An Overview of Biotechnology Product Development: What to do When and Why

>>Hao Wang: Good afternoon. Happy New Year. Thank you very much for
joining the webinar. My name is Hao Wang. I’m the program director for
NINDS CREATE Bio Program. Together with my colleagues
here today, we are very grateful to have the
opportunity to host the webinar by Dr. Zahra
Shahrokh, An Overview of Biotechnology Product
Development: What To Do, When, and Why. Before we start, let me just
say a few words about the CREATE Bio Program. The program funds
research that focus on therapeutic development for
disorders that fall under NINDS mission. The therapeutics can be
about technology products and biologics, which broadly
include peptides, proteins, oligonucleotides, gene
therapies, cell therapies, and emerging entities,
as well as combination products. And the phases can range
from lead optimization to early clinical trials. The program has two tracks:
the discovery track and development track. The discovery track starts
with leads, and ends with a well-characterized
candidate. For example, the leads can
be a monochrome antibody, or a viral vector, or a peptide
that have demonstrated efficacy in relevant
animal models. And during the discovery
track, you can optimize or characterize the leads to
produce the candidate. The development track starts
with a candidate, and funds research that conduct the
IND enabling studies that lead to an IND application
to the FDA, or even small clinical trials. These funding opportunities
are Cooperative agreement mechanisms. And so, we have two
different versions. One is for the FDIR verion through
the U44 mechanisms, and the other one is non-DIR through the
U01 for the discovery track, and through the UH2/UH3,
for the development track. So, although the webinar
will be — so, the webinar will mainly focus on
recombinant proteins, although some of these
aspects will be applicable to other therapeutic
modalities. And the webinar will be
archived on our website. You can also print the
handouts as you go along. And you will be muted during
the webinar, so — but please write your
questions in the chat box. So, if you hover your mouse
over the screen box on top of the screen, you’ll
be able to choose the selection, like chat, and
please send your questions to everyone. And you can ask questions
about our CREATE Bio Program, or questions to
Zahra, regarding the content of this webinar. Your questions will be
answered at the end of the webinar, and just a quick
reminder again, the CREATE Bio application receipt date
this year of February 11th and August 11th. So, with that, it is my
distinct pleasure to introduce Dr.
Zahra Shahrokh. Zahra has deep experience
in developing biotechnology products, with a very
impressive track record of global regulatory approval
of five recombinant proteins, and clinical
development of dozens of therapeutics. She’s currently chief
medical — chief development and officer, and a CMC
consultant at STC Biologics. And prior to that, she was
heading the pharmaceutical and analytical the
development, also the head of CMC program
management at Shire. And she has many other very
impressive credentials that you can read about, but
because of time, I’m going to turn to Zahra and
start the webinar. And thank you very much
again for joining, and Zahra, you should be seeing
that your presenter is shown — just one second.>>Zahra Shahrokh: Yeah. Well, while we’re waiting
for that, hello, everybody. It’s a pleasure
to be presenting. This is work around product
development — yes, thank you. I can see that I’m
a presenter now. Let me see if I can move the
slide up — oops, let’s see. Sorry, I lost the — I
can’t see the slides.>>Hao Wang: Okay, do you
want me to — can you give back to me the presenter
rights, and I’ll project the slides for you, Zahra.>>Zahra Shahrokh: Sure. How did that — let me
[unintelligible] second –>>Hao Wang: There we go. Okay.>>Zahra Shahrokh:
Do you have it? [inaudible] okay,
please go ahead.>>Hao Wang: Okay. There you go.>>Zahra Shahrokh: But I
can’t see the slides from [inaudible]. Are you on page seven?>>Hao Wang: Yes. I’m on page seven.>>Zahra Shahrokh: Okay. All right. There we go. Okay, so the outline of my
presentation today, I’ll walk you through the
different stages, or different phases of
product development. I thought it would be
important for you, even though many of you may be in
the early stage, more on the — sort of the early
research stage, prior to the development stage, to give
you a sense of what, you know, what does it need? What happens, you know, when
you go into development? Or what are some of the
questions that are asked to be able to move to the
development phases? And really that transition
from research to development, there is a
development candidate selection. I think that was — it was
well mentioned how, even in terms of the funding,
there are two stages. You know, once you actually
have a chronicle candidate, then you move forward. That selection of the
clinical candidate, there’s some really key questions
that need to be asked, certain activities
that need to be done. And I’ll walk through that. And then, I’ll describe,
obviously, some of the studies that are involved —
in some animal studies that are involved in developing
in identifying a development candidate requires
conducting Pharmacokinetics and Pharmacodynamics. So, speak to those and how
do you — how do you do them. And actually, show you case
examples of some missteps. Some things that are
sometimes forgotten, or not very well understood early
on because of the discipline is not very well understood
to a number of early investigators. And so, I’m going to try to
walk through that, so as your products move through,
you know, you have a sense for them. Next slide, please. So, before we start some
definitions, TPP, or Target Product Profile, many of
you may have heard this. What this is is, you know,
what is the — it’s really around what is the target,
you know, aspects of your product that
you’re looking for. What should be
the indication? What kind of behavior do you
want to see, you know, of the drug in vivo? How — you know, where is it
made, how do you store it? And basically, when a drug
is approved, there’s a package insert. You know, when you go and
buy drugs, they actually have a package insert. And for approve — for
approving drugs, you need to have all the information
around that package insert. I will describe what you
need to do, in terms of defining target product
profile for early stages. So, we’ll speak to that
during the — in the Webinar. Pharmacokinetics, that I
would abbreviate as P.K., that’s the drug
concentration over time, in vivo, in either animals or
humans, from the time of administration, until it
disappears from the blood compartments. And then, eventually —
typically, you’re talking about blood
Pharmacokinetics, but obviously, drugs may go from
blood into certain tissue compartments, and you could
follow that open tissue of interest. Pharmacodynamics refers to,
now, the impact of the drug. So, you have a certain drug
concentration, but what is the impact in the disease,
either model or in human, during the time that that
drug is hanging around in the body. So, again, I will
refer that to P.D. You — it’s really a sort
of a biomarker, or an early sign of efficacy
of the drug. The concept of relevant
species is an important one. We’ll be talking about it
when we speak to the — to toxicology studies. It’s a species that has the
tissues cross reactivity with the drug that you’re
developing for humans. If it doesn’t cost-react,
then you really don’t have a relevant model, and so,
you cannot predict what [inaudible] in human. And throughout the
therapeutic or therapeutic index, really refers to the
— that — the dose and regimen range that you’re
giving that is efficacious but not toxic. Okay? Next slide, please. So, this flow chart kind of
depicts the state of product development. Again, many of are in
the research phase. That’s when you’re trying to
identify a target that links the disease. Then, when a development
