0:00
Hi everybody.
0:01
We'll introduce our speaker here, Doctor Elena Gontarz.
0:06
She's the Head of Scientific and Technical Affairs at Thermo Fisher Scientific.
0:10
She holds her PhD in biochemistry and has over 10 years of experience in developing biologics for pharma companies.
0:19
Doctor Gontarz is Thermo Fisher Scientific's resident expert in global analytical and process development as well as formulation sciences, and currently she is the head of the Scientific and Technical Affairs team leading the CMC support and biologics programme from phase one to commercial manufacturing.
0:40
So with that, I will hand it over to Doctor Gontarz.
0:44
Thank you for introduction and thank you everyone for coming today.
0:48
I do tend to move a little bit.
0:49
So hopefully it's not going to stop any of the you hearing me, but if something raise the hand and thank you for coming in this early morning and being on this conference and I would like to dive in.
1:06
So when the Thermo Fisher were think was thinking last year about accelerating the biologics, we were thinking about all of the challenges that usually happen during the biologics development and manufacturing.
1:19
So the first thing that we wanted to look into is actual selling development because that's the first step that we encounter during the projects in our shop from the clients.
1:33
Because while Thermo Fisher and everybody knows Thermo Fisher as the actual chemical provider or bioreactor provider or analytical instrumentation provider, it has also contractual manufacturing and CRO services as well.
1:49
So I come from the CDMO portion of the Thermo Fisher.
1:53
So the first thing that we actually do is cell line development and we've had a couple of challenges for the cell line development.
2:02
One of them is that the actual cell line didn't produce enough titer for us to enable our clients to have toxicology studies as well as first clinical studies and also for us to even develop downstream process.
2:17
So for us, we had to run higher scale of bioreactors to get the appropriate amount of material to the clients and for our own process development.
2:28
The other part that also stood out to us is that the stable pool is not quite representative in our currents aligned to comparing to the final clone.
2:38
So we wanted to attack that challenge as well when we were looking at the overall process as well as the last thing that we overall had to look at is that what kind of processes we go through to develop the actual drugs as well as how do we stagger them and how do we approach overall path to IND.
3:02
So while we were discussing it internally, we realised that we actually have to go to external partner to get the new technology in the house for cell line development.
3:14
And thankfully, we found a partner that would provide us with machine learning and AI empowered cut down optimization vector construction as well as we partnered with them because they had transposase technology and CHO-K1 cell line.
3:32
And all of those parts are very crucial to ensure that you have high titers and representative stable pools.
3:39
And the last thing that we have done, we already had beacon technology internally.
3:43
But if you don't have really good cell line, that beacon technology is powerful, but not as powerful.
3:49
So when we brought the new transportation technology with CHO-K1 cell line, we actually saw and we will see in the next slides that Beacon actually enables us to really find the best clones that are there in the very homogeneous already stable pools and provide that high titer to the clients as well as the appropriate product quality.
4:14
With that, I would like to look over the transposase technology, which I hope everybody in the room already knows.
4:21
It's a buzzword for the last couple of years, but transposase technology allows us to move large genes of interest from the vector to the actual gene of the host and it moves it without any kind of issues, fragmentations of the DNA.
4:39
It's really just copy and pasted from the vector to the host and what it enables it is the again high titers, which that's what we are looking for as well as representative product quality.
4:53
And while it's all good and we understood the track record of that the company, we also wanted to make sure that in our actual hands we're able to reproduce their results.
5:03
So we got a couple of clones from them as well as developed our couple of clones on IgG-1s and IgG-4s.
5:11
And this is just snippet with Herceptin example where we compared their couple clones with ours and we saw that during the actual beacon empowered cell and development we do get better clones that actually produce higher titers.
5:31
And again the product quality was a representative.
5:37
So that is what technology we wanted to make sure we're bringing in to empower the path to IND.
5:44
The other parts that we actually thought of even though we have path to iND which is transaction of the final candidate that the clients have all the way to IND.
5:53
We also thought our clients comes come to us with multiple candidates and they need a little bit more data to finalise which candidate they want to go with.
6:04
And that comes from the fact that each different candidate can have different issues with the aggregation or fragmentation or oxidation, deamination, glycation.
6:17
And so all of those attack issues can come up and result in a loss of potency or can result in immune response from human beings.
6:29
And besides all of those kind of hard, harder and questions that we need to think about, they also result in a shorter timeline shelf life if those issues occur during the actual storage or handling of the samples.
6:47
So we actually wanted to look at all of those attributes during the assessment very early on before we even dive in into the downstream development, for example, for manufacturing.
7:01
So what we've done is that we are proposing and to our clients the use of our cell line.
