James Duboff: Genomics 101 - How do pharmaceutical companies use genomic data for drug discovery?

In this explainer episode, we’ve asked James Duboff, Strategic
Partnerships Director at Genomics England, to explain how genomic
data can be used in drug discovery.


You can also find a series of short videos explaining some of the
common terms you might encounter about genomics on
our YouTube channel.


If you’ve got any questions, or have any other topics you’d like
us to explain, feel free to contact us
on info@genomicsengland.co.uk.


You can download the transcript or read it below.


Naimah: How do pharmaceutical companies use genomic data for drug
discovery? Today, I’m joined by James Duboff, a Strategic
Partnerships Director here at Genomics England, to find out
more. 


So James, first of all, what is genomic data, and how does this
relate to our genes? 


James: Let’s start with a simple explanation of what we mean by
genomic data and our genes. So, every cell in our body contains a
complete copy of our genome. Now, genome is kind of a mini
instruction manual that describes exactly how to make you. Now,
those instructions are written in a language called DNA, which is
over 99 percent identical in every single human on the planet, so
you and I are actually genomically very, very similar. The
differences, however, are called variants, and they’re what make
us unique. Now, some of those variants can actually be very
dangerous, and they can code for things like rare genetic
diseases or even cancer. So, we need to read in detail exactly
what’s going on in your DNA and in your genome to see where
changes are and where those variants really are, and we do this
by sequencing the genome. So, if you get a DNA sequence, that’s
effectively an electronic readout of your genomic data, which is
your genome in computational form.


Now, understanding that and working with that is still a
relatively new field, so what we try and do is connect the
genomic data, your genome, with health information, such as
hospital records and what you’re presenting with in clinic, if
you’re in a patient setting, and look at those together to give
context to those variants in the genome. So, genomic research is
actually where we look at how genes and physical outcomes could
be linked. So thinking of, you know, biology and physiology term,
what does a variant exactly do and how might it cause a
disease. 


Naimah: You mentioned both the genome and whole genome
sequencing, and if our listeners aren’t too sure exactly what
they are, they can listen to some of our other explainer episodes
with Greg Elgar, who explains these concepts. So James, next
could you tell me why are pharma and biotech companies interested
in genomic data? 


James: Ultimately, pharma and biotech companies are interested in
genomic data because that really tells them what’s going on
within the blueprint or that mini instruction manual of an
individual. So, pharma and biotech have dedicated research teams
that focus on genomic research, and they look through genetic
databases across the world, such as Genomics England and others,
to really understand the role of the genome in their target
disease areas. By looking at those, that helps them develop new
drugs and tools to specifically diagnose, treat and also even
cure these diseases. 


Naimah: So, how exactly do they do that? Can you explain it in
some simple steps? 


James: I think there are four key areas that they need to focus
on. So, starting with the first, where, whereabouts on a genome
should they focus? Now, the way that a pharma company would do
this, or any researcher really, is by taking two populations of
people. So, you’d take a population who have a known disease, and
you’d compare that to people without. Now if you’re looking at
the genomes of people with the disease and those without the
disease, you can kind of play spot the difference between those
two, and understand whereabouts on the genome variants appear for
the disease population and not for the healthy or undiseased
control group. Now, when you do that, you can kind of pinpoint
exactly whereabouts you see variants only in that patient
population. That helps you identify your target, and that’s known
as target identification, which is essentially pinpointing that
spot on the genome that’s linked only to the disease. Once you
know that, you can use that as a potential target for a new
drug. 


So, once you’ve found that variant, the next step was, what does
that variant do? Is it potentially overproducing something? Is it
activating a promoter and therefore making more and more and more
of a gene product that, you know, might be toxic inside a person
if you have too much? Even too much of a good thing could be a
bad thing. So, is that the case? Or does that variant cause an
underproduction or something to just be not actually made by your
body at all? So, if that variant kind of interferes with a piece
of genetic code, it could stop that gene from producing anything,
and therefore you might be effectively detrimented and deprived
of that particular gene product. And both of those, an
overproduction or an underproduction, could lead to a disease.
So, to understand that in more detail, you might need to look at
gene products as well. 


The next step, once you know whereabouts in the genome you’re
looking and what exactly a variant does, the next step really for
a pharma company is how could you fix that. So, if you’re looking
at too little of something – so, if a variant stops a gene from
actually developing into a gene product then you might need a
drug to boost or to compensate for that, so potentially a
supplement or having some kind of drug that can get the body to
make more of that product. If on the other hand your body is
making too much of something in a way that could be toxic, you
kind of want a drug to reduce those levels, so a drug that could
potentially breakdown that gene product so that you don’t have
too much of it, or stop it from working effectively, so that it
doesn’t seem as if you have too much of it, or otherwise prevent
it being made altogether. 


Now, one example of this prevention is actually a gene silencing
drug, or an ASO, as they’re effectively known, which can be used
as a genetic mask. So, that sits on top of a gene and hides it,
so the body can’t actually make that dangerous varied gene
product. Now, if you’re going to make something like that, you
need to be absolutely sure that masking that entire gene and
stopping even a varied form of it isn’t dangerous, so that last
step really is making sure that your drug is safe and wouldn’t
cause any other issues. So, traditionally, that would have been
done using animal models as kind of a surrogate organism, but now
using genomic databases, you can use human genomics as kind of
real world examples of applying say a genetic mask and hiding an
entire gene or genetic section, and you can look through genetic
databases to have a look for individuals who are alive and
hopefully healthy in the population, who don’t express a certain
gene. So, if you can find people who are healthy, who don’t have
that gene or have variants that stop that gene from being
produced, you kind of can be confident that you can make a drug
to cover that and it would be considered safe. 


