November 16, 2023 • 43 mins
Welcome to the Olink® Proteomics in Proximity podcast!
Below are some useful resources mentioned in this episode:
Olink tools and software
• Olink® Explore 3072, the platform utilized by the UK Biobank to measure ~3000 proteins in plasma: https://olink.com/products-services/explore/
• Olink® Explore HT, Olink’s most advanced solution for high-throughput biomarker discovery, measuring 5400+ proteins simultaneously with a streamlined workflow and industry-leading specificity: https://olink.com/products-services/exploreht/
UK Biobank Pharma Proteomics Project (UKB-PPP), one of the world’s largest scientific studies of blood protein biomarkers conducted to date, https://www.ukbiobank.ac.uk/learn-more-about-uk-biobank/news/uk-biobank-launches-one-of-the-largest-scientific-studies
Research articles
• Dhindsa, R.S., Burren, O.S., Sun, B.B. et al. Rare variant associations with plasma protein levels in the UK Biobank. 2023 Nature, DOI: 10.1038/s41586-023-06547-x
https://www.nature.com/articles/s41586-023-06547-x
• Sun, B.B., Chiou, J., Traylor, M. et al. Plasma proteomic associations with genetics and health in the UK Biobank. 2023 Nature, DOI: 10.1038/s41586-023-06592-6
https://www.nature.com/articles/s41586-023-06592-6
• Ticau S, Sridharan G, Tsour S, et al. Neurofilament Light Chain as a Biomarker of Hereditary Transthyretin-Mediated Amyloidosis 2021 Neurology, DOI: 10.1212/WNL.0000000000011090
https://n.neurology.org/content/96/3/e412.long
• Zannad F, Ferreira JP, Butler J, et al. Effect of Empagliflozin on Circulating Proteomics in Heart Failure: Mechanistic Insights from the EMPEROR Program. 2022 European Heart Journal, DOI: 10.1093/eurheartj/ehac495
https://academic.oup.com/eurheartj/advance-article/doi/10.1093/eurheartj/ehac495/6676779
• Eldjarn GH, et al. Large-scale plasma proteomics comparisons through genetics and disease associations. Nature. 2023 Oct;622(7982):348-358. doi: 10.1038/s41586-023-06563-x
https://www.nature.com/articles/s41586-023-06563-x#Sec44
• [PREPRINT] Carrasco-Zanini et al 2023 Proteomic prediction of common and rare diseases MedRxiv https://www.medrxiv.org/content/10.1101/2023.07.18.23292811v1
• Michaëlsson E, Lund LH, Hage C, et al. Myeloperoxidase Inhibition Reverses Biomarker Profiles Associated With Clinical Outcomes in HFpEF. 2023 JACC. Heart Failure, DOI: 10.1016/j.jchf.2023.03.002
https://www.sciencedirect.com/science/article/pii/S2213177923001257
• Girerd N, Levy D, Duarte K, et al. Protein Biomarkers of New-Onset Heart Failure: Insights From the Heart Omics and Ageing Cohort, the Atherosclerosis Risk in Communities Study, and the Framingham Heart Study. 2023 Circulation Heart Failure, DOI: 10.1161/CIRCHEARTFAILURE.122.009694
https://www.ahajournals.org/doi/abs/10.1161/CIRCHEARTFAILURE.122.009694
Subscribe to the podcast on your favorite player or app:
Apple Podcasts: https://apple.co/3T0YbSm
Spotify Podcasts: https://open.spotify.com/show/2sZ2wxO...
Google Podcasts: https://podcasts.google.com/feed/aHR0...
Amazon Music: https://music.amazon.com/podcasts/d97...
Podcast Addict: https://podcastaddict.com/podcast/409...
Deezer: https://www.deezer.com/show/5178787
Player FM: https://player.fm/series/series-3396598
In case you were wondering, Proteomics in Proximity refers to the principle underlying Olink technology called the Proximity Extension Assay (PEA). More information about the assay and how it works can be found here: https://bit.ly/3Rt7YiY
For any questions regarding information about Olink Proteomics, please email us at info@olink.com or visit our website: https://www.olink.com/
Interested in a specific podcast topic or guest? Reach out to us at PIP@olink.com
WHAT IS PROTEOMICS IN PROXIMITY?
Proteomics in Proximity discusses the intersection of pr
Mark as Played
Transcript
Episode Transcript
Available transcripts are automatically generated. Complete accuracy is not guaranteed.
(00:05):
Welcometo the Proteomics and Proximity Podcast.
Where your co-hosts, Cindy Lawleyand Sarantis Chlamydas from Olink Proteomics,
talk about the intersection of proteomicswith genomics for drug target
discovery, the application of proteomicsto reveal disease biomarkers,
and current trends in using proteomicsto unlock biological mechanisms.
(00:26):
Here we have your hosts, Cindy and Sarantis.
Hey, everyone.
Hello and welcome backto Proteomics in Proximity.
Thanks to our 11 listeners at Sam Ray,Carolina, others.
We are grateful for your attention
and your feedback, and
(00:47):
our listeners have given ussome valuable feedback over time
and they've found us through differentsocial media avenues.
But to make that easier,we're announcing that
we've got actually an email address now,so we'll put this into the show notes.
But it's just P-I-P for Proteomicsin Proximity at Olink.com. [pip@olink.com]
And and we'd be happy to hear from youaround suggestions
(01:11):
you have or any interview recommendationsyou might have.
And with that, today,we are talking to Evan Mills.
Evan, I'll let him introduce himself,but he is
an illustrious pharma executive, here
actually at Olink,and we're excited to talk to him
(01:31):
about how pharma are finding proteomics
super relevant on many different levels.
So with that,let's get on with it.
Hey, Sarantis, how are you?
Hello. I'm fine.Thank you, Cindy. Welcome, Evan.
I'm looking forward to
hear from you all the great news.
Likewise. Good afternoon, Sarantis.
Good early morning to you,Cindy, on the West coast.
(01:53):
It's a little dark over here.A little dark
No, that's all right.I'm really honored to be here.
And I've been wanting to talk abouthow proteomics
and the pharmaceutical industryare aligning for really exciting things.
So very happy to be here.
Can you give us just a little backgroundon your history in this area?
You've been in this for a while.
(02:16):
I have.
I have.
So I was a bench scientist really,you know, passionate about oncology
and neuroscience research.
I did some work at Yale Universityfor awhile
and then I got into the salescommercial side of this world,
started actually in the pharmaceuticalsales industry, which
was exciting because of the opportunityto help patients,
right?
(02:36):
But my real passion was in the scienceand about a decade ago,
there was a very innovativeproteomics company
that caught my attentionand that's where I started this journey,
where I've now been at Olink for overfive years.
And yeah, supporting the most innovative,
ambitious researchers in this multi-omicsspace has just been a phenomenal journey.
So my background is (03:00):
I love science, I want to help people,
I want to have some sort of translationalimpact with the work I do. And
right now at this moment, there's neverbeen more momentum in that direction.
It's really, really exciting.Yes, very exciting.
We've just had here at Olink
(03:21):
three pretty exciting nature papers
come out in
I think it's the online [version on] October 4th.
But the print journal [on] October 11with a beautiful frog on the cover.
It's an exciting timewith those three papers.
So those include a lot of applications
(03:43):
aroundwhy pharma would invest in proteomics.
So I'd love to get your thoughts on why.
Why did 13 pharma cometogether, invest in proteomics?
What's the outcome?
What's the resultthat they see out of that?
Yeah, I mean that's been a real career
(04:05):
highlight, is being able to be involvedin that project from its inception.
And you know, Cindy,with your background in genetics,
there was a previously formedconsortium around whole exome sequencing
in the UK Biobank and then eventuallythe whole genome sequencing.
But there was this idea,and I was having lunch
(04:25):
outside of the Harvard symposium with Dr.
Chris Whelan, very smart geneticistwho was at Biogen at the time.
He's now at Janssen.
And he just asked the question.
