October 17, 2024 • 42 mins
“Our liquid biopsy test alone, we estimate one in four cancer patients in the US are getting it,” Helmy Eltoukhy, cofounder and co-CEO of Guardant Health, tells Bloomberg Intelligence in this episode of the Vanguards of Health Care podcast. Eltoukhy joins BI analyst Jonathan Palmer to discuss the underpinnings of the liquid biopsy revolution and how Guardant has stuck to its technology road map since its founding in 2012, launching products in each of the three major market categories. The conversations also covers the company’s recent approval of its Shield colorectal screening test as well as many of the scientific and commercial challenges to standing up the business.
See omnystudio.com/listener for privacy information.
Mark as Played
Transcript
Episode Transcript
Available transcripts are automatically generated. Complete accuracy is not guaranteed.
Speaker 1 (00:21):
Welcome to another episode of the Vanguards of Healthcare podcast,
where we chat with the leaders at the forefront of
change in the healthcare industry. My name is Jonathan Palmer,
and I'm a healthcare analyst at Bloomberg Intelligence, the in
house research arm of Bloomberg. We're very happy to welcome
Helmi L two K for today's episode. He's the co
CEO and co founder of Garden Health, a leader in
(00:43):
precision oncology. Prior to starting Garden in twenty twelve, he
founded avant Home, which was acquired by Alumna. He completed
his studies at Stanford University, which culminated in a doctorate
degree in electrical engineering, and was a post doctoral fellow
at the university's Genome Technology Thank you so much for
joining us today. How we when we get started with
(01:04):
where Garden fits in the overall world of precision mythology.
Speaker 2 (01:08):
Yeah, no, thank you for having me. Yeah.
Speaker 3 (01:10):
So we Gardens really developed really the first commercially available
what's called liquid biopsy, this idea of being able to
get information from a tumor but through a simple blood test.
And so when we started, we launched the first commercially
available sort of clinical liquid biops in twenty fourteen, and
(01:33):
at that time, really the application of genomics was just.
Speaker 2 (01:38):
Starting to take off to essentially trumor treatment.
Speaker 3 (01:42):
Basically, the idea was, if I could cut a piece
of the tumor sequence the DNA that sent it, it
could give me a lot more information give on college.
It's a lot more information in terms of what best
treatment to apply to those patients. That was taking off,
but there are a lot of challenges there. The average
(02:04):
biopsy long biopsy costs fourteen thousand dollars to perform, as
a nineteen percent complication rate and a one to two
percent mortality rate. So many times physicians were sort of
avoiding doing some of these procedures because they're worried about
the health of the patient, putting them on empirical treatment.
And you know, fast forward today the twenty twenty four
(02:27):
you know, these liquid biopses are now probably forty percent
of all US cancer patients get them as part of
their standard of care. Our liquid biopsy alone, we has
to make one out of four cancer patients are getting
in the in the US, and so it's been really
exciting to see the sort of change that's happened in
(02:48):
a relatively short.
Speaker 2 (02:49):
Time in medicine.
Speaker 3 (02:52):
And so the nice thing is someone can get diagnosed,
have a blood test, and then be put on the
right treatment. And where the field has gone now is
that as those blood tests have gotten more sensitive, and
as we've developed the technology further, we've been able to
apply them to other sort of disease areas and on
college helping cancer survivors essentially with the currents monitoring. And
(03:17):
then most exciting and most recently two months ago, we
got f day approval for the first liquid biopsy that
can be used as a first line sort of screening
methodology for color actual cancer.
Speaker 1 (03:31):
Now, congratulations on that. That's fantastic. And maybe just to
tee off the conversation by rewinding a little bit, you
know when you were at a wumina, you know, when
did you have this realization that you can move these
research tools into the clinic. Was there an aha moment?
Speaker 2 (03:44):
Yeah?
Speaker 3 (03:44):
So you know I was even rewind back from there.
I was at the Stanford Genome Center working at the
tail end of the Human Genome project. First genome, you know,
cost three billion to sequence, and when I was at Alumina,
we got to this really vaunted goal of getting to
(04:05):
one thousand dollars genome, and frankly, we thought, okay, if
we can get to one thousand dollars per genome sequencing costs,
we could essentially have an explosion in terms of solving
human disease. All the information we'd be able to essentially
uncover and reveal would essentially be this sort of you know,
(04:27):
just an amazing sort of like moment in medicine.
Speaker 2 (04:30):
And and so we got there.
Speaker 3 (04:31):
It was, you know, it's around twenty eleven to twenty
twelve we got to the thousand dollars genome, and frankly,
it was more of a whimper than a bang.
Speaker 2 (04:41):
All we got was three.
Speaker 3 (04:42):
Billion letters and we didn't know what to do with them.
And so it's that it's that moment that me and
my co founder just felt like, you know, we could
be working on the five hundred dollars genome and one
hundred dollars gena, you know, trying to use quantum physics
and so on, and you know kind of you know,
these really hard problems, the scaling problems that are solving
(05:03):
or you know, I think sort of one step adjacent,
and you know, we have this amazing new technology.
Speaker 2 (05:10):
It sequence sequences DNA.
Speaker 3 (05:12):
It's really low costs, but you know, why not throw
our hat into actually trying to figure out how to
use this technology to solve human disease. And so that
was really where we sort of got the bug was
at that point, which is we thought there'd be a
lot more value in terms of the application of this technology.
And then you know, cancer specifically. You know, we've had
(05:35):
sort of personal sort of you know, interactions and unfortunately
then you know, delt with you know, cancer in our lives,
family lives, you know, friends' lives, and we ultimately know
you know, it is the sort of you know, the
definition of cancer by some is a disease of the genome.
And so that's really where we felt sort of low
(05:58):
cost genome sequencing could make it big, big difference, especially
if you could access the disease at any time through
a simple, simple blood We thought that it would lead
to an acceleration of progress.
Speaker 1 (06:11):
And you know it, frankly has did the bug catch
you even earlier than that? I mean, you did your
training in electrical engineering, and then this is probably more
of an interdisciplinary art or world, you know, is the
challenge there just more interesting yeah, I would say that.
Speaker 3 (06:25):
You know, growing up, I always wanted to go into medicine,
and then I got to Stanford as an undergrad and
realized it was you know, ten fifteen years.
Speaker 2 (06:35):
I said, well let me, let me, let me go
down to the.
Speaker 3 (06:37):
Engineering roud and I loved math and physics and problem
solving in general. And was it sort of I would say,
a random walk in some ways, because you know, I
was really going for circuit design, building semiconductors. And then
you know, Advisor I talked and said, oh, I have this,
you know, interesting problem where we're using image sensors like
(07:00):
you know, the computer chips and your cell phone that
take pictures and applying them to DNA sequencing. And you know,
I always you know, loved biology. I think, you know,
the back of my sort of mind was this idea
of getting into like medicine maybe through the back door,
and and so that, yeah, that became my PhD project,
and it was a really slippery slope, and that I've
(07:23):
actually became the focus of my.
Speaker 2 (07:26):
First company that it started.
Speaker 3 (07:28):
So yeah, it was was really just sort of getting
sucked into these problems. And you know, I think the
interdisciplinary pieces is something that I really value where frankly,
it was very hard to sort of move the needle
and sort of the mainstream disciplines like you know, circuit design,
(07:48):
all these things well trodden territory, so it's really hard
to innovate. But what I found is if you're gonna
understand two adjacent disciplines well enough, the values in between
those two peaks are often areas of I think opportunity
and really uh you know, I think prime for innovation.
And that really was the first company I did was
(08:11):
intersection of semiconductors and DNA sequencing, and then the second
one here is really an intersection of signal processing, you know,
advanced by the chemistry, and then yeah, a lot of
sort of like AI and machine learning that we're using.
Speaker 1 (08:27):
It's funny, it seems like yesterday to me, but I
know it was a long time ago and we were
talking about all these semi conductor technologies like ion torrent.
I have to ask, is any of that foundational technology
still in use at Illumina today.
Speaker 3 (08:39):
Yeah, the icy system, which is a semi conductor basis,
and that was essentially the fruits of the acquisition, and
then the in the research we did Alumina I think,
you know, in retrospect, you know, if we want to
get sort of more technical on the on the sequencing side,
you know, the semicon your approach sort of kind of
(09:02):
it had a time where it was a better approach,
but as the sort of need for sequencing and like,
you know, the throughput for sequencing grew, it actually wasn't
really a good architecture just because you have your chemistry
fixed to a really expensive substrate, which is a computer.
It's just hard to scale that in the same way
(09:24):
that a sort of removed optical system, you know, sort
of the ones that have won out today.
Speaker 1 (09:30):
I definitely have some questions about kind of the technology
evolution and sequencing and how that's impacting your business.
Speaker 2 (09:35):
But maybe if we talk about.
Speaker 1 (09:37):
The founding of Garden and your first launch of the
three sixty, I mean, can you talk about some of
the challenges in getting that product, you know, up to
scale and actually launched. You know, was it on the
chemistry side, was it on the informatic side, I mean,
what were the chief hurdles to getting that product to fruition?
