0:00
Thank you.
0:02
Good afternoon everyone.
0:03
So I am Claire Huguent, I'm the head of biomarker Services at the Randox Biosciences.
0:08
My talk today is going to be how to channel best basically the discovery biomarkers into an IVD and what's the pitfalls might be along the journey.
0:21
So to start with a very short company overview, I think that Randox is well known to you because of the exposure we've got during the COVID pandemic.
0:30
However, we were founded in 1982 in the Antrim.
0:36
So basically in Northern England.
0:38
We are an IVD manufacturer.
0:40
We are about 3500 employees now and about 1500 scientists in the company.
0:48
5% of the world's populations are diagnosed using Randox products of some sorts and basically we sell those products under our own brand or many others that you may understand.
1:01
So we supply 12% of the world cholesterol test even if probably not less than 10% would be sold under the Randox brand.
1:11
We OEM for a number of companies globally 134 and we manufacture 4.5 billion tests a year on average.
1:23
So Randox has always been very interested in multiplexing and that was the approach that was taken nearly 20 years ago to generate some sort of early stages of discovery for biomarkers for our own sake in the first place and then potentially for developing assays in the, for any kinds of partnerships.
1:45
This is a proprietary technology, which is a ceramic tile.
1:49
You can see the size with the hand picture, which is activated with the chemistry layer.
1:56
And this, the nature of the chemistry depends on the nature of the ligand.
2:00
So we can liaise pretty much any kind of proteins can either of antigens, DNAs, RNAs, drugs and so on.
2:09
Initially the size was limited to 5 by 5 display.
2:14
So we had 25 spots which enabled to test 22 plexes at a time.
2:19
Today we are, we have a 7 by 7 display which leaves 49 spots out of which we still need 3-4 controls that is for proteins.
2:30
And then we can do a 10 by 10, so hundreds of biomarkers for RNAs, DNAs out of which we still need to get 3 spots for controls.
2:40
So out of these basically these technologies being used for fully quantitative assays.
2:46
It is being used today for diagnostics as well as for research.
2:49
We have a number of customised panels which are being used for essentially for my research and the beauty of these technologies that it cannot be operated in very small batches.
3:01
However, obviously for proteomics that is certainly not the tool to go for this.
3:09
The discovery panels we might have at the time was limited to a maximum of a hundreds, promising markers would be restricted to a dozen.
3:19
Maybe the biomarker signature that we would get across would be below the number of five.
3:26
So it was very much obvious we had to scale up and go for somebody else that would be able to multiplex at a much higher level.
3:35
Basically having discovery panels in the thousands, getting hundreds of promising biomarkers, getting biomarker signature that would be in the number of you know dozens 20th and then select a biomarker that then could be taken up to no back to the biochip or to another assay format so that they could be developed into the appropriate IVD format.
3:58
So to do so, we decided to partner with Olink, and I don't believe we need to, I need to explain who Olink are basically, but we have been implementing their full suits about a year ago when we have this ongoing now for research purposes as well with pharma collaborations.
4:16
Basically, we choose them because they were an answer that to the pitfalls we had basically identified in traditional proteomics methods.
4:25
They come with a very high throughput at a reasonable cost for biomarker.
4:31
There is a very high sample volume limitation, so you can run basically 5000 biomarkers from a maximum of 55 microliters.
4:41
You have a very good sensitivity, a very wide dynamic range and exquisite specificity for what we have been able to test so far.
4:52
So the full suits of the Explorers panels, the that's Explorer target and the most customers panel in the Explorer suite.
5:02
Basically the HT gives you access to 5000 proteins, the 3072 and then the Explore 384.
5:13
The bricks basically component composing the 3K panel.
5:18
That's eight panels that can be operated simultaneously or as individual panels.
5:24
They are basically the readout is next generation sequencing.
5:29
They are not fully quantitative.
5:31
They are semi quantitative with an NGS and then and NPX reporting ratio, the Olink target and 48 they are quantitative.
