The latest advancements driving neuroscience drug discovery research
DDW hosted webinar sponsored by ApconiX
In this webinar the following questions will be answered. Where are the breakthroughs in neuroscience research? How are advanced technologies proving beneficial to drug discovery and development? and how is the current unmet need within drug discovery and development for central nervous system, CNS disorders, providing opportunity?
You will hear from Dr Isaac Klein Chief scientific officer at Dewpoint Therapeutics, Dr Sam Clark CEO of Terran Biosciences and Professor Ruth Roberts Co-founder and Director of Safety Science at ApconiX.
Transcript of the Video
1
00:00:09,169 –> 00:00:09,640
[Music]
2
00:00:09,640 –> 00:00:15,360
Hello and welcome to this webinar titled ‘The latest advancements driving neuroscience drug
3
00:00:15,360 –> 00:00:20,040
discovery research’. My name is Megan Thomas and I’m the multimedia editor for
4
00:00:20,040 –> 00:00:27,720
DDW and your host for today’s webinar which is being sponsored by ApconiX.
5
00:00:27,720 –> 00:00:32,920
In this webinar the following questions will be answered. Where are the breakthroughs in
6
00:00:32,920 –> 00:00:38,480
neuroscience research? How are advanced technologies proving beneficial to drug
7
00:00:38,480 –> 00:00:43,240
discovery and development? and how is the current unmet need within drug discovery
8
00:00:43,240 –> 00:00:49,920
and development for central nervous system, CNS disorders, providing opportunity? You will hear
9
00:00:49,920 –> 00:00:58,560
from Dr Isaac Klein Chief scientific officer at Dewpoint Therapeutics, Dr Sam Clark CEO of Terran
10
00:00:58,560 –> 00:01:06,080
Biosciences and Professor Ruth Roberts Co-founder and Director of Safety Science at ApconiX. Over
00:01:06,080 –> 00:01:11,400
to you Isaac. Thank you for having me. My name is Isaac Klein, I’m the chief scientific officer at
12
00:01:11,400 –> 00:01:16,760
Dewpoint Therapeutics and I’m here today to tell you about biomolecular condensates, the Dewpoint
13
00:01:16,760 –> 00:01:25,720
drug discovery platform and how we can apply it for neuroscience drug discovery. So today I’ll
14
00:01:25,720 –> 00:01:31,320
give you a brief overview of condensate biology and how Dewpoint leverages our understanding of
15
00:01:31,320 –> 00:01:35,960
condensate biology to discover a new class of drugs. I’ll tell you a little bit about how
16
00:01:35,960 –> 00:01:40,760
abnormal condensates drive neuro degenerative diseases and the opportunity that gives us to
17
00:01:40,760 –> 00:01:47,160
find new molecules, that could potentially treat some of these difficult to treat entities and
18
00:01:47,160 –> 00:01:53,200
I’ll give you a snapshot of our ALS program, as well as our red syndrome program at Dewpoint. So,
19
00:01:53,200 –> 00:01:58,800
first a bit about condensate biology. The traditional view of the cell is that it is
20
00:01:58,800 –> 00:02:04,960
compartmentalized and organized by membrane bound organelles, things like the mitochondria, nucleus
00:02:04,960 –> 00:02:10,080
and golgi apparatus that we all know well. But over the last 15 years we’ve come to appreciate
22
00:02:10,080 –> 00:02:15,560
that there’s another level of organization in the cell, which is the condensate and these function
23
00:02:15,560 –> 00:02:21,960
as membraneless organelles, or biomolecules like protein and RNA co-ales and concentrate
24
00:02:21,960 –> 00:02:29,440
together in specified areas, to prosecute certain biochemical pathways or reactions.
25
00:02:30,120 –> 00:02:35,000
We now understand that these structures condensates regulate nearly every cellular
26
00:02:35,000 –> 00:02:40,920
process examined. There’s copious evidence that condensates regulate storage of certain entities
27
00:02:40,920 –> 00:02:47,680
in the cell, gene regulation is performed by condensation, both gene suppression as well as
28
00:02:47,680 –> 00:02:54,160
activation, signal transduction, nuclear import, signaling ribosome biogenesis on and on, are all
29
00:02:54,160 –> 00:02:59,840
cellular processes that are now understood to be regulated by condensates. The mechanisms by which
30
00:02:59,840 –> 00:03:05,520
condensates regulate these processes, shown on the right, are diverse and important for how those
00:03:05,520 –> 00:03:11,560
processes occur. Things like accelerating certain reactions and repressing reactions by sequestering
32
00:03:11,560 –> 00:03:16,680
biomolecules required for them into different compartments, as well as organizing cellular
33
00:03:16,680 –> 00:03:23,800
components such that reactions can take place in an efficient step-wise fashion. At Dewpoint, we use
34
00:03:23,800 –> 00:03:29,000
this new understanding to find condensate targets that are associated with certain diseases and
35
00:03:29,000 –> 00:03:34,720
drive those diseases and some examples are shown here. We have several examples, both internally
36
00:03:34,720 –> 00:03:41,360
at Dewpoint and broadly from the academic community studying this space, where condensates can drive
37
00:03:41,360 –> 00:03:47,240
oncogenic transcription that causes cancer.
We have examples where we find that abnormal
38
00:03:47,240 –> 00:03:53,120
condensates cause a toxic gain of function that
causes a cardiomyopathy. Many examples of abnormal
39
00:03:53,120 –> 00:03:59,360
condensates driving neuro degeneration, as well as condensates driving metabolic abnormalities like
40
00:03:59,360 –> 00:04:05,520
diabetes. Taken together, we find that these abnormal condensates or condensate opathies
00:04:05,520 –> 00:04:13,000
can drive many diseases and give us thousands of new targets to work on across therapeutic areas.
42
00:04:13,000 –> 00:04:18,800
We’ve also found that, once characterized, we can identify small molecules that address and
43
00:04:18,800 –> 00:04:24,280
correct these abnormal condensates that cause disease with profound therapeutic consequences.
44
00:04:24,280 –> 00:04:28,920
Some examples are shown here and the important point I’d like to make is, that the molecules
45
00:04:28,920 –> 00:04:34,360
that we discover against these condensates have entirely new mechanisms of action that could not
46
00:04:34,360 –> 00:04:39,800
be anticipated by traditional drug discovery approaches. In the dilated cardiomyopathy
47
00:04:39,800 –> 00:04:44,240
example I introduced on the last slide, where we’ve identified an abnormal condensate that
48
00:04:44,240 –> 00:04:49,160
causes a toxic gain of function and reduced heart function, we’ve discovered small molecules that
49
00:04:49,160 –> 00:04:54,400
dissolve these abnormal condensates and relieve that toxic gain of function conferred on the
50
00:04:54,400 –> 00:04:59,640
cell by the abnormal condensate. In colorectal cancer, we’ve identified a transcription factor
00:05:00,320 –> 00:05:05,720
that can be forced to condense upon exposure to a small molecule, thus sequestering it away from
52
00:05:05,720 –> 00:05:11,680
the genome and preventing anagen activation. In ALS, which is an example I’ll dive into a bit
53
00:05:11,680 –> 00:05:17,800
more deeply later in the talk, we’ve identified an abnormal cytoplasmic condensate that concentrates
54
00:05:17,800 –> 00:05:24,080
a factor from the nucleus and we can find small molecules that remove or departition that factor
55
00:05:24,080 –> 00:05:29,760
from the abnormal cytoplasmic structure and returning it to the nucleus of the motor neuron,
56
00:05:29,760 –> 00:05:37,280
thus restoring normal function. We’ve built a platform that is an end to end process for
57
00:05:37,280 –> 00:05:42,840
condensate identification and condensate modulator discovery and development, to explore this new
58
00:05:42,840 –> 00:05:47,520
space and take advantage of it to find new drugs for patients with diseases that have
59
00:05:47,520 –> 00:05:53,800
high unmet needs. We start with an identification platform component where we use human genetics to
60
00:05:53,800 –> 00:05:59,640
nominate new condensate targets in silico that drive specific diseases or pathogenic processes.