He's like, "Hey, we're thinking aboutwhat makes sense to do next.
We have all this richnessin the genomic data,
but we want to do somethingcloser to phenotype.
(04:46):
Would it even be conceivablefor Olink to run 50,000 samples?"
And this is before,
you know, some of the innovationthat would have made that possible.
And we said, "Yeah,I think we can do that.
I think we can get there,I think we can do that."
So it was just born out of curiosity
and the desireto get closer to phenotype.
(05:06):
So the
goal really of this ambitious project was:
can we both better understand
drug targetsthat have causal links to disease
and can we simultaneously find biomarkersto help the drug development process?
Because obviously with proteomicsyou can do both, right?
It can act as a bit of a filterto tell you which of these
(05:30):
genomic disease associationshave a plausible biological story
and which ones should be pursued,and which ones should perhaps be killed,
but simultaneouslyyou can develop a suite of tools
to determine riskbased on proteomics, to determine disease
progression based on proteomics,and to discover biomarkers,
which are obviously always desired
(05:51):
to aid clinical development. So,
I mean, we're just starting to seeall the publications.
We're starting to understandall the utility that's going
to come from this data set.
And it was just such an ambitious,
smart idea by Chris and then eventuallyMelissa and Linden and Brad,
that countless others who contributedto the project.
(06:14):
Evan, going back to this journey,this amazing journey,
how easy or difficultwas it to convince the genomics community
because you mentioned it was like thisheavy genomics community, right?
Change their mindset in a wayto measure proteins,
how easy or difficult was this process?
It's really hard.
I think it is.
I think it was really hardbecause if you just think about the tools
(06:38):
one would need to developto measure, right?
Our DNA is very nicely organizedinto a helical structure.
There's four bases to measure, and Illuminaand others
now have developed amazing toolsthat can measure that at scale.
Think about the proteome, right?
There's 20 amino acids.
They combinein a myriad of different ways.
(06:58):
I mean, it's just such a formidablechallenge that geneticists would say,
"I'm not so sure.
I'm not so sure the tools exist.
And oh, by the way, yeah, we can measureeverything with genomics.
We can measure everything.
And you're approaching us with somethingthat measures 1500 at the time?"
Right.
And then 3000 of what people assumewould be
(07:21):
maybe 20,000 proteinsthat you could try to capture
in plasma / serum,there's a big debate about that.
So it is challenging,
BUT the obvious central dogmaof being closer to disease
and things
that are reflective of real-time biologyversus your blueprint for your biology
(07:42):
was compelling enough for themto give it a shot, but it was not easy.
So you must have focused onwhat is the near-term
return on investment for pharma,for running a proteomics project.
And I would consider this UK Biobank sort of pQTL developing therapeutic targets.
All of the -
all of those things you've alreadymentioned is more mid- or long-term goals.
(08:05):
How did you - I think it's a great questionSarantis asked -
how do you talk to themabout what you believe is the value?
And I will
also say we - Gary and I - looked ...Gary is our illustrious person
who manages our databaseof over 1400 peer-reviewed publications
and he has seen over 84 ofthose are pharma relevant publication.
(08:28):
So there's a significant numberof publications that have been
that have been put out therethat document
some value to pharma,but that's pretty recent.
How did you approach themwhen you first came to Olink?
Yeah, no, it's a good question.
So we can take a bit of a sidebarfrom the UK Biobank discussion because
(08:51):
really, fundamentally, drug developers
are trying to bring effectivetherapeutics to market faster
and they also invest enormous resourcesinto each program.
And it takes what, 10 to 15 yearson average, you know, to get something approved.
And how many millions of dollars, right?
(09:11):
And patience.
And then what,
90% of clinical trials failI think, or somewhere around there.
Yeah.
I recently had a discussionwith an executive vice
president of research at a major companywho said he would be the world's
best drug developerif he failed 80% of the time.
Isn't that wild?If he could go 80% after failing 90%, he would be the best.
(09:36):
And it's - right?
It's just such a high attrition game.
But those are, Cindy,the way that
a lot of people in the industryare starting to look at
this is (09:46):
with population-scale proteomics or high-throughput proteomics,
you can learn a lot about thingsyou've already invested in.
So let's say that you have a drugthat's approved such as -
I can never say that correctly.
Jardiance, let's go with Jardiance [empaglioflozin].
Yeah.
You know, a very, very effective SLT2
(10:09):
inhibitor used for the treatment of
diabetic control.
They've also noticed, after having itin enough humans in the wild,
that there's significant benefitsto heart failure.
So if you can access - and this is one ofthe publications that you referenced -
if you can access samplesfrom completed clinical trials
(10:32):
and most companies are sitting on these,they're just in their freezers
waiting to be analyzedif they have the exploratory consents.
If you take a
look at proteomics at scale from
lots of humans treatedin the clinical setting, you can learn
a tremendous amount about why certainpeople respond and certain people don't.
(10:53):
Right.
That's the Holy Grail, essentially is: can you proactively know (10:53):
undefined
which patients could go, should go,which therapy.
We often call that "stratifying
patients," just to use the termthat we've used before.
Yep. Yes, absolutely.
And you know, understanding the mechanismof these drugs, right?
Because, you know, you have a target,you have a hypothesis, you tested it in cell-
(11:15):
based models, animal-based models,but you don't really have a chance
to look at scale in a human populationto see how it impacts the human body.
So then with that data,you can A) better understand why there's
this benefit in an indication for whichthe drug was not initially approved.
You can understand - Excuse me,
what other pathways are being
(11:37):
impacted by your therapeutic.Are there repurposing opportunities?
Is there a way to very rapidly
take this thingyou've invested in, this asset,
and figure out that there's more placesthat you could help people,
there's more indications where this drugwould actually be a really good fit.
So that's a very shortwin, short-term win.
(11:57):
And we've noticed multiple clientsbuilding this as a strategy
to take Olink Proteomicsin this case to better
understand already approved drugs,which, in some ways, is counterintuitive.
Right?
I mean, ideallyyou think from the beginning
you would want to know everythingyou can about the drug,
but there's this reverse translation
(12:18):
movement that seems to be bearingquite a bit of fruit for the industry.
That was actually my next-
I'm sorry,
that's certainly my intriguing invite.
This is my next question, Evan, do you seenow this trend of a strategy
in the pharma
because you talk with
the executives, right? And you wouldknow the strategy and you discuss
about this. Do you see this coming?
Do you see that using large-scale proteomics,a big number of data to reposition
(12:42):
a drug, for example, to identifymechanisms of action even in the late stage?
How is your feeling?And why would they ever publish this?
Right.
We think of pharma as needingto hold these things tight.
So yeah, great questions, Sarantis.
That's a good question.
So the answer is yes, in pockets.
I think it's just becoming
(13:03):
a strategyfor the more innovative companies.
Right.
There's always some concern, right,for ongoing trials.
Do we really want to knowthat much at a phase three?
Right.
If we have a candidate compound,do we want to do exploratory research?
Maybe we find something we can't explain.
Maybe we find some safety signals.
(13:24):
So what I'm describing is drugs approved.
Let's extractas much value from that asset as we can.
And there's definitely companiesthat are taking that on as a strategy.
And to your point,I mean, having gotten to know
folks in pharmareally well for the last decade, I mean,
they're great scientists.I think there's this -
(13:45):
I think, not to insult
any of my academic colleagues or people
I've worked with or people that, you know,I've supported over the last 20 years.
I think there's incredibly talented
scientists that see the opportunity
to have a fast path to impact.
And they do want to share.They want to publish.
I mean, look at this consortium.
(14:06):
It was 13 companies that are competitorscoming together
I was complimentingone of the pharma researchers on a hire,
a new hire from academia.
And she was saying they came to mebecause they're a physician,
an M.D., Ph.D., and they said,
I can help one patientat a time in my practice.
(14:28):
But if I come here and do more
broad-based research,I can affect millions.
And I was like, Wow, that'sthat's an interesting perspective.