7:10
Let's look how this cell line produces your molecules, 4 different candidates, 8 different candidates, 12 different candidates and what kind of candidates have what kind of post translational modifications those candidates have.
7:26
As well as look at if we're looking at the purification, is there anything that we can predict that will happen during purification that we can mitigate right now?
7:40
As well as we will look at formulatability of the molecule, what kind of force degraded things we can apply to it and see what kind of liabilities the actual molecule has.
7:52
So with all of the data that comes out of our labs as well as data from the client will allow client to actually narrow down to one final candidate and that will give them the head start to the actual IND.
8:11
So there I gave you kind of the high level example of what we would do.
8:17
But to dive into actual details, what we do and why this workflow is representative of what will happen at your GMP is because when we actually do the cell and development, we will learn stable pools in the actual AMBR 250.
8:34
Our AMBR 250 is scaled down model for all of our scales at GMP as well as we're going to apply our option platform to that AMBR 250 that we will actually keep for your GMP as well.
8:46
So everything that you will see in AMBR 250 with your stable pools should be represented what you were going to see in the actual GMP.
8:54
And I understand product quality will be a representative.
8:57
Titers will not be quite representative because between the stable pool and the final clone, the titers usually increase 2, sometimes three times.
9:05
But the process and how cells behave should with your molecule should be representative of the final process.
9:15
The other part that we look into the purification if the molecule can actually handle low pH and that's more important for more complex molecules because if you are having more complex molecules, a lot of them need detergent VI.
9:31
And we want to know this as early as possible so that we don't have to figure it out during the actual downstream development.
9:41
We know it ahead of time.
9:42
And then during the downstream development, all we have to do is or check couple of parameters for a low pH VI or we actually apply the detergent.
9:51
The one thing that we do only for more complex molecules bispecific FC fusions.
9:55
And we will screen the polishing step resonance just to make sure that we actually ahead of time can have what kind of polishing steps we need to apply to your molecule, especially if it's a heterodimer.
10:09
The other part.
10:09
So that was the process.
10:11
Now we're talking about analytics.
10:12
My favourite part, sorry, it's because I'm come from analytical development formulation sciences.
10:17
But this is my favourite part is because we love our Prometheus, which is high throughput screening system.
10:25
And we wanted to make sure that we're and understand how the molecule behaves during the thermal degradation as well as how it interacts with it with its own molecule.
10:37
So what we've done is that when you have multiple candidates, you will actually screen those candidates for formulation.
10:45
And it's very fast screening.
10:47
It's not your final formulation, it's just to see how what kind of formulation will stabilise that specific candidate.
10:54
So that would be the first step, pre-formulation development.
10:57
Pre-formulation development, sorry.
11:00
And then we will put that specific candidate in its ideal formulation to look at the candidate screening.
11:07
So in that ideal formulation for that specific candidate, how the actual candidates compare to each other.
11:15
And then when we are finalised on the final candidate, we will put the final candidate through liability profiling, which is forced degradation.
11:23
It's a little bit lighter forced degradation than BLA enabling just to empower you to have.
11:29
There are some data on the abilities, but it still gives you quite a lot of understanding what's happening with your molecule under pressure.
11:37
And the other part that's, so that is the actual looking at liabilities.
11:44
The analytical part that we'll look at is we're going to implement the multi attribute method and we implement that because it actually gives us understanding of post translational modifications on the amino acid level.
11:57
And we do that during cell and development, and we do it also during the actual formulate ability assessment as well.
12:05
So again, why do we do this is because we want to make sure that if we have any kind of degradation happening and it could be thermal, can be pH, freeze/thaw, light exposure, any kind of oxidation, what how will the molecule handle itself, how it would degrade and how we can actually translate it to patient mishandling the samples, shipping, handling, packaging, how the samples behave during that as well as downstream development as well and downstream purification.
12:41
And with that is kind of the overview how we wanted to start a little bit earlier with the client and how we will speed up their overall path to IND.
12:53
So that is the first portion of it and that's what we can help with.
12:58
But also we have clients who come to us and they already have that because they already have their own development labs and they have the final candidate and all they need to do is for us to develop their process and do the GMP manufacturing.
13:12
And the ask was how can you do it as fast as possible?
13:16
That was the ask from the clients.
13:18
And we put all of our heads together across all of the different regions of the world because we have manufacturing across the world as well as process development and sat down and had quite a lot of discussions throughout the all 2024.
13:36
How can we actually look at the timelines and the new technologies that we already bought and enable the faster timeline.
13:44
So with that's the first time that we are presenting on the nine month timelines that we get from transfection all the way to patient.