00:06:00,400 –> 00:06:05,680
We then model those condensates in disease relevant cell lines and using an inhouse
62
00:06:05,680 –> 00:06:11,560
platform and bespoke small molecule library, search for chemistry that changes that condensate
63
00:06:11,560 –> 00:06:16,760
in some desirable way. Using a battery of both condensate specific and functional assays,
64
00:06:16,760 –> 00:06:22,560
we then optimize those condensate modulators into drug-like molecules. We translate them
65
00:06:22,560 –> 00:06:29,560
in downstream models and then ultimately turn out drug candidates that can be brought to the clinic.
66
00:06:30,320 –> 00:06:35,640
The pipeline the companies develop using this platform is shown here. We’re primarily focused
67
00:06:35,640 –> 00:06:41,680
internally on oncology and neuro degeneration. Our lead program in colorectal cancer,
68
00:06:41,680 –> 00:06:47,440
which is looking at beta catene condensation, is set to hit the clinic at the end of 2025.
69
00:06:47,440 –> 00:06:52,960
Our next program is looking at disrupting mick condensation for treatment of ovarian cancer
70
00:06:52,960 –> 00:07:00,240
and that comes about a year later. In neuro degeneration, our pipeline is led by ALS, for
00:07:00,240 –> 00:07:05,480
which we’re selecting a development candidate now and targeting the clinic by the end of 2025. We’ve
72
00:07:05,480 –> 00:07:11,280
launched multiple collaborations with large pharma partners, with Bayer and cardiovascular disease,
73
00:07:11,280 –> 00:07:17,120
with whom we are prosecuting four programs in this space and with Novo Nordisk, in the diabetes and
74
00:07:17,120 –> 00:07:22,560
insulin resistance space, where we have multiple programs now in early discovery. In both cases,
75
00:07:22,560 –> 00:07:27,600
these companies have come to us and asked to use our platform to identify novel condensates that
76
00:07:27,600 –> 00:07:34,720
drive these diseases, as well as develop small molecules that address them for the first time.
77
00:07:34,720 –> 00:07:40,560
Briefly about neuro degeneration and condensates. This is a uniquely interesting space for the
78
00:07:40,560 –> 00:07:46,880
intersection of condensates and drug discovery. There are examples throughout the space of neuro
79
00:07:46,880 –> 00:07:52,400
degeneration, of condensates driving disease and some examples of that are shown here and it’s just
80
00:07:52,400 –> 00:07:59,640
a few. There are prominent examples, for example of condensates driving pathology in Alzheimer’s
00:07:59,640 –> 00:08:05,280
disease, in ALS, in several types of myotonic dystrophy, as well as other triplet expansion
82
00:08:05,280 –> 00:08:12,280
diseases like fuches retinopathy, condensates have been implicated in spinocerebellar ataxias,
83
00:08:12,280 –> 00:08:18,680
in Rett syndrome and many other neuro degenerative diseases as well. In an area, where there is a
84
00:08:18,680 –> 00:08:24,000
paucity of novel treatment approaches for these diseases with high unmet need, this is an exciting
85
00:08:24,000 –> 00:08:29,560
area to identify new targets as well as identify new molecules that can address these targets
86
00:08:29,560 –> 00:08:35,240
for the first time. At Dewpoint specifically, we’ve launched programs throughout this space.
87
00:08:35,240 –> 00:08:39,920
We have a lead program in ALS that I’ll tell you a bit about, as well as a program a bit further
88
00:08:39,920 –> 00:08:46,080
behind in Rett syndrome. We also have programs internally in myotonic dystrophy, Fewets retinopathy,
89
00:08:46,080 –> 00:08:54,600
as well as frontal temporal dementia. So a bit about our ALS program first. ALS is a difficult
90
00:08:54,600 –> 00:09:00,600
to treat disease and it’s caused by a seemingly desperate array of both genetic, environmental
00:09:00,600 –> 00:09:07,320
fact and environmental factors. What we’ve discovered is that nearly all cases of ALS, 97% of
92
00:09:07,320 –> 00:09:14,440
them regardless of the ideology of this disease, are caused by abnormal condensates that are formed
93
00:09:14,440 –> 00:09:20,200
in the cytoplasm of motor neurons and you can see those here. These condensates are specifically
94
00:09:20,200 –> 00:09:27,240
characterized by a concentration of a key splicing factor called tdp43, that in the disease state,
95
00:09:27,240 –> 00:09:31,960
is sequestered in these abnormal condensates,
as opposed to being in a nucleus where it should
96
00:09:31,960 –> 00:09:38,640
normally reside. This causes two things; one, there’s a toxic gain of function by inappropriate
97
00:09:38,640 –> 00:09:43,600
sequestration of this protein in the cytoplasm and two, there’s a loss of function because the
98
00:09:43,600 –> 00:09:49,880
protein is not where it should be in the nucleus. We modeled this in our laboratory and went to
99
00:09:49,880 –> 00:09:56,320
search for small molecules that move tdp43 out of the inappropriate condensate and back to the
100
00:09:56,320 –> 00:10:02,840
nucleus where it should be and that’s exactly what we found, this series of molecules that from which
00:10:02,840 –> 00:10:07,920
we are nominating a development candidate just in the next few months, the consequences of it are
102
00:10:07,920 –> 00:10:16,320
quite profound and shown on the next slide. First, across now 30 IPSC derived motor neuron models
103
00:10:16,320 –> 00:10:23,720
from patients with ALS, we can see a restoration of normal tdp43 function. We can see that in terms
104
00:10:23,720 –> 00:10:30,880
of normal splicing of tdp43 regulated genes like pole dip 3 and we can can see it in the expression
105
00:10:30,880 –> 00:10:38,800
of key ALS associated genes that are also driven
by tdp43 function, like statman 2. When we examine
106
00:10:38,800 –> 00:10:44,880
motor neuron survival as well as degeneration in the laboratory, we can see that our molecule
107
00:10:44,880 –> 00:10:52,440
condensate modulators is able to rescue neuro degeneration of IPSC drive motor neurons from
108
00:10:52,440 –> 00:10:59,240
ALS patients and remarkably in mouse model of ALS, we’re able to eliminate the expression
109
00:10:59,920 –> 00:11:06,080
of neurofilament light chain in the CSF of these mice. Neurofilament light chain is one
110
00:11:06,080 –> 00:11:13,640
of the only clinically validated biomarkers for ALS that is accepted by the FDA on for the basis
00:11:13,640 –> 00:11:19,520
of an approval of a new molecule. So this is particularly exciting as this has not been seen
112
00:11:19,520 –> 00:11:26,440
to our knowledge with any other therapies that are currently on the market. Briefly for Rett
113
00:11:26,440 –> 00:11:31,440
syndrome, I wanted to share some progress we’ve made in this disease. Rett syndrome is caused by
114
00:11:31,440 –> 00:11:37,960
an array of mutations in a protein called mecp2 but our platform nominated this is potentially
115
00:11:37,960 –> 00:11:43,240
being associated with an abnormal condensate behaviour. Indeed what we found, is when this
116
00:11:43,240 –> 00:11:49,840
protein mecp2 is mutated instead of localizing to the heterochromatin where it should be,
117
00:11:49,840 –> 00:11:56,520
it concentrates in a nucleolar condensate where it should not be, and this similar to ALS causes
118
00:11:56,520 –> 00:12:03,560
two things. One, is by abnormally localizing in the nucleolar condensate, it disrupts nucleolar
119
00:12:03,560 –> 00:12:10,840
function and by not being properly located in the heterochromatin compartment, it loses its ability
120
00:12:10,840 –> 00:12:17,440
to regulate normal gene expression. We modeled this in the laboratory and went to look for small
00:12:17,440 –> 00:12:23,720
molecules that reverse this condensate behavior and indeed we found chemistry that can pull the
122
00:12:23,720 –> 00:12:30,320
mutant mecp2 out of the nucleolus and restore its normal localization on the heterochromatin and the
123
00:12:30,320 –> 00:12:37,000
consequences of that molecule in cellular models are shown here. First on the left, you can see
124
00:12:37,000 –> 00:12:45,160
a normalization of mutant mecp2 localization out of the nucleolar compartment on the left. In the
125
00:12:45,160 –> 00:12:51,520
middle slide, you can see that not only is this mutant mecp2 removed from the nucleolus but it
126
00:12:51,520 –> 00:12:57,400
actually restores it, its localization in the heterochromatin is restored and on the right,
127
00:12:57,400 –> 00:13:04,080
you can see that nucleolar function is normalized upon exposure to this condensate modulator,
128
00:13:04,080 –> 00:13:09,840
as examined by phosphorylation of the S6 subunit of the ribosome which
129
00:13:09,840 –> 00:13:18,120
is a key measure of nucleolar function and health. Plenty more to see on our website.