I like that. And it lines upwith what you're saying.
I think a great exampleof reverse translation that you've talked about.
I think one of the
examplesyou've talked about in the past of
(14:52):
of this, you know, taking samplesthat are sitting in the freezer
where a massive investment has beenmade is the one from Simina Ticau.
And
Paul Nioi. Paul is, of course,
also on the UK Biobank
flagship paperthat came out last week / this week,
(15:13):
whichever online footprintyou want to reference.
Can you tell us about that example?
Yeah,this is a really, really interesting story
and that this originated aboutfive years ago and was published in 2019.
So it's a bit dated, but I think the pointis incredibly powerful.
So, you know, hereditarytransthyretin-mediated amyloidosis is
(15:36):
a genetically defined disease
that really has -
And can I just say that you can pronounce that,
but empaglioflozin is pretty
darn easier to say than -
I'm sorry. It just seems funny.
[Empaglioflozin is] Jardiance, butanyway, you know, back to back to hATTR.
No,
you know,
I've probably told that storymore times than -
(15:58):
Yeah,
my gosh, it's hard.
So no, but, hATTR is a really,
really debilitating diseasewith a variable rate of onset.
So if it's the hereditary form,it runs in your family.
Right.
You can be screened to knowif you're carrier
and know if you're at riskfor developing the disease.
(16:20):
Alnylam developed a drug, patisiran,
that is an siRNA - excuse me,
RNAi-based therapeutic where they are
very effective at slowing the symptomsand helping these patients.
However, even with thisgenetically defined population,
it was hard to know when the diseasewas becoming active,
(16:40):
when these patientswere a good candidate for treatment.
So they ran a retrospective study.
This is before Olink hadan NGS readout, so it only measured
like 1100 proteins and they discoveredneurofilament light [NFL],
which is a very ubiquitousbiomarker for neuronal damage.
(17:01):
But they found thatthis biomarker, this neurofilament light, was
A) indicative of disease progression,was also a biomarker of efficacy.
so after patients were treatedwith patisiran, it dropped significantly
and it was a diseasebiomarker, it was 4-fold elevated
in the patients versus healthy controlsthat they measured in the study.
(17:23):
And so now what's really interestingis there's a protein-based assay
that could give treatment decisioninformation, right?
So it's being validated
and it's only a single biomarkerand it's a ubiquitous biomarker.
But in this subset, you know, proteomicsis giving you some actionable insights
in a genetically defined populationwhere they're now
(17:45):
developing cutoffs to try to see, hey,if you come to your clinician
and NFL is measured,and once you hit a certain cutoff,
that might actually indicate,even though you don't have symptoms,
the disease process has startedand you are a candidate for treatment.
So it's great for the patient.
It's obviously great for Alnylam.
So they can, you know, justifypatients getting on their therapy.
(18:08):
And it was where a proteomic screen, right?
they didn't know what to look for.
They didn't have this hypothesis.
They just wanted to see whatwhat's changing
in these patients after treatment,what's changing over time.
And I think that's a powerful waythat unbiased proteomics
can point us in the directionof actionable
(18:30):
biomarkersto help patients and clinical development.
So yeah, that wasthat was a really interesting story.
Great.
Actually, I would like to go way back
because you mentionedabout pharma and academia and then we know
at the beginning it was really difficultto communicate, right?
The two little worlds,they were like separated:
academic research versus pharma research.Do you see this changing?
(18:53):
And do you see a benefit of this change?
Yeah, absolutely.
So we just got off the phone.
Cindy and I were just on a callwith a really, really impressive
academic researcherwho mentioned that she's on the board
for two very large important studiesthat are being run by pharma companies.
(19:13):
Right?
She's an expert in her fieldand she's advising on how they should
spend, you know, their research dollarsto best move,
you know, very important therapiesthrough the clinic.
I see it happening all the time.
I mean, so,
our team focuses on primarilypharma and large population cohorts.
Right. And there's
(19:35):
incredible connections between the two.
Right?
Because if you think about it,if I'm a pharma company
and I'm interested in atopic dermatitis,for example,
it would behoove me to really profilewith all these new omics technologies
as many patients from the best cohortsin the world that have atopic dermatitis.
(19:56):
You could do thatthrough a population cohort
and you knowthere's going to be some subset.
What's probably more efficient is to workwith, you know, KOLs in the field.
Yeah.And then they've collected the samples.
Yeah. You provide the resources
and then with thatright from the protein side
you could discover, yeah, are theredisease progression biomarkers, are there
(20:19):
endo types,
are there sub phenotypes where there'sslightly different molecular drivers
that we could then approachwith different molecular entities
that we either haveor that we could develop
to have a higher rate of successin the clinic. Precision medicine.
So absolutely.
No. Yeah, Yeah.
And that's been a term that's been reallykind of reserved for oncology.
(20:39):
Right? Primarily.
And, you know, I think that
that's because the tools have existed
at the genetic level and obviously canceris a very genetically driven disease.
But if you look at, you know,some of the more
multi-system diseases that,you know, in the cardiometabolic space,
(21:00):
in the autoimmune space,you know, proteins
I think will be the next big thingin terms of finding
signals that can differentiatesubtypes of patients
and then give them better, bettertreatment options in the future.
You talk about cardiometabolic.Would you consider
(21:20):
like a blockbuster kind of disease,do you see
pharma investing more on theseor expanding on this research?
Because for me, seeing pharma,they are moving far away
without of course leaving behind thetraditional type, if we can say
a disease like cancer, I see that nowpharma is going to rare disease,
they're going to cardiometabolic disease,they're going to PCT disease.
(21:44):
What is your feeling? What
do you see in the upcomingyears with pharma?
I mean,
without getting too philosophicalabout why,
you know, the GLP1, GIP1 that
you know, the otherLilly and Novo competition and others,
you know, there's Pfizer and a lot ofother companies are getting involved.
(22:07):
Right.
There'sjust a huge societal issue with obesity
and there's enormous amounts of investmenthappening in that field.
I do think that there's a bit
of a gold rush right now,but scientifically,
what's really interesting is,you know, it's not just about obesity.
I've been fortunate enough to talk tosome of the leadership at these companies
(22:30):
who are really trying to develop the next,
you know, Mounjaro, the next Semaglutide,
and what they're noticingis there's so many knock on benefits
and there's so many benefitsto multimorbidity
that they want to both understandat the molecular level
what's driving that, but also understand,you know, are there patients
(22:52):
who have a more aggressive form of obesityfor lack of a better term?
Right.
Is there a subtype of patientsthat really need
30% weight loss or 40% weight loss?
So it's a fascinating effortand I mean, given
the reality that it'sa very environmentally driven condition,
(23:12):
proteomics, I think, willwill be an indispensable tool.
I mean, again, the other day, talkingto, you know, a KOL in this space saying
these companies and
society generally says, well,let's do genomics first, right?
Like, we have all these samples.
We're goingto just do whole genome sequencing
and see if there's some sort of signalin the genetics
that's going to help usanswer these questions.
(23:35):
And they're starting to say,hey, wait a second, there's these
proteomic tools now that don'tyou think it makes more sense in obesity
to look at the proteinsand they're dynamic and you can look
at multiple timepoints and see what'schanging post-treatment, etc., etc.
So it's just an interesting side note that
in this field I think proteomics is goingto be particularly valid.
(23:57):
And I just want to definea couple of terms. Okay.
KOL as key opinion leader.
We use that a lot at Olink around here.
People that aredriving and influencing
decisions that are happeningout in the field particularly or,
you know, we are thinkingin terms of genetics and proteomics
and then the the semaglutideand these GLP1 agonists
(24:18):
that Evan mentionedare not only relevant in obesity,
but they're actually beingalmost prescribed
where people pay out of pocketin some offerings.
So I've met people that arereally keen to be on them
or are on them and who have hada lot of success in reducing their,
(24:41):
maybe not in the,you know, obese category,
but an overweight categorywhere again, you can expect
based on what we've seen, health benefitsthere as well.
