Tuesday, May 21

Influencing ADC Performance through Linker Design

Webinar Q&A

Webinar Transcript 5.21.2024

0:03
Hello everyone, Thank you for attending today’s webinar Influencing ADC Performance Through Linker Design presented by Vector Labs.

0:14
My name is Joan Wiseman and I will be hosting today’s webinar.

0:18
I’d like to start by introducing today’s speakers.

0:22
Doctor Doughty Jackson is a consultant from Jackson Consulting Group, and Matthew Geesey is the Senior Scientist at Vector Laboratories.

0:34
You can read their full BIOS on the left side of your window by selecting the Speakers tab.

0:40
Just a few notes before I hand it off and we begin.

0:45
There are additional resources available with today’s presentation.

0:49
You can click the Handouts tab on the left side of your screen and download them from there.

0:56
If you would like closed captions, they are available from the bottom right corner of the video player.

1:05
You can turn them on there and this webinar is being recorded so you will be able to watch it on demand within 24 hours after we finish here today.

1:15
Finally, we would love to hear from you.

1:18
If you have a question at any point during the presentation, feel free to submit it using the Q&A tab on the left side of your screen, time permitting, will conclude with AQ and a session and answer as many as we can.

1:33
All right, we’re ready to begin.

1:35
Docs.

1:36
Doctor Jackson, please go ahead.

1:39
Great.

1:39
Thank you for the introduction.

1:41
What I’m going to do, I’m going to start off with this slide in particular and this is the slide that kind of encompasses everything we’re going to talk about in this webinar today, both Matt and I, we’re going to talk about the types of linkers that you want to consider for developing your ADC payloads.

1:58
You’re going to talk about their, their the congregation sites because the linkers consists of two parts that’s the part of the antibody and they also there’s the payload part as well.

2:08
And we’re going to talk a little bit about some of the things you want to consider when you start developing your AD CS in terms of the payload and hydrophobicity and also what you want to consider with regards to looking at drug antibody ratios.

2:21
But to start, we see that there is essentially 2 different types of linkers.

2:26
There’s cleavable linkers and there’s non cleavable linkers.

2:29
Within that.

2:29
We also have linear or branch linkers and you have combinations of both.

2:34
And then I know you’ll hear more about this during mass presentation.

2:38
On the antibody side, as I mentioned before, we have cystines, lysines, non natural amino acids and carbohydrates that you can use to conjugate your your linker pillows to the antibody.

2:51
Within that whole category also you have site specific Adcs and you have stochastic or random conjugation methodologies and the random methodologies are primarily conjugated to cysteines or lyses.

3:04
While the site specifics can utilize either site specifically located cysteines, non natural amino acids or the carbohydrates that I had mentioned previously now show specific examples of these different types of methodologies in my part of this presentation and measure Matt would do the same thing as well.

3:23
Once we go beyond just conjugating to the to the to the antibody and your payload, we really have to take into consideration some of the CMC properties with your AD CS as well.

3:34
I’m not going to go through every single last point, but one of the key points you’ll hear from my part of the presentation and math part of the presentation is related to minimizing aggregation.

3:44
And this is going to be important because This is why you use different types of liquid chemistries to mitigate the the the polarity of the payloads to minimize aggregation.

3:56
So taking consideration liquor stability and also the pharmacokinetic properties of the ADC to help us maximize the therapeutic index.

4:10
So in this slide it’s just an overview of what an antibody drug conjugate is.

4:15
We know that for antibody drug conjugates consists of three parts, there’s the antibody part, there’s the linker and then there’s a cytotoxic drug or payload that’s used as well.

4:26
We know from the antibody side of things, right, there’s different types of antibodies that were used as IgG ones, IgG Twos, IgG, Fours.

4:35
I’m going to not going to talk about those specifically, but it would you will see them on some of the slides moving forward.

4:41
With regards to the linkers, there’s different types of linkers as I mentioned before, there’s protease cleavable linkers, there’s non protease cleavable linkers, there’s a, there’s APH or or acid labile linker, glutathione and sulfatase linkers.

4:55
All these equivable linkers that take that take in consideration some of the some of some of the key features of functionalities within the tumor cell that will also elicit the actual release of the palate within the tumor within the tumor cells themselves.

5:12
As I mentioned before.

5:13
Also there’s linear and there’s branch linkers as well.

5:15
And we’ll hear more about now how the AD C’s work.

5:19
We know the AD C’s that she bind to the cell surface antigen on tumor cells and step one they’re internalized and upon internalization they make their way from the endosomal compartment to the lysosomal compartment where they actually are actually released.

5:35
Now in the lysosomal compartment you have a lot of different features here.

5:38
You have low pH environment and you’re going to have proteases as well which which facilitate the release of the cleavage of the payload into the intracellular compartment.

5:50
And then depending on the mechanism of action of the payload or either inhibit DNA synthesis or inhibit tubulin synthesis synthesis which results in the induction of apoptotic cell death in the tumor cell.

6:05
In addition to this, we know that some of the payload would actually be released and if the payload is released, some of those payloads will actually go into nearby neighboring tumor neighboring cells that don’t express the tumor estrogen but were also killed by the payload through apoptotic cell death.

6:23
And this phenomenon is known as bison or activity.

6:26
So as you go through the process of developing your AD CS and you have to make a determination of whether or not you want bystander activity or not and that’s going to, that’s going to help you with regards to selection of your of the appropriate linker for your ADC.

6:42
In this slide it really highlights why you want to develop an ADC.

6:46
So there’s two, there’s, there’s two specific reasons and as to why you want to develop your ADC.

6:51
One is that the ADC helps you deliver the payload directly to the tumor cell.

6:57
So you get specific two killing of that tumor cell and you have reduction of the systemic toxicities that are commonly associated with small molecule inhibitors.

7:07
In addition to the ADC improves the pharmacokinetic properties of the payload as you can see here as with the small molecule inhibitor, what you want to do overall, regardless of whether it’s a small molecule inhibitor or an ADC, you want to maximize the duration of response at the site of the tumor.

7:25
One way in which you could do this as by maximizing your AUC or your exposures for for for the for the small molecule inhibitor or the ADC, you want to remain above the the sub therapeutic exposures because you want to stay in this Green Zone which is where you have your best effect.