130
00:13:18,120 –> 00:13:22,560
There’s a QR code on the left to the Dewpoint website, which has more information about our lead
00:13:22,560 –> 00:13:29,520
programs, our platform and our company and on the right, is a QR code that links to condensates.com,
132
00:13:29,520 –> 00:13:34,400
which is our educational website. More detail about condensate science, research, updates
133
00:13:34,400 –> 00:13:39,640
from our company in the field, as well as expert insights from those that study this broadly. Thank
134
00:13:39,640 –> 00:13:47,000
you. Thanks very much for that presentation, it was very enlightening. Next up, we’ll hear from
135
00:13:47,000 –> 00:13:54,880
Dr Sam Clark CEO of Terran Biosciences. Over to you Sam. All right, thanks. Today we’re going to
136
00:13:54,880 –> 00:13:59,160
talk about pro drug chemistry, breakthroughs in the schizophrenia space, focusing on how
137
00:13:59,160 –> 00:14:05,400
medicinal chemistry has been utilized to address unmet medical need in drug development and making
138
00:14:05,400 –> 00:14:12,960
better therapeutics in the neuroscience space. Now, first thing is how medicinal chemistry fits
139
00:14:12,960 –> 00:14:18,240
into a drug discovery engine. Typically when we think about drug discovery, we’re thinking
140
00:14:18,240 –> 00:14:25,640
about the development of molecules that have never been discovered before. The issue there is that it
00:14:25,640 –> 00:14:32,760
presents a high risk, in that most molecules are found to be toxic before they ever get into humans
142
00:14:32,760 –> 00:14:40,240
and then most don’t make it through phase one. So very few truly new molecules ever actually get
143
00:14:40,240 –> 00:14:47,320
into the target population with the disease. Then when they do, there’s the question of efficacy and
144
00:14:47,320 –> 00:14:55,200
that’s so why from such a large pool of molecules, so few make it through. There is a second way to
145
00:14:55,200 –> 00:15:02,240
address unmet medical need and that is through improving existing drugs, that have critical
146
00:15:02,240 –> 00:15:07,680
pharmacological issues that limit their efficacy. In this case, you have drugs that are potentially
147
00:15:07,680 –> 00:15:12,800
highly effective, could already address an unmet medical need but are not doing so due
148
00:15:12,800 –> 00:15:19,880
to pharmacokinetic limitations. What that means is that these molecules which may have been approved
149
00:15:19,880 –> 00:15:26,800
before, have problems getting absorbed into the body or metabolized creating either challenges for
150
00:15:26,800 –> 00:15:34,360
patient populations or limiting their efficacy, due to the way they’re metabolized in the body.
00:15:34,360 –> 00:15:42,640
Medicinal chemistry comes in, in it seeks to design and develop molecules that are targeted
152
00:15:42,640 –> 00:15:48,760
towards specific indications or development paths and, in this case, it involves designing, then
153
00:15:48,760 –> 00:15:53,840
synthesizing the molecules in the lab, putting them into animal studies and then eventual GMP
154
00:15:53,840 –> 00:16:02,920
scale up and human studies. So let’s talk about how that plays out. So a pro drug approach has
155
00:16:02,920 –> 00:16:08,120
been utilized to approve some of the leading anti psychotics on the market today and we’ll talk
156
00:16:08,120 –> 00:16:13,120
about the breakthroughs in schizophrenia around that. So first we need to cover what is a pro
157
00:16:13,120 –> 00:16:20,840
drug. Now pro drug is an inactive molecule that’s metabolized into an active form of the drug in the
158
00:16:20,840 –> 00:16:28,240
body and this molecule is engineered to provide advantages over the original drug to overcome
159
00:16:28,240 –> 00:16:34,880
pharmacokinetic limitations and let’s talk about how that’s been applied to the neuro psych space.
160
00:16:34,880 –> 00:16:41,560
Two of the most effective antipsychotics and most widely used on the market, are Aripiprazole which
00:16:41,560 –> 00:16:51,160
was Abilify and paliperidone which is Invega. Now, Abilify was originally invented as a once
162
00:16:51,160 –> 00:16:57,320
daily oral pill and it wasn’t until a number of years later that it was developed into a long
163
00:16:57,320 –> 00:17:04,240
acting injectable antipsychotic. That provides several benefits for patients in that patients
164
00:17:04,240 –> 00:17:10,560
with schizophrenia, often have poor medication compliance and when they are not taking the pills
165
00:17:10,560 –> 00:17:16,800
daily it can lead to reductions in the effective blood levels of the drug and a worsening of the
166
00:17:16,800 –> 00:17:21,920
disease. With diseases like schizophrenia we actually know that antipsychotics are disease
167
00:17:21,920 –> 00:17:28,080
modifying, patients that don’t receive sufficient doses of antipsychotics have a worse course of
168
00:17:28,080 –> 00:17:35,160
the disease over time and worse outcomes. So it’s very critical to have medications that
169
00:17:35,160 –> 00:17:41,720
have pharmacokinetics that work well with these patients and in this case Abilify, Aripripazol,
170
00:17:41,720 –> 00:17:49,480
was later improved via a pro- drug approach into this long acting injectable Aristata. Paliperidone
00:17:49,480 –> 00:17:54,880
similarly was also improved to make a long acting antipsychotic that first lasted 1 month,
172
00:17:54,880 –> 00:17:59,240
then the 3 month was developed and now they’ve developed the six-month injectable.