So I just wanted to throw that in. Really interesting space, right?
And yeah, maybe also on that mode,a lot of these drugs
(25:02):
and lot of these inhibitors,as you mentioned, Evan, there are influencing
more than one disease, right?There are targeting
more than one.
And I think that's sortof where some of them left off.
You have a drug for morethan one disease
is like my feeling or have you seen thishappening from your perspective?
Yeah, for sure.
I mean, it's
then that's where the deeper understandingof the mechanism of these drugs.
(25:25):
Right. Which, you know, Yes.
There are great model systemsthat if you using a cyno [cynomolgus macaques]
model, monkey model,you know, eventually mouse models.
Rat models.There's all kinds of models.
And you can get a good senseof how your drug's behaving.
But you know, often with these phaseone or phase two studies,
(25:46):
the amount of patients is fairly small.
You can get an idea,but that's why I do,
I believe that, you know, companiesare investing significant
resourcesto look at the bigger studies.
Right.
Because you can just see,you just get more statistical power.
You have a better chanceof really understanding how
my drug's impactingmultiple pathways, multiple organ systems.
(26:10):
And then once you have that knowledge,it just makes you so much better
informed for new therapeutic ideas,
even just repurposingthe existing therapeutics.
So, yes, Sarantis,
the more indications, the better,
I mean, just from a simple pragmaticbusiness perspective.
But having the molecular,you know, justification
(26:33):
I think is what, as a society,we should all ask for.
I mean, just seems to be what they -
Obviously when we're consenting
and all of us where you knowparticipants in these sorts of trials.
I think another promise hereand we're going into ASHG soon [ASHG = American Society of Human Genetics]
and we've got 25 different posters of folksleveraging it, leveraging
(26:55):
some proteomics from Olink,which is really exciting to see.
This is a genetics conference and clearlythere's this value of layering
the genetics onto -or, the proteomics onto the genetics.
There's also seven talks
that doesn't include the talksthat we're sponsoring.
So I think
(27:16):
in this in this environment,
I guess I'm wondering:
what are you mostexcited about, Evan?
Sorry.
No, that's fine.
I mean, it's a hard question.
Let's let's just talkabout this environment being
(27:38):
the fact that there are three publicationsin Nature, right?
Three publications that just dropped
about the promise of population proteomics.
Right.
So, I mean,
I just think it's the beginning, right?
So 50,000 samples from a largely northernEuropean cohort has led to a treasure
(27:59):
trove of insights, 14,000 associations,80%-plus of which were novel.
People can dig into that for a
long time. And referenceback to it with their own studies to
- Yeah, that's the way forward -
corroboratethe signals that they're seeing.
I think we've talked about thatbefore.
Yeah. Go ahead, Evan.
And that's super exciting right,because that will provide a bit
(28:21):
of a backbone to understand causality
and give us insightsinto drug targets and biomarkers.
That's great.
You know, but it's just a small
subset of the world's available resourcesfrom a cohort perspective.
So there's enormous benefitto going bigger
as the AstraZeneca rarevariant paper shows.
(28:43):
Right.
To capture these rare variants.
And this is what the RegeneronGenetics Center has done for years, right?
They're doing genomics on all of thesevery large populations, these founder
populations, to find these signalsthat really come out when you go big.
That will happenat the protein level as well.
I think going to different partsof the world,
(29:04):
there's just going to be enormous richnessas we go from that. Diversity
Without question,everyone wants to do that.
But if you say the thingI'm most excited about, to be honest,
is proteomic risk scores and the potential
for a whole suite of tools to help
(29:26):
perhaps, you know, consumersone day, certainly drug developers,
perhaps health insurance companies,who knows where this all goes.
But, you know, speaking to Ben Sunand some of the head analysts from the UK
Biobank project, they, with just 50,000samples and machine learning,
and I'd say algorithms, are ableto pick up on these patterns
(29:48):
right out of sometimesa small number of proteins.
I believe Claudia Langenberg and RobertScott had a paper where
it was between like five and 20 proteins
could distinguish your riskof a large number of common diseases.
I think
once those are validatedand those are refined,
that is a game changerbecause then I'm a drug developer,
(30:11):
I can apply these algorithmsto all my clinical trials
and better understand,"Hey, are we on the right track
and what other impactsare we having on a wide range of diseases?"
I mean, to me that's incredibly exciting.
And it's not without its challenges,right?
I mean, you have to validate these thingsand sufficiently
(30:33):
havestatistically powered studies,
but one could imagine that there could be
a suite of tools in the future basedon, you know,
a manageable number of measurementsthat could be used clinically.
And that's where I thinkthe next big evolution will be
is taking this datathat's been generated by either
(30:57):
academic funding, pharmafunding, government funding to really
look at a lot of diseasesat the protein level, at scale,
using these new proteomic technologies
and then whittling it down to thingsthat are clinically actionable
that you would have never foundif you didn't take a broader view.
Right. I think that's the difference.
And just to double click on those authors,so there's Ryan Dhindsa
(31:20):
on this rare variant paper.
He's the first author.
He's at Baylor workingalso with AstraZeneca, where Slavé
Petrovski is the the PI on that paper.
There's Ben Sun who you mentionedand Chris Whelan paper.
That's our [UK Biobank] flagship paper.
We consider it sort of
the broadest group from the UKBiobank Pharma Proteomics Project
(31:44):
and then, of course, theI want to also just touch on Grimur
Eldjarn and Kari Stefansson's
paper.
Well if only to to highlight
something Kari saidabout proteomics in general
(32:05):
and that was along the linesof what you're describing, that proteomics,
that an algorithm they've been ableto develop with proteomics, can predict
all-cause mortality in any individual.
So how many years doesone have left to live? Right.
So if I go into a clinical trial
and I've got a prediction of 30 yearsleft to live,
(32:25):
and then I go onto this drug and part waythrough that trial, or maybe three
quarters of the way through that trial,you look at my proteomics score
on my prediction, onhow long do I have to live.
This is a way to have very shortclinical trials
that actually are representativeof a longer period.
(32:48):
I mean, imagine a depression trial.
I remember there wasthere was one trial on depression.
It was something like six weeks. Right.
If you're talking about major depressivedisorder, a six-week
window is a hardone to draw conclusions from.
And we do the best we can.
But having something like thisthat is a reflection in the future
of what this is doing to your proteins,I think is very exciting.
(33:16):
I mean,
yeah, I'm thinking about data that's new.
Let's say there's an era of proteins,versus big data generation
for biomarker discovery,then what is coming next?
The in vitro diagnostics era is booming,for example? Then some of these biomarkers
be like customized andused for clinical diagnosis.
So how do you see this road map?I know this is difficult to predict, but
(33:39):
what do you see this comingactually from your perspective?
I think so. I think so.
And Cindy's point, I think is really- so to sort of touch on that real quick
and then and I'll touch on that, Sarantis,because they're certainly connected.
But they're slightly different in my view.
So this idea of having a risk scoreto help, you know, shorten a trial, right.
(33:59):
Give you some sort of a surrogateend point or some sort of early read.
I mean, I remember,you know, Kari [Stefansson] in a
presentationhe gave mentioning that, you know,
if you could apply this,you know, risk score,
you could
cut the time of the cardiovascularoutcomes trial, you know, significantly,
I think by more than halfand save hundreds of millions of dollars.
(34:23):
Right.
And I think broadly thatwould help everybody
because the
the companies developing therapeuticswould not have to spend so much money,
it would be less expensiveand the right patients would get,
you know, the right drug
because they're at higher riskif you use an enrichment strategy.
So I think that's absolutely coming.
There's no doubt about it. But then,
(34:43):
you know the
real end game, I think, Sarantis, was
what you've referred to in terms of thisin vitro diagnostics piece.
You know, so I was recently visiting RocheDiagnostics, you know, in Basel.
And they're, you know, worldleaders in diagnostic tests
and by and large, today it's a single-plexassay.