7:43
What you want to do also is stay out of this red zone as much as possible.

7:47
This is where you’re going to have your adverse effect exposures.

7:50
This is where you have your toxicities associated with your molecule.

7:54
You want to minimize this as much as possible, so that’s why the ADC in particular does both stops.

8:00
It minimizes the the adverse exposures.

8:04
It maximizes the the the the time above the sub therapeutic exposure.

8:10
You mean the giving your ADC the best time to to carry out its function which is to kill tumor cells Now within the AD CS themselves, what we want to do is maximize the therapeutic window not just between the small molecule inhibitor ADC, but also between AD CS as well.

8:25
And maybe there’ll be some questions in the audience related to the maximizing the therapeutic window.

8:36
So in this slide, this slide show, this slide because I want to talk about you know what what the current state of the state-of-the-art is for for linkers as it relates to and also linker catabolism as it relates to AD CS that are already approved.

8:49
So all these AD CS have already been shown to be efficacious and to be safe that’s why there’s 13 have been approved globally, 11 have been approved by the FDA.

8:59
As you can see looking at the highlighted linker portion that there’s a there’s a variety of different types of linker options available.

9:06
There’s the MC which is your Malamite linker and this is the non creepable linker is to conjugate it to assisting you also have your DI peptide Lakers shown here in your file.

9:18
Alabelle sit linkers for example, those are also conjugated to Cystis and they’re all cleavable linkers.

9:23
They’re protease cleavable linkers as I mentioned before previously in addition to the DI peptide Lakers you also have the four peptide Lakers shown here which is the liquor that’s primarily used for in her two.

9:35
We’ve heard a lot about a her two and the success that it’s experienced in the clinic, in the clinic right now we know this is cleavable linker conjugates to cystines as well as using a topo somerase inhibitor.

9:45
What I’m going to do is to first start off talking about Mylitard.

9:49
Mylitard was the very first ADC that was approved and was approved for treating AML cancer patients who are a little bit on the older side 60 plus.

10:00
So with this, with the with Mylitard, you can see the design it’s using.

10:04
It’s utilizing A Hydrozone linker which is Aphla Bio linker and that is one ’cause.

10:10
That is that only occurs once the once the ADC recess the lysosomal compartment and therefore initiates, begins to initiate the process of releasing the payload from the linker.

10:22
After hydrolysis, the glutathione comes into play and glutathione will will cleave this disulfide bond here, thus resulting in the actual release of the fully functional Collegiomycin payload.

10:37
A couple of other things I want to point out for this ADC in particular is that it’s an average dollar between 2:00 and 3:00, so the average of two or three drugs per antibody and this 50% of the antibody is unconjugated.

10:51
So this is unusual because of the simple fact that 50% of your antibodies is unconjugated.

10:56
That means 50% of your antibodies going to be working against your ADC.

11:01
And we also noticed here there are no space and there’s and there’s no pegs.

11:05
So conjugation conditions using this type of technology really are going to play an important role of maximizing the effectiveness or the the CMC properties of your ADC.

11:21
That brings me to the next slide where I’m talking more about one of the one of the members of the RSAT in class which is MMAE.

11:29
As you can see this paper design is different from the first one whereby you have this melamite conjugation loyalty that conjugates the systems.

11:37
You have now this dipeptide linker, the Val SIT linker and you also have this Pepsi spacer.

11:45
This Pepsi spacer is there because it gives the cotepsin B the a phrase cleavable enzyme, the ability to come into this space, cleave the dipeptide linker, thus resulting in the release of this active molecule which is MMAE.

12:03
MMAE is very high membrane, high membrane permeability.

12:06
So as I mentioned before, this is one of these molecules where where you’re able to get bystander activity.

12:12
Now some of the other features that we see routinely now using this, this and other types of conjugations is that we have between 75 to 9590% conjugation efficiency.

12:23
Remember monotile was only 50% so we’ve gone, we’ve we’ve improved that substantially there is between 9095% greater monomeric content.

12:34
So you’ve reduced the aggregation issues associated with some of the other conjugations for over these types of AD CS.

12:42
The traditional ADC if we’re using stochastic conjugation is Darf 4, so four drugs per antibody.

12:49
If you try to go above DAR 4 using this type of linker payload, you’re going to run it, then you’re going to run into aggregation issues.

12:56
And I know Matt has in his presentation some technology using very, very interesting liquor chemistry which can get you around this potential issue.

13:05
So you can get develop AD CS both with a higher than DAR 4 for MMAE.

13:12
With that being said, there is the other member of this of the family and that is the MMAF.

13:20
This is also an Rs status monomethyl RSAT and F it’s using the MC or the malamite linker conjugates a cysteine as you notice that there’s there’s there’s no dipeptide and there’s no pap pap spacer associated with this one.

13:36
And what happens is that the way this this payload is actually released, this is the active payload here this cyst MCMMAF.

13:43
So this, this result, this this needs antibody degradation in the lysosomal compartment which leaves you dust with the cysteine conjugated to the MC, the malamite linker, and the active warranty which is which is MMAF.

13:58
This molecule has low membrane permeability, which means once it gets inside the tumor cell, it stays inside the tumor cell.

14:05
So you’re not going to get bystander activity with this type of linker payload system.

14:17
Moving forward-looking at as we as as the, as the AD CS develop the chemistry associated with developing the AD CS and the linker becomes a little bit more intricate.

14:29
And I’m showing you here, this is, this is an ADC for against TRUMP 2 using the SN 38 topoisomerase inhibitor payload.

14:38
The linker here is CLCO 2A.

14:41
But you notice here is that you begin to see the incorporation of PEG solubilizing groups because of the hydrophobic nature of the payload.

14:51
Next you have to add this because you want to increase your solubility.

14:55
So in some cases you might be able to make chemical modifications of the payload to address some of the hydrophobic nature, some of the hydrophobicity issues associated with the payload.

15:06
Sometimes you’re not.

15:07
In cases where you’re not able to do that because you’ll decrease the potency of the payload, Now you have to really focus on the chemistries of your linker, which is why they added this PEG seven group here to recruit and improve the solubility for your conjugation.

