0:00
Good morning everyone, I hope you're all sufficiently caffeinated after the last break.
0:05
Great to see so many faces in here, some of which are familiar, and I've spoken to over the last few days, but some of which are new, so I hope you learn a little bit more about us during this session.
0:15
By background, I worked at MedImmune slash AZ as an IHC scientist and actually my old boss is in the room right now, not to embarrass him or anything.
0:27
So hopefully I do justice, Arthur.
0:30
And then I moved into the commercial field.
0:31
I did a number of years Leica Biosystems working in pathology there and have recently in the last year joined diagnostic analytics where I'm helping establish our pathology services model.
0:44
And that's what we're going to be talking to you a little bit about today.
0:48
So to set the scene and you'll see this narrative occurring a few times throughout the presentation.
0:55
But I think that this analogy that I'll go into really quite describes what we're seeing in the space pathology can be a little bit of a storm and translational research and clinical trials is a bit of a storm in itself.
1:09
We've got storm clouds, you've got wind, you've got rain, you've got a mixture of different endpoints, different protocols, you've got different indications, different programmes all going on at the same time.
1:21
And most organisations have limited resource to deal with all of these things.
1:27
And the ones that do have the resource tend to find bottlenecks somewhere.
1:31
So throughout this presentation, I'll be referring back to this analogy of this storm in a teacup and very much how pathology can help guide the way in some of these instances.
1:40
So the reality today and the things that we're seeing crop up time and time again is within translational research and clinical trials specifically related to pathology, we're seeing multiple sites.
1:52
So that's study sites also research sites and global organisations, multiple vendors.
1:59
So you've got Central Labs, CROs, you've got hospital sites, you've got people like ourselves, and then you've got a number of different internal stakeholders too.
2:09
There's a burden of oversight on the CROs.
2:12
So everyone working with CROs and central labs generally always very happy with the stuff that comes out of that.
2:18
But someone needs to be checking the QC of all of those things that come through.
2:23
And that leads to the next point, which is sort of an inconsistent QC.
2:26
So some of the clients we've been talking to are telling us that the things that are coming through, the staining looks good one day, maybe not the next day.
2:34
And that could be from an internal source, an external source.
2:37
And there needs to be a ground truth regional variability.
2:42
And this is both from a lab perspective, but also from a regulatory perspective.
2:46
So for clinical trials that span globally, you've got different regulatory bodies and different variations of different things that you need to be addressing all at the same time and incorporating that all into the study plans and the protocols.
3:01
On top of that.
3:02
Another layer is multiple endpoints.
3:04
So you may be looking at patient enrolment, then you've got pre/post treatment biopsies, a whole bunch of different things going on at once, a whole bunch of different tissue sections or tissue biopsies coming at different time points and the need to manage that process.
3:18
We then layer in another element, which is the lack of pathologists.
3:22
So this is something very well known within the NHS in the clinical space, but the number of pathologists that are working in the field at the moment is dwindling.
3:31
So when you've got all of these needs and you need specialist pathologists and there's not many of them, it's a storm in a teacup.
3:39
And then the final layer to that is everyone wants to achieve more data with less sample.
3:45
So I'm having lots of conversations with people at the moment about we've only got one or two tissue sections, what can we do with it?
3:51
We want to get as much information out of that as possible.
3:53
What's going on with the immune cells?
3:55
Where's the fibrosis?
3:56
Can we quantify this?
3:57
Can we quantify that?
3:58
Everyone wants more information out of less and when you've got all of these other factors at play at the same time, there's a lot to consider.
4:08
And the impact of that, and this is not exclusively pathology feeding into the stats on the right hand side of the slide, but pathology is a factor that impacts this.
4:15
So 80 to 90% of translational programmes never make it to a clinical trial.
4:21
So we've got quite a high failure rate and then of those that do make it into trials, 85 to 90% of those never make it as a drug on the market.
4:30
So we're in a very high stakes, high pressure environment.
4:33
Everyone wants these compounds to get to patients and to help make a difference with diseases at the end.
4:39
But there's a whole bunch of things from a pathology aspect that go into that and from multiple other aspects, pathology being a small part of that.
4:47
So what are we hearing first hand?
4:49
So marketing have very nicely given me some storm clouds to follow my narrative.
4:54
But these are some direct quotes that we've heard over recent weeks.
4:58
So we don't have enough in house pathology expertise.