candidate is selected, we’re ready to do development
effort, and the phases are basically a number of pre-
clinical studies to identify what should be the dose
and regimen, the P.K./P.D. as I mentioned, and then
toxicology studies. Then, the phase clinical
study is really — phase one and phase two, referred to
what’s called a human proof of concept. So, you’ve done animal proof
of concept in early stages in pre-colorful phases, and
then there’s the human proof of concept, where you’re
showing, in phase one, what would be the doses that are
[inaudible] safety, so what — how high can you go,
while it’s still safe in humans, and then phase two,
you’re looking at now — starting to look at either
biomarkers or some safety — some sorry, some efficacy
science, and you’re really identifying the dose and
the frequency of dosing. And really, the efficacy
— the true efficacy trial [inaudible] is
the phase three. And then, when the drug is
ready for submission for the agency for approval, it’s
either a biological license application, which referred
to [unintelligible] biologics, that’s
what you file. Or actually [unintelligible]
it’s called NDA, the New Drug Application. And then, it — the work
doesn’t end at that point. There are a number of
post-marketing commitments, and additional, you know,
post [unintelligible], I’m sure, of what’s called phase
four, where maybe you want to expand label, you want
to do pediatric studies or other additional safety —
accumulation of safety data, and so forth. So, it’s an ongoing program,
once you have a product commercialized. Next slide, please. So, what are some of
the — what do you for? What are some of
the questions? What do you do in these
different phases? Particularly in that
development candidate selection phase. So, first, in research, as
mentioned, you’re trying to link some sort of a targeted
disease, and you’re stringing different
compounds that affects that target. These are — typically,
you start in vitro, right? You’re looking at maybe
binding to a particular receptor, or blocking
a particular antigen. Typically, you’re dealing
with microgram scale, maybe milligram, of you know,
preparation of compounds in this phase. And the compounds are being
tested in animal models, in animal efficacy models. So, it’s really a
[unintelligible] animal proof of concept. At that point, now you have
— the phase showed some success. Now, you need to actually
generate something that is a clinical candidate. So, let’s say in early phase
maybe you have a mouse model to actually generate
as a mouse protein. Now, you actually have to
have a suitable stem line for something that you’re
going to put into humans. So, a lot of — what’s
called CMC activities begin at this stage. CMC stands for Chemistry
Manufacturing and Control, and it refers to everything
about the physical product. You know, how you make it,
how you keep it stable, what formulation, what analyses? So, those are some work that
begin at this stage to show that you have something you
can — you can actually take later on into the clinic. The material, the compound
at this stage, you need to show that it has adequate
stability or solubility. Let’s say, you know, if the
compound — you’ve been able to make it fresh and have
done some animal studies — but if you can’t keep it
refrigerated for, you know, eventually, for two years,
which is what ultimately you need, well, then that’s not
going to be a development candidate. So — and that can speak
to some of examples of [unintelligible] that you
need to do during this phase. This is when you need to do
P.K., and try to correlate the in vivo drug levels to
the action of the drug, the P.D. — activity of the drug,
based on which you define the dose and the regimen and
the route of administration. This is what one would do
to sequester activity to identify the relevant animal
models, and you’re scaling the material to presume at
this point, you’re going to get into the hundreds of
[unintelligible], and maybe even gram levels. And then, you’re ready to do
what’s called IND enabling work. So, before going to humans,
a investigation new drug — IND stands for Investigation
New Drug, application form has to be filed with
whatever regulatory agency that you’re — whichever
country that you’re going to be doing those studies. And for that, what’s called
GOP toxicology studies are needed. So, you can do some initial
toxicology studies, say, you know, in N.A. with a lab, or even
in your own lab. But when it is — when it’s
IND enabling, it’s done under what’s GLP stands for,
Good Laboratory Practice, which means that it’s
documented, and certain — there’s certain requirements
around dose solution analysis and dose stability,
and you know, there are guidance’s around that. So, those toxicology studies
enable the — going into human. The manufacturing process —
so, all the CMC work has to be done now, and the
manufacturing process is scaled up, typically to,
again, many, many grams. It’s not kilos, depending
on the — for example, antibodies. Sometimes, you need a lot
for doing early work. And early in the development
phase, you’re looking into what should be — what will
be projected market design should be — and therefore,
what kind of scale do I need? What kind of cost of
good can I tolerate? So, basically, you’re
talking about financial feasibility of
developing the program. It’s beyond technical
feasibility. And of course, you have to
have a clinical design and clinical protocol to be able
to then move forward into human proof of concept. Okay? Next slide, please. So, with regards to target
product profile, if you just look at the — basically,
here’s — it’s allowing to have a directed thinking
about the information you need to register
the drug eventually. So, for early stages, this
is really a living document. It’s going to change over
time as you get more information. At the beginning, you
may know very little. You only know some in vitro
information, but really, it’s not until, you know,
as you go through the — through the pre-clinical
phase, you’re going to collect more information,
and you’ll have what’s listed on the right side,
which is the packet insert information — lots and lots
of information around the use of indication, and the
contraindictions and the warnings, and so forth. But early stage, on the left
side of the slide that the pre-clinical stage, the
first thing that you need to decide is what is the
primary indication that you’re going to
take this drug into. What is the target
population? Should they need to be —
should they be adults? Can you use them
in early ages? A specific — a condition,
they may or may not have, what would you actually
have to include? How long — you know, what
is [inaudible] and how long? What is the duration
of administration? Is this something acute? You’re going to [inaudible]
acute, or it’s chronic, and they have to — you have
consider a [inaudible] administration because it’s
for lifelong — you know, it’s a lifelong disease. The regimen is the frequency
of administration. Again, you know,
weekly, monthly. Is it a bolas [phonetic sp]? Is it an effusion? Bolas being [inaudible]
quick, or is it a — like a sustained effusion? And what would
your end point? What kind of — what kind of
clinical data do you need to collect? So, those are all efficacy
related — what are called efficacy related. Then, CMC related, which, as
I mentioned, is about, you know, what is the actual
physical product? What does it look like? Is it liquid? Do you need to
[unintelligible] it? Is it going to be in a vial,
or a pre-filled syringe? So, those are the questions
you need to ask, you know, the moment you have
what looks promising. In an in vitro proof of
concept, you need to ask these questions: what
do I need to develop? And of course, from a safety
perspective, what kind of risk — it’s really
a risk assessment. What does this drug — what
would it potentially — what kind of legal risks
would it have? Is it going to have
[unintelligible] properties? Is it going to have certain
[unintelligible] perhaps [unintelligible] or lack
of, for that matter? Those are the information