15:26
The last slide I’m going to show with regards to talking about the different types of linkers, this this one where this is ACD 1980 C it’s an IGT one and you have see in this case you have DPEG 8 as your as, as, as part of the linker component.

15:44
Why?

15:44
Because it improves solubility, reduces aggregation.

15:47
The second part is that there’s a Val Ala linker Estelle of a Val sit linker.

15:53
As you can see the the Val Ala Laker has lower hydrophobicity than the Val sit linker.

15:59
So that helps with aggression, improving the solubility of your of of your Laker payload.

16:04
There’s also the path spacer as well.

16:07
As I mentioned before, this type of payload, the PBDS are known to be very hydrophobic payloads.

16:14
With that being said, we know that you can also get bystander activity with this type of a payload that you that we could not get with the MMAF payload.

16:26
As I mentioned before, I talked about the therapeutic index slide, I specifically talked about exposure as opposed to dose because it’s the exposure, the pharmacokinetic properties that’s going to drive your efficacy and it’s also going to play an important role with regards to your toxicology findings as well.

16:44
So we know that with naked antibodies or unconstitated antibodies, they have a really good normal IgG profiles in terms of pharmacokinetics.

16:54
But with the ADC themselves, you begin to see a loss of the ADC over time.

17:00
You get this deconjugation of the of the of the payload away from the ADC.

17:05
So you have more instability with this type of conjugation methodology with the if you as you increase the stability of the linker, you increase the the stability of the ADC, thus giving you more pharmacokinetic properties that more in line with you see with the total antibody and you see less deconjugation.

17:27
This is going to be important.

17:28
As I mentioned before, what you want to do is to have more time of the more time on target.

17:33
So the more time you can get the ADC in in onto the tumor, get more drug release into the tumor, the better efficacy you’re going to get.

17:42
But that being said, if you look at stochastic conjugation methodology shown here at the bottom, you can see that as you increase the DAR, the number of drugs for antibody, you also increase the clearance rate.

17:53
So This is why it’s really going to be important to find the optimal DAR for your for your ADC.

17:59
So you want the best of both worlds.

18:00
You want to make sure you’re you have enough stability to get to get the ADC, the drug to the tumor.

18:06
And you also don’t want to have the liability of having increased toxicology by having a premature release of the of the payload.

18:16
So this improved stability leads to an improved therapeutic index.

18:25
So what’s next coming for for AD CS?

18:28
I’ve talked about what we are, what we already know about AD CS and the linkers, but what’s coming next and really you fall from the two categories, one is a site specific conjugation and the other is using branch linkers to accommodate dual payloads for site specific conjugation.

18:43
I’m going to show some specific examples using primarily Agenomoto’s license site specific conjugation technology because I think this is interesting.

18:53
There’s about 92 licenses that are available for conjugation using the stochastic, the stochastic method and most of the time for site specific reason existing.

19:03
So this is a little bit different take on the site specific conjugation.

19:07
We also have non natural amino acids.

19:09
I mentioned Anabrax as the leader for for this technology.

19:12
We also have some some exciting conjugation technologies coming to the forefront from sinus fix and alpha map because they’re using glyco what is called glyco conjugations, so using the carbohydrates on the antibody themselves as points of of of contact for or connection conjugation for the for the Laker payload.

19:31
The advantages of of the of the of a Jinomoto’s and Synfax and Alpha Maps technology is that doesn’t require antibody engineering.

19:40
So you can just take an antibody off the shelf that you’re using and use these type of technology as opposed to using something like Amberxis technology that’s going to require antibody engineering and that you you have to find the right site to actually incorporate your non natural amino acid to get the stability and the efficacy you’re looking for.

20:01
With regards to branch Lakers and dual payloads.

20:04
Some of the advantages potential advantages are one is you can combat against drug resistance mechanisms and the other one is that you might be able to find the right combination of the two payloads that give you.

20:15
Synergistic distribution of tumor growth, this is something I think we’re all looking forward to see how this, how how this plays out into the in, in the clinic in this slide we’re just looking at the overall schematics of these different types of technologies and I want to highlight a few, a few, a few key points.

20:32
One is with the Jinomoto’s technology, the site matters where you get where you conjugate.

20:40
So if you look at Lysing 248 versus Lysing 288, your overall yields are going to be much higher with using lysine 248 as opposed to using lysine 288.

20:51
So that not only does the cut the linker payload matter but also the sites where your conjugate matters as well in terms of your yields.

20:59
With regards to alpha map, their technology, they are using A1 enzyme technology relative to A2 enzyme methodology that you see for Cinephyx.

21:08
Both are using modifications of the of carbohydrates to elicit their site specific conjugation technology and the last one as I mentioned before this is anthraxis technology.

21:20
We’re using incorporation of non natural amino acids using the stable oxing bond and also they’re also using the MMAF payload as you mentioned.

21:28
As I mentioned before doesn’t really have bystander activity.

21:31
So you have several options available to you with regards to looking at site specific conjugations.

21:38
With that being said, so what’s the advantage?

21:40
So the advantage of site specific conjugations is that you know the site specific conjugations can improve your pharmacokinetic properties and hopefully with that being said, they can also it gives you a better therapeutic inning compared to your traditional stochastic conjugations on the panel.

21:58
A.

21:59
This is the example I showed before.

22:00
What your ADC looks like using traditional stochastic conjugation methodologies in panel B is the example of what it looks like using site specific where you have reduced deconjugation of your ADC.

22:14
What Agenda Moto has shown as well is that that the the stochastic conjugation methodology results in this and also what the results in decreased stability of your ADC while the site specific version of their ADC results in almost identical or pharmacokinetic properties to the naked antibody.

22:36
Second thing is the DAR, the number antibody number of drugs for antibody, What’s the site specific Uridar 1.6?

22:44
What’s the sarcastic Uridar 4?

22:47
So essentially for for half the DAR you get better from kinetic properties, but how does this translate into efficacy and also toxicity where they perform an in vitro efficacy study in the NCIN 87 tumor model.

23:02
And in this tumor model you can see that the site specific ADC performs just as good as the stochastic ADC with the DAR 4.

23:11
So that tells you that although you have less drug per antibody, the activity is still similar to what you get with the stochastic conjugation methodology.

23:22
There are other examples not shown here where you can see that the site specific at ADC gives you better efficacy compared to the stochastic conjugation method but efficacy is 1 sided equation in terms of your therapeutic index.