5:02
We may have clients that they have two or three pathologists that work with them, but they're not experts in the tissue type or the therapeutic area.
5:11
So they're doing what they can with what they've got, but they can't drill into the detail enough to answer some of the scientific questions that are coming up.
5:19
Next up, we don't want to micromanage our CROs, like our CROs do a really great job.
5:24
We need to make sure the QC's there, but we don't want to micromanage them.
5:27
How do we handle that?
5:29
Also, our CRO is fantastic, but they don't have the specific disease experience.
5:33
This is something that we've heard quite a few times and I'll come on to later, like how we can help overcome that.
5:39
But specific like specifically like the big CROs don't always have the niche experience.
5:46
Particularly in rare disease indications, endpoints aren't consistently scored.
5:51
So particularly where you've got multiple pathologists working on long trials over a number of years, as many of you may know, pathologists don't always agree, and they don't always have the same scoring methodology.
6:06
Despite the amount of training that they get on the scoring methodology, there's always variances.
6:11
So endpoints aren't always consistently scored, particularly over these long term studies.
6:17
Likewise, we have a quality management issue.
6:19
So we've seen this with a couple of clients with in-house staining across different geographies.
6:25
There's variability and the quality management is a problem for them, particularly when there's only two or three tissue sections available and every stain matters.
6:35
And finally, this is one that we get quite a lot, but our pathology protocol is suboptimal.
6:40
Can you help us fix it?
6:41
We get this quite a lot, particularly at a later stage when something started to go wrong on the trial with the pathology outputs and then we're almost parachuted in to try and help fix it.
6:51
We prefer to be involved more upfront, but these are the types of problems that we're hearing time and time again from some of our clients.
6:59
So how can expert pathology help unlock some of this clarity and help with this storm and what's going on?
7:07
Endpoint definition of feasibility is super important.
7:10
So the upstream part, before we even start thinking about what we're doing with the tissue downstream, it's super important that we define that and we build that in the first instance.
7:20
And whether that's done by an organisation like us or internally or a combination of the two, this is really critical that moves into the protocol review and the manual development.
7:30
So we work in a space where we develop these pathology manuals.
7:35
So every stakeholder on a project is super informed on what needs to happen at what point and what that looks like. Really crucial step that we've found a number of biotech and pharma companies are missing or the detail is there and the pathology protocols there, but it doesn't account for any issues that can arise.
7:57
As we all know, things derail very quickly and if there aren’t those contingency plans in place, there's going to be problems.
8:04
Embedded pathology leadership, and this is something that I'll come on to again later.
8:07
It's really important to get across that this is exactly what we're trying to offer as a service but embedding expert pathologists with the relevant tissue type and disease experience within study teams so that all of these decisions that are made by people that are best informed and able to support with that trial and try and support its success later downstream.
8:30
It goes without saying but slide review and consensus review is a really important part of pathology and really important part of translational programmes and clinical trials.
8:39
It's kind of the bread and butter of pathologists’ work is looking at the slides and doing the analysis and what's going on in the scoring, but also the consensus part being really important.
8:49
So I mentioned earlier that pathologists don't always agree with each other and there's usually particularly in rare disease indications, some variance in scoring methodology.
9:00
So providing consensus review super important.
9:03
So whether that's a two or three pathologist read, so like dual read models with an adjudication process in place, if there's ever any discordant cases.
9:13
And finally having oversight of CROs and labs, whether that's internal or external, but providing that QC and being that bridge between the sponsor and the CRO to ensure that everything that goes through is relevant to the study and is accurate.
9:31
So our mission, coming back to my storm analogy, we want to be the lighthouse in that storm.
9:37
And for a number of our existing clients, we are the lighthouse in the storm.
9:41
So we embed directly into the workflows that are already existing within the teams.
9:47
So that might be in a precision medicine group, that might be in a clinical biomarker team, that might be in a companion diagnostic development group, a mixture of all of them, but also within the CROs and central labs and the study sites.
10:00
So we've become that embedded pathology leader with the point of contact of what's this, what do I do with that?
10:06
This has gone wrong; I haven't got enough tissue here.
10:08
How does that look like?
10:09
We become that central service and whatever the gap or shortfall may be, we provide a service to fit that.
10:17
So we like to go with the term of bespoke services.