173
00:17:59,240 –> 00:18:05,680
So imagine instead of having to take pills every day, the patients just come in for a once,
174
00:18:06,360 –> 00:18:13,200
twice a year injection. So let’s just talk about that a little bit. The way they did that was,
175
00:18:13,200 –> 00:18:20,560
they attached pro drug moieties and what these are, are inactive esters on the side here with
176
00:18:20,560 –> 00:18:28,880
long carbon chains and this the palmitate esther here and this is the lauroxil esther here. Those
177
00:18:28,880 –> 00:18:35,960
provide lipophilicity to the molecules enabling them to be turned into long acting injectables and
178
00:18:35,960 –> 00:18:43,720
thus last a long time in the intramuscular Depot injection. There’s a second big advantage where
179
00:18:43,720 –> 00:18:51,880
pro drugs can address an unmet medical need and that is that they are able to take advantage of
180
00:18:51,880 –> 00:18:59,360
the FDA’s accelerated regulatory pathways. So for example, with Aristata, they were able to utilize
00:18:59,360 –> 00:19:07,840
the 505 B2 pathway and what that means is, instead of having to go all the way back to phase 2 phase,
182
00:19:07,840 –> 00:19:13,200
multiple phase three efficacy studies, which can take many years to get a drug developed and all
183
00:19:13,200 –> 00:19:20,440
the while patients are waiting, the FDA recognizes that pro drugs which are ultimately releasing the
184
00:19:20,440 –> 00:19:26,680
active molecule that was already shown to be safe
and effective. In this case, for Aripripazole laroxil, the
185
00:19:26,680 –> 00:19:33,280
active molecule had previously been approved as Abilify and so what the FDA requires for these
186
00:19:33,280 –> 00:19:40,360
improved versions, is pharmacokinetic bridging studies, to bridge the data to the old effective
187
00:19:40,360 –> 00:19:46,640
levels. This represents a major breakthrough for patients because patients with these pathways,
188
00:19:46,640 –> 00:19:52,200
you can get the drugs approved much faster with minor bridging studies, that are quick to do and
189
00:19:52,200 –> 00:19:57,520
often just one pivotal bridging study is all that’s required. In that sense, it’s a big
190
00:19:57,520 –> 00:20:02,920
win for patients because they don’t have to wait all of the extra years that are typically in play
00:20:02,920 –> 00:20:10,280
for a full phase 3 development program. Now let’s talk about one of the biggest breakthroughs right
192
00:20:10,280 –> 00:20:19,320
now in the schizophrenia space a new drug, first new mechanism in 50 years which is called KarXT,
193
00:20:19,320 –> 00:20:25,160
it’s a combination of these two molecules, xanomeline and trospium. Now the reason why this is such a big
194
00:20:25,160 –> 00:20:33,760
breakthrough is that xanomeline has a new mechanism of action, in fact it represents the first new
195
00:20:33,760 –> 00:20:42,120
mechanism of antipsychotics in over 50 years. All of the current antipsychotics on the market either
196
00:20:42,120 –> 00:20:48,360
directly affect the dopamine type 2 receptor and sometimes they also affect the serotonin 2A
197
00:20:48,360 –> 00:20:56,200
receptor and they work by blocking those receptors, but that mechanism has been rehashed over multiple
198
00:20:56,200 –> 00:21:01,960
different antipsychotics with different degrees of potency and binding over the last 50 years.
199
00:21:01,960 –> 00:21:09,320
Xanomeline acts on a completely different receptor system, the muscarinic receptor system, and in
200
00:21:09,320 –> 00:21:16,200
doing so that represents a big step forward, the first new way to treat psychosis and
00:21:16,200 –> 00:21:25,280
schizophrenia in half a century. However, xanomeline has some side effects because it activates the muscarinic
202
00:21:25,280 –> 00:21:31,000
receptors, which are beneficial in the brain for
patients with schizophrenia, it’s also activating them in
203
00:21:31,000 –> 00:21:37,000
the peripheral nervous system; in the stomach, in the intestines and activation of the muscarinic
204
00:21:37,000 –> 00:21:45,840
system there can cause problems like diarrhea, nausea, vomiting, sweating and so it’s been combined
205
00:21:45,840 –> 00:21:51,800
with another drug trospium. Trospium blocks the muscarinic receptors but it doesn’t cross the
206
00:21:51,800 –> 00:21:58,240
blood brain barrier and so the combination of the two represents a really unique mechanism where
207
00:21:58,240 –> 00:22:04,800
xanomeline is able to activate the receptors in the brain helping patients with schizophrenia. In trospium
208
00:22:04,800 –> 00:22:10,800
is able to block those receptors in the rest of
the body reducing side effects. Xanomeline has also
209
00:22:10,800 –> 00:22:16,080
previous been in trials where it’s shown efficacy in Alzheimer’s and other conditions so it really
210
00:22:16,080 –> 00:22:24,080
represents a paradigm shift for psych disorders. However, there is an issue with xanomeline and trospium
00:22:24,080 –> 00:22:30,440
and well first, before we’ll go into that, we’ll talk about that as part of the clinical studies.
212
00:22:30,440 –> 00:22:36,560
It’s been through three phase three clinical trials, shown a big reduction in schizophrenia
213
00:22:36,560 –> 00:22:43,600
symptoms and it’s up for potential approval
just next month in September. So very exciting.
214
00:22:43,600 –> 00:22:50,000
The downside is for all of these trials they have to use a twice daily oral formulation. Now most
215
00:22:50,000 –> 00:22:56,440
antipsychotics are at least once daily so to move to a twice daily represents a potential burden
216
00:22:56,440 –> 00:23:04,880
for patients. Now, with twice daily, the reason why is because the drug has issues with absorption and so
217
00:23:04,880 –> 00:23:11,760
the pharmacokinetic limitations are because these two molecules are very old, for example xanomeline
218
00:23:11,760 –> 00:23:19,160
was originally invented in the 1990s and trospium was invented in the 1960s, so as old molecules as
219
00:23:19,160 –> 00:23:25,480
you can imagine, they have not the pharmokinetics has not have not been improved since then. However,
220
00:23:25,480 –> 00:23:31,440
it still represents a big breakthrough and addressing an unmet medical need for a new
00:23:31,440 –> 00:23:36,160
mechanism that treats schizophrenia, without the side effects that typically are associated with
222
00:23:36,160 –> 00:23:42,720
classical anti-psychotics which can cause parkinsonian like symptoms in patients. Xanomeline and trospium don’t
223
00:23:42,720 –> 00:23:48,280
have those effects so it’s a big breakthrough for patients but it comes with a downside of a
224
00:23:48,280 –> 00:23:54,600
relatively, potentially difficult dosing paradigm, of having to take two pills a day every day to
225
00:23:54,600 –> 00:23:59,840
maintain efficacy and in a patient population like schizophrenia, where there are other
226
00:23:59,840 –> 00:24:06,200
options like long acting injectables, this poses a potential problem. Now, if we think about classic
227
00:24:06,200 –> 00:24:12,200
drug development such as you know developing an entirely new molecule, that could make patients
228
00:24:12,200 –> 00:24:20,520
wait a long time for a new therapeutic and it also comes with the potential for less efficacy. So far
229
00:24:20,520 –> 00:24:26,600
other muscarinic drugs in the space have not been able to reach the same efficacy and right now
230
00:24:26,600 –> 00:24:33,120
there are no other muscarinic close to approval but in clinical trials other muscarinic drugs, that
00:24:33,120 –> 00:24:40,280
have shown efficacy, have not matched the level of efficacy of xanomeline and trospium, the karXT
232
00:24:40,280 –> 00:24:47,800
efficacy. So, as you can see, sometimes developing a new drug isn’t the best option. So this is an
233
00:24:47,800 –> 00:24:55,600
example where medicinal chemistry can address that large unmet medical need and for example, creating
234
00:24:55,600 –> 00:25:09,960
pro drugs of xanomeline and trospium to fix the absorption issue and thus create a once daily form. This is where medicinal chemistry has been utilized in, for example, a terXT where they’ve
235
00:25:09,960 –> 00:25:15,640
created the decanoate pro drugs, here is what I’m showing you, where just similar to what we saw with
236
00:25:15,640 –> 00:25:25,760
Invega or Aristata, attaching fatty acids to the molecules it changes the properties and then once
237
00:25:25,760 –> 00:25:33,520
in the body the that body’s natural enzymes will cleave these, allowing normal xanomeline and trospium to
238
00:25:33,520 –> 00:25:38,640
produce its effects. That’s very exciting for patients because in that sense you can get the
239
00:25:38,640 –> 00:25:45,320
best of both worlds. You can get the better dosing regimen, the once daily oral and the long acting
240
00:25:45,320 –> 00:25:52,720
multi-month injectable, but also have the improved
241
00:25:52,720 –> 00:26:00,400
matched by any other more selective muscarinic drug in development. Just to illustrate
242
00:26:00,400 –> 00:26:08,080
what typically goes into a comprehensive prodrug program, these are very in-depth designed for terXT,
243
00:26:08,080 –> 00:26:13,880
involved more than 10,000 pro drugs, synthesizing over a thousand pro drugs with more than 200
244
00:26:13,880 –> 00:26:21,360
full-time chemists. Then in vitro studies, more than 500 of those and another 500 in vivo studies and
245
00:26:21,360 –> 00:26:29,600
PK modeling. So medicinal chemistry programs are no less intensive than classic novel drug discovery
246
00:26:29,600 –> 00:26:38,160
programs but they can result in enormous benefits for patients, cutting the weight time down by a
247
00:26:38,160 –> 00:26:44,280
large degree for patients waiting for new drugs and improving drugs that have efficacy and fixing
248
00:26:44,280 –> 00:26:52,680
the problems that may limit their use and that basically sums it up. Ready for any questions.