23:38
What about toxicity?

23:40
In terms of toxicity what we see is that the four at the 40 makes per kick you can stochastic conjugation methodology.

23:47
You have a significant loss in body weight and with the site specific conjugation you don’t see that right.

23:55
And then of course looking at outside of body weight as a surrogate for toxicity there are specific markers of toxicity you can you can look you can use to in your evaluation as well.

24:06
One is ASTAST is a marker commonly commonly associated with liver toxicity and plate.

24:12
The loss of platelets is commonly commonly associated with thrombocytopenia.

24:17
What you can clearly see is with the site specific conjugation you can see that the level of AST are low almost to the same levels as your of of your buffer or your control while the site while they start.

24:30
While the sarcastic conjugation gives you high levels of AST, the same trend is also seen here for platelets where if you take a look at the platelets the the site specific does not result in a significant decrease in platelets while the while the sarcastic conjugation methodology does.

24:48
So these data tells you that you can probably get better therapeutic index at least preclinically and hopefully clinically from using not only this type of technology but perhaps site site specific technologies in general.

25:05
So if site specific technology is also good, what does, what does it look like in terms of the clinical landscape.

25:11
We know that for site specific AD CS they’re primarily in in in early clinical development phase one already phase two and very few are in phase three right now.

25:22
But the ones that are in phase three I’ve highlighted here in this table and what you’ll the take home message from this slide is that what you see is that with regards to liquor and payload they’re using liquors and payloads that are commonly associated with the approved AD CS.

25:39
So this is the effort to try to de risk your your ADC development.

25:44
You can go with a novel novel approach with a novel ADC novel technology and and and novel link or payload.

25:52
But that adds a significant amount of risk for your development going to something that’s already been clinically approved or clinically active trying to help you de risk that moving forward.

26:01
And this is what many of them, many of the companies using site specific technologies have have the approach they’ve taken.

26:11
Now with regards to the last part of my presentation when we’re talking about the utility of branch Lakers to to facilitate the incorporation of dual payloads.

26:21
So with this technology, essentially what they’ve done is to use a the MMTGAS to mediate the liquor conjugation to the antibody and they have orthogonal click chemistry that they’re using to incorporate MMAE and MMAF to the ADC, right.

26:41
This is a 1 pot reaction and I’m pretty sure Matt’s going to talk about this a little bit more in his part of this presentation.

26:46
But the take home message is that you can get more bang for the buck.

26:50
You can get 280, you can get two liquor payloads can’t get to your ADC giving you a dollar of between 4:00 and 2:00, two for the MMAF, 4 for the MMAE.

27:05
So the question is So what is what’s the advantage of doing this?

27:08
What they show here in the review with efficacy studying this ACC 1954 tumor model.

27:14
What they show is that the site specific AD CS with the dual payloads gives you better activity compared to just taking the tretuzumab ADC that has a DAR four and the tretuzumab ADC that has a DAR and it has an MMAF of a DAR 4 as well and adding them together.

27:33
So this to the addition of 280 CS with higher DAR doesn’t doesn’t necessitate better activity compared to the DAR that that the dual that has a dual payload that’s both seen in the tumor volume and also with regards to the the tumor weight.

27:52
And with that being said, I think the last thing I want to make sure go back, the last thing I want to make sure we we see you pay attention to is that it’s really good to have a do have have a an an ADC toolbox available to you where you have different antibody formats, different linker payloads to choose and optimize to give you the best ADC moving forward.

28:15
And with that, I’m going to acknowledge that thank Vector last for giving me the opportunity to to share with share with you did some of my thoughts on developing it linkers and payloads for AD CS Fierce Farmer for hosting this and also Beacon by Hanson Wade for giving me access to their ADC database which is really good for competitive intelligence.

28:36
Now I’ll turn this over to Matt now.

28:39
All right.

28:40
OK.

28:40
Thank you Doctor Jackson for that synopsis of some of the important features of ADC Design and Performance Trust.

28:48
Everybody else on the other side of the screen is having an outstanding day today.

28:53
Doctor Jackson mentioned that the linker was one of the tools in the toolbox of ADC design.

28:59
And so for this last section, I’d like to give some examples from the literature where scientists and researchers have used linker payload design, specifically the incorporation of PEC component to improve ADC performance.

29:14
Bio Design leverages results such as these to provide linkers that you can use for optimizing your ADC design.

29:21
So I’ll finish up with some examples of some of those products that we sell.

29:27
OK, so first we’ll get into how incorporating a PEG into the Link or payload can help you build the ADC and improve physical chemical properties such as hydrophobicity and aggregation.

29:43
So Doctor Jackson had shown us the structure of Zylanta and had mentioned that the PEG 8 spacer was important to reduce aggregation and improve solubility for the PVDADC.

29:55
And these are some more examples of that by spirogen Cgen and AstraZeneca.

30:00
And what we have here is payloads on the left, linker payloads in the middle, AD CS over on the right.

30:06
In the top we have a DAR two ADC without a PEG spacer and that linker pay or that payload C log P is 2.91.

30:14
That’s a measure of hydrophobicity.

30:16
In the middle they have put the Val Ala cleavable trigger on and the Malium in reactive group That resulted in an increase in the C log P of that payload.

30:25
That linker payload is called Tallarine.

30:27
Conjugation of Tallarine to an antibody gives us a DAR two ADC with 13% aggregate and 51% yield in the example in the middle.

30:38
Now a PEG 8 spacer has been incorporated into the linker payload design.

30:43
The payload is shown over there on the left.

30:46
It has AC log P of 2.98.

30:48
The structure is manipulated a little bit compared to the above example, but hydrophobicity has not changed much in the over to the right they have put the Val Ala trigger, the PEG 8 Spacer and the Malium and Reactive Groupon.

31:01
This linker payload is known as Tessarine and it has a lower C log B.

31:05
Conjugation of Tessarine to an antibody gives us a DAR 2.5 ADC with only 3% aggregation and 86% yield.

31:14
So what we can see here is that incorporating that peg spacer resulted in an improvement in aggregation profile and an improvement in conjugation efficiency.