5:42
They are qPCR reads.
5:44
They are restricted in the number of proteins they can read.
5:47
So that's basically a few hundreds.
5:50
And then the flex and focus panels, they are custom panel that can be developed on purpose for any kind of specific design.
5:58
And they are up to 21 biomarkers which are fully analysed quantitatively and can go down to one single Plex.
6:06
So of course this technology comes in the number of steps you need to operate it in labs which are fully equipped with a variety of different types of instrumentation.
6:18
One single run is spread across 2 days and that requires specifically significant skill sets in your labs.
6:27
So you need to make sure that what you're reporting matches what was initially reported by the Uppsala lab in Olink.
6:36
Very recently on top of the training that they provide to every lab, they have instituted a concordance testing and we have been private to becoming one of the beta testers.
6:46
And basically this concordance testing, it is the split sample analysis of 50 samples between the Upsala reference lab and the Randox labs or any lab that would be coming to this concordance certification process.
6:59
So they are looking at the number of calls you get, your sensitivity, the correlation with your own data.
7:05
The full report is provided to you upon submission of the data.
7:09
Basically what it is really is a proficiency testing applied to proteomics, which is very unique and certainly the first of its kind.
7:20
We came certified from the Explore panel in January.
7:25
Certainly we are looking at getting the same types of certification for targets we have submitted.
7:32
We're waiting for the results.
7:34
But more importantly, we want to use this certification further on as a transition towards the next steps for validation.
7:42
And so why these concordance matter specifically, you do not always have the luxury to be revalidating 3000 biomarkers in a row.
7:52
And if you have to do so, the validation costs are absolutely huge.
7:58
So basically you basically use this concordance testing to qualify that your lab is operating.
8:04
And of course the qualification comes with each of your techs operating.
8:08
So you should have assurance that each of the lab’s techs have been certified.
8:14
You basically also become available to use the validation data that have been published and generated initially by Olink, which is limiting the need to revalidate those panels right away.
8:27
Specifically for the non-quantitative assays of the very large panels.
8:33
You can use this to harmonise basically many labs contributing to these to the same cohort.
8:39
And that's very precious, specifically if you want to be bridging samples with countries that do not would not export samples such as China, for example.
8:49
And then this overall facilitates the transition between your highly complex initial high panel to a single plex or a little plex, assay.
9:03
So getting back to how this links into our CDx capabilities, I would say CDx or any kind of IVD development capabilities.
9:12
If you want to be successful in this pathway.
9:14
Then basically bringing your attention the on the early days and how the identification of the biomarker has been, you know initially absolutely matters because from the high multiplex technology we have used initially to get you know your biomarker signature, then you need to derive your biomarker selection.
9:35
You then get an RUO that might be Randox developing the RUO that can be somebody's assay we inherit that can be a by analytical assay that can be an in house assay of some sort.
9:48
And this RUO basically for lack of labelling from a regulatory standpoint is not something which is fully validated.
9:58
Then to run in a clinical trial assay CTA, you need to have your optimization stage locked down.
10:07
And this is essential because you will be basically measured down the road with a bridging study between your CTA, which is basically supporting your drugs final efficacy performance and the final IVD that might be used as a CDx and the bridging studies depending of how closely your CTA matches the IVD criteria.
10:31
I will come back to the optimization stage, which is absolutely essential.
10:34
But then you get to scale up, do your clinical validation, making sure that your clinical decision points are within the acceptance criteria that you may have not been able to set yourself.
10:47
Sometime the regulatory authorities set this for yourself and then you have to abide with the corresponding challenges.
10:56
And obviously once everything is final from a technical standpoint, you still need to register everything, and this registration effort starts years before the assay is finished. And supporting the logistics of the assay, but also potentially supporting the logistics of the sampling specific in rare disease matters and bringing in sample collection kits and registering the sample collection kits in the local corresponding locations.
11:23
Also falls within the scope of what we can do.
11:28
And then I will not expand on that.