(35:04):
Yeah.
You measure one thing.and there's a lot of reasons for that
you know it's challenging to havemultiplexed assays validated to the level
that today we're used tobeing required from the FDA and others.
But biologically and just
you know, if you just think about
(35:25):
the complexity of disease, single marketis probably not the best thing to do.
So I do think that's comingand I hope in the rest of my career
I have a role to play in that
because if we can have very predictive
multi marker tests to be usedin the diagnostics space,
(35:46):
that to mewill be the biggest societal benefit
that can come from all of the amazing workthat's happening right now.
I think that that's where this all goes.
And you can just imagine a futurewhere there's much more resolution
to your personal risk for disease,your personal response
to therapies that we just don't see today.
(36:07):
So yeah,I think that's where it goes.
It's a hard road.
Well, but we're already seeing multi-gene testing in cancer
and stratifying and diagnosingto help better serve cancer patients.
So I think there's still a lotto be done there and I think you know that
pan-cancer study that
came out of Mathias Uhlén's team,which we've talked about on the podcast
(36:32):
before, is a great placewhere proteomics is making inroads.
So, yeah, fantastic.
Also,
to add onto that, as we said,one biomarker is not enough.
We have a lot of examples in papers
where you see the
additive value of having morethan one biomarker
that are really great.Erik Michaëlsson, Mathias Uhlén,
(36:54):
and you know, there are plenty of papersand so there is a divide.
And I think that if onlythe community start realizing that
having more than one biomarker willincrease the value of their work
Yeah yeah.
And it really just depends on
who you're talking to in termsof what they think the next big thing is.
Right? You asked for my opinion,I gave you my opinion.
You know, someone else could say, "Hey,I just want to measure
(37:18):
ten, you know, million samplesand then we're going to get much
richer insightsinto the next best drug targets.
And then that's going to createmore efficient pipelines and a better,
you know, drug development universein the next 50 years."
And yes, I think that'll happen, too.
(37:38):
But but yeah, there's just on allends of the drug development spectrum,
these innovations,you know, that Olink and others have made
I think are really,really going to be transformative.
And they already are.
They already are. But it's so early.
I'm sorry, not to go on a tangent,but it really is early.
Yeah.
You know, and part of withwhole genome sequencing,
(38:02):
the cost dropping has really enabled things.
And that's an important point.To be honest,
it really is.
You know, if we're just going to befrank and honest about,
you know, the opportunitiesto help as many people as possible,
if a tool is
prohibitively expensive,it's never going to have broad adoption.
(38:22):
And when I joined Olink
things cost a certain amount of moneyand now things cost less.
There's certainly -
We get more from it. Yeah, yeah, yeah.
Yeah, exactly. Yeah.
There's more datacoming out for a lower all over cost.
Right.
And that's what themarket has expected.
That's what peopleare demanding and again
(38:47):
that's very hard. Innovation, ittakes a lot of innovation.
But, you know, that'sI believe I'm excited to be here
because I know that the missionis the democratization of proteomics
to just get it out there,get it in the hands of the best and
brightest analysts out there. Right.
All of the great big data
folks who have developed such great toolsin that genetic space.
(39:09):
And I'll also say, you know,when you were talking to Chris Whelan
well before this whole UK-PPP project
came to fruition, there was
no guarantee that Olink was going to bethe chosen technology.
It's such an honor that the tools
and the priorities that we
thought were important - specificity,all that -
(39:32):
that those were also importantand continue
to be very important to pharma andand then I'm going to also
just point out that we're now at just,as of this year, at about 5400 proteins
and a really increased
streamlined workflow,increased throughput capability,
which is very exciting to see, too. Any
(39:56):
last comments? Yeah.
Please go ahead, Evan.
No, no, I will.
And I hope this I hope I can say thisbecause, you know, I'm an Olink employee
and this is a Proteomicsin Proximity podcast, right?
So I do think that there's going to be
multiple tools eventually that are goingto answer these questions, right?
(40:17):
I mean, I'm not so myopic as to thinkthat Olink is the only tool out there,
I think we
have some really compelling attributes
for the large scale projectsand for these large clinical analyzes.
But I get excited about continuedinnovation across,
you know, the earlier side of the researchspectrum where there could be tools
(40:39):
that can rapidly tell youabout all these different proteoforms
and phosphorylation states.
And yeah, it's a community, right,that's coming together.
And I think that,
there's just somuch has happened in the last
decadethat I've really been focused in the space
and it's goingto continue to evolve. And
(41:02):
I'm grateful thatwe've gotten
13 companies togetherto do something really big.
We continue to be integrally
involved in the strategy ofdrug development
from a large number of the world'sbest companies.
And I just think that it's all leadingto a more efficient process.
(41:25):
I mean,I have X number of years on this planet.
I want my time to be spentmaking a difference
for my kids and their kids.
And I truly believe that this kind of workis going to enable that.
So thank you for having me on.
Yeah, it was great. Sarantis,
any last words from you?
I mean, it was great.It was great to hear
(41:47):
your perspective and I agree with you.
I think that proteomicsis the major research
from now on. And you're going to see
a lot of papers.And it's only the beginning.
And we're looking forward to the upcomingprojects. Fantastic.
Well that's it for us today.
Again, thank you, Evan, for joining us.
(42:08):
Thank you very much. Thank you.
I think there are a couple of authorsthat we may not have said clearly,
and that was Faiez Zannadwho was integral in this.
And Milton Packer,I don't think we mentioned
Milton, who both were integralin really understanding and repurposing,
identifying, repurposing opportunitiesand their
(42:33):
empagliflozin [Jardiance].
And I think there's an "a" in there.
Empagliflozin.
Yeah, so I just wanted to click on thisand we'll put those into the show notes
as well.
Thanks as always to my co-host, Sarantis.
Thank you, of course.If you enjoyed listening
to Proteomics in Proximity, pleaseshare it with a friend or a colleague
(42:56):
who you think might also enjoy it, maybewe'll get more than 11 listeners, we'll see.
And remember, you can reach out to usat Proteomics in Proximity
at PIP@olink.comand you know, anything,
any feedback, positive, negative,
who should we interview?
We would be grateful for the suggestionsand the feedback.
(43:17):
And with that, we'll close.
Thanks, everyone. Thank you.
Thank you for listening to the Proteomicsin Proximity podcast
brought to you by OlinkProteomics. To contact the hosts
or for further information,simply email info@olink.com.
Welcometo the Proteomics and Proximity Podcast.
Where your co-hosts, Cindy Lawleyand Sarantis Chlamydas from Olink Proteomics,
talk about the intersection of proteomicswith genomics for drug target
discovery, the application of proteomicsto reveal disease biomarkers,
and current trends in using proteomicsto unlock biological mechanisms.
(00:26):
Here we have your hosts, Cindy and Sarantis.
Hey, everyone.
Hello and welcome backto Proteomics in Proximity.
Thanks to our 11 listeners at Sam Ray,Carolina, others.
We are grateful for your attention
and your feedback, and
(00:47):
our listeners have given ussome valuable feedback over time
and they've found us through differentsocial media avenues.
But to make that easier,we're announcing that
we've got actually an email address now,so we'll put this into the show notes.
But it's just P-I-P for Proteomicsin Proximity at Olink.com. [pip@olink.com]
And and we'd be happy to hear from youaround suggestions
(01:11):
you have or any interview recommendationsyou might have.
And with that, today,we are talking to Evan Mills.
Evan, I'll let him introduce himself,but he is
an illustrious pharma executive, here
actually at Olink,and we're excited to talk to him
(01:31):
about how pharma are finding proteomics
super relevant on many different levels.
So with that,let's get on with it.
Hey, Sarantis, how are you?
Hello. I'm fine.Thank you, Cindy. Welcome, Evan.
I'm looking forward to
hear from you all the great news.
Likewise. Good afternoon, Sarantis.
Good early morning to you,Cindy, on the West coast.