31:23
Now we’ll note the above example, they conjugated tallerine to an antibody, an engineered antibody at Cysteine 239 and that proceeded in 96% yield and only 2% aggregate.

31:36
So there are other ways to achieve this goal.

31:38
As Doctor Jackson mentioned, conjugation site matters and that’s the case there.

31:42
But that example in the middle shows that incorporating a PEG spacer can also help you achieve this goal.

31:47
And that can be important if you don’t have access to antibody engineering or if you just can’t find a site that helps you achieve the desired goal.

31:56
The example on the bottom shows that you can also apply this approach to hired our AD CS.

32:01
The payload is shown over there on the left.

32:03
It’s structure has been manipulated further to reduce the hydrophobicity.

32:07
Even more, it did reduce the potency of that payload, which is now why you’re having to compensate that with a higher DAR ADC.

32:15
Putting the Val Ala trigger, PEG 8 spacer, and MALAMID reactive group on that payload reduces the C log P even further.

32:22
And conjugation of that linker payload to an antibody gives us a DAR 7.7 ADC with only 5% aggregate and 99% yield.

32:32
So these examples show that the combination of that PEG spacer, the cleavable trigger, and the payload can impact ADC conjugation efficiency and yield.

32:43
These examples here show that you can also apply this strategy to other payloads.

32:48
This was some work done by Millipore Sigma and they’re using the Topa 1 inhibitor SM38.

32:54
Topa 1 inhibitors are all the rage right now, and there’s several examples of Daray AD CS with Topa 1 inhibitors that don’t need a PEG to improve their properties.

33:05
Inheritu is the best known example.

33:07
AstraZeneca has reported some in the literature, Zymeworks has reported some in the literature, so it’s not always needed.

33:13
But Doctor Jackson did show us the structure of Trodelvy, and Trodelvy has an SM38 payload and they didn’t need a PEG 8 spacer to improve its properties.

33:23
And these are some examples that are that also need that help.

33:27
The top example shows a DAR four and a DAR 8 Valsit SM38 AD CS with a standard linker.

33:35
Over on the right we can see the C log P of that linker payload is 3.81.

33:40
This results in a DAR 480C that’s 20% aggregate and a DAR 880 C that’s nearly completely aggregated.

33:48
In the example on the bottom, we have some DAR four and DAR 8 Valsit SM38AD CS Now they’ve incorporated the hydrophobic DDCO moiety to use click chemistry, very hydrophobic reactive group, but they’ve offset that by putting a PEG 12 spacer in between there.

34:06
Over on the right, we see that they did indeed reduce the C log P of that linker payload and this provides a DAR 480C with only 8% aggregate and a DAR 880 C that’s only 15% aggregate.

34:18
That’s a significant improvement from the example on the top.

34:21
So these examples show that PEG spacers can reduce linker payload, hydrophobicity and ADC aggregation if you have a problematic combination of cleavable trigger and payload.

34:33
These examples here show a similar thing with the popular MC Valsa MMAE linker payload.

34:40
This is also known as Vidotin.

34:42
Doctor Jackson showed us some examples of this and mentioned it’s problematic above a Darfur and these examples kind of support that.

34:50
The example on the top, it’s the MC Valcit MMAE linker payload.

34:55
They’ve incorporated the hydrophobic DBCO reactive group, tried to offset the hydrophobicity with the PEG 12 spacer.

35:03
But over on the right you can see that this wasn’t really enough to do the trick and it’s still 72% aggregated.

35:09
So the example on the middle there, what they’ve done is replaced the PEG 12 spacer with a PEG 24 spacer.

35:16
And over on the right we see that that did indeed reduce the hydrophobicity of the linker payload and now this DAR ADC is only 9% aggregated.

35:27
There’s more to aggregation than hydrophobicity, It’s a simple calculation to do.

35:31
So we use that.

35:32
But some chemical moieties are just prone more prone for self association.

35:37
Insulin Liceborough is a great example of that.

35:39
So this example in the bottom kind of shows that it’s that exact same combination of DBCO, reacted group, MC Valcit MMAE.

35:48
But now what they’ve done is they’ve taken the PEG spacer and put it in an orthogonal position rather than between the payload and the antibody.

35:55
So what this has done is it kind of Shields that trigger and payload.

35:59
And in this case it’s only a PEG 8 rather than a PEG 24 and PEG 12 like the examples above.

36:06
And you can see that’s reflected in the C log P over on the right, which is actually higher than the above 2 examples.

36:13
But the DAR 880 C is only 7% aggregated, which is lower than the above 2 examples.

36:19
They further went on to show that you can put a PEG 12 and a PEG 24 on the side and it reduces your aggregation even further.

36:26
But this just goes to show that PEG spacer length and the positioning of that PEG can also influence linker payload hydrophobicity and ADC aggregation.

36:35
And those two aren’t necessarily correlated.

36:38
And if you have a really problematic combination of reactive group, cleavable trigger and payload, you might have to optimize both of those variables.

36:51
Doctor Jackson mentioned the use of branch linkers to accommodate multiple payloads.

36:55
If you’re looking at developing ADC with a dual mechanism of action, this example like his uses both MMAF and MMAE.

37:03
So you have ADC with mixed bystander activity.

37:07
But what CGEN did here is they put a PEG 24 spacer in that orthogonal position and they capped it off of that carboxylic acid group.

37:15
So this improves the hydrophilicity and water solubility of that linker payload.

37:20
Putting MMAF and MMAE on there then results in a DAR 16 ADC with mixed bystander activity and it’s 98% monomer.

37:29
And that’s a pretty impressive feat for a DAR 16 ADC.

37:34
These branch linkers can also be important for increasing the DAR when you have limited sites available for conjugation.

37:41
And you can encounter this when you’re using certain engineered antibodies or using certain site specific conjugation technologies.

37:48
You just don’t have the number of reactive groups there.

37:51
So here’s an example where they took a multi functional linker and installed it at Q 95 via microbial transglutaminase.

37:59
So even though you still only have two conjugation sites on that antibody, the branched linker allowed you to put 4 reactive groups on that antibody and now you can come at it with this single payload linker.

38:12
Do your bio orthogonal click chemistry and you end up with a DAR 4 ADC with two payloads per conjugation site and less than 1% aggregate.