10:21
I think everyone kind of uses that as a tagline at the moment, but what we really truly do is get to the pain points of our clients and figure out what the gaps are and where the shortfalls are.
10:31
And we help fill those gaps.
10:32
Whether that be data scientists to analyse multimodal data sets, whether that's AI engineers to build algorithms bespoke to a clinical or a translational programme so that the downstream CDX can be manufactured and developed much faster because the AI is done upfront.
10:51
All of these different things.
10:52
When there's a gap, we can address it.
10:55
And finally, it's a completely flexible engagement model.
10:58
So I like to say you can spin us up and spin us down as and when you need us.
11:03
A number of our clients will use this for multiple projects at any given time.
11:08
But we also get the weird and wonderful requests of we're really not sure what to do with this.
11:12
We only want to work on this project.
11:13
What can you do?
11:15
We spin up a service.
11:16
We bring on board the pathologists and the scientists that are required.
11:19
We deliver the body of work and then we can scale it back down.
11:22
There's no commitment or maximum, minimum requirements of working with an organisation like ours.
11:28
And this brings me on to the importance of this integration piece.
11:32
I've mentioned a few times this embedded pathology leadership and really it's about becoming that integration piece that's critical for success.
11:43
So the diagram on the screen is a massive oversimplification of what it actually looks like in practise.
11:49
But essentially what it means is we can become that overarching support for all of the downstream labs and applications of the services.
11:58
So whether that's the CROs, the central labs, the test sites, we can produce pathology guided documentation on best practise for taking biopsies for the clinical trials so that we know we're getting enough tissue in the right area so that the biopsy is meaningful.
12:15
It saves the patients from undergoing multiple biopsies to get the correct data sets out for the trials, things like that.
12:21
But really what this means is there's full project management oversight of what's going on.
12:26
The sponsor can have, if chosen one single point of contact for all pathology related matters, everything's managed under one roof. Regular touch points with a single point of contact so that any question from the study team or the PI at any given point can be addressed very quickly.
12:44
That embedded pathology resource, essentially we're seen as an extension of pathology teams, whether they're existing already or an organisation doesn't have a pathology group.
12:54
We can become that like a virtual pathology department.
12:57
It gives confidence in the pathology outputs.
13:01
What we're really strong on is providing quality pathological insights into what's going on.
13:06
It's really critical and it's really important.
13:10
And also touching on regulations, but everything needs to be regulator ready, whether that's FDA, MHRA, whoever, everything needs to be GCP audit ready and ready to go so that when a clinical trial has been successful, these primary, secondary endpoints are ready to go.
13:26
The data sets there and ready to be audited at any given moment.
13:31
So what does this mean for sponsors?
13:34
It means that you get clear endpoints with clear data sets, reduced variability.
13:39
So whether we're implementing AI image analysis, pathologist reads, dual pathologist reads the reduction in variability is quite strong.
13:50
The data is regulator ready.
13:52
As I've mentioned, there's fewer deviations from study plans and translational programmes because we do the upfront work on understanding what the scientific questions are, and we build the pathology protocol around that and also faster timelines.
14:07
So I mentioned that we embed ourselves in the team, but we do have to give some context 250 pathologists that work with us globally.
14:17
And because of that, the turnaround of these data sets is incredibly fast.
14:21
So whether that is a 24 hour report turn around for patient enrolment or whether that's just the standard pretreatment biopsy and you need the reports back within five days, we can adhere to that and we can make sure that the timelines are met at every given moment.
14:38
So I thought I would bring in how this actually works in practise and what this means.
14:43
And considering I'm not under NDA with most of the people in this room, it's fairly anonymized.
14:49
So there's not a huge amount of detail in here, but it does go into some detail on a study that we've been working on.
14:55
So we had a discussion with a client on with regards to a phase 2b trial.
15:02
This was in an autoimmune disease, and they said to us, we only know of 1 pathologist globally that has experience in this rare disease, and we don't know what exactly we're doing.
15:12
Our CRO have been brilliant, but they also don't know too much about the disease type.
15:17
They can do all the staining, but they don't know what staining to do for us and we're a little bit lost.
15:21
Oh, and by the way, we want to enrol patients in three months’ time.
15:25
So it was a little bit of a high pressure situation.
15:28
But what we did essentially, and there's some info on the left that you can have a read of.
15:33
But our approach to this was let's sit down and understand exactly what it is that you require.