you need to collect early on. Next slide, please. Yep. There we go. So, I tried to summarize as
a kind of simplified — a pre-clinical, the key
clinical steps towards development, and there
are really six steps. So, the first one — again,
I’m sure you’re all doing that, is you have to
identify a target that is related to a disease. It’s either through
literature — in some cases, it’s actually
natural history. For example, I was
[unintelligible] I was involved with
[unintelligible]. And a number of times, you
cannot really do — you’re really looking at the
natural history of the disease to then predict, you
know, what kind of clinical trial you’re
going to design. It’s not like — it’s not
always animal models, or things you can do, other
than collecting natural history. Or you would do — this
is your own studies, and develop your own
disease model. Then, you then identify a
ligand’s, you know, for that target, that’s
your compound. And that — you need
to show that that has a [unintelligible] effect. If it’s not [unintelligible]
dependent, you cannot [unintelligible] the
specificity, right? You want something that
is specific, and you can control the dose
activity and safety. So, for example, it’s
binding to something — you know, either as an
agonist or antagonist. And then, in vivo, it has
some sort of an efficacy in an animal model. The next one — the
next one, if you — so, typically, what you’ll do
is an in vitro — in an in vitro [unintelligible]
binding [unintelligible] you’ll look for
the IC50 and IC90. IC50 — it’s really telling
you about the efficacious dose right? So, you know you need to be
at least that — you know, about that at side of action
for it to be effective. IC90 is a really useful one
— parameter, because when you don’t, you know, the
drug is going — and I’ll show you in a couple of
slides, the drug will disappear over time,
right, in the blood. But you want to then decide
when do I dose again so the activity continues. And that’s — that IC90
actually finds your trough level — what’s the minimum
level before you dose again? And I’ll show you with
graphics later on. Then, the next step is you
want to, as mentioned, identify irrelevant species
so you’re going to get, you know, maybe many, many, you
know, 30, 40, 50 different types of tissues from
different animal models, from rodents and a number of
times, for biologics, you actually will use
non-human primates. And you show — do I have
[unintelligible] compound show — cause reactivity,
you know, compared to the human tissue, does it show
a similar reactivity to the animal tissues or not? And that’s how you select,
you know, what should be your efficacy — or your
safety models, actually — safety animal models. Next? You have to click, one more? Yeah. So, at this point, you will
conduct a single dose P.K. So, but it’s a
built escalation. You start with certain dose
that you know it’s going to — that’s from your in vitro
studies, you know it’s going to show activity, and
you can keep going up. And basically, you’re
looking for a region where your P.K. is linear — and I’ll
describe what that means. But also, what would be the
— eventually, you want to identify what’s the
maximum tolerated dose. What’s the maximum dosage
you can get — sometimes, it’s maximum feasible dose. For example, if you can
never get enough in solution, for a lot of
biologics, you actually don’t reach MTD or they
could be pretty safe at the — you know, at the hundreds
of mix that you, you know, that you give them in
animals, in which case, it’s basically maximum feasible,
whatever you can actually make and put into solution. Now, in terms of P.K.,
you’re looking for a region where — for the
[unintelligible] P.K. where the area under the
curve — AUC stands for area under the curve, right, of
the dose — of the dose time effect, is actually reaching
saturation, meaning that whatever receptors —
whatever it has to bind to [unintelligible] a tissue
that it thinks it’s binding to [unintelligible]
saturated them. Now, your P.K. is actually linear with
dose, and that defines, basically, your sort of your
minimum dose, vivo, that you want to go with. So, from the single dose
P.K., now you do multiple dose P.K., and here, you
actually do them in a diseased model, where you’re
trying to use that to establish P.K./P.D. relationship, right? The relationship between
blood level and activity in vivo. That helps you define what
should be the frequency of dosing, as mentioned, you
know, you try to — as soon as you reach IC90, you want
to actually dose again. But you want also — you
want show there’s no accumulation, and it doesn’t
— accumulation does not occur when you’re
in the linear P.K. range of dosing. And here’s where you start,
now, getting a sense for “therapeutic window” right? You’re seeing efficacy, but
at what point are you going to see toxicity? So, these are really the six
key activities you’re going to conduct in a pre-clinical
phase, you know, to get them into — to have a
development candidate, and proceed to an IND
enabling study. Next slide, please. So, some fundamentals
around P.K.s and P.D.s. So, the first one, I’m
showing you a single dose of what’s called two
compartment models. If you see — look at the —
look at the blue line, or the one that is coming from
say about 100 — so the Y-axis is a concentration
of drug in the blood. The X-axis is the time. And in blue is where the
drug is dropping very rapidly. And then, you actually see,
there are two phase, right? And then, it’s slowing down. So, you can actually take
this data and fit it into what’s a called two
compartment model. You see a very fast
elimination in — initial fast distribution phase, and
then it’s eliminating what’s called the data phase at a
different — at a slower rate. This is typical of biologics
— it’s at least two compartments, if
not more, actually. The next slide. And this [unintelligible]
all this data with. So, the next one shows
the — on the Y-axis, is actually the clearance,
which is a calculated value of how many — how much, you
know, blood basically — what is the clearance rate,
right, per kilo, body weight, as opposed to —
as a function of dose. And what you see is that
it’s — it has a saturation. So, its clearance actually
is maybe very high at the beginning because there are
many things that are — it’s a high clearance. It’s faster clearing because
there are many tissues, there are many things that
it’s reacting to, the drug is interacting with. But at some point, you
saturated everything, and now, there’s no more sink,
and now, it’s no longer dose dependent. And so if it’s no longer
dose dependent, if you actually convert it on the
right side, that’s your studies — basically your
serum levels, right? [unintelligible]
as a saturation. And that’s why you want to
take, actually — you want to be the region where
you’re past the non-linear phase to start
dosing humans. Next one shows what happens
with a multiple dose P.K. Now, if you actually have
selected a region where you’re not — you actually
aren’t saturating, then with repeat dosing, you actually
will not see accumulation. So, each of those curves
going up and down is the blood level. You dose, blood level — you
know, it goes up, and then it disappears to
reach a trough level. And then, you’re dosing
again, and it disappears. So, typically, you want to
do this multiple dose P.K. after you’ve done a single
dose, and look for the, you know, the timing of the
trough, and the level — and then adjust, you know,
adjust how to dose — how frequently you want to dose. Next is this relationship
between Pharmacokinetics, the blood level of
drugs, and activity. So, what’s plotted is on
the Y-axis is the percent response. So, this could be either a
therapeutic effect — for example, you know,
efficacy marker. Or it could actually
be toxicity. So, whatever that
percent response is. And on the X-axis is
actually the dose level, right? So, what you see on the
leftmost graph is the efficacy of signals. So, for example, as you keep
dosing more and more, you get to see more and more
either number of animals responding, or higher