Naimah: Okay, so that’s really interesting. So, what you’re
saying is, by using human genomic data, we can test the impact
and safety of gene targeting drugs directly in humans.


James: Yes, exactly. So, you can ask that question of would
hiding that gene entirely cause any other health issues or any
adverse effects really from a drug that hides it. And the really
useful thing about that is that we’d know the impact of a gene
targeting drug before you’d say start a clinical trial, so that
really stacks the odds in your favour of the drug working safely,
which is really powerful for a drug company that would otherwise
invest a lot of money in a clinical trial that could be a risky
endeavour for the company and also for participants. So, this is
very useful for patients, and also fundamentally it’s a lot more
useful for a company to be assessing safety using humans and
human genomics directly as opposed to using a surrogate organism
like a mouse, which many people would argue is not a good
reflection of what would happen in humans.


Naimah: Can you tell me briefly if genomics can be applied to
other stages of the drug delivery pipeline? 


James: Yes, in fact genomics can be applied all along the drug
discovery and development R&D pipeline. So, as an example,
biomarker identification. A biomarker is a biological product or
a chemical signal that’s associated with a disease, that you can
find and monitor inside the 


body. So, you can look at an increase in that biomarker or a
decrease in that to monitor whether a drug is working as you’d
expect. Is the drug increasing levels of something being
produced, or is it decreasing that product being produced? And
you can use that to understand whether it’s possible to
potentially develop that treatment, would that treatment actually
work. So, that’s really important in monitoring drug impact and
also understanding clinical endpoints for a trial. 


You can also look at biomarker identification to look for genes
and variants that are associated with a disease that could help
you understand who best to enrol in a clinical trial. So,
clinical trial recruitment is another key area, where if you
involve the genome in your enrolment criteria, you can
essentially just recruit the most suitable people where you know
the drug will work best, and also you’re sure that the drug would
be most safe and effective at treating their condition. And then
actually to go a step further on the clinical trial point,
clinical genomic datasets are actually really useful, if you
think about it, in the opportunity to recontact participants too
where they’ve consented. So, what I mean by that is, a pharma
company could directly find and recruit optimal patients with
either a rare disease or a cancer where their drug would help
most, based solely on their genome, and that’s a really, really
exciting point, because that offers the opportunity for pharma to
both develop a drug based on that genomic dataset, but also then
deliver the drugs to treat those same exact people. 


Naimah: So, how do pharmaceutical companies access this
data? 


James: Well, there are different datasets, and each different
dataset has a different population within those, and each of them
have their own consent models and governance rules on how that
data can be used and who can access it, and how they access it.
So, some of these datasets just hold genomic data, while others
would have additional biochemical data and also health
information potentially on participants. So, depending on the
different types of data, there’ll be different access limitations
and restrictions. So, some entities and some datasets can be
simply downloaded, and that could be very useful for pharma and
biotech companies, because that means that they could use them
inhouse. Other datasets and groupings of genomic data and
libraries of sorts would operate a TRE or an SDE model, so that
stands for a trusted research environment or a secure data
environment, and these are essentially – you could consider them
as libraries, like a reading library, where you can come in and
read the books but not take out those books, or genomes in this
case. 


Naimah: Can you tell me, what impact does the use of genomic data
for drug discovery have on the public or patients? 


James: Oh, there’s huge impact on drug discovery, and ultimately
genomic research really helps drug companies make better
treatments for patients and the public. So, we’ve already seen
the benefits of genomics used in drug discovery, and I think we
will do more and more as DNA sequencing is used more in clinic,
and also that’s going to keep happening the more cost keeps
dropping, which is making genomic medicine really and genomic
healthcare increasingly feasible at scale. So, 20 years ago, it
cost over £100 million and it took years to sequence a genome,
but today you can sequence a genome within a few hours for under
£1,000. 


Naimah: What are the benefits of having your genomes sequenced in
a healthcare setting? 


James: Ultimately, genomics enable a faster and more accurate
diagnosis. That enables early intervention, which can really
maximise the treatment impact and improve outcomes. So, what I
mean by early intervention, if you can give a drug before someone
shows symptoms then you could prevent them ever getting the
disease, so that’s moving towards preventative medicine, which is
really exciting and absolutely enabled through genomics. So,
genomics really help pharma companies make also better drugs and
target the underlying disease directly rather than just
addressing symptoms, so this helps them make more effective and
safe treatments to really improve overall outcomes for
patients.  


Another thing to think about is that some drugs are already on
the market but used for different reasons. Genomics can help
pinpoint the root cause of that disease within a genomic setting,
so that can highlight repurposing opportunities for existing
drugs. Now, existing drugs are those that have already been
proven safe in humans and approved for use, albeit potentially in
a different setting. Now, if a drug could be shown by genomic
research to be targeting the same root cause within the same
biological pathway, they could very easily be repositioned and
applied in an entirely new disease. 


So, I guess to finish, through genomics, drug development can
help us move towards precision healthcare, and by that I mean
making targeted treatments for specific patients. That will be
far more effective and have significantly fewer side effects. In
the case of participants in clinical genomics sequencing
programmes open to researchers, that also means matchmaking
opportunities for companies to diagnose and treat unique
patients. In the case of ultra rare conditions, that means they
can create a treatment specifically for that patient and then
work with their doctors to deliver the brand new drug just to
them, to ultimately save lives. 


Naimah: That was James Duboff explaining how pharmaceutical
companies can use genomic data for drug discovery. If you’d like
to hear more explainer episodes like this, you can find them on
our website at www.genomicsengland.co.uk. Thank you for
listening.