249
00:26:52,680 –> 00:26:58,400
Thank you for that presentation Sam. Next we’ll hear from Professor Ruth Roberts, Co-founder
250
00:26:58,400 –> 00:27:06,000
you Ruth. Thank you for the introduction and today
251
00:27:06,000 –> 00:27:12,400
I’m going to talk to you about iSLA, which is an integrated in vitro seizure liability assay which
252
00:27:12,400 –> 00:27:21,160
uses human induced pluripotent stem cell neurones or hiPSC and human ion channels. So let’s look
253
00:27:21,160 –> 00:27:26,360
at the problem we’re working on today. In drug discovery and drug development, as many of you
254
00:27:26,360 –> 00:27:32,520
will already know, safety is the main reason why drugs fail and you can look at this excellent
255
00:27:32,520 –> 00:27:39,240
paper from Cook et al in nature reviews, that looks at the AstraZeneca data set, showing that safety in beige
256
00:27:39,240 –> 00:27:45,880
brown there, is the main reason for failure. When we dig into the data further on in that paper,
257
00:27:45,880 –> 00:27:54,080
you can see that CNS arrowed here is top of the clinical failure table. So the main reason drugs
258
00:27:54,080 –> 00:28:01,120
fail in the clinic is due to CNS issues. It’s also an important reason for failure preclinically
259
00:28:01,120 –> 00:28:08,240
as we can see here and of those CNS failures, a significant number of those, are due to seizure
260
00:28:08,240 –> 00:28:12,720
which is unwanted electrical activity in the brain that crops up during the pre-clinical
261
00:28:12,720 –> 00:28:20,320
animal testing or even more devastatingly in clinical development as we can see here. So why
262
00:28:20,320 –> 00:28:26,520
does this happen? Well essentially our current approaches for screening for seizure don’t work
263
00:28:26,520 –> 00:28:33,880
and this is a rough schematic adapted from Alison Easter’s paper from 2009, where we can see during
264
00:28:33,880 –> 00:28:40,480
the important GLP toxicology phase in the center of the screen, that precedes first time in humans,
265
00:28:40,480 –> 00:28:48,200
we conduct a standard regulatory package. Within that, we can also look at rodent behavior and EEG
266
00:28:48,200 –> 00:28:55,520
to confirm any seizure type activities that may have been seen in the animal testing in GLP tox
267
00:28:55,520 –> 00:29:02,760
and back in 2009, Alison’s paper was suggesting that we could screen to prevent these seizures
268
00:29:02,760 –> 00:29:09,040
as a pre-screen by using rat blown brain slicers or perhaps zebra fish assays, to avoid going
269
00:29:09,040 –> 00:29:16,120
into animal testing with compounds that had the potential to cause these seizure like activities.
00:29:16,120 –> 00:29:21,640
Those tests were introduced but in my view they haven’t really made that much of a difference to
271
00:29:21,640 –> 00:29:28,040
the failure rate as you can see. So we really need to improve preclinical seizure screening
272
00:29:29,520 –> 00:29:35,440
because essentially it currently, it’s very hard to detect, the current methods are very low throughput,
273
00:29:35,440 –> 00:29:41,840
they’re expensive and if they’re done in rats and dogs or in rat brain slices or zebra fish, we have
274
00:29:41,840 –> 00:29:48,120
poor translation to humans. So we thought it was a good time to rethink and redesign our approach to
275
00:29:48,120 –> 00:29:56,920
screening for seizure. How did we do this? Well we started by putting an idea to an SOT workshop way
276
00:29:56,920 –> 00:30:05,560
back in 2019 co-chaired by Jennifer Pearson from HESI and described in our paper from 2019.
277
00:30:05,560 –> 00:30:11,280
We put to this committee, or this workshop, could we actually approach screening for seizure based
278
00:30:11,280 –> 00:30:17,080
on human cellular models and human ion channels and this was discussed at this workshop in great
279
00:30:17,080 –> 00:30:26,280
detail and subsequently published in these papers you see here. What we put to this workshop was, we
00:30:26,280 –> 00:30:34,120
could screen for seizure using a micro electrode array abbreviated to MEA. This describes the MEA
281
00:30:34,120 –> 00:30:40,000
assay, so on the left hand side you’re looking at neurons plated in the micro electrode array plates
282
00:30:40,000 –> 00:30:46,920
shown here and each plate has a number of wells and each well has 16 electrodes in the bottom of
283
00:30:46,920 –> 00:30:54,040
the well. We then look at electrical traces shown here, from each of the 16 electrodes numbered 1
284
00:30:54,040 –> 00:31:02,480
to 16 on the day five pattern and over the time course between day five to day 23, these plated
285
00:31:02,480 –> 00:31:09,720
human neurons settle down into a nice harmonous, peaceful firing pattern as you can see in day
286
00:31:09,720 –> 00:31:15,760
- So they need to make connections in order to behave like normal neurons but by day 23, we’re
287
00:31:15,760 –> 00:31:23,880
seeing a nice normal firing pattern. We then tested these neurons with a variety of compounds known to
288
00:31:23,880 –> 00:31:30,840
cause seizure. shown on the left hand side. So as you can see in this graph, we’re looking at the
289
00:31:30,840 –> 00:31:36,920
alterations in the micro electro parameters with the number of seizurogenic compounds compared to,
00:31:36,920 –> 00:31:43,720
at the bottom the negative control paracetamol or acetamenophen. We see a whole different range
291
00:31:43,720 –> 00:31:51,280
of impacts on the MEA parameters with compounds known to cause seizure. So essentially in this
292
00:31:51,280 –> 00:31:58,160
human neuronal assay in vitro, the majority of seizuregenic compounds increase network burst frequencies,
293
00:31:58,160 –> 00:32:04,200
decreased duration and decrease the number of spikes per network bursts. So we’re able to
294
00:32:04,200 –> 00:32:09,080
see the phenotypic manifestation of seizure in this human invitro
295
00:32:09,080 –> 00:32:17,760
assay. As well as looking at a phenotypic assay, which is human neurons, we wanted to also look at ion
296
00:32:17,760 –> 00:32:23,240
channels because we know they’re associated with seizure. So we know there are variety of
297
00:32:23,240 –> 00:32:29,840
epilepsies there are human ion channel mutations, so we listed all of the potential ion channels
298
00:32:29,840 –> 00:32:35,000
that could be associated with seizure and as you can see at the bottom, we started with hundreds
299
00:32:35,000 –> 00:32:40,760
of ion channels and then based on expression profile literature on their function and the
00:32:40,760 –> 00:32:48,640
pharmacology associated with them, we narrowed it down to 15 seizure related ion channels. This is
301
00:32:48,640 –> 00:32:54,400
the same compound you saw on the MEA data across the top but down the left hand side are our 15
302
00:32:54,400 –> 00:33:00,920
ion channels associated with seizure. What you can can see here, is with CNS active therapies
303
00:33:00,920 –> 00:33:07,440
associated with seizure particularly like, for example amoxapine chlorpromazine, you see a high
304
00:33:07,440 –> 00:33:14,200
impact in other words inhibition at under 30 micromolar of signaling at these ion channels
305
00:33:14,200 –> 00:33:20,480
which is associated with seizure. So we now have a really elegant assay where we’re able to see the
306
00:33:20,480 –> 00:33:26,960
seizurogenic phenotype and induce human pluripotent stem cell neurons in vitro and then we can look at
307
00:33:26,960 –> 00:33:32,920
the mechanistic basis of that seizure, looking at the different ion channels implicated in seizure
308
00:33:32,920 –> 00:33:39,400
and looking at the profile of each compound against those ion channels. So we published
309
00:33:39,400 –> 00:33:45,800
this work back in 2023, where we described this integrated approach. All the data I’ve shared
00:33:45,800 –> 00:33:50,760
with you today are in this paper. Please feel free to take a look or email me and I can send you the