response rate — let’s say, if this is a tumorous ring
carriage [phonetic sp] or whatever the efficacy marker
is, you know, you basically have this curve, which would
have, you know, which would have, you know, basically
a saturation level of dose response. So, typically, you take your
ED5 [inaudible] — sorry, ED10. So, at 10 percent a response
when you start doing your dose escalation P.K. study. And then, on the right side,
on the right hand curve, the red curve, is the toxicity. So, you may not see any
toxicity, say, in this case, like above — let’s see
maybe that’s 20, 30 [unintelligible], but then,
all of the sudden, you’re going to start seeing
toxic effects, right? And typically, that’s very
steep, usually, you know, and the level of drug below
which you have no observable events, the annual E.L. level is where, typically,
in your IND enabling toxicity, you want to show
that I’ve gone this high, and I have no observable
adverse events or toxic effects. And NTB is basically is
somewhere on that curve, pretty low on that curve,
where it’s maximally tolerated. You’re not killing
[unintelligible] toxic, but it’s, you know, maximally
— how should I say it? [inaudible] some toxicity,
but obviously beyond which, you know, the animals are
dying, or it’s very severe. So, the therapeutic —
sorry can you go back? So, the safety factor, or
the therapeutic index, is the difference between
those two curves, right? It’s that window between
those two curves, okay? Yeah. Next slide, please. So, I want to show you a
couple of examples — of case examples of things
that could happen to you. So, here’s a case of a
protein that — what if you can’t find actually any —
detect any tissue binding of the drug? And this is from a
real case, by the way. Well, it so turns out that,
you know, when you’re doing — when you’re looking at
tissue binding, you’re actually — generally,
are preserving, right? You’re fixing using
formaldehyde-soaked tissue and so forth. And proteins may denature,
so then they actually lose their epitope, and you won’t
necessarily detect, you know, [inaudible] so you
wouldn’t — you would not find relevant species. And so, one solution for
that is if you could click — yeah, one solution is
actually avoid doing tissue binding [inaudible]
protein cannot tolerate. You instead can do
bio-distribution studies using other methodologies. You can do autoradiography
sorry, can you go back? You can have a — yeah,
can you go back to the [inaudible] yeah. You could have a radioactive
compound and look for that, right, in cross sections, or
what’s called [inaudible] where you could have a label
approaching with an outside emitter, and follow,
actually, non-invasively, in vivo, where the drug went. Next slide, please. Here is an example of when
you’re trying to do, you know, look for binding, and
get some, you know, binding [unintelligible] response
curve with a drug and [unintelligible] receptor. And I’ve seen that a
lot, actually, being a consultant, where the
hook effect is missed. And so, let me explain that. So, in a typical
— say, P.K. You’re doing a P.K. — if you look at the
schematic on the right, you have a plate. You have a captured antibody
that captures the drug, and then you have a readout
antibody that is, let’s say, for example, HRP labeled, or
something that, you know, provides a signal
for readout. As you increase the amount
of drug at — in this case, the increased
[unintelligible] drug added to the plate, [inaudible]
should keep going up, right? But I’ve seen where people
have gone — if you go too far up, actually,
nine [unintelligible] relationship is beyond
what’s called the hook effect, where you keep
adding more and more drug, but you actually get
less and less signal. And the way you know this
is actually the other way around. You’ve added a drug. You do different
dilution series, right? And you see that as you’re
diluting it, the signal is actually going up as
opposed to going down. And you know if that
happens, you’re on the right side of the hook effect,
and you have to keep going. You have to keep diluting
until you reach the other side, where as you dilute
the response goes down, and it’s linearly —
it’s proportional. If you’re diluting twofold,
the respond should go down twofold. If it’s not, it’s in the
non-linear range, and the numbers don’t make sense. You actually would have the
wrong [inaudible] wrong whatever measurement
that you did in that [unintelligible]. And I’ve seen that a
lot, where you actually underestimate, you know,
the actual blood level. Next slide. Here is another case where,
which is also pretty common, where you have, actually, a
circulating blood antigen or ligand. Now, what do you do, right? That is going to interfere
with your drug connotation because it’s
going to be bound. So, typically, in this case,
you have to have assays that measure the bound
and the “free” drug. And to measure the “free”
drug, you have to have — typically, if you’re using
— if it’s a — you know, for biologics, for protein,
you have antibodies, right? To detect the drug, you have
to find an antibody that’s non-blocking. [unintelligible] more, so
here’s, let’s say, for example, a case — let’s
say you’re developing an antibody — and actually,
a human antibody. How are you going
to do that readout? It’s very complicated
because — a great example is anti-HER2. Patients are shedding
antigens from HER2 itself, so if you want to measure
anti-HER2, you have two issues. First, you’re trying to
measure something that is an antibody to a human
antibody, right? And also — sorry,
to a human antigen. But you also have
antigen in the blood. So, first, you have to find
an antibody is not blocked by trastuzumab-bound HER2,
and the other one is that your readout has to be an
anti-human IgG1 because you’re measuring
a human antibody. And then, to measure the —
so that allows the — to measure the bound, right? Sorry, so to measure a
bound, you would capture the bound with anti-HER2 that is
not blocking, and you read out with anti-human IgG1
to measure the free, you actually capture with
HER2 itself, right? The antibody is not bound to
the antigen, you captured on the HER2 antigen itself. Okay? So, here’s an example. Another example is we’re
working with a client that’s working with eotaxin they’re
developing an anti-eotaxin and it’s an IgG4. And that was kind of
complicated because blood levels and certain
information [unintelligible] have eotaxin in blood. So, what we have to do is
that you would have to capture the eotaxin with
the antibody that we’re developing, and then use
a readout, monoclonal, antibody, that does
not detect human IgG4. Because the blood has a lot
of IgG4, so if I use any general, you know,
anti-IgG4, it would react to my drug. So, I just identify an
antibody that does not react to human IgG4, only to the
drug, and then we ended up having to dilute the serum
enough that the eotaxin eventually
[unintelligible] right? Eventually does not
interfere with the assay. So, it’s really complicated,
developing with this P.K. assay. Next slide, please. Now, what about what’s
called ADA assay. These are — again, these
are things you need to do pretty quickly after you
have a direct, you know, a direct — what you think
is a you have a direct candidate in order to be
able to develop the program. So, ADA stands for
Anti-Drug Antibodies. So, for biologics, when you
inject some [inaudible] inject the antibody with the
— or the biologic, the body develops antibodies to them. And you need to know — so,
it’s a safety concern, the immunogenicity or developing
an anti-drug antibody is a safety concern
for biologics. And so, one needs to have
assays that detects do you actually these antibodies in
the blood, particularly — yeah, if you could click on
the next one — yeah, next — yep, particularly because
it could actually neutralize the activity of the drug. You know, an antibody
combined to the site of action or the site of
activity of the drug, and therefore, neutralize it. It could obviously result
in anaphylactic reaction. One of the things that are
sometimes ignored is when you make a drug in a