Speaker 3 (09:54):
Yeah, there were a lot, I mean even just product
definition and believe it or not, was sort of big
open question. There really wasn't anything out there that was
the sort of similar and so the idea was, you know,
do you go really hot spot and just look at,
you know, a couple genes, do you look at like
a whole panel, do you look at you know, the
(10:15):
whole exome and and so there was just a lot
of sort of white space that that was there that
we could really use to sort of carve.
Speaker 2 (10:24):
The that that was one.
Speaker 3 (10:26):
Two it wasn't clear that you know, this actually works,
that you are doing, you know, looking at blood, that
the DNA that you're getting there that seems to be
coming from tumors is actually coming from the relevant part
of the tumors. There was a, i think the leading
hypothesis that oh, these are just dead cells that you know,
(10:47):
aren't fit parts of the tumor, that aren't representative of
the actual tumor that you want to treat. And so
there there was I think some fundamental science questions of
whether this approach one was relevant and two could be
sensitive enough to actually replace a tissue biopsy. And so
that's where we were fortunate that we worked with i
(11:09):
think five or six sort of leading cancer centers that
had bank samples and matched tissue biopsies, and so we
did a lot of i would say clinical validity, validation
of the science very early on. So yeah, that was
that was a second one. And then yeah, and there
were a lot of i think informatic challenges.
Speaker 2 (11:29):
If you think about it.
Speaker 3 (11:30):
We were trying to detect DNA at a fraction that
was probably one tenth or one one hundredth of what
has what had been detected for the noise of sequencing
is really high.
Speaker 2 (11:45):
It's at the one percent level, and the signals.
Speaker 3 (11:49):
We were trying to detect our one part in one thousand,
so so ten times ten times lower.
Speaker 2 (11:56):
So yeah, there were a lot of sort of both
chemistry challenges.
Speaker 3 (12:00):
The chemistry that was used for sequencing wasn't very efficient
and the conversion of molecules. We wanted to leave no
molecule left behind in a tube of blood becoming with
the tumor, so we had to improve the efficiency actually
by a factor of ten, and that.
Speaker 2 (12:16):
Coupled with improving the sort of err.
Speaker 3 (12:18):
Rate by you know, ten or one hundred or one
thousand sort of was a secret sauce to getting this
to work for the first time.
Speaker 1 (12:26):
And maybe just on the evolution of the product. You know,
can you talk a little bit about you know, you've
expanded I think to it's called a couple of different
flavors of three sixty and can you maybe tie that
back to the technology advancements as well.
Speaker 3 (12:39):
Yeah, So we have now maybe twelve official updates to
three sixty over them last ten years, and most recently
we really had a sort of major technology shift to
this what we call smart liquid biopsy. But you know
those previous eleven or sort of upgrades. Essentially, every ten thousand,
(13:03):
fifty thousand samples we sequenced, we learned a lot more
about sort of the idiosyncrasies of the chemistry, how to
improve that, and then frankly, much more importantly, we learned
a lot about the sort of signal part of it,
in terms of the signal seeing from blood and one
of the things that it wasn't just about lowering the
(13:25):
noise levels so we could see sort of every molecule
that was there and get to that single molecular sensitivity.
But it turns out there's biological noise as well. There's
these processes called chip or clonal hematopolysis and determined potential.
It's these mutations that happen in your blood cells that
essentially could look like cancer. But is this sort of
(13:48):
I would say, interfering symbol a signal that's there. And
so the more data that we're able to collect, the
more we're able to sort of use the machine learning
and classif you know, some of these kind of non
tumor drive mutations that are there there in blood. And
so we were just getting better at essentially separating signal
(14:10):
from noise and sort of multiple dimensions technically is biological
noise and so on, and that was really I think
a testament to the sort of number of patients who were.
Speaker 2 (14:19):
Able to help.
Speaker 3 (14:19):
And the same just the huge database we have now
petabytes and petabytes of information, some like fifty petabytes.
Speaker 1 (14:27):
Now, maybe talk about that back in piece a little
bit more, because if I think about machine learning and
where I've heard it used in application as opposed to
you know, on the research side, you guys are definitely
one of the first spaces that you know, we've seen
this really come to the four.
Speaker 3 (14:43):
Yeah, Like one of one of the things we used
to talk about was when we first got in the field,
and it still is I think largely I would say
one of the biggest challenges when we talk about data
and healthcare is that it's frankly just data starved. You know,
we think we have all this information you know, mars
and you know, some of the blood tests that people
take exams, but it's actually very sparse information and not
(15:07):
not very high quality frankly. And you know, when we
when we think about like even one blood test that
we have, like you know, three sixty we're generating I think,
you know, something like you know, tens, if not hundreds
of gigabytes of information for blood tests and so, you
know me obviously, the last year has been like you know,
(15:30):
truly exciting from a technological innovation point of view in
terms of the sort of like large language models that
are there and a lot of that is really you know, predicated,
and the fact that you know, for the last thirty
to forty years, you know, billions of human beings have
put sort of every thought into the Internet.
Speaker 1 (15:48):
And we have you know, we have a quintalient.
Speaker 3 (15:51):
Pages of information that's there, and so there was a
sort of critical juncture in terms of you know, having
that data and we view the blood tests that essentially
providing especially you know, as our screening test sort of
gets to sort of full force and we're we're helping
millions of individuals every year. That data deluge is you know,
(16:12):
we're generating data at a rate that's sort of unprecedented
in healthcare. Just if you think about longitudinal information and
every patient across millions of markers and the blood and
seeing how they change in a minute level. We think
that is a sort of like you know, first step
that's needed to get to a sort of new era
(16:33):
of very proactive medicine. And so that's that's you know,
base Camp for us is getting shield our early detection
sort of out there. And initially, you know, it's a
correctal cancer test. I'm sort of jumping you know ahead here,
but like it's a correct it's a corective cancer test.
(16:54):
But you know, we're we're seeing signals for dozens of
other cancers with the same test. It's it's just a
matter of turning the software on, you know, learning those
signals and then and then turning it on through over
the air sort of update, just like you know Tesla
does with it with its cars, and and then we're
also seeing signals and the you know, cardiovascular space and
(17:18):
inflammatory disease and so on.
Speaker 2 (17:20):
So it's truly.
Speaker 3 (17:21):
Exciting what what these blood tests may be able to
do over the next ten years.
Speaker 1 (17:25):
That's fascinating. You know, I think you know, most of
us are pretty familiar in the marketplace with the oncology
use case, but what are some of the cardiology or
neurology use cases that you potentially see coming to the.
Speaker 3 (17:37):
Fore Back to the beginning of the conversation about genomics
sort of being a little bit underwhelming.
Speaker 2 (17:41):
In some ways.
Speaker 3 (17:44):
You know, there's trillions of cells in our body all
have the same three you know, six billion, six billion
letters that we've inherited from our parents, and and yet
our cells are completely different. Your liver cells look different
than your skin cells, and you know I cells and
so on. And why is that is because you know,
(18:04):
there's a sort of epigenetic layer, the software layer that's
on top of the DNA, and so that's the technology
we sort of developed for early cancer detection, which is
looking at that software layer, that epigenetic layer, and when
we see those state of those switches, we not only
know what kind of cell it originated from, was it
you know, colon cell, liver cell, and so on, But
(18:27):
we can see the specific type of cancer that's there,
and that technology is general purpose, so we can see
you know, like heart tissue that's in the blood.
Speaker 2 (18:38):
We can see you.
Speaker 3 (18:38):
Know, signs of inflammatory disease diabetes, like you know, rheumat arthritis,
And so essentially, as long as we can read that
software layer and we have enough data to decode what
the state of the switch is actually, like mean, from
a normal tissue type, from a disease tissue type, from
the pathology of the disease, we can essentially apply this
(19:02):
to sort of any disease in the body as long
as some of those cells shed into the bloodstream. And
so that's what it was truly exciting about this is
genomics made a huge amount of progress in cancer, infectious diseases,
and rare diseases, but those are really the only diseases
where the genome changes. Most other diseases, it's just that
(19:25):
pigenetic layer that changes.
Speaker 1 (19:27):
And maybe just skipping back, you know, rewinding again, but
when you started the company, did you see these three
pillars to the stool or three legs of the stool
around screening treatment and MRD. Was that apparent early on
or did it grow over time.
Speaker 2 (19:41):
Yeah, that was very interesting.
Speaker 3 (19:43):
Like even our first PowerPoint presentation that you know, when
we started the company very clearly talked about those stepping stones.
Speaker 2 (19:51):
We have a.
Speaker 3 (19:53):
Concept that we like to call called the technology compounding,
where you know compounding. Obviously in finance, we're familiar with
very very strong force. It's the same thing in technology.
We sort of have the illusion that, you know, revolutions
happened really quickly, or paradigm shifts, you know, happened quickly,
but you know that happened slowly at first, and then
(20:15):
the gain of gain speed and and that's because they're
essentially exponentially building on the previous layer. And that's exactly
what we thought. In our goal was develop a blood
test and you know, so that annual physical you could
detect cancer early. That was always in our goal from
day one.