(01:53):
It's a little dark over here.A little dark
No, that's all right.I'm really honored to be here.
And I've been wanting to talk abouthow proteomics
and the pharmaceutical industryare aligning for really exciting things.
So very happy to be here.
Can you give us just a little backgroundon your history in this area?
You've been in this for a while.
(02:16):
I have.
I have.
So I was a bench scientist really,you know, passionate about oncology
and neuroscience research.
I did some work at Yale Universityfor awhile
and then I got into the salescommercial side of this world,
started actually in the pharmaceuticalsales industry, which
was exciting because of the opportunityto help patients,
right?
(02:36):
But my real passion was in the scienceand about a decade ago,
there was a very innovativeproteomics company
that caught my attentionand that's where I started this journey,
where I've now been at Olink for overfive years.
And yeah, supporting the most innovative,
ambitious researchers in this multi-omicsspace has just been a phenomenal journey.
So my background is (03:00):
I love science, I want to help people,
I want to have some sort of translationalimpact with the work I do. And
right now at this moment, there's neverbeen more momentum in that direction.
It's really, really exciting.Yes, very exciting.
We've just had here at Olink
(03:21):
three pretty exciting nature papers
come out in
I think it's the online [version on] October 4th.
But the print journal [on] October 11with a beautiful frog on the cover.
It's an exciting timewith those three papers.
So those include a lot of applications
(03:43):
aroundwhy pharma would invest in proteomics.
So I'd love to get your thoughts on why.
Why did 13 pharma cometogether, invest in proteomics?
What's the outcome?
What's the resultthat they see out of that?
Yeah, I mean that's been a real career
(04:05):
highlight, is being able to be involvedin that project from its inception.
And you know, Cindy,with your background in genetics,
there was a previously formedconsortium around whole exome sequencing
in the UK Biobank and then eventuallythe whole genome sequencing.
But there was this idea,and I was having lunch
(04:25):
outside of the Harvard symposium with Dr.
Chris Whelan, very smart geneticistwho was at Biogen at the time.
He's now at Janssen.
And he just asked the question.
He's like, "Hey, we're thinking aboutwhat makes sense to do next.
We have all this richnessin the genomic data,
but we want to do somethingcloser to phenotype.
(04:46):
Would it even be conceivablefor Olink to run 50,000 samples?"
And this is before,
you know, some of the innovationthat would have made that possible.
And we said, "Yeah,I think we can do that.
I think we can get there,I think we can do that."
So it was just born out of curiosity
and the desireto get closer to phenotype.
(05:06):
So the
goal really of this ambitious project was:
can we both better understand
drug targetsthat have causal links to disease
and can we simultaneously find biomarkersto help the drug development process?
Because obviously with proteomicsyou can do both, right?
It can act as a bit of a filterto tell you which of these
(05:30):
genomic disease associationshave a plausible biological story
and which ones should be pursued,and which ones should perhaps be killed,
but simultaneouslyyou can develop a suite of tools
to determine riskbased on proteomics, to determine disease
progression based on proteomics,and to discover biomarkers,
which are obviously always desired
(05:51):
to aid clinical development. So,
I mean, we're just starting to seeall the publications.
We're starting to understandall the utility that's going
to come from this data set.
And it was just such an ambitious,
smart idea by Chris and then eventuallyMelissa and Linden and Brad,
that countless others who contributedto the project.
(06:14):
Evan, going back to this journey,this amazing journey,
how easy or difficultwas it to convince the genomics community
because you mentioned it was like thisheavy genomics community, right?
Change their mindset in a wayto measure proteins,
how easy or difficult was this process?
It's really hard.
I think it is.
I think it was really hardbecause if you just think about the tools
(06:38):
one would need to developto measure, right?
Our DNA is very nicely organizedinto a helical structure.
There's four bases to measure, and Illuminaand others
now have developed amazing toolsthat can measure that at scale.
Think about the proteome, right?
There's 20 amino acids.
They combinein a myriad of different ways.
(06:58):
I mean, it's just such a formidablechallenge that geneticists would say,
"I'm not so sure.
I'm not so sure the tools exist.
And oh, by the way, yeah, we can measureeverything with genomics.
We can measure everything.
And you're approaching us with somethingthat measures 1500 at the time?"
Right.
And then 3000 of what people assumewould be
(07:21):
maybe 20,000 proteinsthat you could try to capture
in plasma / serum,there's a big debate about that.
So it is challenging,
BUT the obvious central dogmaof being closer to disease
and things
that are reflective of real-time biologyversus your blueprint for your biology
(07:42):
was compelling enough for themto give it a shot, but it was not easy.
So you must have focused onwhat is the near-term
return on investment for pharma,for running a proteomics project.
And I would consider this UK Biobank sort of pQTL developing therapeutic targets.
All of the -
all of those things you've alreadymentioned is more mid- or long-term goals.
(08:05):
How did you - I think it's a great questionSarantis asked -
how do you talk to themabout what you believe is the value?
And I will
also say we - Gary and I - looked ...Gary is our illustrious person
who manages our databaseof over 1400 peer-reviewed publications
and he has seen over 84 ofthose are pharma relevant publication.
(08:28):
So there's a significant numberof publications that have been
that have been put out therethat document
some value to pharma,but that's pretty recent.
How did you approach themwhen you first came to Olink?
Yeah, no, it's a good question.
So we can take a bit of a sidebarfrom the UK Biobank discussion because
(08:51):
really, fundamentally, drug developers
are trying to bring effectivetherapeutics to market faster
and they also invest enormous resourcesinto each program.
And it takes what, 10 to 15 yearson average, you know, to get something approved.
And how many millions of dollars, right?
(09:11):
And patience.
And then what,
90% of clinical trials failI think, or somewhere around there.
Yeah.
I recently had a discussionwith an executive vice
president of research at a major companywho said he would be the world's
best drug developerif he failed 80% of the time.
Isn't that wild?If he could go 80% after failing 90%, he would be the best.
(09:36):
And it's - right?
It's just such a high attrition game.
But those are, Cindy,the way that
a lot of people in the industryare starting to look at
this is (09:46):
with population-scale proteomics or high-throughput proteomics,
you can learn a lot about thingsyou've already invested in.
So let's say that you have a drugthat's approved such as -
I can never say that correctly.
Jardiance, let's go with Jardiance [empaglioflozin].
Yeah.
You know, a very, very effective SLT2
(10:09):
inhibitor used for the treatment of
diabetic control.
They've also noticed, after having itin enough humans in the wild,
that there's significant benefitsto heart failure.
So if you can access - and this is one ofthe publications that you referenced -
if you can access samplesfrom completed clinical trials
(10:32):
and most companies are sitting on these,they're just in their freezers
waiting to be analyzedif they have the exploratory consents.
If you take a
look at proteomics at scale from
lots of humans treatedin the clinical setting, you can learn
a tremendous amount about why certainpeople respond and certain people don't.
(10:53):
Right.
That's the Holy Grail, essentially is: can you proactively know (10:53):
undefined
which patients could go, should go,which therapy.
We often call that "stratifying
patients," just to use the termthat we've used before.
Yep. Yes, absolutely.
And you know, understanding the mechanismof these drugs, right?
Because, you know, you have a target,you have a hypothesis, you tested it in cell-
(11:15):
based models, animal-based models,but you don't really have a chance
to look at scale in a human populationto see how it impacts the human body.
So then with that data,you can A) better understand why there's
this benefit in an indication for whichthe drug was not initially approved.
You can understand - Excuse me,
what other pathways are being
(11:37):
impacted by your therapeutic.Are there repurposing opportunities?
Is there a way to very rapidly
take this thingyou've invested in, this asset,
and figure out that there's more placesthat you could help people,
there's more indications where this drugwould actually be a really good fit.
So that's a very shortwin, short-term win.
(11:57):
And we've noticed multiple clientsbuilding this as a strategy
to take Olink Proteomicsin this case to better
understand already approved drugs,which, in some ways, is counterintuitive.
Right?