particular cell line, the whole cell protein that
are actually made — [unintelligible] with the
protein with the product builds and generate
antibodies in patients upon repeat dosing. And those could actually
— and it used to be, very early on, in some — in
some yeast cell lines that actually were anaphylactic
reactions, unfortunately, in patients, to the whole
cell proteins that [unintelligible]
with the protein. So, it’s really important —
the immunogenicity of the drug is really,
really important. Or what comes with the
drug itself is important. The other thing
is that the P.K. would be unusual, right? So, let’s say you dose —
the first two doses, you have beautiful linear P.K. response, and then all of
the sudden, your — you know, you see that, you know
there are no blood levels, even though you’re injecting
the same amount of drug. Well, it still turns out
that if an antibody binds to it, it’s going to
get cleared faster. So, here’s an example
in this figure. Let me walk you through. So, basically, on the top
one, the top straight line is where, basically, if you
have — if you don’t — if you have an antibody that
— or you have no anti-drug antibody, you should
pretty much get the same [unintelligible]
right, in a drug? Your drug level should
be the same with [unintelligible] injection. But if you generate
antibodies to a drug, then you’ll see that with
increased [unintelligible], you’ll have less and
less levels, actually. So, the drug level
is decreasing. Your dose the next time
around is actually not going up as high, and it’s
dropping, you know, more. And depending on where that
antibody is reacting to, you could have, you know,
faster clearance. For example, in this case,
the drug is an antibody, and the anti-drug, the ADA,
was actually against S.C. And that cleared of —
actually, not as fast as it would if you actually had
the antidrug, antibody against the fat portion. And it really depends
on the product. So, you could really —
you could really have very unusual P.K. profile. So, understanding the
level and the formation of neutralizing — or of
antibodies in blood helps to interpret the P.K. data as well. Okay? Next slide, please. Now, developing an ADA,
actually, itself, is also — could be pretty complicated. So, typically, you have
— you lay down the drug, right, on the plate. You add the serum — for
example, the serum from animals or patients. So, the — if you have any
antidrug antibody, it should bind to that, right? To the plate. And then, you read it out
with the anti-IgG — the majority of it — of the
antibody’s ADAs are actually IgG, and most of them
are actually IgG1. In some cases, you could
have IGE, in which case you should — there are anti-in
general, anti-I.G.s, not necessarily IgG. So, if you can
click, please. So, the biggest problem
becomes where are you going to get an — in order to
develop an assay, you have to have an
anti-drug antibody. Well, where are you going
to get that from, right? And so, you’re starting
from a program you don’t — there’s nothing right? So, it’s actually really
important to think early — for biologics to think
early, and develop, you know, immunize animals, and
try to get some antibodies to your drugs. And I mean, [inaudible]
because it’s eventually going to need that
for your top studies. The other thing is the
sensitivity of that assay has to be about correcting
[unintelligible] has to be around, you know, half a
microgram per mil, 200 to 500 nanograms per
mil so that antidrug [unintelligible] needs to
be pretty good — has to be pretty sensitive. It also has to be pretty
specific because if it recognizes other IgGs in
your human serum, you know, then you know, you have
way too high of a level. It’s [unintelligible] had
clients where they were in phase three, and all of the
sudden, you know, they were trying to develop and assay
and all of the sudden, they were seeing what looked
like a huge signal. They were panicking. You know, what do we do? And it turned out to be all
[unintelligible] it was really cross-activity to
other things in blood, and it took quite a long time to
figure out what those other things are. And it wasn’t their drug. I mean, luckily enough,
it was not the drug. So, moving forward, let’s
say you’ve done all this. Now, you have to develop
— you have two toxicology studies. And then, as I mentioned,
you have to have a — for toxicology studies, you
have to a relevant species. So, you do in vitro studies
IC50 of the animal cells, right, is the same as the
human cells, or similar? Close enough? You would do tissue cross
reactivity from different organs and find —
specifically, this is a good time to find what kind of
organs specific toxicity, you know, would you —
could potentially occur. And then, there is an ICH
guideline S6, which, by the way, you can go to and type in ICH, and you can read a number
of ICH guidelines for, you know, once you start getting
to — close to development phase, you know, I highly
recommend you read them. There are the
non-clinical ones as S6. There is, you know, all
kinds of CMC quality guidelines you should read
and be familiar as what it is that the
agency looks for. So, for recombinant
proteins, you have to have two relevant species,
and this makes it really complicated. Let me just give you an —
as an example, in small molecule drugs or drugs. Typically, you do like a —
maybe a rat or some rodent, and then a lot of times, you
use a dog because the rodent usually is your
non-relevant, and maybe it’s just a general tox and the
dog is a good one because you can do, you know, oral
— it’s good for, actually, for oral — mimicking
oral uptake. Finding two relevant
species is really tricky. And so, sometimes, we —
maybe you can justify say, look, you’ve tried, and
you can only find one. Like, rodents being
sometimes not being relevant at all. And sometimes, you can
actually argue not using a non-relevant species —
particularly European agencies are actually really
open to not using a lot of animal studies — not
killing a lot of animals. So, in that regard, it’s
really good to come and say, look, I’ve identified, you
know, [unintelligible], a number of tissues and organs
and species, and here is my relevant species and
therefore, I’m going to do toxicities on that. And as mentioned, again, my
experience has been that a number of times, the best
relevance issues for biologics or for recombinant
proteins is monkey. I’ve had an example where
either cyno or rhesus didn’t even work. So, unfortunately, had to do
the work in a chimpanzee, and that’s — or a
[unintelligible] animal. And that’s obviously, very
expensive and very — you know, it’s unethical if you
don’t — if you can avoid it. Next. And in tox studies, you’re
doing dose escalation. You’re doing multiple doses,
and you go factors — you know, many factors, three,
five, 10 times higher, until you reach MTD, and then
you have a recovery phase. And then you stop dosing and
you make sure that whatever safety signals is it
actually reversible or not. Those are the questions you
have answer based on which you designed your
clinical program. Next slide, please. Yeah, I mentioned to you
some of the small molecules, you know, typically, you do
non-relevant and a relevant model. So, it’s different
for biologics. Now, what if you really
cannot find a cross reacting species? I mean, that could happen. If you can click, please. Yeah, click — on more? Yeah. In this case, you wouldn’t
be able to predict toxicity, so you actually have — and
I’ve had a situation like this, where we actually had
to make an animal version of the protein, to — because
the human protein was not, you know, cross-reacting
with the animal tissues. So, we had to do the
toxicity in the animal version. Or an alternative is you
actually generate — if you can click, please —
generate transgenic, you know, animals. So, or you make like — you
make a humanized mouse, you know, to do tox studies, or
you have knock-out and so forth. And then, you put the — you
put the human gene in there. You knock it in, and
then develop that. And those are rare. That’s just something
