the very exciting field of imaging and spatial biology.
1:53
So currently within spatial biology, we have two platforms on the market.
1:57
So on the left here you see the UltraMicroscope Blaze.
2:01
This is our 3D imaging system.
2:03
It's a light sheet-based microscope designed for imaging full organs or even small organisms in three dimensions with cellular resolution.
2:15
And then on the right here you have our topic for today's talk, which is the MACSima platform.
2:21
If you want to see one in real life, we have it at the booth.
2:23
So feel free to go check it out and talk to the team there.
2:28
This is a very powerful and flexible platform designed for imaging tissue sections with really hundreds of markers possible in subcellular resolution.
2:46
So the engine that kind of drives what we do on the MACSima is this MICS approach. MACSima imaging cyclic staining.
2:57
And with this concept that we can stain with up to 3 or 4 markers at a time, then we can image it.
3:03
And you know the ROI's that you select for your experiment and then we erase those stains and basically repeat the cycle as you can see over and over again.
3:13
And in principle, you know, you can repeat this countless times.
3:16
We have data sets of over 200 plus markers showing that this is possible.
3:22
I think if you were here in one of the previous talks earlier, you all saw this slide example of someone using this for 50 markers on single tissue sections. Kind of top level what the workflow looks like for how we do same section multiomics on the MACSima platform.
3:42
So this is all driven by the MICS technology.
3:44
You can start with virtually any FFPE tissue sample.
3:49
Then you go through some sample preparation.
3:51
You're working with a section here, that goes on to the instrument where we perform this MICS process.
3:57
We typically start with RNA detection followed by protein detection on the same sample, generate a lot of data and that data can then be analysed and visualised in our MACS IQ view software.
4:10
And here you can do all kinds of spatial biology analysis.
4:13
You know things such as cell phenotyping, cellular neighbourhood analysis, looking at different cell to cell interactions, functional states of cells and you know things like that.
4:29
Specifically for RNAsky, the way this works, zooming in a little bit on the sample prep and the target detection proportions of the workflow you start with the FFPE tissue on this high res slide.
4:41
So as the name suggests, this is a consumable that we offer that allows you to take high resolution images for either RNA protein or true multiomic data sets.
4:53
You combine your sample with the RNAsky panel, you go through the sample preparation system that consists of hybridization, ligation, amplification and that goes on to the instrument for detection and imaging.
5:05
So here we stain or label up to four RNA per cycle.
5:10
We image them and erase them and complete this MICS cycle.
5:13
And basically we're doing a non decoding based detection here with direct labelling.
5:23
This is a little bit of how some of the data looks.
5:26
I hope it looks nice enough.
5:28
Also on this projection here, so you can see just here in the same ROI on the same tissue, you can see some example of RNA o 3 transcripts are here with a different coloured dots.
5:41
See in the middle protein from the same tissue and then you can see the data merged or IQ view analysis software.
5:49
They're obviously nice pretty images, but you can also see things for example here that we have pretty good correlation between RNA and protein detection shown some validation also of the workflow.
6:08
Target applications, I think everyone at this conference knows that there's a lot of target applications for transcriptomics and multiomics.
6:16
We see a huge benefit of combining RNAsky with Hi-plex protein analysis.
6:22
You can do things like cross validate, upstream screening methods.
6:27
You can look at things that are difficult to look at with just protein, such as chemokines or cytokines, do RNA protein co-expression analysis, look at different signalling pathways.
6:38
And the truth is, you know, sometimes you may have studies where certain antibodies perform poorly or they're simply not available on the market.
6:47
And transcriptomics is a great way to kind of bridge that gap and fill it in.
6:55
So yeah, having shown how the technology works, you know, I'm happy to also present here our first curated panel for RNAsky.
7:05
It's a 24 Plex called the IO Explore.
7:09
It's designed to work broadly across different cancer systems.
7:13
It's an immune oncology panel.
7:14
So we have immune cell profiling, tumour characterization, different cell state markers and of course some controls, housekeeping and negative controls in here.
7:27
I want to get into a little bit of the benchmarking data that we have for this panel.
7:33
So we have evaluated reproducibility so intra-samples.
7:38
So from the same block when we look at for example serial sections, we get very strong correlations between results from this RNA sky panel when we look between samples as well very strong correlations and very positive results here as well.
7:53
And when we validate against other technologies such as bulk RNA sequencing, single cell RNA sequencing also over here, we see very positive results and strong correlations across different technologies.
8:11
Also looked at it in multiple different tissue types and you can see here different tissues.
8:17
So we know that the technology works across tissue types.
8:20
We get a high density of detections per cell specificity we evaluate with a false discovery rate analysis.
8:28
So we get also here pretty low FDRs that are on par or even you know surpassing some competing technologies.
8:38
Sensitivity also done by comparing the counts that we get out of RNAsky versus single cell sequencing and what we see is the data shows that we are more sensitive than a single cell RNA sequencing here and finally to you know show the point that this is all done with sub cellular resolution.
8:57
So really when we zoom in, this is an analysed ROI here.
9:03
And what you can see is each dot here represents a detected transcript.
9:06
The white lines here are cell boundaries.