311
00:33:50,760 –> 00:33:57,240
paper and we were very excited, on the back of this paper, when it was mentioned in a publication from
312
00:33:57,240 –> 00:34:03,200
the FDA, where they were looking at new approach methodologies in the nonclinical assessment of
313
00:34:03,200 –> 00:34:10,560
pharmaceuticals and our paper is cited in this paper from FDA, Avira et al, as a new approach methodology
314
00:34:10,560 –> 00:34:16,600
with a context of use around screening for seizure. So please take a look at that paper too, it’s very
315
00:34:16,600 –> 00:34:25,520
interesting reading. So what we’re saying is, that iSLA, the invitro seizure liability assay, is an
316
00:34:25,520 –> 00:34:31,000
innovative new approach to screen for seizure and essentially what we’re using it to do, is to
317
00:34:31,000 –> 00:34:38,720
replace the in vitro or ex vivo models here at brain slice and zebra fish, to screen very early
318
00:34:38,720 –> 00:34:44,920
for the compound’s ability to potentially cause seizure, but can also be used if seizures
319
00:34:44,920 –> 00:34:50,840
are seen in animal studies in the GLP tox phase or in humans in the clinic, to go back and look
00:34:50,840 –> 00:34:57,160
at the mechanistic basis of those potential seizures, looking at ion channels. So we’re really
321
00:34:57,160 –> 00:35:04,120
saying that this essay has got multiple uses in understanding seizure and it really is a complete
322
00:35:04,120 –> 00:35:09,800
redesign of our thinking. We’re not trying to replace a particular animal test with an invitro
323
00:35:09,800 –> 00:35:16,360
assay. We’re saying let’s step back, look at what we’re trying to do to ensure patient and volunteer
324
00:35:16,360 –> 00:35:23,040
safety and then design something that’s human specific, that can be translated directly to human
325
00:35:23,040 –> 00:35:31,760
risk benefit in the clinic. So how is iSLA used? As I’ve already alluded to, there are two main modes
326
00:35:31,760 –> 00:35:38,280
of use of iSLA. The first is the proactive which is the predictive, in other words we’re looking
327
00:35:38,280 –> 00:35:45,560
at MEAs for a chemical series to prioritize which is the more safe series regarding seizure or early
328
00:35:45,560 –> 00:35:51,720
hazard identification and redesigning compounds that may have a seizurogenic potential, via
329
00:35:51,720 –> 00:35:57,760
interaction at an ion channel and then later on,
as I already alluded to, when issues have been seen
00:35:57,760 –> 00:36:03,160
either in the animal tests or in clinical trials, we can do an MEA study to understand the human
331
00:36:03,160 –> 00:36:08,440
relevance of the seizures seen in animals and an ion channel panel to then understand the molecular
332
00:36:08,440 –> 00:36:15,720
basis. I have a couple of case studies to share with you which are very recent data of work
333
00:36:15,720 –> 00:36:22,840
we’ve been doing with clients. In this example, there was a very difficult situation where the
334
00:36:22,840 –> 00:36:31,920
compound caused convulsions in the dog but not in other preclinical species. So was the convulsion
335
00:36:31,920 –> 00:36:37,320
seen in the dog relevant for humans is of course the really important question. When we looked at
336
00:36:37,320 –> 00:36:43,840
this on the MEA assay, there was a dog specific metabolite that caused seizurogenic phenotype. This
337
00:36:43,840 –> 00:36:49,800
metabolite was not present in the rat and is not present in humans, so it looks like this is dog
338
00:36:49,800 –> 00:36:56,600
specific. Then when we went to an ion channel panel, we saw a hit on an ion channel panel that
339
00:36:56,600 –> 00:37:03,160
gave us mechanistic insight into the molecular basis of the dog specific seizures that were seen with
00:37:03,160 –> 00:37:10,640
this compound and this data accompanied the IND, when that was submitted shown here. In the second
341
00:37:10,640 –> 00:37:17,040
example, there was seizure seen with a chemical series and the challenge here was, we need to
342
00:37:17,040 –> 00:37:24,400
identify a chemical series with low seizurogenic risk, so we see here chemical series through A, B, C, D
343
00:37:24,400 –> 00:37:32,960
and E and when we looked in MEA, we can see here that series A had a low risk whereas series E had
344
00:37:32,960 –> 00:37:39,040
a high risk and then a smattering of scores in between the two. So we guided this client towards
345
00:37:39,040 –> 00:37:46,440
series A, which had distinct structural features from the other compounds being considered in lead
346
00:37:46,440 –> 00:37:53,080
optimization. So we were able to design from the very beginning, a compound with low seizurogenic risk
347
00:37:53,080 –> 00:37:57,800
compared to those that may have a risk in either preclinical or clinical development,
348
00:37:57,800 –> 00:38:03,080
which is a really great outcome and saves a lot of time and effort later on, trying to understand
349
00:38:03,080 –> 00:38:10,800
issues that can be avoided. So to conclude what I’ve shared with you today, seizures remain an
00:38:10,800 –> 00:38:16,480
issue in preclinical and clinical development. So we still see a high incidence of seizures
351
00:38:16,480 –> 00:38:25,400
in animal tests in the geop animal studies and then also in the clinic. iSLA uses human induced
352
00:38:25,400 –> 00:38:32,880
pluripotent stem cell neurons and micro electrode array in a bespoke ion channel pane,l to detect
353
00:38:32,880 –> 00:38:38,880
and understand the molecular basis of seizures when they occur or to predict and prevent them. We
354
00:38:38,880 –> 00:38:44,800
have great translation to humans because these are human cells and human ion channels. As well, we
355
00:38:44,800 –> 00:38:50,720
have really great three R’s benefits so it’s been an interesting discussion, I have been asked which
356
00:38:50,720 –> 00:38:57,560
test are we replacing? Are we reducing animal use or are we replacing animal use? Yes we are in a sense,
357
00:38:57,560 –> 00:39:04,960
that only those compounds that go forward into GLP toxicology testing in animals, have only
358
00:39:04,960 –> 00:39:10,360
those compounds that have been already screened for seizure liability will progress so we’re not
359
00:39:10,360 –> 00:39:15,200
wasting valuable resources and time testing compounds that we could have predicted, would
00:39:15,200 –> 00:39:21,600
not make it through to the clinic. As I’ve already described, this work has been used for the
361
00:39:21,600 –> 00:39:28,440
proactive mode which is the ideal place to be, to design compounds, select between the lead series and
362
00:39:28,440 –> 00:39:34,920
optimized molecules and then, in problem solving where we are looking to explain observed
363
00:39:34,920 –> 00:39:42,960
seizures and their mechanistic basis. Finally I’d like to acknowledge the team who were behind this
364
00:39:42,960 –> 00:39:49,160
work. Here we see the whole ApconiX team at a recent science day, top left, particularly I’d like to
365
00:39:49,160 –> 00:39:55,720
acknowledge the ion channel lab and MEA team led by Mike Morton. The work was done primarily by
366
00:39:55,720 –> 00:40:00,720
Kim Rockley, shown bottom left and also the rest of the lab team; Karen, Emily, Magalie,
367
00:40:00,720 –> 00:40:07,080
Hannah, Beth, Katie and Louisa, all of whom have contributed to this extensively. I’d also like
368
00:40:07,080 –> 00:40:12,240
to acknowledge our collaborators Bristol Meers Squib, who provided some of the rat brain slice
369
00:40:12,240 –> 00:40:19,160
data for our comparisons and Axion Biosciences, who provided the maestro machine that the MEA
00:40:19,160 –> 00:40:24,560
recordings were done in and Fujifilm Cellular Dynamics for the human induced pluripotent stem
371
00:40:24,560 –> 00:40:30,880
cell neurones. And finally, just a happy to say we won the Kings Award recently. So those lovely