Speaker 2 (20:35):
But we always had a.
Speaker 3 (20:36):
Sort of a three step approach, which was, let's start
with late stage disease stage four, you know, cancer patients
where the significant fire and blood we you know, it's
maybe easier to develop either to commercialize and then learn
by doing like you know, the thousands, tens of thousands,
hundreds of thousands of blood tests that we processed at
(20:57):
that time we could learn from and so then yeah,
MRD was the sort of second stepping stone stage one
cancer patients, and then and then early detection was the third.
And you know, so the thirty five thousand foot view
was sort of eerily similar, like eerily sort of unchanged
from what we actually accomplished. But you know, obviously the
(21:19):
details are very different.
Speaker 1 (21:21):
There's a lot of ways to get to the same place, right.
Speaker 3 (21:23):
Yeah, And we thought, like you know, we would just
use genomics and detect it at lower and lower level,
and you know that obviously hit a wall just the
sensitivity is not there.
Speaker 1 (21:34):
You know, we've been we've been talking a lot about
I guess the science side of it, you know, I
think the other aspect from a business perspective is the
commercial you know, and reimbursement and diagnostics. I don't have
to tell you has been a challenge.
Speaker 2 (21:46):
Yeah, forever.
Speaker 1 (21:48):
Can you talk a little bit about the pathways, you know,
maybe we'll start with Gardener three sixty and segue into
some of the others like shield, you know, what is
the pathway or roadmap look from a reimbursement escape.
Speaker 3 (22:01):
Yeah, so, especially in cancer Medicare is typically the first
milestone getting covered for those over sixty five. And so
the government, you know, and kind of the sort of
Medicare jurisdictions that are there, they have a pretty i
would say, at least like straightforward approach in terms of
working with them to get a coverage decision for that
(22:24):
specific area put out there, and then obviously working with
them in terms of publications that are required to show
the clinical validity of the approach and some clinical utility.
And so they typically tend to be the first step.
And the private payers, you know, typically don't cover anything
unless Medicare is so you know, everyone is waiting for
(22:48):
everyone else. And the other challenge is that even though
Medicare is a sort of reasonable un ramp, it can
still take two to three years at minimum for a
completely new product to get reimbursed. I mean, i'll give
you bring in perspective, we launched Garden through sixty and
twenty fourteen and didn't get Medicare coverage probably until late
(23:11):
twenty seven. It was only last year that we sort
of got over three hundred covered lives in terms of
actually getting most patients covered with with Garden through sixty,
so it was nearly a ten year journey, and you know,
if we weren't sort of well financed, you know, we
obviously this wasn't like a big area. We've probably been
(23:32):
very difficult to commercialize any of these sort of new
technologies that are that are out there. So it's it's
a a joke that diagnostics, you know, or at least
starting a diagnostics company, I wouldn't wish on my worst
sort of enemy, just because there's.
Speaker 2 (23:51):
So much that is.
Speaker 3 (23:54):
Taking risks in terms of launching auts, getting the data,
investing before you see a sort of penny of return,
and I think that's really the hard part. That's also
why it's very hard for I would say, companies that
aren't purely diagnostic companies to sort of straddle different different
industry like tools companies. I'd much rather be in the
(24:16):
tours tool space.
Speaker 1 (24:18):
You build it, you ship, but you actually get paid
for it, no reiborsement risk.
Speaker 2 (24:23):
Yeah, timelines are short, you know.
Speaker 1 (24:25):
In conjunction with that, how has the market at least
on the therapy selection side. I mean, there's a lot
more players now today versus when you start. And I
guess maybe thinking about that from a couple of different lenses,
you know, how do you compete against the other players
out there? And then you know, I also look at
some of the other service spaces as an analyst, and
we've seen a lot of consolidation on the oncology MSO side,
(24:49):
and I'm just wondering how that might impact some of
the players in this space if they start to standardize
what oncologists are actually doing from a treatment or practice perspective.
Speaker 2 (25:00):
Yeah.
Speaker 3 (25:00):
No, I mean I think you have you bring up
really good points. And so, you know, we have three
areas of competition. There's a therapy selection side and there's
some sort of fairly you know, well capitalized companies in
that space. And there's the m r D side and
you know there's a couple of other players there, and
(25:21):
and then obviously screening and early detection there's some big,
big players there. So we we always feel like an underdog. Frankly,
We're working in you know, sort of all these three
areas and there's some you know, obviously, you know, I
think large, large, large players there. But you know, I
think that it's a double edged sword. The there's so
(25:43):
much excitement around sort of cancer testing, cancer screening and
so on, that it has been a sort of well
funded space, which I think has caused really the sort
of competitive makeup you see where you have multiple large
players that are continuing to do well, which I think
(26:03):
in most other spaces you would have seen consolidation much
much earlier. And so I think consolidation will happen eventually.
It just doesn't make sense for multiple companies to have
sort of huge sales forces be selling very similar products
in some ways.
Speaker 1 (26:19):
And what do you think is going to be the
catalyst for that?
Speaker 2 (26:21):
You know, I think I think eventually.
Speaker 3 (26:25):
There'll be certain companies that continue to innovate and continue
to sort of be at the forefront of you know,
sort of commanding kind of the right premium, bring gross
margins on their products as a result, and so on,
and then other companies that you end up being much
more commoditized. And so I think you'll see sort of
multiple separation in the space. And then at some point, yeah,
(26:48):
it'll make sense to you know, for certain players to
consolidate because why why sort of respend the three dollars
on the same sales and marketing across different players.
Speaker 2 (26:58):
But it'll take some time.
Speaker 3 (27:00):
I think there's still a lot of excitement in terms
of you know investors, you know, unfortunately, but yeah, eventually,
like all these spaces can consolidate, especially the commodity players.
Speaker 1 (27:13):
Got it, you know, maybe just shifting gears back to
shield because that was a very big win for you
guys this year. You know, listening to the ad COM,
it seemed like there was a lot of focused on,
you know, the advanced adenoma piece, and I guess maybe
as a way person, you know, I'm trying to wrap
my head around like why that piece was so important
for them, because if I think about screening tests with
(27:35):
my wayman's knowledge, I'm thinking about the negative predictive value,
not necessarily the positive predictive value of that piece, which
I think would have to be extremely high or the
sensitivity would have to be extremely high to get a
really good positive predictive value. Is that Is that the
right way of thinking about it? Yeah?
Speaker 3 (27:51):
I mean I think it was a sort of letting
perfect be the enemy of like really damn good kind
of like what we achieve. And you know, I would
say that you know that most of the tests out
there don't have great sensitivity for as you know, fit
is like in the twenty plus percent, and so we're
sort of comparing like you know, sort of a metric
(28:15):
that you know, it actually doesn't have like a great
sort of comparitor.
Speaker 2 (28:20):
That's that's out there.
Speaker 3 (28:21):
And and one where the science I think is less certain.
We only we know that you know, less than five
percent of advance at nom has actually become cancer. You
get sort of mini shots and goal, you know, what
would you wanted to tell you would you want to
have a test that had one hundred percent sensitivity for
a of course, But I think we were sort of
(28:43):
missing the point that we've we have, you know, a
bunch of good tests that are out there today with
only around like sixty percent net compliance.
Speaker 2 (28:53):
And so we have.
Speaker 3 (28:55):
Essentially fifty thousand people dying of colorrectal cancer every single year,
seventy six percent of those coming from the unscreened population,
from the fifty million Americans who essentially are not up
to date with screening with colon ask Beer's tool based testing.
And and so this is really where I think we're
(29:15):
super excited about the bending of mortality curves that we
could potentially engender. You know, by shield, we're seeing compliance
rates that adherence rates in the ninety plus percent, and
when you take this sort of participation, patient participation piece
out of the sort of patient's hands and it just
(29:36):
becomes a sort of blood routine. Blood test really changes
the nature of access.
Speaker 2 (29:43):
To this technology.
Speaker 3 (29:44):
And so yeah, we're yeah, we're I think we're very
pleased that in the end sort of I think the
science and you know, I think that the true benefits
of this test sort of one out and and yeah,
and we're very excited with the traction we're seeing so far.
Speaker 1 (30:04):
What do you have to do from an organizational perspective
just for the shield launch? You know, do you have
to invest in the labs, do you have to build
out the salesforce? You know, how does that?
Speaker 2 (30:13):
Yeah, we're all of the above.
Speaker 3 (30:15):
So, you know, I think we're fortunate that, you know,
this is the sort of third product that you know,
like a big franchise. We've launched a company in terms
of testing, so operationally, I think we're very well equipped
to sort of get the through put you know up. Yeah,
the salesforce you know, will be a sort of over
(30:37):
one hundred by end of this year, and then eventually
over the next few years, wouldn't have to build it
up to a you know, six seven hundred person sales
force and maybe even larger over time since there are
two hundred thousand primary care physicians out there, a lot
of marketing materials, you know, have a pretty bold marketing
(30:58):
campaign as well. So yeah, it's it's just it's really
going from zero to sixty and sort of two seconds.