to be aware of. Next slide, please. So, early tox studies. You find a relevant species,
you do tissue reactivity, you know, to identify the
age and physiological space, and define that, if it’s
for — if you’re doing, you know, studies in pediatrics,
you have to do actually juvenile animals. Makes it more complicated. You can, early on, use
non-naive animals, not for GLP-enabling tox, but early
tox, you know, you know the — a lot of drugs, you
know, clear very fast. So, if you’ve waited beyond
the recovery phase and the drug has cleared — the
previous drugs have cleared, you can use the non-naive
cheaper, it’s okay for early studies. Typically, you take maybe
three, three doses per arm. It’s easier to start
with male animals. Just male animals, you
don’t have complications. You know, hormonal
complications. Later on, for — again, for
IND enabling, you will have to show — if the drug is
typically used in both males and females, you have to do
both genders — tox studies. But in early trials, you may
be take a low — a med, you know, that is kind of
efficacious on the higher — high dose that’s near MTD. And typically, you go —
let’s say you start dose esclatation and you can
start with the very low level and then do either
factoids of 10 or three — again, depending on what
grains you have to play with. As mentioned, you typically
find the no observable adverse effect level and
maximum feasible dose. And from that, then you
design your clinical. So, in a tox study, you’re
doing clinical observations — you know, your weight and
height and — sorry, your weight and just the
general appearance. And then you do blood
chemistry, and then histopathology. And of course, for
biologics, you have to do immunogenicity, which you
don’t do in small molecule. Next slide, please. Now, here’s an example of —
I mean, you may think it’s silly, but it’s
a real example. It occurred at a big pharma. And in an early stage, the
— basically, they were doing — they had seen that
a particular drug was using — they using a drosophila
model, and they saw eye loss actually, in that model. And it’s like, okay, this
is drosophila who cares. They went and they did
pre-clinical studies, and pre-clinical studies —
obviously, not in drosophila and it was pretty good. But in patients, they
started seeing blindness. And instead of going back
and trying to go back and say, “Well we can actually
see drosophila, you know, and so maybe there’s
maybe something to it.” Well, it turns out that
there is — there is an impact — this particular
drug has an impact on the eye — I mean, the
program died, obviously. But you know,
donate more data. The data was like,
basically ignored. It will be good to really
understand, if there is a toxicity, what
does that mean? Is it real, and really
follow up on it, and not just ignore it. Okay? Next slide. So, soliciting CMC experts
early on in the program. This is the one — again,
being a person who’s more on the — as I said, on the
CMC, on the actual, physical part of the product, that’s
the last thing that people think about, right? Everybody’s very busy trying
to — trying to come up with something that is effective
with an animal model, you know, coming up with all
kinds of, you know, maybe different drugs. But what do CMC
folks bring to you? They — obviously, they —
you know, those are the folks who know how to make
cell lines that are stable. You have to have a stable
cell line for a clinical program. They would know how to grow
them such that the titer is high, and it’s
reproducibly high. At some point, a drug may
not be financially feasible because it cannot make it. And unfortunately — that
would be unfortunate, right? If you had this wonder drug,
right, early on, but you cannot actually make it. They would know how to
purify it to an appropriate purity and reproducibly —
and again, I’m using this word reproducibly because a
number of times, in early research, you just need to
make something once, and you know, show something. But you know, now, if
you have a developmental candidate, you need to be
able to make it, over and over and over, with the
same characteristics. Another thing that is often
forgotten, and I’ll give you a couple of examples, is,
again, your CMC folks would know what kind of buffer
excipients should they put the protein in? Can you free it, even? You know, how do you
keep it happy, right? So, whatever you have today,
a month from now, is the same thing. The other thing that’s
around characterizing protein structure and
relating it to activity with analytical tools, like HPLC,
like mass-spec and looking at, for example
glycosylation is important, and so forth. Next slide. So, the link between
pre-chronical and CMC, here’s — I’m showing you
the activity, and also sort of the time links
between them. So, early on, obviously CMC
folks are not involved. You’re trying to identify
a compound that impacts certain clinical target,
diseased target. And you’re doing in
vitro binding study. But this is the time where
now you need to — start making compounds, you’re
— early on, maybe you’re making different versions of
those compounds to do P.K. and P.D. dose regimens; it’s really
important to characterize the structure-function
relationship. People do that with small
molecules very readily, right; it’s a
common thing to do. For biologics it’s not that
often done, because if you find one, you can test
it and run with it. Understanding what is
critical about that protein; you know, does it, for
example, maybe it needs to be dimeric, in
solution, to be active. If you don’t know that,
and you always purify the monomer, or you’ve
haphazardly purified the dimer once, but the next
time around you don’t have the dimer, and you’ve lost
efficacy but you don’t know why. Does it have
reasonable solubility? Is it reasonably stable? Which sequence to select? You can look for certain hot
spots, certain degradation sites, on the protein;
actually, if you want, you can manipulate and take it
out, you can mutate it and take it out to have a
better clinical candidate. Next slide, please. So, here’s an example,
unfortunately, of IND-enabling development
“No-Go”, unfortunately. There was a drug that — it
was an anti-cancer drug, it showed beautiful
effectiveness in animal models. In the primate tox
studies, it showed severe nephrotoxicity;
the program died. And actually I was
personally involved in this one, where I was, as I said,
I was on the CMC side — I was doing
biophysical studies. And I showed that this
particular cytokine actually trimerizes the
human receptor. It was required for
bioactivity; that actually was a collaboration
with research. The study was done in
monkeys, but it actually dimerized the
monkey receptor. So, later on we found out
that dimerization of the monkey receptor actually
triggered a cellular toxicity that you won’t
see with a human cytokine. So when we were using the
human cytokine on human cells, right, it was not
dimerizing; when we were using the monkey cytokine
on monkey cells, it was not dimerizing, but using the
human cytokine on the monkey cell, it was dimerizing,
and it was therefore being toxic. What’s weird is that the
difference between the human and the monkey cytokine was
one amino acid, and that affected how it trimerized,
and it was the cause of toxicity. So, we went back, actually,
with the program; we actually made the monkey
cytokine, we showed that it was absolutely clean
in toxicology studies. The program came back to
life and we proceeded into clinical development. But here’s where it — it’s
really that, the people who understand formulation and
clinical characterization could help, you know, rescue
a program when you see something weird like this,
so that collaboration is really important. Next slide. Here’s another example of
what was basically an, unfortunately, uninformed
“Go” decision. So, this was an enzyme, it
showed great activity in animal models. The program moved into IND
stage and the molecule came to the process development
group, but the molecule we made in process development
actually did not show any animal efficacy, and it also