372
00:40:30,880 –> 00:40:36,640
pictures top right, is me smiling away at Windsor Castle going into receive The King’s Award for
373
00:40:36,640 –> 00:40:43,080
the great work we’ve done at ApconiX so thank you. Thank you for that presentation Ruth and
374
00:40:43,080 –> 00:40:48,240
thank you to everyone who has presented today. Before we continue to the Q&A,
375
00:40:48,240 –> 00:40:53,520
I’d just like to remind everyone that this webinar will be available to access on demand on the DDW
376
00:40:53,520 –> 00:41:03,560
website www.ddw-online.com. We’ll now move on to the questions. Isaac, this first question is
377
00:41:03,560 –> 00:41:10,440
for you. Can the same condensate drive more than one disease, can cmods developed for a particular
378
00:41:10,440 –> 00:41:15,880
target be used in other diseases? This is one of the reasons we’re excited about condensates
379
00:41:15,880 –> 00:41:21,720
for drug discovery broadly and ALS actually provides a really nice example of this. So
00:41:21,720 –> 00:41:30,000
abnormal TDP 43 condensates in the cytoplasm, well established to dry almost 100% of ALS cases, but we
381
00:41:30,000 –> 00:41:35,640
also now understand that it drives about half cases of frontal temporal dementia and may also
382
00:41:35,640 –> 00:41:41,480
be associated and drive other neurodegenerative diseases, like chronic traumatic brain injury and
383
00:41:41,480 –> 00:41:47,720
even a certain percentage of patients diagnosed with Alzheimer’s. So we’ve already shown in
384
00:41:47,720 –> 00:41:52,920
cellular models, that our molecule can be potentially used to treat frontal temporal
385
00:41:52,920 –> 00:42:00,320
dementia. We’re exploring other indications now and this pattern of one condensate driving multiple
386
00:42:00,320 –> 00:42:07,080
diseases is also seen in the oncology space and so short answer is, yes a molecule discovered
387
00:42:07,080 –> 00:42:12,520
by our platform for a particular condensate, could potentially be applied to other diseases
388
00:42:12,520 –> 00:42:18,880
driven by the same structure. Thanks for that answer. Isaac, the next question I have here is,
389
00:42:18,880 –> 00:42:26,000
can condensates be used as biomarkers and can they be used to identify patient populations that would
00:42:26,000 –> 00:42:33,520
benefit from treatment with a specific cmod? Yes, so one of the many advantages of condensate based
391
00:42:33,520 –> 00:42:39,120
drug discovery, is that they’re highly translatable entities. ALS here provides another outstanding
392
00:42:39,120 –> 00:42:44,920
example, where the same condensate abnormality can be seen in cell line models, animal models as well
393
00:42:44,920 –> 00:42:50,400
as in patient biopsies. So that provides us a nice chain of translatability all the way from
394
00:42:50,400 –> 00:42:58,720
discovery to the patient. In terms of biomarker use, that is certainly an area that we’re excited
395
00:42:58,720 –> 00:43:04,360
about. We’ve already demonstrated that condensates can be used as biomarkers or specifically in terms
396
00:43:04,360 –> 00:43:11,000
of patient selection, both in the neurodegenerative space as well as in the oncology space. Also
397
00:43:11,000 –> 00:43:16,240
importantly the downstream consequences of abnormal condensate function, once understood, can
398
00:43:16,240 –> 00:43:22,320
also be used as biomarkers for patient selection and potentially even to monitor the response to
399
00:43:22,320 –> 00:43:28,240
therapy with a condensate modulator. That was a great answer Isaac thank you. The last question
00:43:28,240 –> 00:43:33,880
I have for you here is, what role do condensates play in more common neurodegenerative diseases
401
00:43:33,880 –> 00:43:40,400
like Alzheimer’s or Parkinson’s? Could the Dewpoint platform address these needs? Absolutely,
402
00:43:40,400 –> 00:43:45,720
they do play a role in those more common diseases like Alzheimer’s disease, Parkinson’s disease and
403
00:43:45,720 –> 00:43:50,800
others. We simply haven’t tackled them because there are areas that aren’t quite bite-sized
404
00:43:50,800 –> 00:43:56,200
for a small biotech to tackle from a development standpoint but we’re excited to, once we reach that
405
00:43:56,200 –> 00:44:00,840
stage and certainly if there were any partners, that are interested in leveraging our platform
406
00:44:00,840 –> 00:44:06,160
to find and develop small molecules for those larger indications, we’d be very excited to tackle
407
00:44:06,160 –> 00:44:13,120
them sooner than we had anticipated. That’s great Isaac thank you for that answer. Sam I now have a
408
00:44:13,120 –> 00:44:20,440
question for you. How does a pro drug approach to improving existing therapeutics compare to
409
00:44:20,440 –> 00:44:29,280
designing new therapeutic molecules? So when you take a pro drug approach you’re very much focused
00:44:29,280 –> 00:44:36,520
on one target that you already know. So you have a drug that you already know works and you already
411
00:44:36,520 –> 00:44:42,320
know that drug is safe, so those are two massive benefits. There are big benefits for patients
412
00:44:42,320 –> 00:44:47,920
in addressing an unmet medical need. In a classic drug discovery program you’re starting with an
413
00:44:47,920 –> 00:44:55,960
unknown. Perhaps you used modelling to try to design a number of new structures or classically you know
414
00:44:55,960 –> 00:45:01,480
had people come and design them those by hand. But either way, you’re working with new structures that
415
00:45:01,480 –> 00:45:07,760
have never been tested in humans, have never been shown to be safe and more than 90% get knocked out
416
00:45:07,760 –> 00:45:14,240
before phase one and/or post phase one and then by then, you still have the efficacy issue. Even
417
00:45:14,240 –> 00:45:18,920
if they’re found to eventually be safe and most aren’t, then the question of whether they’ll be
418
00:45:18,920 –> 00:45:24,000
effective is a big question. That’s what we’ve seen with the newer muscarinic, this a normally was
419
00:45:24,000 –> 00:45:30,440
invented in the 1990s but it’s still, to date, the most effective muscarinic in clinical trials so
00:45:30,440 –> 00:45:36,440
far to treat schizophrenia. The newer ones developed since then, in the 2000s and the last
421
00:45:36,440 –> 00:45:42,720
decade, have not been able to match the efficacy of the original. So with medicinal chemistry you start
422
00:45:42,720 –> 00:45:49,240
with the molecule you know is safe and effective and then you just fix a few problems on it to
423
00:45:49,240 –> 00:45:55,280
improve its pharmacokinetics, its absorption and potentially even reduce side effects. So it’s very
424
00:45:55,280 –> 00:46:01,800
exciting by starting with a known component. However similar to classic drug development, it still can
425
00:46:01,800 –> 00:46:09,240
be extremely in-depth program involving thousands of designs. Thanks for that answer Sam. The next
426
00:46:09,240 –> 00:46:14,240
question I have here for you is, how would changing the pharmacokinetics of previously
427
00:46:14,240 –> 00:46:21,040
approved therapeutics contribute to addressing the unmet medical need for CNS disorders? So
428
00:46:21,040 –> 00:46:27,240
there’s actually a couple ways. One of which that we talked about earlier in the talk, was
429
00:46:27,240 –> 00:46:31,800
simply through improving the dosing paradigm. With patients like schizophrenia or patients with
00:46:31,800 –> 00:46:39,840
Alzheimer’s disease, taking pills multiple times a day represents a burden and a potential for
431
00:46:39,840 –> 00:46:47,520
where these the patients may not be compliant with the medications and thus not get the amount
432
00:46:47,520 –> 00:46:53,120
of the drug that’s effective. With a pro drug approach, you can improve the pharmacokinetics
433
00:46:53,120 –> 00:47:00,640