38:22
So in this case the linker design allowed them to increase the DAR on an antibody with limited conjugation sites by increasing the number of reactive groups per conjugation site.

38:33
This is an alternative approach.

38:35
This is Lego Chembio that did this here.

38:37
What they did is they took a linker and installed it at an engineer tag at the C terminus of the antibody, the affarnaceal transferase.

38:46
So now even though there’s still only two reactive groups per antibody, they’re reactive in bio orthogonal carbonyl groups.

38:55
Now you can use this multi payload linker with the oxene chemistry into your bio orthogonal oxene ligation and again you can end up with a Darfur ADC with two payloads per conjugation site and no mention of issues with aggregation.

39:09
So in this case a branch linker design enabled them to increase the dollar of an antibody with limited conjugation sites by the use of these multi payload linkers.

39:20
So that was some examples of how you can incorporate a PEG spacer into your linker payload design to help you build the AD CS and improve their physical chemical properties.

39:30
For the next two sections, we’ll look at some in vivo properties and I will just focus on PK and Tox.

39:38
Doctor Jackson had a nice slide showing the effect of conjugating A payload to an antibody on the clearance.

39:44
Generally when you conjugate an antibody with a payload, the clearance of the ADC has increased relative to the naked antibody and this is correlated with DAR.

39:56
The higher the DAR, the faster the clearance and this was established by the O, GS, C, Gen.

40:02
and Immunogen back in the day with oristatin payloads and matansinoid payloads.

40:08
Here I’ve got some examples with some PEG 8 spacers and PBD payloads showing how you can try to mitigate that that effect.

40:16
The examples on the top are DAR 2AD CS.

40:19
The one on the left is a Valala PVDADC without a PEG spacer, and that DAR two ADC has a half life of 2.4 days.

40:29
The example on the right is a Valala PVDADC and now a PEG 8 spacer has been incorporated into that design.

40:36
And that DAR two ADC has a half life of 7.2 days.

40:41
So this is an example where that linker payload design, changing it up resulted in a three fold increase in the half life of the antibody.

40:48
And I know there’s some caveats with this head to head comparison, the antibodies are different and that can matter.

40:54
We don’t know the half lives of the naked antibodies, so you can’t really look at the direct comparison there.

41:00
But I think we can probably agree that a half life of 2.4 days is a lot less than we’d like to see for an ADC, whereas a half life of 7.2 days is more in that acceptable range.

41:11
I’m also not claiming that this fixes everything.

41:13
This exact same payload over on the right was in robot T, which failed in clinical trials.

41:19
It was in ADC T3 O1 which failed in clinical trials.

41:23
So we’re just looking at half lives here, and if that’s a concern, this is a way to address it.

41:28
The example on the bottom shows the effect of an increasing DAR on AD C’s half life and here we have some DAR 24 and eight Valala PVD 80 CS, again with that PEG spacer in there.

41:42
In this case we do know the half life of the unconjugated antibodies 13.4 days.

41:48
And then beneath that I show the half lives of the DAR, two DAR four and DAR 880 CS.

41:53
And what we see is that even though there is a decrease relative to the naked antibody, it’s not completely surprising.

42:00
We can see that there is no dependence on DAR.

42:03
The AD CS have the same half lives for DAR 24 and 8:00, so this is an example where PEG spacers can help you improve the half lives of your AD CS.

42:13
Again, at least for PVD payloads.

42:15
And this is an important caveat.

42:17
A lot of people have had success with linear PEG spacers and their payloads, but C Gen.

42:23
did some nice work showing that this is not a universal solution.

42:27
They had some ADC’s with the glucuronide trigger and the MME payload dar 8 ADC’s, and when they put a PEG 24 spacer in there, they found that ADC was actually cleared faster and was less efficacious than the ADC without any PEG at all.

42:45
So it actually had the opposite effect.

42:47
What they also showed in that Seminole work is that if they took the PEG and put it orthogonal to the payload and antibody rather than between the two, that result in DAR 8 ADC was cleared slower and was more efficacious than the other two ADC’s.

43:04
So I’m not going to cover that initial result.

43:07
But this is a follow up study that they did where they look more closely at the effect of that orthogonal PEG spacer on the PK of the ADC.

43:17
So the structure of the ADC is shown there in the middle and they vary the number of ethylene oxide units in that side chain.

43:23
The rat PK is shown over there on the right and what we can see is that for the AD C’s with the PEG 8, the PEG 12 and the PEG 24 modifiers, the pharmacokinetics and the exposure are real similar to the naked antibody.

43:38
So that’s a pretty neat finding.

43:40
I also put the HIC retention times below that.

43:42
This is from a different study that they did, but I just thought it was good to see this comparison.

43:48
We like measuring the HIC retention times of AD C’s to gauge their hydrophobicity and then we use that to infer PK liabilities downstream.

43:58
Later attention times means more hydrophobic means cleared faster, but here we can see that’s clearly not the case.

44:05
The unconjugated parent antibody has a retention time of 3.7 minutes and those 380 CS, those DAR 880 CS with the PEG 8/12/24 spacers have retention times that are nearly three times as great.

44:19
So we would think more hydrophobic cleared faster.

44:22
But based on the PK studies above we see that that’s clearly not the case.

44:25
There’s exposures are about the same and a number of other groups I’ve noticed this now as well.

44:31
So just use caution when you’re trying to use HIC results to extrapolate to PK liabilities.

44:38
It’s not not always a correlation.

44:44
So finally I’d like to get to the great Slayer of AD CS in the clinic tox.

44:49
This is by far the number one reason that most AD CS fail in the clinic is off target tox for failure to show efficacy at the maximum tolerated dose.

44:58
And I’d just like to show some examples of how this can be reduced with pegs in the linker payload design.

45:06
So again, this is another example by the Mighty C Gen.

45:08
They’re looking at the effect of that orthogonal PED modifier on the tolerability as assessed by loss of body weight.

45:16
They’re varying the number of ethylene oxide units in that side chain.

45:20
The chart over on the right shows the percent body weight change, and what we can see is that these DAR 8 ADC’s, their tolerability windows that 17 times the efficacious dose is similar to vehicle for the PEG 8, the PEG 12, and the PEG 24 modifiers.

45:36
So that’s a pretty neat finding.