Speaker 1 (31:07):
That's a good problem to have.
Speaker 3 (31:08):
It's a problem to have, and you know, despite the competition,
you know, we think we have over a two year
head start in terms of any other blood tests that
would be sort of medicare approved for screening. So yeah,
we want to sort of make the best of it
and make sure we get as many patients as possible.
Speaker 1 (31:28):
You just touched on something that I wanted to ask about,
which is how meaningful you think that that first mover
advantage is going to be in this marketplace.
Speaker 2 (31:35):
It's it's always it's always important.
Speaker 3 (31:37):
I mean, this is something that in many case studies
on in the pharma space in terms of what first
mover advantage in the drug space like means. And yeah,
you see considerable separation even having a year sort of
head start in the therapeutic space, and you know, we believe,
you know, having you know, two two plus years is enormous,
(31:58):
not just because I sort of getting to more patients,
getting to them quickly, but also in terms of this
tech compounding that I talked about. You know, today it's
a correctal test, you know, tomorrow it's looking at dozens
of cancers, and then maybe you know, you know, further
down the road, it's looking at dozens of different diseases.
(32:19):
And and I think the head start is not just
about traction with the with the test, but it's also
traction with the sort of innovation curve that we're going
to have with that with that same test, And.
Speaker 2 (32:30):
Yeah, it's very easy.
Speaker 3 (32:31):
Like we see people comparing sort of single cancer approaches
to SHIELD and and yeah that that may look relevant today,
but not in sort of you know, in a couple
of years, and we're looking at, yeah, dozens and dozens
of cancers with that with that same test.
Speaker 1 (32:49):
And so just out of curiosity, why is that single
cancer approach not as applicable? Is it just the underlying technology?
Speaker 3 (32:58):
It's just like apples to orange. It's like, you know,
if I start comparing my landline to my smartphone, because
you know, they share one function, it's it's just a
different platform that's there. And so you just the utility
function is very different. If you're a primary care physician,
you know, you have something that does really well, and
(33:20):
does you know one hundred different things you're probably going
to do that versus you know, one basically tacking on
you know, one hundred or sort of different different tests
that it's there at it. You know, I think the
idea is, you know, how good does that sort of
artisanal like butcher have to be good?
Speaker 2 (33:41):
Have to going to the supermarket? Right?
Speaker 1 (33:45):
I like that analogy. Maybe going back to the technology
piece of it. You know, we've touched on this a
little bit, but you know, can you talk about the
the kind of the digital piece of your business, the
underlying data and you know there's a competitor out there
that's monetizing that with clinicians and pharma customers, and I
guess what's your thought process around a digital piece of
(34:06):
your business?
Speaker 2 (34:08):
Yeah? I know we're seeing a lot of traction there.
It's an area that.
Speaker 3 (34:13):
Our methodology was the data is going to be important,
but once there's enough of it and the right kinds
of it, And so our viewpoint was, let's focus on
sort of creating the data with all the tests we're
doing and getting there as quickly as possible. And so
that has really been the first sort of decade of
(34:33):
sort of development of gardens is building that testing layer,
that data acquisition layer, so to speak. And now we're
with Garden and Form, we're really starting to I think
see a lot of sort of exciting progress there and insights.
We work with I think dozens of pharma companies now
just in the data set, we work with one hundred
and eighty former companies on the testing side, and we've
(34:57):
been able to essentially take our data and because a
lot of it is longitudinal, it's the first time I
really see the evolution of disease sort of like in
response to therapeutic pressure and all these things. And so
it's it's it's a really unique database where you can
see with the sort of exquisite detail and resolution exactly
(35:19):
the effect certain drugs have.
Speaker 2 (35:21):
On the you know, the.
Speaker 3 (35:23):
Genetics of the disease, a sort of time course of disease.
And and then now with the epigenetic layer that is
sort of a backbone to our tests, we're just seeing
a different, you know, sort of side of cancer. I'll
give you an example like PARP inhibitors are obviously very
important for those with a sort of bracketing mutation. It
(35:46):
turns out that the sort of epigenetic silencing or promoter
methylation of BRACA is actually twice as prevalent as the mutations,
and yet no one has been sort of testing for that.
No one is detecting it because we haven't, you know,
we haven't. We don't have the technologies before our test
(36:08):
to be able to look there. And so now these
are a whole bunch of sort of patients that are
immediately candidates for these important class of therapies that sort
of no one knew about, and so there were this
is just many of like dozens or you know, sort
of applications it in day one. If you can see
the epigenetics of the disease, you can essentially put patients
(36:31):
on better therapies. You can design better therapies. We're seeing
sort of cancers that you would think are all the
same because they have the same mutation. When you look
at them with the epigenetic lens, you can actually see
subclasses of those of the diseases and actually, you know,
see very different response rates in terms of subtype one
(36:53):
versus subtype two of the same sort of gene mutation
or gene mutated disease, and so it's it's we think
this is going to essentially spur I think much more precise.
Speaker 2 (37:07):
Drug development. Really, I think put us into.
Speaker 3 (37:09):
A higher gear for what we thought precision medicine would
give us.
Speaker 1 (37:14):
Now, do you have to build a salesforce around that
component or are the drug companies coming to you and saying,
what might what data might you have around PARPs that
sort of.
Speaker 2 (37:22):
Thing working with them. We have a pretty large business too.
Speaker 1 (37:27):
You know, we've only got a couple more minutes here.
And one of the things I just wanted to touch
base on was, you know, giving your background and sequencing
and technology, just how has the cost curve or maybe
not even the cost curve the evolution of sequencing technology.
It's a more competitive space than it probably ever was.
You know, we've got Aluminum watching the novase x and
their five base genome. You know, how meaningful is this
(37:49):
for your business that the underlying I guess technology moves forward.
Speaker 2 (37:55):
Yeah, I mean it's been I think relatively meaningful. I
think when we.
Speaker 3 (38:01):
Started, you know, I think we're just getting two thousand
dollars genome maybe a couple of thousand dollars genome and yeah,
now you're you have companies have been a little talking
about a hundred dollars two hundred dollars genomes, and essentially
we look at that as bandwidth. When you think about, like,
you know, companies that are in the communications space, if
(38:21):
you know, they can get more bandwidth, they can get
sort of more juice in the pipes. It just means
they can do more the same thing for us. You know,
the the more we can look in you know, various
areas of the genome, the better a signal has got,
our signal to noise. We detect cancer earlier, we potentially
(38:41):
other types of diseases. I mean, we see a future
that you know, eventually the cost of extracting sort of
an infinite amount of data from your blood will go
down to zero or will tend down to zero. When
you think about the ten dollars genome, and you know,
there have been a lot of exciting advancements and proteomics,
some of these technologies that allow us to look at
(39:04):
you know, between five thousand to ten thousand proteins in
the blood and when that comes down to very low costs,
it's it's going to be like, you know, fantastic, just
the amount of like data, the sort of the amount
of signals we can see you in a tube of
blood and see that sort of you know, longitudinal sort
(39:26):
of way. I mean, I would say that, you know,
the classic example I gave is baseline temperature, like very
few people actually know, like there we're told the fairy
tale it's six, but it can be anywhere from you know,
sort of ninety seven to ninety nine. And and so
if something like temperature we don't measure longitudinally today, I mean,
(39:49):
just think about like how much progress we can make
when we're looking at sort of billions of markers every year.
Speaker 2 (39:56):
Got it? You know?
Speaker 1 (39:57):
I like to wrap up these conversations with a lesson
from the guest about something that drives them and their
day to day. Is there is there one thing if
you think about your past, that that drives you, know,
your your mission or how you inform your team about
what you want to do with with your time.
Speaker 3 (40:11):
Yeah, I would say that it's you know, for us,
our cultural drives from being patient first. And you know
in our quality policy, you know, all our materials talk
about you know, really putting yourself in the shoes of
the patient, imagining there they're like you know, loved one,
(40:32):
your mother or your father, brother, sister. And what we
found is that, you know, this is a hard space.
It's it's you know, the decisions aren't very clearly black
and white or sort of left or right in terms
of word, and so when you use that lens of
you know, putting yourself in the shoes of the patients,
(40:55):
it really is very clarifying and sort of these especially
difficult junctures have as a as a company. And so
there's been many times where you know, the sort of
patient first rubric, you know, maybe has led us to
something that maybe in the short term isn't sort of
as profitable or you know, isn't lucky the return ROI
(41:17):
isn't there. But it's always been absolutely the right decision,
you know, both obviously for patients, but both for the
company as well and the medium to long term. And
so it's really been something that we take too heart.
Speaker 1 (41:29):
That's great, So I think we'll wrap it up here.
That's Helming l two K, CEO of Garden Health, Thank
you so much for joining us for our latest episode.
For our listeners, Please make sure to click the follow
button on your favorite podcast, app or website so you
never miss a discussion with the leaders in healthcare innovation.