11:30
But we have developed the full sample traceability portal which is also something we are available for licencing.
11:39
Basically, we have labs which are ISO 13485 accredited.
11:43
We develop of the shelf assays, which are our own assays.
11:49
We can develop on a variety of formats whichever types of technology would be available because we OEM on those platforms and we can also produce our antibodies and proteins that we use widely for QC and calibrators developments.
12:05
So what is it that we bring typically between the CTA and the final CDx?
12:12
Well, your CTA assay brings you know it’s usually has been developed quickly and it might be coming from a high multiplex assay, or it can be by analytical.
12:23
Usually comes in multiple components which are not kitted means it cannot be operated commercially.
12:30
The elements are manually prepared on the day of use which requires additional skill sets and also prevents the widespread of the corresponding technique.
12:40
It is not registered and it's use is restricted to the lab which has been developing the corresponding test.
12:47
The CDx or the IVD basically has to be kitted.
12:52
All the critical elements must have been thought through to be ready for use.
12:56
So they come in the appropriate dilution that the reagents have to be stable for at least 12 months to guarantee effectiveness in the distribution.
13:05
It has to be registered in every relevant country.
13:09
It means it has to be scaled up in terms of production depending of your study population and obviously it has to be made available to as many labs as needed in the world for the diagnostics to be used. Stemming from the CTA to the IVD means you have to make early decisions.
13:32
And this is where basically you inherit the test, which is merely a research use assay, and you have to make early decisions with pretty much no visibility in what you have to do downstream.
13:44
So specifically you need to decide on whether your matrix, the one you inherit for your assay is kept.
13:52
Specifically, do you keep an a matrix, a serum, plasma, urine matrix for your QCs and calibrators?
13:59
Or you do move to alternative options.
14:03
Do how many sets of QCs do you want to get?
14:06
How many calibrators do you want to get?
14:08
This is impacting your throughputs because obviously the more calibrators, the more precise your curve, but the less samples you can run.
14:17
Your dilution scheme has to be established at the time and whether you want to go for automation.
14:22
All of this is very important.
14:25
You need to make this at the time you have many more many unknowns, which was resource commitment site, the time allocations and course.
14:34
I know I have 5 minutes only, so I need to accelerate a little bit.
14:38
But basically making the decision of the critical components, making sure that the clinical samples are available to run your course correlation and that you have everything in our control with the project management system is essential.
14:55
You need to get your biomarker, your early biomarker assays as closely as possible to what you want to get at the end.
15:04
So that's why getting to a highly validated system such as the one we inherit now from Olink makes sense.
15:12
You need to make sure that you have as much information upstream to avoid duplicating your experiments and you need to make sure that you have secured your IP on critical reagents.
15:24
You cannot dwell on inheriting an assay that is meant you know to be running on the polyclonal.
15:32
You need to make sure that your monoclonal, if it's not your own production, is still available and available worldwide.
15:39
There is no local distribution that might be secured otherwise and so on.
15:47
You need to secure the appropriate biomarker selection and do the early validation.
15:53
Obviously the choice on which platform you operate from is depending on your population.
15:59
I said that you could use a multiplexed assay initially and down to a single Plex.
16:04
It's an overkill to be using a high Plex system for a single Plex assay.
16:08
It is still nonetheless doable and desirable sometimes when you want to fully automated the assay and not change the format and keep your data on.
16:17
And that is something that we do for point of care applications, for example.
16:22
So all of these can be centralised in lab services.
16:24
Obviously, we can also ensure the full regulatory registration of those.
16:30
And you know, many countries, not only the ones listed here, Brazil, Japan or in the list as well.
16:36
And definitely we certainly support commercialization by helping with market awareness, education of the prescribers, building a dedicated IT platform distribution in the countries and making sure that we ensure the appropriate raw material developments as needed.
16:57
OK, I hope that you get a sense of what we can do.
17:02
Basically we if there is any need for basically assay development, I just give me a ring and we'll try to see what we can do for you.
17:12
Thank you.