45:38
I don’t present any of the results here, but they did some follow up studies that are worth looking into.

45:42
They also show that increasing that PEG length is correlated with a reduction in off target uptake by liver, spleen and bone marrow.

45:50
And they also show that it’s correlated with improved clinical pathology profiles.

45:55
So this is an example where these PEG modifiers can be used to improve tolerability, probably by reducing off target uptake and these two examples just so this same strategy can be applied for different combinations of cleavable triggers and payloads.

46:11
In that top example is Adari ADC has the same MMAE payload as in the previous example, but now the cleavable trigger is changed out to Val ICE, and in this case there’s only a 5% decline in the body weight of mice dosed at 20 times the efficacious dose.

46:26
In the example on the bottom, it’s another DAR 8 ADC.

46:30
The cleavable trigger is still Val lice, but now they’ve changed out the payload to pass the Taxol, and in this case that DAR 8 ADC only show well, essentially no change in body weight for mice dosed at 10 times the efficacious dose.

46:43
So these examples show that incorporation of PEG spacers and modifiers can improve the tolerability of these high DAR AD CS for a variety of different combinations of cleavable triggers and payloads.

46:55
So in conclusion, as you can see choosing A linker for your conjugate is easy in theory, but it’s difficult in practice.

47:05
What’s good for the goose is not always good for the gander and bio design is a from Vector Laboratories can help deliver your therapeutics potential by harnessing proprietary linker architectures, incorporating chemical biology expertise and leveraging 20 plus years of linker design to provide a toolbox of linkers that you can use for optimizing your AD CS performance.

47:31
So I just want to open up that toolbox a little bit so you can see what we have.

47:36
First, we have different linker architectures.

47:38
We have linear branched and sidewinders.

47:41
They can accommodate different numbers and different types of payloads, and they can control the orientation of those components on the linker payload like that orthogonal linear designation.

47:54
2nd, we have different reactive groups for conjugating your payload and your cleavable trigger.

47:59
If you have your own proprietary cleavable trigger, you’re probably going to want a carboxylic acid or an active Ester.

48:05
If you’re using the standard dipeptide cleavables, those can be provided as the Val Sit and Valala PMP carbonates.

48:13
We have other groups available for attaching payloads, but right now most common payloads in the clinic are attached via amines and active esters, so I’ve just restricted to that for this webinar 3rd we have a number of reactive groups for conjugation to the antibody.

48:31
To keep it simple, most of the examples that I’ve put in this webinar are conjugation to hinge cysteines.

48:37
So we have a number of options available for conjugating to Sulfhydro groups and hinge cysteines.

48:42
As Doctor Jackson mentioned, there’s a lot of site specific and bio orthogonal conjugation technologies that are garnering interest.

48:49
We do have those groups as well, the DBCOS, the TCOS, the tetrazines, the aminoxes, so those options are available as well.

48:59
We also have different end capping for different property modification.

49:03
These include methoxy, hydroxy and carboxylic acids.

49:07
They can help change the polarity and or charge in physiological environments.

49:12
The only example in this slide deck was C Gen’s DAR 16 that had the carboxylic acid group at the end.

49:19
But Mersana’s synthomer scaffold is also making use of features such as this.

49:28
And this is an example of a more sophisticated custom structures that we can provide for for you with our Bio design service.

49:36
We can work with you to really help fine tune the Physico chemical properties of your ADC, optimize the DAR, we can control steric congestion around that trigger and payload to protect metabolic liabilities or modulate payload release rates.

49:51
The structure on the left we refer to as the body armor architecture.

49:55
We had a bio conjugate chemistry paper in 2022 that details some of that.

50:00
The structure on the right will be analogous to Mersana’s symptom or scaffold and I would recommend that you go and look at some of their work with that.

50:07
It’s pretty interesting.

50:09
And then these would finally be some examples of kits that we can provide through the Bio design service.

50:15
In this case, there’s three component kits that could facilitate the construction of dual MOAADCS.

50:22
Step one, you activate your first payload.

50:24
Step 2, you activate your second payload.

50:26
Step three, you put them on the carrier.

50:28
Step 4, you conjugate them to the antibody.

50:30
So this would really help overcome resource constraints, especially if you’re chemistry limited.

50:36
Let us handle some of the chemistry and it can also maximize your development diversity by allowing you rapid access to different combinations of payloads and linker variables.

50:50
So in summary, linkers play a critical role in multiple aspects of an AD CS performance, from how you build them to their physical chemical properties to their indivo performance.

51:01
It’s therapeutic window, right?

51:02
You want to get more of this ADC where you want and keep it away from places that you don’t.

51:07
ADC performance can be optimized with the right linker design, and bio design from Vector Laboratories is a collaborative developmental process to help you achieve this goal.

51:18
I’d like to conclude with a quote from the father of modern medicine.

51:22
Medicine is a science of uncertainty and an art of probability, and drug development is certainly no different, and antibody drug conjugates are certainly no exception.

51:32
The road to developing a successful ADC is filled with uncertainty and challenges, and it’s definitely littered with failed attempts.

51:41
It is exciting to see all the different ways that developers are trying to overcome these challenges, from new payloads that work with old mechanisms of action, to new payloads with new mechanisms of action, to different cleavable triggers, to different conjugation chemistries.

51:56
And we’d like to throw linker design into that bag of tricks for ADC developers to use.

52:02
And that concludes this part.

52:04
I’d like to thank all our webinar registrants for tuning in.

52:07
Hopefully it was informative.

52:08
And let’s keep making some medicines.

52:10
Yeah.

52:10
And with that, we’ll open up for some questions.

52:15
Fantastic.

52:15
As we dive into the Q&A.

52:18
Quick reminder to the audience, you still have time to submit your questions.

52:21
Just use the Q&A tab on the left side of your screen.

52:26
We’ll cover as many as we can and the team will reach out and answer your questions offline if we don’t get to them.

52:35
So do submit your questions.

52:37
Our first question says would you consider re bridging conjugation site specific conjugation, Yes, I’ll I’ll take that question.

52:47
I think the short answer is yes, and it’s the reason why I answered yesterday questions because site specific conjugation methodologist covered two different points.

52:56
One is that you want a homogeneous DAR is thyrobridge technology gives you that.

53:01
The second one is related to the stability of the ADC.