9:08
So we're really taking multiple transcripts for a cell.
9:11
And yeah, it's a full subcellular resolution.
9:18
Here's some more images showing different tumour microenvironments through the lens of RNAsky.
9:24
So we have breast cancer here, squamous cell carcinoma, colorectal cancer.
9:31
You can see really the unique tumour micro environment for each of these.
9:36
If we zoom in over here on the carcinoma in the middle, you can see just more detail here, different detections.
9:48
I'm not a cell biologist or pretending to be a cell biologist up here, so I won't go into too much detail of pretending to know what everything is, but you can definitely see strong spatial patterns.
10:01
And yeah, so definitely looking good.
10:07
And then before we get into the case study, also just to share that RNAsky is compatible with downstream H&E staining.
10:15
So the same tissue section that goes through multiomics on the Miltenyi workflow on the MACSima and also then be stained with H&E.
10:25
And that data can be then combined into our MACS IQ view software.
10:30
So you can combine the data and you know, do different visualisations and different overlays and analysis.
10:38
Shifting gears a little bit now to talk more about this case study that we have here.
10:44
So this was a study of colon cancer-associated fibroblasts led by a scientist in Germany, David Agorku, who collaborated with Emily Neil's team who sits here in Waltham.
10:58
One of my colleagues in the study, they sought to, you know, characterise this colon cancer micro tumour environment and developed a custom 48 flex RNA sky panel and combine that with a 45 flex protein panel.
11:17
A little bit about why the study was done.
11:19
So in general, cancer associated fibroblasts or calves are known to play a role in cancer prognosis.
11:28
They've been well characterised in a lot of cancer systems such as lung, breast or pancreatic cancer, but their effects and the role that they play in colorectal cancer has not been studied.
11:41
So that was a gap that was identified by the team at the time.
11:43
So that was really a motivation for looking into this, trying to find something new here with this multi omics workflow.
11:52
So to kick off the study, the team started with tumour tissue and normal tissue, took it through a tissue dissociation workflow, isolated the fibroblasts and did single cell RNA sequencing of the pre enriched fibroblasts.
12:08
And what they found was that there were in total 11 populations of fibroblasts that were enriched, but three of them were unique to the cancer sample.
12:19
So three of them fell into this calf category.
12:23
The rest of them were present in both in the normal and the tumour tissue.
12:30
So the team wanted to look closer at these novel CAFs and wanted to evaluate their interaction with T cells.
12:38
So they designed an in vitro assay here where they cultured these novel CAFs with T cells, then took the cultures and performed flow cytometry analysis on them.
12:50
And what they found here was that the CAF suppressed activated T cells in this in vitro study and that led basically to the next steps here.
13:00
So, seeing that these CAFs were inhibiting CT cells, they named these TinCAFs or T cell inhibiting CAFs.
13:08
And obviously it's now seeing this interaction in vitro.
13:11
They were particularly interested to see what this relationship looked like in situ.
13:17
So the challenge with doing this with just proteomics alone is that there's general lack of antibodies for T cell subtyping.
13:27
And also for all these different CAF populations, the protein expression is relatively broad.
13:33
So the sky panel came in to fill that gap.
13:37
And so they designed this custom 48 Plex RNA sky panel.
13:42
Part of it was targeting 4 different CAF populations.
13:46
So one control here.
13:49
And then the three unique population that were found in the cancerous tissue, including this novel TinCAF, they combine that with 45 Plex proteomics panel to really assess the immune cell dynamics into patients.
14:07
So here you can see again, this is really the raw data coming out of the workflow for the two patients in this study.
14:14
Overall, this approach is able to resolve over 40 cell populations.
14:20
You can see here stromal cells and structure in kind of the dark red and in particular the CAFs here.
14:27
So the RNA detection and cyan in both of these samples here.
14:36
So the spatial result are nice, but then of course they want to do quantification.
14:41
So they pulled all this data into our MACS IQ view software where you can, you know, characterise and divide cells according to, you know, really countless metrics.
14:52
They were able to quantify here the different types of immune cells and quantify the different CAF populations here 4 CAF populations quantified including this novel TinCAF population.
15:09
So then now we're looking at actually process data.
15:12
So this is I believe the last data slide that I have.
15:18
So yeah, now looking at how these TinCAF and T cells were kind of localised within the tissue.
15:25
And you know, reflecting back on the in vitro study, what they did find eventually was that the TinCAFs shown here in cyan where in fact kind of spatially Co localised with, you know, immunosuppressive tumour promoting and exhausted T cells.
15:41
And I know that David who ran the study in Germany was extremely excited about this result.
15:52
I think that's kind of the end of the talk here.
15:55
So I think just to kind of reiterate, we have this same section multiomics workflow available on the MACSima RNA sky, you know, gives you everything all the way from sample preparation and support kits we have the IO Explore panel currently available.
16:13
We are also now starting to offer custom panels for customers with really any target we offered in two formats, 12 Plex and 24 Plex.
16:23
And it's really end to end workflow support.
16:26
So all the way from sample prep through you know true multiomics data analysis in our MACS iQ View Software at the end of the talk.
16:38
Yeah, thank you everyone.
16:40
Any questions I have to answer them.