I'm Jonathan Palmer, and you've been listening to the Vanguards
of Healthcare podcast by Bloomberg Intelligence. Thank you, Take care,
(42:01):
pass
Speaker 2 (42:17):
Us bases.
Welcome to another episode of the Vanguards of Healthcare podcast,
where we chat with the leaders at the forefront of
change in the healthcare industry. My name is Jonathan Palmer,
and I'm a healthcare analyst at Bloomberg Intelligence, the in
house research arm of Bloomberg. We're very happy to welcome
Helmi L two K for today's episode. He's the co
CEO and co founder of Garden Health, a leader in
(00:43):
precision oncology. Prior to starting Garden in twenty twelve, he
founded avant Home, which was acquired by Alumna. He completed
his studies at Stanford University, which culminated in a doctorate
degree in electrical engineering, and was a post doctoral fellow
at the university's Genome Technology Thank you so much for
joining us today. How we when we get started with
(01:04):
where Garden fits in the overall world of precision mythology.
Speaker 2 (01:08):
Yeah, no, thank you for having me. Yeah.
Speaker 3 (01:10):
So we Gardens really developed really the first commercially available
what's called liquid biopsy, this idea of being able to
get information from a tumor but through a simple blood test.
And so when we started, we launched the first commercially
available sort of clinical liquid biops in twenty fourteen, and
(01:33):
at that time, really the application of genomics was just.
Speaker 2 (01:38):
Starting to take off to essentially trumor treatment.
Speaker 3 (01:42):
Basically, the idea was, if I could cut a piece
of the tumor sequence the DNA that sent it, it
could give me a lot more information give on college.
It's a lot more information in terms of what best
treatment to apply to those patients. That was taking off,
but there are a lot of challenges there. The average
(02:04):
biopsy long biopsy costs fourteen thousand dollars to perform, as
a nineteen percent complication rate and a one to two
percent mortality rate. So many times physicians were sort of
avoiding doing some of these procedures because they're worried about
the health of the patient, putting them on empirical treatment.
And you know, fast forward today the twenty twenty four
(02:27):
you know, these liquid biopses are now probably forty percent
of all US cancer patients get them as part of
their standard of care. Our liquid biopsy alone, we has
to make one out of four cancer patients are getting
in the in the US, and so it's been really
exciting to see the sort of change that's happened in
(02:48):
a relatively short.
Speaker 2 (02:49):
Time in medicine.
Speaker 3 (02:52):
And so the nice thing is someone can get diagnosed,
have a blood test, and then be put on the
right treatment. And where the field has gone now is
that as those blood tests have gotten more sensitive, and
as we've developed the technology further, we've been able to
apply them to other sort of disease areas and on
college helping cancer survivors essentially with the currents monitoring. And
(03:17):
then most exciting and most recently two months ago, we
got f day approval for the first liquid biopsy that
can be used as a first line sort of screening
methodology for color actual cancer.
Speaker 1 (03:31):
Now, congratulations on that. That's fantastic. And maybe just to
tee off the conversation by rewinding a little bit, you
know when you were at a wumina, you know, when
did you have this realization that you can move these
research tools into the clinic. Was there an aha moment?
Speaker 2 (03:44):
Yeah?
Speaker 3 (03:44):
So you know I was even rewind back from there.
I was at the Stanford Genome Center working at the
tail end of the Human Genome project. First genome, you know,
cost three billion to sequence, and when I was at Alumina,
we got to this really vaunted goal of getting to
(04:05):
one thousand dollars genome, and frankly, we thought, okay, if
we can get to one thousand dollars per genome sequencing costs,
we could essentially have an explosion in terms of solving
human disease. All the information we'd be able to essentially
uncover and reveal would essentially be this sort of you know,
(04:27):
just an amazing sort of like moment in medicine.
Speaker 2 (04:30):
And and so we got there.
Speaker 3 (04:31):
It was, you know, it's around twenty eleven to twenty
twelve we got to the thousand dollars genome, and frankly,
it was more of a whimper than a bang.
Speaker 2 (04:41):
All we got was three.
Speaker 3 (04:42):
Billion letters and we didn't know what to do with them.
And so it's that it's that moment that me and
my co founder just felt like, you know, we could
be working on the five hundred dollars genome and one
hundred dollars gena, you know, trying to use quantum physics
and so on, and you know kind of you know,
these really hard problems, the scaling problems that are solving
(05:03):
or you know, I think sort of one step adjacent,
and you know, we have this amazing new technology.
Speaker 2 (05:10):
It sequence sequences DNA.
Speaker 3 (05:12):
It's really low costs, but you know, why not throw
our hat into actually trying to figure out how to
use this technology to solve human disease. And so that
was really where we sort of got the bug was
at that point, which is we thought there'd be a
lot more value in terms of the application of this technology.
And then you know, cancer specifically. You know, we've had
(05:35):
sort of personal sort of you know, interactions and unfortunately
then you know, delt with you know, cancer in our lives,
family lives, you know, friends' lives, and we ultimately know
you know, it is the sort of you know, the
definition of cancer by some is a disease of the genome.
And so that's really where we felt sort of low
(05:58):
cost genome sequencing could make it big, big difference, especially
if you could access the disease at any time through
a simple, simple blood We thought that it would lead
to an acceleration of progress.
Speaker 1 (06:11):
And you know it, frankly has did the bug catch
you even earlier than that? I mean, you did your
training in electrical engineering, and then this is probably more
of an interdisciplinary art or world, you know, is the
challenge there just more interesting yeah, I would say that.
Speaker 3 (06:25):
You know, growing up, I always wanted to go into medicine,
and then I got to Stanford as an undergrad and
realized it was you know, ten fifteen years.
Speaker 2 (06:35):
I said, well let me, let me, let me go
down to the.
Speaker 3 (06:37):
Engineering roud and I loved math and physics and problem
solving in general. And was it sort of I would say,
a random walk in some ways, because you know, I
was really going for circuit design, building semiconductors. And then
you know, Advisor I talked and said, oh, I have this,
you know, interesting problem where we're using image sensors like
(07:00):
you know, the computer chips and your cell phone that
take pictures and applying them to DNA sequencing. And you know,
I always you know, loved biology. I think, you know,
the back of my sort of mind was this idea
of getting into like medicine maybe through the back door,
and and so that, yeah, that became my PhD project,
and it was a really slippery slope, and that I've
(07:23):
actually became the focus of my.
Speaker 2 (07:26):
First company that it started.
Speaker 3 (07:28):
So yeah, it was was really just sort of getting
sucked into these problems. And you know, I think the
interdisciplinary pieces is something that I really value where frankly,
it was very hard to sort of move the needle
and sort of the mainstream disciplines like you know, circuit design,
(07:48):
all these things well trodden territory, so it's really hard
to innovate. But what I found is if you're gonna
understand two adjacent disciplines well enough, the values in between
those two peaks are often areas of I think opportunity
and really uh you know, I think prime for innovation.
And that really was the first company I did was
(08:11):
intersection of semiconductors and DNA sequencing, and then the second
one here is really an intersection of signal processing, you know,
advanced by the chemistry, and then yeah, a lot of
sort of like AI and machine learning that we're using.
Speaker 1 (08:27):
It's funny, it seems like yesterday to me, but I
know it was a long time ago and we were
talking about all these semi conductor technologies like ion torrent.
I have to ask, is any of that foundational technology
still in use at Illumina today.
Speaker 3 (08:39):
Yeah, the icy system, which is a semi conductor basis,
and that was essentially the fruits of the acquisition, and
then the in the research we did Alumina I think,
you know, in retrospect, you know, if we want to
get sort of more technical on the on the sequencing side,
you know, the semicon your approach sort of kind of
(09:02):
it had a time where it was a better approach,
but as the sort of need for sequencing and like,
you know, the throughput for sequencing grew, it actually wasn't
really a good architecture just because you have your chemistry
fixed to a really expensive substrate, which is a computer.
It's just hard to scale that in the same way
(09:24):
that a sort of removed optical system, you know, sort
of the ones that have won out today.
Speaker 1 (09:30):
I definitely have some questions about kind of the technology
evolution and sequencing and how that's impacting your business.
Speaker 2 (09:35):
But maybe if we talk about.
Speaker 1 (09:37):
The founding of Garden and your first launch of the
three sixty, I mean, can you talk about some of
the challenges in getting that product, you know, up to
scale and actually launched. You know, was it on the
chemistry side, was it on the informatic side, I mean,
what were the chief hurdles to getting that product to fruition?
Speaker 3 (09:54):
Yeah, there were a lot, I mean even just product
definition and believe it or not, was sort of big
open question. There really wasn't anything out there that was
the sort of similar and so the idea was, you know,
do you go really hot spot and just look at,
you know, a couple genes, do you look at like
a whole panel, do you look at you know, the
(10:15):
whole exome and and so there was just a lot
of sort of white space that that was there that
we could really use to sort of carve.
Speaker 2 (10:24):
The that that was one.