53:06
So the stability of pharmacokinetic properties using thyrobridge accomplishes that as well.

53:11
So I think that short answer is yes, it does fantastic.

53:15
Yeah, I can.

53:16
I can add to that a little bit too.

53:19
Conjugating to hinge disulfides, you’re still conjugating that same region, but you’re doing in a defined ratio now and you are saturating that whole area, so you’re getting a a defined R that is stable.

53:32
So I would agree.

53:33
And the other thing, a lot of those next generation Maliomids, the Thio bridges, the Dibromo Maliomids, the Dibromo Perizi diones, they are kind of hydrophobic and a lot of them do see some benefits from incorporating a little deep egg spacer in there to help out with the yields and the properties of the linker.

53:51
Alright, how does branched linker affect the PKPD of ADC?

54:00
Yeah that’s that’s that’s another good question.

54:03
I think the the the answer is is complicated, but I think in short with the with the limited number of examples we do have, it shows that the branched linkers in in of themselves are provides the same type of stability as the unconjugated antibodies.

54:19
So you have good pharmacokinetic properties, but I think that’s really dependent upon multiple factors that we talked about today.

54:25
It’s going to be dependent upon what type of linker you’re going to use, what type of payloads you’re going to use incorporating into that whole structure.

54:31
With regards to pharmacodynamic properties, we don’t know.

54:35
The answer is still the jury’s still out for that.

54:37
But I think the dual payloads also gives you more complexity part of your pharmacokinetic assay.

54:44
So now you have two different payloads you have to monitor as opposed to one specific payload that you have to use before.

54:50
So you ask, so moving for your assay you develop are going to be more complicated.

54:58
Is there I may not be pronouncing this right, is there AC log P range ideal for linker payload?

55:05
Yeah, I’ll take this one.

55:06
This is kind of a a chemistry thing.

55:09
You you could look at Lipinski’s rule of five for ideal C log P for orally bioavailable drugs, which I think is somewhere between maybe two and five, three and five, certainly less than five.

55:22
But I would say the short answer is no.

55:26
We did have a slide in there showing that you can still have a linker payload with a high C log B that does not aggregate if you have the DPEG position correctly.

55:35
You have to remember that you can only do so much to your payloads hydrophobicity before.

55:40
It’s not going to get across cell membranes, it’s not going to reach its target, it won’t undergo endosomal escape.

55:47
So you do need to maintain a certain level of hydrophobicity.

55:51
So yeah, it’s tough to say there’s an ideal C log P And again, maybe Lipinski’s rule of fives are the it’s the best guide, but you can only go so far and then you’re going to have to try to compensate with other things on that antibody or check what are the cases where linear or orthogonal linkers would be used.

56:13
I’ll take that one again.

56:15
Thanks, Matt.

56:16
Not again, no.

56:19
No one-size-fits-all answer here.

56:22
But I would say if you have a payload that you cannot manipulate much to get its properties into a reasonable area, then you’re probably going to want to use an orthogonal type linker.

56:36
Otherwise you have this greasy payload out on the end of this long chain like a greasy tetherball and they’re just going to want to glom onto everything.

56:45
Non specific interactions with issues and vivo aggregation with each other.

56:51
If you have a payload where you can manipulate the payload, Astrazeneca’s done a lot of nice work with this things.

56:59
You can lower the C log P of that payload to get into a nice reasonable range.

57:03
Then maybe you want to use a little short linear linker just to nudge its properties into a you know, a right range orthogonal linker might help with properties.

57:12
But you are going to affect the enzyme access to that payload.

57:15
You’re going to affect other things.

57:17
So you kind of have to look at the package deal.

57:19
So I would say if you have a payload that is super hydrophobic, even look at people are putting protax on now you can’t really do much with those.

57:27
Those are firmly in the beyond rule of five category.

57:30
Maybe you want to think about an orthogonal anchor because you can only add so much peg if you can manipulate it add.

57:37
I’ve seen Glycos did some nice stuff with some glucurini pro drugs as well.

57:42
You know, if you can get AC log P down that way, then maybe a linear is just fine.

57:45
Yeah.

57:48
Why is there an increase in hydrophobicity as determined by HIC when the PEG linker is used?

57:55
Dow, do you want me to take that one?

58:00
Yeah, go for it.

58:05
So PEG has an interaction with an HIC column and with the HIC resin.

58:10
And when you were on hydrophobic interaction chromatography.

58:14
So PEG is amphiphilic.

58:16
It’s not actually hydrophilic, right?

58:17
The reason it’s so hydrophilic is each ethylene oxide unit is strongly associated with three molecules of water.

58:25
That’s why it’s hydrophilic.

58:26
When you load a HIC column, you’re in what, one molar, 1, 1/2 molar sodium or ammonium sulfate?

58:34
You’re kind of forcing everything to shed its hydration layers.

58:38
That’s not really an accurate depiction of what happens in V1.

58:43
So especially with PEG when you force it to shed its hydration layer, it interacts with the column and that’s kind of neat reflected on what you see in vivo.

58:52
I’ve seen some interesting studies that actually Agensis did this a while ago.

58:56
They were looking at hydrophobicity by the ANS assay and they could put more and more PEG on there and they actually decreased the hydrophobicity of the ADC by that assay.

59:07
So is it, is it the assay you’re using?

59:09
Is it the starting point?

59:12
So I just don’t think that that maybe HIC chromatography is the best way to, you know, look at the hydrophobicity of that ADC.

59:21
If you’re wondering how it gave in vivo, PEG is just gonna stick to that column in that situation.

59:27
But when you throw it in vivo, I don’t think you’re gonna see those problems.

59:30
No, you won’t.

59:33
Yeah, well, fantastic.

59:35
We’re actually coming to the end of our time, but we’ve had so many really great questions and since we couldn’t get to all of them, the team will be doing their best to get back to everyone who submitted personally after the webinar.

59:51
This webinar has been recorded so you can access the recording within 24 hours just using the same link sent to you earlier.

1:00:02
There is an exit survey, so if you would, please stay logged on just for a quick minute to complete a brief survey.1:00:09
Thank you so much for joining this Fierce Pharma webinar today.

1:00:14
We look forward to seeing you at future events.

Explore BioDesign