13:55
And I want to emphasise that's to patient and I'll show you why as well as 11 month timeline for clients that are not as open to as much risk.
14:06
And we understand that.
14:07
So that's why we have two options as well as 13 months for bispecifics and FC fusions.
14:15
And all of this is enabled because this nine months, 11 and 13 months because we actually templated quite a bit of the process and we have all of documentation ready for the actual project.
14:28
And this is the overall timeline for the nine months I'm going to.
14:33
And if you want to learn a little bit more about the 11 month timeline, we have a booth where we can actually discuss with you the 11 month timeline and how it differs from this, the nine month timeline.
14:45
If what we had to do is look at the cell and development and how can we actually speed up the cell and development even though cells need time to grow, what are the actual critical steps that we need to go through in order to produce the final clone? As well as we produce the toxicology material from the stable pool.
15:06
Because now the new technology allows us to have the stable pool being representative of the final clone, have the representative toxicology material as well of the final clone.
15:18
So the stable pool goes into the toxicology badge and that is ready for client at the beginning of month 7.
15:27
As well as what we did on the back end after manufacturing is that we actually moving the drugs substance to drug product to packaging and labelling facilities all under quarantine.
15:37
And if you follow a little bit more where the actual drug substance and drug product and packaging and labelling the green line is actually released.
15:48
They're all released within days of each other and making sure that the information flows from one QP to another through that.
15:56
The one thing that I mentioned, but why it's not it's path, it's called path to IND, but it also can be called path to patient because in the New Zealand and Australia you actually don't need the IND filing to in order to screen for first patient for your clinical trials.
16:16
So when you actually get to the Australia and New Zealand, you release your drug product, you show them that drug product is fully released and that's all you need to start screening your first patient.
16:28
And even though we are from CDMO, we do have CRO services in Australia as well and they can this could be a seamless transition within the actual Thermo family from the CDMO to CRO.
16:43
The other timeline that I would like to look at is the bispecific FC fusions and explain why it's so much longer comparing to the 9 month timeline for IgG-1s, IgG-4s.
16:54
While with the nine month timeline we can go quite fast and we have all of the parameters pretty much finalised for the downstream, we still do DOE for the downstream CEX steps just because that's the very important step in order to provide the correct product quality for the client in the bispecific FC fusion antibodies.
17:18
You really we wanted to still spend quite a lot of time on the downstream development.
17:23
We want to make sure that the product that we give to the patient is actually very pure.
17:30
And especially whenever it comes to heterodimers, we want to make sure that the homodimers are not present there or present in very small amounts.
17:39
And that was the key for us is just to make sure that all of the breadth of FC containing molecules are covered under this timeline.
17:47
So it's not just one specific scaffold of the bispecific, it's any kind of FC containing molecule can fit within this timeline.
17:57
So that's why we're spending quite a lot of time still on the downstream process development and formulation development. With the toxicology badge being available at the start of the month 11.
18:09
And the toxicology badge is from the final clone with your finalised process on the downstream.
18:15
And again, the same idea here on the back end of the production is that drug substance or drug product to packaging labelling is all inside of the Thermo Fisher and all shipped under quarantine.
18:30
The one thing that I want to think about through here is that all of our process development happens in the United States and that one, the cell and development all the way through tech transfer protocols and then our tech transfer protocols are site agnostic.
18:47
That's how we decided to run our process development because we have multiple drug substance sites.
18:54
The other part that we wanted to emphasise is that because of both of those timelines, the nine month timeline and 11 month timeline, they're so fast.
19:04
We want to allocate the actual specific sites in the system that do this kind of work.
19:10
So the people actually do the same thing over and over again and they know exactly what they need to do when it comes to their portion.
19:19
So that's why all of the drug substance, drug product, packaging and labelling, it's all within the European Union.
19:26
We have drug substance in the Netherlands, we have drug product in Italy and packaging and labelling in Germany.
19:32
So all of those sites are allocated specifically for that.
19:36
Yes, they do other projects, but they know that they need to do this specific timelines and they need to make sure that all of the drug products delivered on time to the clinic because you can dose the first patient so early.
19:53
And at this point, I want to just point out that we have a booth.
19:57
So if you have any questions, please let us know.
20:00
We're there and available as well as we have two posters.
20:05
One of them is about our upstream process identification on the upstream that we're already implementing here in path to IND.
20:12
So we actually want to intensify our process to deliver when higher titers.
20:17
As well as the second poster is about by specific FC fusion approach here I just presented the timeline and sometimes people would like to know a little bit more about how did you get there.
20:29
So we wanted to have a poster that goes with this talk.
20:33
And at this point, I thank you for the attention.