had significant solubility issues; we couldn’t even
reach the higher level. So it took about, quite a
few months to go back and forth between research and
development, and we talked, wondering what’s
going on, right. As you might imagine, there
was a lot of angst at — for the program, and it turned
out — if you can click on the bullet, it turned out
that the efficacy studies were done with
a mouse enzyme. Of course, the development
folks were making the human enzyme. The human enzyme was
significantly different; it had different charge
variants, the glycosylation was different in the human
enzyme, and it didn’t show activity to the mouse cells
that were used in vitro in the animal model. So, unfortunately, we
actually — this took a year and a half and $1.5 million
spent, and this was a real program, unfortunately,
to show that this program wouldn’t work the way it was
designed, and at least the human protein is not going
to work the way you thought it did in animals. I don’t know if, since then,
if the program has been picked up in other ways, but
it’s a lot of time and money wasted, unfortunately. As a poor basis of
development, a “go decision,” without
any kind of analytical characterization. Here’s an example of a drug
we that had to develop for an infant — it was an
infant [unintelligible] therapy, it was to be
delivered into the brain. So because, as you know,
many proteins — actually almost no proteins, pass the
brain barrier, except for certain
receptor-mediated ones. So from the clinical we
needed to get — basically they told us to dose, we
wanted to dose no more than once a month. They wanted to deliver, of
course, only a limited, less than one mL; it’s in
children, so we needed some sort of device, and the drug
has to be filtered before administration. I want to use this example
as one where, again, your CMC folks and the clinical
folks, pre-clinical, clinical, really
need to talk. So, the implication of that
clinical demand was that we had to have 100 mg/ml,
soluble and stable protein. We were very limited in the
formulation composition — basically, the only approved
drug is in salt and water that you, you know,
can put into the brain. So, how are you going to
formulate something to be stable and soluble,
particular proteins, where you usually need buffers,
you cannot live with salt and water alone, and the
endotoxin limit is much lower than IV. So again, in terms of
manufacturing, we had to be much, much more careful, and
we had to have a device, some sort of intrathecal
device, that is small enough, and which grows with
children, and so forth. We actually — again, this
was developed, it’s a real case example, it was
developed, and — I’m not sure; next slide? I’m not sure if I give the
specific example — Yeah. So, what we did is that,
just to add to the hurdle, the enzyme in vivo — if you
look at the blood level, it’s only two or
three day’s half-life. So we started asking the
question should we engineer a molecule that has a longer
half-life, because we can’t dose, we can’t do
intrathecal dose, like, daily, or every three days. We were able to, we actually
had to come up with — we didn’t modify the product —
but we did come up with a particular, as I mentioned,
we needed to come up with a 100 mg/ml soluble enzyme,
and it was only 15 mg/ml at maximum solubility. So we had to do a tox
study, pilot tox study, for different formulations to be
able to find a formulation that allowed delivery of the
enzyme at the high level, and actually had to do
develop, develop the device ourselves, so we actually
had to go and manufacture, work with manufacturers to
develop a device that would grow with, you
know, patients. And this turned out to be
one where you can actually implant the device under the
skin in the abdomen region, and have a catheter that
goes into the intrathecal space and it’s sutured in
there, and the catheter can grow, it has enough length
that as a kid grows, it actually grows with them,
it doesn’t come out. Next example. Here’s where a cytokine was
to be developed and there were all kinds — we really
needed the CMC knowledge, otherwise we just
wouldn’t make it. So here’s an example where,
during the animal studies, the pre-clinical efficacy
studies, there were many different variants,
different versions of this enzyme, made. The efficacy results were
inconsistent, so — meaning that you would see efficacy
once, and the other times you wouldn’t see it. So basically we got
involved, development folks got involved, and started
asking the question, “So how you — what are you
doing with the protein?” The protein was being
refrigerated until dosing; we didn’t know anything
about the stability of the [unintelligible]
around stability. The animals were dosed with
Alzet mini-pumps; if you know, these are little
aquatically-driving [phonetic sp] mini-pumps
that you can actually put Sub-Q — or under the skin
and in the mouth — and you can load them with a 100
microliter, 200 microliter, drug, and it’s delivered
over weeks, or months actually, some times. So the drug is sitting at
— close to 37 degrees, and this is in PBS formulation,
which is typically used in early research for 20 days. Well, it turned out that the
protein heavily aggregates in PBS at 37, so it
explained, you know, the lack of efficacy. And actually, it turned out,
with doing pH stability studies, that pH5 is best,
so we had to design a new formulation. And then we looked at
structural, structural stability of the different
variants, and actually found that some of them
are just not stable. We actually ignored them
because they’re not stable. We were able to move into
IND-stage development, IND-enabling stage, through
the right candidate selection and the
right formulation. Next. So, I think here I’ve
summarized — it’s the lessons learned. It’s that to do the
pre-clinical studies with, so that big D development
line is in sight. You want to know — so after
you — use characterization tools to understand the
molecule; to understand not only it’s biology, but
understand, also, its structure. What expression system do
you use; is it something developable — otherwise
you’re going to lose a lot of time. When you’re making materials
— is it scalable, or reproducible? And in terms of P.K., P.D. — really assess the
question of developability, if it’s something that is
disappearing very fast, it may not be a
clinical candidate. Next slide, please. And I think I also mentioned
enough of the common missteps in product
development, where the “R” folks, the research folks,
and the “D” development folks are disconnected
because they’re not, typically not in the same
department, they’re not reporting to the same
structure; they’re not even in the same
country sometimes. So, we’ve seen the wrong
molecule selected as the basis for the “go” decision,
we’ve seen little to no formulation development,
very little analytics or characterization, and these
are some of the steps in product development. And, I think — next slide. Next slide? Oops — yep. So I think that’s
it with the seminar. I hope that — hopefully you
found that you learned some new things, and I’ll
be happy to answer any questions.>>Hao Wang: All right. Zahra, would you
mind — okay. Thank you very much, Zahra,
for the wonderful webinar; and I think those case
studies are extremely helpful. I think, because of the
technical difficulties in the beginning, we’re running
out of time, but I think if anyone has questions, we’re
willing to stay here longer and to answer
your questions. So Zahra, are you available
for a few minutes to answer questions?>>Zahra Shahrokh: Yes.>>Hao Wang: Sure; okay. So we’re going to go
through some questions. I think, Zahra, on your
webpage you should be able to see the chat, and
look at the questions. And to all the participants
— I just want to mention that my contact information
is right here, and the CREATE Bio Program webpage
link is underneath this slide. If you have any further
questions that’s not answered by this webinar,
please feel free to contact me about the program, and —
or if you are interested in applying to the program,