and thus improve the experience for patients and ease of dosing and thus the potential outcome of
434
00:47:00,640 –> 00:47:05,920
taking that medication. A second way, though that we haven’t gotten into as much, is that you can
435
00:47:05,920 –> 00:47:12,360
potentially reduce side effects by changing the pharmacokinetics. Many side effects of drugs are
436
00:47:12,360 –> 00:47:18,280
often of immediate release drugs the ones that are not on an extended release, are often due
437
00:47:18,280 –> 00:47:25,360
to the spikes in plasma concentrations due to the immediate effect of the drug. Now with a pro
438
00:47:25,360 –> 00:47:32,400
drug approach enabling longer acting formulations, it smooths out the pharmacokinetic profile of the
439
00:47:32,400 –> 00:47:38,680
drug, reducing plasma spikes that could cause side effects and thus potentially improving things like
00:47:38,680 –> 00:47:45,400
nausea, vomiting and diarrhea and other spike related incidences, with a nice smooth long-
441
00:47:45,400 –> 00:47:51,200
acting profile. It further addressing the unmet medical need by providing patients with drugs
442
00:47:51,200 –> 00:47:57,120
with better safety profiles as well. So in this sense, we have two ways that medicinal chemistry
443
00:47:57,120 –> 00:48:05,320
and thus the pro drug approach has been used to really address this unmet medical need, with on a
444
00:48:05,320 –> 00:48:11,600
fast timeline using the FDA’s accelerated pathway and pro drugs now have been used to improve many
445
00:48:11,600 –> 00:48:20,120
neuroscience drugs to date, such as, for example, a Gabapentin, other ones and as well as,
446
00:48:20,120 –> 00:48:26,760
drugs in the antibiotic space and across all the different systems. So it’s a widely used strategy
447
00:48:27,360 –> 00:48:34,040
that results in big breakthroughs for patients. Thanks Sam. A final question for you here. What
448
00:48:34,040 –> 00:48:41,440
role does medicinal chemistry play in advancing drug discovery? With medicinal chemistry it’s
449
00:48:41,440 –> 00:48:48,920
not just making small modifications, it those modifications that shed a lot of light on structure
00:48:48,920 –> 00:48:54,960
activity relationship and that’s help push the field forward in the development of understanding
451
00:48:54,960 –> 00:49:02,400
how structural modifications to molecules, can create large improvements in those molecules.
452
00:49:02,400 –> 00:49:08,720
So, in addition medicinal chemistry often utilizes a lot of novel synthetic roots
453
00:49:08,720 –> 00:49:15,160
to make these molecules in a way that would be safe for human consumption and often GMP
454
00:49:15,160 –> 00:49:20,400
processes and so medicinal chemistry in addition to improving the molecules is also pushing forward
455
00:49:20,400 –> 00:49:27,080
the chemical field and our depth of what’s possible in chemistry because there may be
456
00:49:27,080 –> 00:49:30,840
molecules out there that could potentially be improved by pro drugs but no one knows
457
00:49:30,840 –> 00:49:35,680
how because they’re considered difficult molecules. With new approaches in medicinal
458
00:49:35,680 –> 00:49:41,480
chemistry’ been able to improve upon that and thus widen in the space and this has
459
00:49:41,480 –> 00:49:47,240
been utilized in neuroscience, as well as many other systems now to improve a lot of existing
00:49:47,240 –> 00:49:52,400
molecules and and even push through classic drug discovery based on the discoveries from medicinal
461
00:49:52,400 –> 00:50:01,160
chemistry. Great thanks for those answer Sam. The next question I have here is for you Ruth.
462
00:50:01,840 –> 00:50:11,040
Why is CNS safety such a concern? Yes, that’s a really great question. CNS safety is a big
463
00:50:11,040 –> 00:50:17,440
concern for two reasons. One is because it’s the primary reason for clinical failure and having
464
00:50:17,440 –> 00:50:22,840
a drug fail by the time it gets to the clinic
is really bad because you’ve invested so much
465
00:50:22,840 –> 00:50:29,800
resource by that time. The other thing is one of the simple ways that people say to avoid
466
00:50:29,800 –> 00:50:38,760
CNS liability or CNS issues, is to avoid blood brain barrier penetration. Which is fine if your
467
00:50:38,760 –> 00:50:44,320
compound isn’t intended to get into the brain but for CNS therapeutics of which many people
468
00:50:44,320 –> 00:50:49,520
are involved in delivering and developing CNS therapeutics for anxiety, for Parkinson’s,
469
00:50:49,520 –> 00:50:56,800
for depression etc, these compounds need to get into the brain so we need to understand early
00:50:56,800 –> 00:51:04,120
a mechanistic level how to distinguish the efficacy from the unwanted side effects at
471
00:51:04,120 –> 00:51:10,400
seizure related ion channels. Thanks for that answer Ruth. The next question I have for you is,
472
00:51:10,400 –> 00:51:20,280
is this assay designed only for CNS targeting? Yes, so again I touched on that before, is that
473
00:51:20,280 –> 00:51:27,160
this assay is vital for anyone developing CNS targeting therapeutics because they are
474
00:51:27,160 –> 00:51:32,920
intended to penetrate the blood brain barrier and therefore, if they have any interaction with these
475
00:51:32,920 –> 00:51:38,760
ion channels, they are likely to be associated with a CNS risk. However, if you look back at
476
00:51:38,760 –> 00:51:45,720
the AstraZeneca data set, although CNS targeting therapies are a hot spot for CNS toxicities as
477
00:51:45,720 –> 00:51:51,040
you would expect, there are also many compounds in other therapeutic areas such as oncology,
478
00:51:51,040 –> 00:51:57,760
rheumatoid arthritis, anti-infectives, that also have a seizurogenic risk and that’s presumably
479
00:51:57,760 –> 00:52:03,800
because in order to get the efficacy required, those compounds happened to have an interaction
00:52:03,800 –> 00:52:08,840
with ion channels in the brain. But knowing now what we know about ion channels and their
481
00:52:08,840 –> 00:52:14,680
associated with seizure, we can work to design that out early not just for CNS therapeutics but
482
00:52:14,680 –> 00:52:23,160
for other therapeutics as well. Thanks Ruth and a last question for you. Why are current methods not
483
00:52:23,160 –> 00:52:31,880
able to detect these issues? I think the primary reason is translation. So we are dependent on two
484
00:52:31,880 –> 00:52:38,080
species toxicology testing in the GLP phase, which would more often than not be rat and dog.
485
00:52:38,080 –> 00:52:45,760
It would seem that the seizurogenic mechanisms do not translate from the rat or the dog into humans.
486
00:52:45,760 –> 00:52:55,200
So we can’t assume that testing in those tox species will detect human risk. Having said that,
487
00:52:55,200 –> 00:53:01,800
we do have some interesting in vitro models of course, zebra fish and rat brain slicers, that
488
00:53:01,800 –> 00:53:07,080
are not using whole animals but again there’s a translation challenge. So I think it’s primarily
489
00:53:07,080 –> 00:53:14,400
around translation and some of the tests in the past have aim to replace whole animal studies with
00:53:14,400 –> 00:53:21,280
maybe in vitro tests done on zebra fish or rat cells. So we essentially have gone back to basics
491
00:53:21,280 –> 00:53:27,480
and said this needs to be a human-based system if we’re going to improve at help’s testing so again
492
00:53:27,480 –> 00:53:34,760
I think the primary reason is lack of translation to humans. That brings us to the end of this
493
00:53:34,760 –> 00:53:41,200
webinar. I’d like to thank everyone for tuning in and a big thanks to our presenters. If you would
494
00:53:41,200 –> 00:53:50,320
like to listen to the webinar again it will be available on demand at www.ddw-online.com. Thank
495
00:53:50,320 –> 00:54:00,240
you for the sponsor of today’s webinar ApconiX and thank you to the audience for listening.