Speaker 3 (10:26):
Two it wasn't clear that you know, this actually works,
that you are doing, you know, looking at blood, that
the DNA that you're getting there that seems to be
coming from tumors is actually coming from the relevant part
of the tumors. There was a, i think the leading
hypothesis that oh, these are just dead cells that you know,
(10:47):
aren't fit parts of the tumor, that aren't representative of
the actual tumor that you want to treat. And so
there there was I think some fundamental science questions of
whether this approach one was relevant and two could be
sensitive enough to actually replace a tissue biopsy. And so
that's where we were fortunate that we worked with i
(11:09):
think five or six sort of leading cancer centers that
had bank samples and matched tissue biopsies, and so we
did a lot of i would say clinical validity, validation
of the science very early on. So yeah, that was
that was a second one. And then yeah, and there
were a lot of i think informatic challenges.
Speaker 2 (11:29):
If you think about it.
Speaker 3 (11:30):
We were trying to detect DNA at a fraction that
was probably one tenth or one one hundredth of what
has what had been detected for the noise of sequencing
is really high.
Speaker 2 (11:45):
It's at the one percent level, and the signals.
Speaker 3 (11:49):
We were trying to detect our one part in one thousand,
so so ten times ten times lower.
Speaker 2 (11:56):
So yeah, there were a lot of sort of both
chemistry challenges.
Speaker 3 (12:00):
The chemistry that was used for sequencing wasn't very efficient
and the conversion of molecules. We wanted to leave no
molecule left behind in a tube of blood becoming with
the tumor, so we had to improve the efficiency actually
by a factor of ten, and that.
Speaker 2 (12:16):
Coupled with improving the sort of err.
Speaker 3 (12:18):
Rate by you know, ten or one hundred or one
thousand sort of was a secret sauce to getting this
to work for the first time.
Speaker 1 (12:26):
And maybe just on the evolution of the product. You know,
can you talk a little bit about you know, you've
expanded I think to it's called a couple of different
flavors of three sixty and can you maybe tie that
back to the technology advancements as well.
Speaker 3 (12:39):
Yeah, So we have now maybe twelve official updates to
three sixty over them last ten years, and most recently
we really had a sort of major technology shift to
this what we call smart liquid biopsy. But you know
those previous eleven or sort of upgrades. Essentially, every ten thousand,
(13:03):
fifty thousand samples we sequenced, we learned a lot more
about sort of the idiosyncrasies of the chemistry, how to
improve that, and then frankly, much more importantly, we learned
a lot about the sort of signal part of it,
in terms of the signal seeing from blood and one
of the things that it wasn't just about lowering the
(13:25):
noise levels so we could see sort of every molecule
that was there and get to that single molecular sensitivity.
But it turns out there's biological noise as well. There's
these processes called chip or clonal hematopolysis and determined potential.
It's these mutations that happen in your blood cells that
essentially could look like cancer. But is this sort of
(13:48):
I would say, interfering symbol a signal that's there. And
so the more data that we're able to collect, the
more we're able to sort of use the machine learning
and classif you know, some of these kind of non
tumor drive mutations that are there there in blood. And
so we were just getting better at essentially separating signal
(14:10):
from noise and sort of multiple dimensions technically is biological
noise and so on, and that was really I think
a testament to the sort of number of patients who were.
Speaker 2 (14:19):
Able to help.
Speaker 3 (14:19):
And the same just the huge database we have now
petabytes and petabytes of information, some like fifty petabytes.
Speaker 1 (14:27):
Now, maybe talk about that back in piece a little
bit more, because if I think about machine learning and
where I've heard it used in application as opposed to
you know, on the research side, you guys are definitely
one of the first spaces that you know, we've seen
this really come to the four.
Speaker 3 (14:43):
Yeah, Like one of one of the things we used
to talk about was when we first got in the field,
and it still is I think largely I would say
one of the biggest challenges when we talk about data
and healthcare is that it's frankly just data starved. You know,
we think we have all this information you know, mars
and you know, some of the blood tests that people
take exams, but it's actually very sparse information and not
(15:07):
not very high quality frankly. And you know, when we
when we think about like even one blood test that
we have, like you know, three sixty we're generating I think,
you know, something like you know, tens, if not hundreds
of gigabytes of information for blood tests and so, you
know me obviously, the last year has been like you know,
(15:30):
truly exciting from a technological innovation point of view in
terms of the sort of like large language models that
are there and a lot of that is really you know, predicated,
and the fact that you know, for the last thirty
to forty years, you know, billions of human beings have
put sort of every thought into the Internet.
Speaker 1 (15:48):
And we have you know, we have a quintalient.
Speaker 3 (15:51):
Pages of information that's there, and so there was a
sort of critical juncture in terms of you know, having
that data and we view the blood tests that essentially
providing especially you know, as our screening test sort of
gets to sort of full force and we're we're helping
millions of individuals every year. That data deluge is you know,
(16:12):
we're generating data at a rate that's sort of unprecedented
in healthcare. Just if you think about longitudinal information and
every patient across millions of markers and the blood and
seeing how they change in a minute level. We think
that is a sort of like you know, first step
that's needed to get to a sort of new era
(16:33):
of very proactive medicine. And so that's that's you know,
base Camp for us is getting shield our early detection
sort of out there. And initially, you know, it's a
correctal cancer test. I'm sort of jumping you know ahead here,
but like it's a correct it's a corective cancer test.
(16:54):
But you know, we're we're seeing signals for dozens of
other cancers with the same test. It's it's just a
matter of turning the software on, you know, learning those
signals and then and then turning it on through over
the air sort of update, just like you know Tesla
does with it with its cars, and and then we're
also seeing signals and the you know, cardiovascular space and
(17:18):
inflammatory disease and so on.
Speaker 2 (17:20):
So it's truly.
Speaker 3 (17:21):
Exciting what what these blood tests may be able to
do over the next ten years.
Speaker 1 (17:25):
That's fascinating. You know, I think you know, most of
us are pretty familiar in the marketplace with the oncology
use case, but what are some of the cardiology or
neurology use cases that you potentially see coming to the.
Speaker 3 (17:37):
Fore Back to the beginning of the conversation about genomics
sort of being a little bit underwhelming.
Speaker 2 (17:41):
In some ways.
Speaker 3 (17:44):
You know, there's trillions of cells in our body all
have the same three you know, six billion, six billion
letters that we've inherited from our parents, and and yet
our cells are completely different. Your liver cells look different
than your skin cells, and you know I cells and
so on. And why is that is because you know,
(18:04):
there's a sort of epigenetic layer, the software layer that's
on top of the DNA, and so that's the technology
we sort of developed for early cancer detection, which is
looking at that software layer, that epigenetic layer, and when
we see those state of those switches, we not only
know what kind of cell it originated from, was it
you know, colon cell, liver cell, and so on, But
(18:27):
we can see the specific type of cancer that's there,
and that technology is general purpose, so we can see
you know, like heart tissue that's in the blood.
Speaker 2 (18:38):
We can see you.
Speaker 3 (18:38):
Know, signs of inflammatory disease diabetes, like you know, rheumat arthritis,
And so essentially, as long as we can read that
software layer and we have enough data to decode what
the state of the switch is actually, like mean, from
a normal tissue type, from a disease tissue type, from
the pathology of the disease, we can essentially apply this
(19:02):
to sort of any disease in the body as long
as some of those cells shed into the bloodstream. And
so that's what it was truly exciting about this is
genomics made a huge amount of progress in cancer, infectious diseases,
and rare diseases, but those are really the only diseases
where the genome changes. Most other diseases, it's just that
(19:25):
pigenetic layer that changes.
Speaker 1 (19:27):
And maybe just skipping back, you know, rewinding again, but
when you started the company, did you see these three
pillars to the stool or three legs of the stool
around screening treatment and MRD. Was that apparent early on
or did it grow over time.
Speaker 2 (19:41):
Yeah, that was very interesting.
Speaker 3 (19:43):
Like even our first PowerPoint presentation that you know, when
we started the company very clearly talked about those stepping stones.
Speaker 2 (19:51):
We have a.
Speaker 3 (19:53):
Concept that we like to call called the technology compounding,
where you know compounding. Obviously in finance, we're familiar with
very very strong force. It's the same thing in technology.
We sort of have the illusion that, you know, revolutions
happened really quickly, or paradigm shifts, you know, happened quickly,
but you know that happened slowly at first, and then
(20:15):
the gain of gain speed and and that's because they're
essentially exponentially building on the previous layer. And that's exactly
what we thought. In our goal was develop a blood
test and you know, so that annual physical you could
detect cancer early. That was always in our goal from
day one.
Speaker 2 (20:35):
But we always had a.
Speaker 3 (20:36):
Sort of a three step approach, which was, let's start
with late stage disease stage four, you know, cancer patients
where the significant fire and blood we you know, it's
maybe easier to develop either to commercialize and then learn
by doing like you know, the thousands, tens of thousands,
hundreds of thousands of blood tests that we processed at
(20:57):
that time we could learn from and so then yeah,
MRD was the sort of second stepping stone stage one
cancer patients, and then and then early detection was the third.