I mean, ideallyyou think from the beginning
you would want to know everythingyou can about the drug,
but there's this reverse translation
(12:18):
movement that seems to be bearingquite a bit of fruit for the industry.
That was actually my next-
I'm sorry,
that's certainly my intriguing invite.
This is my next question, Evan, do you seenow this trend of a strategy
in the pharma
because you talk with
the executives, right? And you wouldknow the strategy and you discuss
about this. Do you see this coming?
Do you see that using large-scale proteomics,a big number of data to reposition
(12:42):
a drug, for example, to identifymechanisms of action even in the late stage?
How is your feeling?And why would they ever publish this?
Right.
We think of pharma as needingto hold these things tight.
So yeah, great questions, Sarantis.
That's a good question.
So the answer is yes, in pockets.
I think it's just becoming
(13:03):
a strategyfor the more innovative companies.
Right.
There's always some concern, right,for ongoing trials.
Do we really want to knowthat much at a phase three?
Right.
If we have a candidate compound,do we want to do exploratory research?
Maybe we find something we can't explain.
Maybe we find some safety signals.
(13:24):
So what I'm describing is drugs approved.
Let's extractas much value from that asset as we can.
And there's definitely companiesthat are taking that on as a strategy.
And to your point,I mean, having gotten to know
folks in pharmareally well for the last decade, I mean,
they're great scientists.I think there's this -
(13:45):
I think, not to insult
any of my academic colleagues or people
I've worked with or people that, you know,I've supported over the last 20 years.
I think there's incredibly talented
scientists that see the opportunity
to have a fast path to impact.
And they do want to share.They want to publish.
I mean, look at this consortium.
(14:06):
It was 13 companies that are competitorscoming together
I was complimentingone of the pharma researchers on a hire,
a new hire from academia.
And she was saying they came to mebecause they're a physician,
an M.D., Ph.D., and they said,
I can help one patientat a time in my practice.
(14:28):
But if I come here and do more
broad-based research,I can affect millions.
And I was like, Wow, that'sthat's an interesting perspective.
I like that. And it lines upwith what you're saying.
I think a great exampleof reverse translation that you've talked about.
I think one of the
examplesyou've talked about in the past of
(14:52):
of this, you know, taking samplesthat are sitting in the freezer
where a massive investment has beenmade is the one from Simina Ticau.
And
Paul Nioi. Paul is, of course,
also on the UK Biobank
flagship paperthat came out last week / this week,
(15:13):
whichever online footprintyou want to reference.
Can you tell us about that example?
Yeah,this is a really, really interesting story
and that this originated aboutfive years ago and was published in 2019.
So it's a bit dated, but I think the pointis incredibly powerful.
So, you know, hereditarytransthyretin-mediated amyloidosis is
(15:36):
a genetically defined disease
that really has -
And can I just say that you can pronounce that,
but empaglioflozin is pretty
darn easier to say than -
I'm sorry. It just seems funny.
[Empaglioflozin is] Jardiance, butanyway, you know, back to back to hATTR.
No,
you know,
I've probably told that storymore times than -
(15:58):
Yeah,
my gosh, it's hard.
So no, but, hATTR is a really,
really debilitating diseasewith a variable rate of onset.
So if it's the hereditary form,it runs in your family.
Right.
You can be screened to knowif you're carrier
and know if you're at riskfor developing the disease.
(16:20):
Alnylam developed a drug, patisiran,
that is an siRNA - excuse me,
RNAi-based therapeutic where they are
very effective at slowing the symptomsand helping these patients.
However, even with thisgenetically defined population,
it was hard to know when the diseasewas becoming active,
(16:40):
when these patientswere a good candidate for treatment.
So they ran a retrospective study.
This is before Olink hadan NGS readout, so it only measured
like 1100 proteins and they discoveredneurofilament light [NFL],
which is a very ubiquitousbiomarker for neuronal damage.
(17:01):
But they found thatthis biomarker, this neurofilament light, was
A) indicative of disease progression,was also a biomarker of efficacy.
so after patients were treatedwith patisiran, it dropped significantly
and it was a diseasebiomarker, it was 4-fold elevated
in the patients versus healthy controlsthat they measured in the study.
(17:23):
And so now what's really interestingis there's a protein-based assay
that could give treatment decisioninformation, right?
So it's being validated
and it's only a single biomarkerand it's a ubiquitous biomarker.
But in this subset, you know, proteomicsis giving you some actionable insights
in a genetically defined populationwhere they're now
(17:45):
developing cutoffs to try to see, hey,if you come to your clinician
and NFL is measured,and once you hit a certain cutoff,
that might actually indicate,even though you don't have symptoms,
the disease process has startedand you are a candidate for treatment.
So it's great for the patient.
It's obviously great for Alnylam.
So they can, you know, justifypatients getting on their therapy.
(18:08):
And it was where a proteomic screen, right?
they didn't know what to look for.
They didn't have this hypothesis.
They just wanted to see whatwhat's changing
in these patients after treatment,what's changing over time.
And I think that's a powerful waythat unbiased proteomics
can point us in the directionof actionable
(18:30):
biomarkersto help patients and clinical development.
So yeah, that wasthat was a really interesting story.
Great.
Actually, I would like to go way back
because you mentionedabout pharma and academia and then we know
at the beginning it was really difficultto communicate, right?
The two little worlds,they were like separated:
academic research versus pharma research.Do you see this changing?
(18:53):
And do you see a benefit of this change?
Yeah, absolutely.
So we just got off the phone.
Cindy and I were just on a callwith a really, really impressive
academic researcherwho mentioned that she's on the board
for two very large important studiesthat are being run by pharma companies.
(19:13):
Right?
She's an expert in her fieldand she's advising on how they should
spend, you know, their research dollarsto best move,
you know, very important therapiesthrough the clinic.
I see it happening all the time.
I mean, so,
our team focuses on primarilypharma and large population cohorts.
Right. And there's
(19:35):
incredible connections between the two.
Right?
Because if you think about it,if I'm a pharma company
and I'm interested in atopic dermatitis,for example,
it would behoove me to really profilewith all these new omics technologies
as many patients from the best cohortsin the world that have atopic dermatitis.
(19:56):
You could do thatthrough a population cohort
and you knowthere's going to be some subset.
What's probably more efficient is to workwith, you know, KOLs in the field.
Yeah.And then they've collected the samples.
Yeah. You provide the resources
and then with thatright from the protein side
you could discover, yeah, are theredisease progression biomarkers, are there
(20:19):
endo types,
are there sub phenotypes where there'sslightly different molecular drivers
that we could then approachwith different molecular entities
that we either haveor that we could develop
to have a higher rate of successin the clinic. Precision medicine.
So absolutely.
No. Yeah, Yeah.
And that's been a term that's been reallykind of reserved for oncology.
(20:39):
Right? Primarily.
And, you know, I think that
that's because the tools have existed
at the genetic level and obviously canceris a very genetically driven disease.
But if you look at, you know,some of the more
multi-system diseases that,you know, in the cardiometabolic space,
(21:00):
in the autoimmune space,you know, proteins
I think will be the next big thingin terms of finding
signals that can differentiatesubtypes of patients
and then give them better, bettertreatment options in the future.
You talk about cardiometabolic.Would you consider
(21:20):
like a blockbuster kind of disease,do you see
pharma investing more on theseor expanding on this research?
Because for me, seeing pharma,they are moving far away
without of course leaving behind thetraditional type, if we can say
a disease like cancer, I see that nowpharma is going to rare disease,
they're going to cardiometabolic disease,they're going to PCT disease.
(21:44):
What is your feeling? What
do you see in the upcomingyears with pharma?
I mean,
without getting too philosophicalabout why,
you know, the GLP1, GIP1 that
you know, the otherLilly and Novo competition and others,
you know, there's Pfizer and a lot ofother companies are getting involved.
(22:07):
Right.
There'sjust a huge societal issue with obesity
and there's enormous amounts of investmenthappening in that field.