you’re encouraged to contact us and talk about your
project; we’re very interested in learning
about your project. So, I’m actually going to go
through the chat to see the questions I can answer.>>Female Speaker: I just want
to remind everyone, as well, that if you — when you go
to your chat and you’d like to write a question, please
send it to everyone. We’re at different
locations, and that way we’ll all be
able to read it. And I think while some of
those questions are coming in, we’ve had a few
that have come in. Hao, potentially you
could address some. There was a couple
questions: What are some funding opportunities to
support IND-enabling studies –>>Hao Wang: Yes, I
can address that. Let me go back to one of my
early slides and — just one second. Okay, so I think this
slide might be helpful in answering your questions
about what are the funding opportunities to enable
IND-enabling studies. And I assume you’re asking
for developing therapeutics that are within scope of an
NINDS mission, and this is biologics and biotechnology
products, which include peptides, proteins,
oligonucleotides, gene therapies, cell therapies —
and that’s the scope of our program. And you can see here on
the slide the CREATE Bio Program, the goal of the
CREATE Bio Program is to enable P.I.s to develop
therapies, and IND-enabling studies are a critical
step in that process. The CREATE Bio Program is
divided into two tracks — we have a discovery track,
and a development track. At the discovery stage, you
are — you can start with a lead, and you optimize that
lead into a candidate, to the extent that you are
almost ready to pull the trigger and do the
IND-enabling study. The actual IND-enabling
study, and funding, is in the development track, where
you will be funded to get ready for the IND-enabling
study, and actually conduct the IND-enabling studies,
which are typically GLP, toxicology, and related
studies, and also conduct small clinical trials. So, hopefully, that
answered your question. The details are in the
Funding Opportunity Announcement, and the
program announcement numbers are on this slide.>>Female Speaker: And, Hao,
just while you’re on that slide, another question
was, “What is the extent of collaboration for these
cooperative agreement applications? Do they need inside, like
NINDS collaborators, or what does that mean, as a
cooperative agreement?”>>Hao Wang: Okay. The cooperative agreement
mechanism is an assistance to the investigator
[unintelligible] his studies, so technically
the investigators are the primary driver for the
study, and the program staff are more involved than the
regular R01, but you don’t have to have an
NINDS example in the neuro-collaborators. By cooperative agreement,
basically the program staff here, for example, the
CREATE Bio Program, will be getting substantially
involved in the program, for example, in finalizing the
milestones, providing input, providing technical
assistance, also making funding decisions, reviewing
the progress report — the annual progress report. We may have mid-year calls,
annual calls, to discuss about your project.>>Female Speaker:
Thank you, Hao. Zahra, I think we have
a couple questions that perhaps you can help with. One question is, “What are
the toxicity studies for a molecule for the brain?”>>Zahra Shahrokh: Ah, okay. So, for brain delivery
basically you need to do both exposure — typically,
even if you inject into the brain, like Epstein
biologics, they actually come back into the
blood compartments. So you have to do tox
studies not only delivering to the brain, but also
include IV arm in there, as well. Otherwise the concept is the
same; you have to do dose escalation, you know,
multiple dosing, dose escalation, repeat
those things. The principles are the same;
it’s just the question, now you have do on
both compartments.>>Female Speaker: Okay, and
then I think another one. What about blood substitutes
for fluorocarbons that are regulated as biologics?>>Zahra Shahrokh:
What is that about?>>Female Speaker: That’s a
program question, right?>>Zahra Shahrokh: Ah, okay.>>Hao Wang: So I guess the
question is, are these in scope for the
CREATE Bio Program? The answer is that these
are in scope, pretty much. We use a few examples in
our Funding Opportunities Announcement, but pretty
much anything that is not small molecule, and not
devices, can fit into the CREATE Bio Program. Otherwise, it’s under an
NINDS mission in developing therapeutics, and that
stage-wise, it’s within the scope.>>Female Speaker: All right. We had another question. Even though this webinar
was focused on recombinant proteins, we did have a
question just more related to cell therapies. Zahra, perhaps you
can help address it. The question was, “What type
of doses should be explored for cell therapies?” and the example was that
they have using FDA-approved neural stem cells that are
already in clinical trials for some indications. Maybe, can you discuss
dosing for a cell therapy?>>Zahra Shahrokh: Okay,
that’s a good one. I have to think about that. Obviously, I don’t have — I
haven’t done — personally I haven’t developed
cell therapies. So again, in selecting dose
what I would look for is what have you seen,
what have you done in pre-clinical studies, right. It’s the same thing, in
terms of if you’ve seen certain effects — because
these are typically it’s a sub cell, so you typically
don’t need what [phonetic sp] in your rating
over long-term impact. So the question is dose
response — is there a minimum number of cells you
need to inject beyond what you’ve saturated. I’m not sure if I can
explain more than that, but, you know, it really depends
on your pre-clinical studies. The other aspect, by the
way, is also around safety, right. How much — if you did a
certain level of cells, is that going to, what’s that
going to do locally, you know, upon the site of
injection, or is there a safety concern in there? I don’t know if that helped?>>Hao Wang: I think
that’s helpful, Zahra. I think the principle is
actually very similar; it really depends on the
specific cases, and the doses are determined by
efficacy and safety, so in principle it’s similar. But if you want to talk
with us further about your project, in the context of
your project, we’ll be happy to speak to you about
your specific case.>>Female Speaker: Is
there another question?>>Female Speaker: We have
another question, I think, how — related
to our program. If a drug has been approved
by the FDA for one indication, but
investigators are using it to treat a different
indication, how would they apply for U01
to develop this?>>Hao Wang: Yeah, so
repurposing is allowed in this program, and we
actually have specific language in the FOA,
the Funding Opportunity Announcement,
about repurposing. There are some additional
considerations that you might want to read about. If you need additional —
so if you have efficacy, demonstration of efficacy,
of your agent that’s repurposing from a, that’s
repurposing for a compound or a biologic that has been
demonstrated efficacy in other indications,
you can apply here. But it should be clear about
what kind of additional IND-enabling studies
you might need. For example, if you have a
pediatric population, or you have a different dose range,
that the previous indication does not cover the
population or the dose range, that IND, additional
IND-enabling studies are needed, those can be
supported by the CREATE Bio Program.>>Female Speaker: I think
we — there’s one more question. It’s just related to, again,
stem cells, and whether they qualify for our program and,
Hao, I think you’ve addressed that before — that cell
therapy is in-scope, as long as it’s in our mission.>>Hao Wang: Right, right. And, for your specific case,
just please contact us for your specific project.>>Female Speaker: Okay. Anything else? All right. So thank you very much for
joining our webinar, and we hope that we have more
opportunities to interact with you. Hopefully we will be able
to organize additional webinars, or just a
question-and-answer period with our CREATE Bio team. So just stay tuned,
and Happy New Year. Thank you very much.