And you know, so the thirty five thousand foot view
was sort of eerily similar, like eerily sort of unchanged
from what we actually accomplished. But you know, obviously the
(21:19):
details are very different.
Speaker 1 (21:21):
There's a lot of ways to get to the same place, right.
Speaker 3 (21:23):
Yeah, And we thought, like you know, we would just
use genomics and detect it at lower and lower level,
and you know that obviously hit a wall just the
sensitivity is not there.
Speaker 1 (21:34):
You know, we've been we've been talking a lot about
I guess the science side of it, you know, I
think the other aspect from a business perspective is the
commercial you know, and reimbursement and diagnostics. I don't have
to tell you has been a challenge.
Speaker 2 (21:46):
Yeah, forever.
Speaker 1 (21:48):
Can you talk a little bit about the pathways, you know,
maybe we'll start with Gardener three sixty and segue into
some of the others like shield, you know, what is
the pathway or roadmap look from a reimbursement escape.
Speaker 3 (22:01):
Yeah, so, especially in cancer Medicare is typically the first
milestone getting covered for those over sixty five. And so
the government, you know, and kind of the sort of
Medicare jurisdictions that are there, they have a pretty i
would say, at least like straightforward approach in terms of
working with them to get a coverage decision for that
(22:24):
specific area put out there, and then obviously working with
them in terms of publications that are required to show
the clinical validity of the approach and some clinical utility.
And so they typically tend to be the first step.
And the private payers, you know, typically don't cover anything
unless Medicare is so you know, everyone is waiting for
(22:48):
everyone else. And the other challenge is that even though
Medicare is a sort of reasonable un ramp, it can
still take two to three years at minimum for a
completely new product to get reimbursed. I mean, i'll give
you bring in perspective, we launched Garden through sixty and
twenty fourteen and didn't get Medicare coverage probably until late
(23:11):
twenty seven. It was only last year that we sort
of got over three hundred covered lives in terms of
actually getting most patients covered with with Garden through sixty,
so it was nearly a ten year journey, and you know,
if we weren't sort of well financed, you know, we
obviously this wasn't like a big area. We've probably been
(23:32):
very difficult to commercialize any of these sort of new
technologies that are that are out there. So it's it's
a a joke that diagnostics, you know, or at least
starting a diagnostics company, I wouldn't wish on my worst
sort of enemy, just because there's.
Speaker 2 (23:51):
So much that is.
Speaker 3 (23:54):
Taking risks in terms of launching auts, getting the data,
investing before you see a sort of penny of return,
and I think that's really the hard part. That's also
why it's very hard for I would say, companies that
aren't purely diagnostic companies to sort of straddle different different
industry like tools companies. I'd much rather be in the
(24:16):
tours tool space.
Speaker 1 (24:18):
You build it, you ship, but you actually get paid
for it, no reiborsement risk.
Speaker 2 (24:23):
Yeah, timelines are short, you know.
Speaker 1 (24:25):
In conjunction with that, how has the market at least
on the therapy selection side. I mean, there's a lot
more players now today versus when you start. And I
guess maybe thinking about that from a couple of different lenses,
you know, how do you compete against the other players
out there? And then you know, I also look at
some of the other service spaces as an analyst, and
we've seen a lot of consolidation on the oncology MSO side,
(24:49):
and I'm just wondering how that might impact some of
the players in this space if they start to standardize
what oncologists are actually doing from a treatment or practice perspective.
Speaker 2 (25:00):
Yeah.
Speaker 3 (25:00):
No, I mean I think you have you bring up
really good points. And so, you know, we have three
areas of competition. There's a therapy selection side and there's
some sort of fairly you know, well capitalized companies in
that space. And there's the m r D side and
you know there's a couple of other players there, and
(25:21):
and then obviously screening and early detection there's some big,
big players there. So we we always feel like an underdog. Frankly,
We're working in you know, sort of all these three
areas and there's some you know, obviously, you know, I
think large, large, large players there. But you know, I
think that it's a double edged sword. The there's so
(25:43):
much excitement around sort of cancer testing, cancer screening and
so on, that it has been a sort of well
funded space, which I think has caused really the sort
of competitive makeup you see where you have multiple large
players that are continuing to do well, which I think
(26:03):
in most other spaces you would have seen consolidation much
much earlier. And so I think consolidation will happen eventually.
It just doesn't make sense for multiple companies to have
sort of huge sales forces be selling very similar products
in some ways.
Speaker 1 (26:19):
And what do you think is going to be the
catalyst for that?
Speaker 2 (26:21):
You know, I think I think eventually.
Speaker 3 (26:25):
There'll be certain companies that continue to innovate and continue
to sort of be at the forefront of you know,
sort of commanding kind of the right premium, bring gross
margins on their products as a result, and so on,
and then other companies that you end up being much
more commoditized. And so I think you'll see sort of
multiple separation in the space. And then at some point, yeah,
(26:48):
it'll make sense to you know, for certain players to
consolidate because why why sort of respend the three dollars
on the same sales and marketing across different players.
Speaker 2 (26:58):
But it'll take some time.
Speaker 3 (27:00):
I think there's still a lot of excitement in terms
of you know investors, you know, unfortunately, but yeah, eventually,
like all these spaces can consolidate, especially the commodity players.
Speaker 1 (27:13):
Got it, you know, maybe just shifting gears back to
shield because that was a very big win for you
guys this year. You know, listening to the ad COM,
it seemed like there was a lot of focused on,
you know, the advanced adenoma piece, and I guess maybe
as a way person, you know, I'm trying to wrap
my head around like why that piece was so important
for them, because if I think about screening tests with
(27:35):
my wayman's knowledge, I'm thinking about the negative predictive value,
not necessarily the positive predictive value of that piece, which
I think would have to be extremely high or the
sensitivity would have to be extremely high to get a
really good positive predictive value. Is that Is that the
right way of thinking about it? Yeah?
Speaker 3 (27:51):
I mean I think it was a sort of letting
perfect be the enemy of like really damn good kind
of like what we achieve. And you know, I would
say that you know that most of the tests out
there don't have great sensitivity for as you know, fit
is like in the twenty plus percent, and so we're
sort of comparing like you know, sort of a metric
(28:15):
that you know, it actually doesn't have like a great
sort of comparitor.
Speaker 2 (28:20):
That's that's out there.
Speaker 3 (28:21):
And and one where the science I think is less certain.
We only we know that you know, less than five
percent of advance at nom has actually become cancer. You
get sort of mini shots and goal, you know, what
would you wanted to tell you would you want to
have a test that had one hundred percent sensitivity for
a of course, But I think we were sort of
(28:43):
missing the point that we've we have, you know, a
bunch of good tests that are out there today with
only around like sixty percent net compliance.
Speaker 2 (28:53):
And so we have.
Speaker 3 (28:55):
Essentially fifty thousand people dying of colorrectal cancer every single year,
seventy six percent of those coming from the unscreened population,
from the fifty million Americans who essentially are not up
to date with screening with colon ask Beer's tool based testing.
And and so this is really where I think we're
(29:15):
super excited about the bending of mortality curves that we
could potentially engender. You know, by shield, we're seeing compliance
rates that adherence rates in the ninety plus percent, and
when you take this sort of participation, patient participation piece
out of the sort of patient's hands and it just
(29:36):
becomes a sort of blood routine. Blood test really changes
the nature of access.
Speaker 2 (29:43):
To this technology.
Speaker 3 (29:44):
And so yeah, we're yeah, we're I think we're very
pleased that in the end sort of I think the
science and you know, I think that the true benefits
of this test sort of one out and and yeah,
and we're very excited with the traction we're seeing so far.
Speaker 1 (30:04):
What do you have to do from an organizational perspective
just for the shield launch? You know, do you have
to invest in the labs, do you have to build
out the salesforce? You know, how does that?
Speaker 2 (30:13):
Yeah, we're all of the above.
Speaker 3 (30:15):
So, you know, I think we're fortunate that, you know,
this is the sort of third product that you know,
like a big franchise. We've launched a company in terms
of testing, so operationally, I think we're very well equipped
to sort of get the through put you know up. Yeah,
the salesforce you know, will be a sort of over
(30:37):
one hundred by end of this year, and then eventually
over the next few years, wouldn't have to build it
up to a you know, six seven hundred person sales
force and maybe even larger over time since there are
two hundred thousand primary care physicians out there, a lot
of marketing materials, you know, have a pretty bold marketing
(30:58):
campaign as well. So yeah, it's it's just it's really
going from zero to sixty and sort of two seconds.
Speaker 1 (31:07):
That's a good problem to have.
Speaker 3 (31:08):
It's a problem to have, and you know, despite the competition,
you know, we think we have over a two year
head start in terms of any other blood tests that
would be sort of medicare approved for screening. So yeah,
we want to sort of make the best of it
and make sure we get as many patients as possible.
Speaker 1 (31:28):
You just touched on something that I wanted to ask about,
which is how meaningful you think that that first mover
advantage is going to be in this marketplace.
Speaker 2 (31:35):
It's it's always it's always important.