15:38
Let's build the pathology protocol around the scoring methodology that's needed and let's define that first.
15:46
Then once we know that we can figure out exactly how many pathologists you need, what that looks like and how we can support with that.
15:53
And of our pathologist network, we actually had five that were experts in the disease area that were required globally.
16:01
So our sponsor immediately had access to five more pathologists that they didn't have before the conversation with us.
16:08
We then coordinated all of the specimen logistics for this specific for the specific trial and ensured secure digital review.
16:17
So I've mentioned our pathologist network being global and there being 250 of them.
16:22
The reason we can still keep the turnaround time very quick is because we do everything digitally.
16:27
So the central labs and the CROs that were on this study would scan all of the slides in, would send securely all of those images to us and we would use our platform to push them out to the pathologist for the reporting.
16:41
That process is almost instantaneous.
16:43
Those images go through AI QC on our side and then they go to the pathologist for review.
16:48
So it keeps that turn around super quick, keeps things moving fast.
16:53
We implemented a dual reader model on this study, so we were very aware that it's a rare disease.
16:59
The scoring method that was chosen had some interesting nuances about it.
17:04
So the sponsor was quite keen that we had more than one pathologist read it.
17:08
So this model was 2 pathologists read in the first instance.
17:12
If the score comes out the same, great.
17:14
If it didn't, we would then host an adjudication meeting with the PI and then the two pathologists would be able to see what each other have scored.
17:22
They discuss it.
17:23
If they can reach agreement, great, if not, then we bring in a third pathologist.
17:28
And that process worked very well.
17:31
And then finally on that point, all of the pathologists on the study, including backup pathologists, which is quite important in case someone gets sick or there's someone that can't support with the turnaround times for any given time.
17:44
We provided full training and calibration for all of those pathologists on the study.
17:48
So everyone was aware of what was coming, when it was coming, what the nuances of the disease was, what the sponsor was trying to achieve, what the endpoints were supposed to be.
17:57
They had the full training prior and then everyone was briefed.
18:01
So if anyone dropped out for any given reason, there was someone to step back in immediately.
18:06
And the impact of this for the sponsor was huge.
18:09
So they got a standardised audit ready workflow ready to go.
18:14
We reduced the variability in scoring.
18:17
So there were no cases in the end that we couldn't reach consensus with less than with more than three pathologists.
18:24
We never needed an additional work to go to failsafe mode, and it freed the sponsor from the operational burden of the regulator ready output.
18:34
So we provide everything in a fully holistic report style, and everything is archived and ready for audit and review by the FDA.
18:46
So to conclude and to sort of bring everything back together with my nice storm and lighthouse analogy, Diagnexia Analytix is a pathology partner.
18:57
We work with any CRO, any central lab, any sponsor, biotech, you name it, completely agnostic in that sense.
19:05
We embed ourselves as pathology leaders to support and guide and work with study teams.
19:12
We provide access to our network of pathologists.
19:15
We develop, design and implement the pathology manuals and any other central lab facing guidance that's required.
19:22
We provide central reporting and slide review, and we can support with external quality assurance and assay validation support as well.
19:30
So even where labs are building assays in translational phases for trials, we can support with that for specificity and ensure that the staining is correct before it goes any further.
19:42
And finally, a nice point, given the exhibit hall downstairs being full of really cool spatial biology companies, we're also able to correlate the pathology data with spatial multiomic datasets too including things like liquid biopsies.
19:59
So where there's an interest in seeing whether what's going on in the tissue matches what's going on in the blood or any of these really cool spatial techniques, and you want to know exactly where it is in the tissue and have that pathological interpretation of that, we can pull all of that data set together as well.
20:16
So to conclude, I'm not saying that we're going to change the weather and stop the storms from coming.
20:21
I don't think anyone's got a magic wand to do that, unfortunately.
20:25
But what I'm saying is pathology doesn't have to be a weak link in that regard.
20:29
Pathology can be a strength and really what we do here is act as the lighthouse guiding sponsors on their ships to harbour safely.
20:39
That's the idea anyway.
20:41
So we won't stop the choppy water, but we do try our very best to guide you into harbour.
20:47
So thank you for that very much.
20:49
I hope you found that interesting.
20:51
Please do visit our booth if you've got any questions or if we've got time now, I'll be happy to take anything.
20:57
So yeah, thank you all for your time and I hope that was useful and interesting for you guys.