Speaker 3 (31:37):
I mean, this is something that in many case studies
on in the pharma space in terms of what first
mover advantage in the drug space like means. And yeah,
you see considerable separation even having a year sort of
head start in the therapeutic space, and you know, we believe,
you know, having you know, two two plus years is enormous,
(31:58):
not just because I sort of getting to more patients,
getting to them quickly, but also in terms of this
tech compounding that I talked about. You know, today it's
a correctal test, you know, tomorrow it's looking at dozens
of cancers, and then maybe you know, you know, further
down the road, it's looking at dozens of different diseases.
(32:19):
And and I think the head start is not just
about traction with the with the test, but it's also
traction with the sort of innovation curve that we're going
to have with that with that same test, And.
Speaker 2 (32:30):
Yeah, it's very easy.
Speaker 3 (32:31):
Like we see people comparing sort of single cancer approaches
to SHIELD and and yeah that that may look relevant today,
but not in sort of you know, in a couple
of years, and we're looking at, yeah, dozens and dozens
of cancers with that with that same test.
Speaker 1 (32:49):
And so just out of curiosity, why is that single
cancer approach not as applicable? Is it just the underlying technology?
Speaker 3 (32:58):
It's just like apples to orange. It's like, you know,
if I start comparing my landline to my smartphone, because
you know, they share one function, it's it's just a
different platform that's there. And so you just the utility
function is very different. If you're a primary care physician,
you know, you have something that does really well, and
(33:20):
does you know one hundred different things you're probably going
to do that versus you know, one basically tacking on
you know, one hundred or sort of different different tests
that it's there at it. You know, I think the
idea is, you know, how good does that sort of
artisanal like butcher have to be good?
Speaker 2 (33:41):
Have to going to the supermarket? Right?
Speaker 1 (33:45):
I like that analogy. Maybe going back to the technology
piece of it. You know, we've touched on this a
little bit, but you know, can you talk about the
the kind of the digital piece of your business, the
underlying data and you know there's a competitor out there
that's monetizing that with clinicians and pharma customers, and I
guess what's your thought process around a digital piece of
(34:06):
your business?
Speaker 2 (34:08):
Yeah? I know we're seeing a lot of traction there.
It's an area that.
Speaker 3 (34:13):
Our methodology was the data is going to be important,
but once there's enough of it and the right kinds
of it, And so our viewpoint was, let's focus on
sort of creating the data with all the tests we're
doing and getting there as quickly as possible. And so
that has really been the first sort of decade of
(34:33):
sort of development of gardens is building that testing layer,
that data acquisition layer, so to speak. And now we're
with Garden and Form, we're really starting to I think
see a lot of sort of exciting progress there and insights.
We work with I think dozens of pharma companies now
just in the data set, we work with one hundred
and eighty former companies on the testing side, and we've
(34:57):
been able to essentially take our data and because a
lot of it is longitudinal, it's the first time I
really see the evolution of disease sort of like in
response to therapeutic pressure and all these things. And so
it's it's it's a really unique database where you can
see with the sort of exquisite detail and resolution exactly
(35:19):
the effect certain drugs have.
Speaker 2 (35:21):
On the you know, the.
Speaker 3 (35:23):
Genetics of the disease, a sort of time course of disease.
And and then now with the epigenetic layer that is
sort of a backbone to our tests, we're just seeing
a different, you know, sort of side of cancer. I'll
give you an example like PARP inhibitors are obviously very
important for those with a sort of bracketing mutation. It
(35:46):
turns out that the sort of epigenetic silencing or promoter
methylation of BRACA is actually twice as prevalent as the mutations,
and yet no one has been sort of testing for that.
No one is detecting it because we haven't, you know,
we haven't. We don't have the technologies before our test
(36:08):
to be able to look there. And so now these
are a whole bunch of sort of patients that are
immediately candidates for these important class of therapies that sort
of no one knew about, and so there were this
is just many of like dozens or you know, sort
of applications it in day one. If you can see
the epigenetics of the disease, you can essentially put patients
(36:31):
on better therapies. You can design better therapies. We're seeing
sort of cancers that you would think are all the
same because they have the same mutation. When you look
at them with the epigenetic lens, you can actually see
subclasses of those of the diseases and actually, you know,
see very different response rates in terms of subtype one
(36:53):
versus subtype two of the same sort of gene mutation
or gene mutated disease, and so it's it's we think
this is going to essentially spur I think much more precise.
Speaker 2 (37:07):
Drug development. Really, I think put us into.
Speaker 3 (37:09):
A higher gear for what we thought precision medicine would
give us.
Speaker 1 (37:14):
Now, do you have to build a salesforce around that
component or are the drug companies coming to you and saying,
what might what data might you have around PARPs that
sort of.
Speaker 2 (37:22):
Thing working with them. We have a pretty large business too.
Speaker 1 (37:27):
You know, we've only got a couple more minutes here.
And one of the things I just wanted to touch
base on was, you know, giving your background and sequencing
and technology, just how has the cost curve or maybe
not even the cost curve the evolution of sequencing technology.
It's a more competitive space than it probably ever was.
You know, we've got Aluminum watching the novase x and
their five base genome. You know, how meaningful is this
(37:49):
for your business that the underlying I guess technology moves forward.
Speaker 2 (37:55):
Yeah, I mean it's been I think relatively meaningful. I
think when we.
Speaker 3 (38:01):
Started, you know, I think we're just getting two thousand
dollars genome maybe a couple of thousand dollars genome and yeah,
now you're you have companies have been a little talking
about a hundred dollars two hundred dollars genomes, and essentially
we look at that as bandwidth. When you think about, like,
you know, companies that are in the communications space, if
(38:21):
you know, they can get more bandwidth, they can get
sort of more juice in the pipes. It just means
they can do more the same thing for us. You know,
the the more we can look in you know, various
areas of the genome, the better a signal has got,
our signal to noise. We detect cancer earlier, we potentially
(38:41):
other types of diseases. I mean, we see a future
that you know, eventually the cost of extracting sort of
an infinite amount of data from your blood will go
down to zero or will tend down to zero. When
you think about the ten dollars genome, and you know,
there have been a lot of exciting advancements and proteomics,
some of these technologies that allow us to look at
(39:04):
you know, between five thousand to ten thousand proteins in
the blood and when that comes down to very low costs,
it's it's going to be like, you know, fantastic, just
the amount of like data, the sort of the amount
of signals we can see you in a tube of
blood and see that sort of you know, longitudinal sort
(39:26):
of way. I mean, I would say that, you know,
the classic example I gave is baseline temperature, like very
few people actually know, like there we're told the fairy
tale it's six, but it can be anywhere from you know,
sort of ninety seven to ninety nine. And and so
if something like temperature we don't measure longitudinally today, I mean,
(39:49):
just think about like how much progress we can make
when we're looking at sort of billions of markers every year.
Speaker 2 (39:56):
Got it? You know?
Speaker 1 (39:57):
I like to wrap up these conversations with a lesson
from the guest about something that drives them and their
day to day. Is there is there one thing if
you think about your past, that that drives you, know,
your your mission or how you inform your team about
what you want to do with with your time.
Speaker 3 (40:11):
Yeah, I would say that it's you know, for us,
our cultural drives from being patient first. And you know
in our quality policy, you know, all our materials talk
about you know, really putting yourself in the shoes of
the patient, imagining there they're like you know, loved one,
(40:32):
your mother or your father, brother, sister. And what we
found is that, you know, this is a hard space.
It's it's you know, the decisions aren't very clearly black
and white or sort of left or right in terms
of word, and so when you use that lens of
you know, putting yourself in the shoes of the patients,
(40:55):
it really is very clarifying and sort of these especially
difficult junctures have as a as a company. And so
there's been many times where you know, the sort of
patient first rubric, you know, maybe has led us to
something that maybe in the short term isn't sort of
as profitable or you know, isn't lucky the return ROI
(41:17):
isn't there. But it's always been absolutely the right decision,
you know, both obviously for patients, but both for the
company as well and the medium to long term. And
so it's really been something that we take too heart.
Speaker 1 (41:29):
That's great, So I think we'll wrap it up here.
That's Helming l two K, CEO of Garden Health, Thank
you so much for joining us for our latest episode.
For our listeners, Please make sure to click the follow
button on your favorite podcast, app or website so you
never miss a discussion with the leaders in healthcare innovation.
I'm Jonathan Palmer, and you've been listening to the Vanguards
of Healthcare podcast by Bloomberg Intelligence. Thank you, Take care,
(42:01):
pass
Speaker 2 (42:17):
Us bases.
Host
Jonathan Palmer
Popular Podcasts
24/7 News: The Latest
The latest news in 4 minutes updated every hour, every day.
Therapy Gecko
An unlicensed lizard psychologist travels the universe talking to strangers about absolutely nothing. TO CALL THE GECKO: follow me on https://www.twitch.tv/lyleforever to get a notification for when I am taking calls. I am usually live Mondays, Wednesdays, and Fridays but lately a lot of other times too. I am a gecko.
The Joe Rogan Experience
The official podcast of comedian Joe Rogan.