I do think that there's a bit
of a gold rush right now,but scientifically,
what's really interesting is,you know, it's not just about obesity.
I've been fortunate enough to talk tosome of the leadership at these companies
(22:30):
who are really trying to develop the next,
you know, Mounjaro, the next Semaglutide,
and what they're noticingis there's so many knock on benefits
and there's so many benefitsto multimorbidity
that they want to both understandat the molecular level
what's driving that, but also understand,you know, are there patients
(22:52):
who have a more aggressive form of obesityfor lack of a better term?
Right.
Is there a subtype of patientsthat really need
30% weight loss or 40% weight loss?
So it's a fascinating effortand I mean, given
the reality that it'sa very environmentally driven condition,
(23:12):
proteomics, I think, willwill be an indispensable tool.
I mean, again, the other day, talkingto, you know, a KOL in this space saying
these companies and
society generally says, well,let's do genomics first, right?
Like, we have all these samples.
We're goingto just do whole genome sequencing
and see if there's some sort of signalin the genetics
that's going to help usanswer these questions.
(23:35):
And they're starting to say,hey, wait a second, there's these
proteomic tools now that don'tyou think it makes more sense in obesity
to look at the proteinsand they're dynamic and you can look
at multiple timepoints and see what'schanging post-treatment, etc., etc.
So it's just an interesting side note that
in this field I think proteomics is goingto be particularly valid.
(23:57):
And I just want to definea couple of terms. Okay.
KOL as key opinion leader.
We use that a lot at Olink around here.
People that aredriving and influencing
decisions that are happeningout in the field particularly or,
you know, we are thinkingin terms of genetics and proteomics
and then the the semaglutideand these GLP1 agonists
(24:18):
that Evan mentionedare not only relevant in obesity,
but they're actually beingalmost prescribed
where people pay out of pocketin some offerings.
So I've met people that arereally keen to be on them
or are on them and who have hada lot of success in reducing their,
(24:41):
maybe not in the,you know, obese category,
but an overweight categorywhere again, you can expect
based on what we've seen, health benefitsthere as well.
So I just wanted to throw that in. Really interesting space, right?
And yeah, maybe also on that mode,a lot of these drugs
(25:02):
and lot of these inhibitors,as you mentioned, Evan, there are influencing
more than one disease, right?There are targeting
more than one.
And I think that's sortof where some of them left off.
You have a drug for morethan one disease
is like my feeling or have you seen thishappening from your perspective?
Yeah, for sure.
I mean, it's
then that's where the deeper understandingof the mechanism of these drugs.
(25:25):
Right. Which, you know, Yes.
There are great model systemsthat if you using a cyno [cynomolgus macaques]
model, monkey model,you know, eventually mouse models.
Rat models.There's all kinds of models.
And you can get a good senseof how your drug's behaving.
But you know, often with these phaseone or phase two studies,
(25:46):
the amount of patients is fairly small.
You can get an idea,but that's why I do,
I believe that, you know, companiesare investing significant
resourcesto look at the bigger studies.
Right.
Because you can just see,you just get more statistical power.
You have a better chanceof really understanding how
my drug's impactingmultiple pathways, multiple organ systems.
(26:10):
And then once you have that knowledge,it just makes you so much better
informed for new therapeutic ideas,
even just repurposingthe existing therapeutics.
So, yes, Sarantis,
the more indications, the better,
I mean, just from a simple pragmaticbusiness perspective.
But having the molecular,you know, justification
(26:33):
I think is what, as a society,we should all ask for.
I mean, just seems to be what they -
Obviously when we're consenting
and all of us where you knowparticipants in these sorts of trials.
I think another promise hereand we're going into ASHG soon [ASHG = American Society of Human Genetics]
and we've got 25 different posters of folksleveraging it, leveraging
(26:55):
some proteomics from Olink,which is really exciting to see.
This is a genetics conference and clearlythere's this value of layering
the genetics onto -or, the proteomics onto the genetics.
There's also seven talks
that doesn't include the talksthat we're sponsoring.
So I think
(27:16):
in this in this environment,
I guess I'm wondering:
what are you mostexcited about, Evan?
Sorry.
No, that's fine.
I mean, it's a hard question.
Let's let's just talkabout this environment being
(27:38):
the fact that there are three publicationsin Nature, right?
Three publications that just dropped
about the promise of population proteomics.
Right.
So, I mean,
I just think it's the beginning, right?
So 50,000 samples from a largely northernEuropean cohort has led to a treasure
(27:59):
trove of insights, 14,000 associations,80%-plus of which were novel.
People can dig into that for a
long time. And referenceback to it with their own studies to
- Yeah, that's the way forward -
corroboratethe signals that they're seeing.
I think we've talked about thatbefore.
Yeah. Go ahead, Evan.
And that's super exciting right,because that will provide a bit
(28:21):
of a backbone to understand causality
and give us insightsinto drug targets and biomarkers.
That's great.
You know, but it's just a small
subset of the world's available resourcesfrom a cohort perspective.
So there's enormous benefitto going bigger
as the AstraZeneca rarevariant paper shows.
(28:43):
Right.
To capture these rare variants.
And this is what the RegeneronGenetics Center has done for years, right?
They're doing genomics on all of thesevery large populations, these founder
populations, to find these signalsthat really come out when you go big.
That will happenat the protein level as well.
I think going to different partsof the world,
(29:04):
there's just going to be enormous richnessas we go from that. Diversity
Without question,everyone wants to do that.
But if you say the thingI'm most excited about, to be honest,
is proteomic risk scores and the potential
for a whole suite of tools to help
(29:26):
perhaps, you know, consumersone day, certainly drug developers,
perhaps health insurance companies,who knows where this all goes.
But, you know, speaking to Ben Sunand some of the head analysts from the UK
Biobank project, they, with just 50,000samples and machine learning,
and I'd say algorithms, are ableto pick up on these patterns
(29:48):
right out of sometimesa small number of proteins.
I believe Claudia Langenberg and RobertScott had a paper where
it was between like five and 20 proteins
could distinguish your riskof a large number of common diseases.
I think
once those are validatedand those are refined,
that is a game changerbecause then I'm a drug developer,
(30:11):
I can apply these algorithmsto all my clinical trials
and better understand,"Hey, are we on the right track
and what other impactsare we having on a wide range of diseases?"
I mean, to me that's incredibly exciting.
And it's not without its challenges,right?
I mean, you have to validate these thingsand sufficiently
(30:33):
havestatistically powered studies,
but one could imagine that there could be
a suite of tools in the future basedon, you know,
a manageable number of measurementsthat could be used clinically.
And that's where I thinkthe next big evolution will be
is taking this datathat's been generated by either
(30:57):
academic funding, pharmafunding, government funding to really
look at a lot of diseasesat the protein level, at scale,
using these new proteomic technologies
and then whittling it down to thingsthat are clinically actionable
that you would have never foundif you didn't take a broader view.
Right. I think that's the difference.
And just to double click on those authors,so there's Ryan Dhindsa
(31:20):
on this rare variant paper.
He's the first author.
He's at Baylor workingalso with AstraZeneca, where Slavé
Petrovski is the the PI on that paper.
There's Ben Sun who you mentionedand Chris Whelan paper.
That's our [UK Biobank] flagship paper.
We consider it sort of
the broadest group from the UKBiobank Pharma Proteomics Project
(31:44):
and then, of course, theI want to also just touch on Grimur
Eldjarn and Kari Stefansson's
paper.
Well if only to to highlight
something Kari saidabout proteomics in general
(32:05):
and that was along the linesof what you're describing, that proteomics,
that an algorithm they've been ableto develop with proteomics, can predict
all-cause mortality in any individual.
So how many years doesone have left to live? Right.
So if I go into a clinical trial
and I've got a prediction of 30 yearsleft to live,
(32:25):
and then I go onto this drug and part waythrough that trial, or maybe three
quarters of the way through that trial,you look at my proteomics score
on my prediction, onhow long do I have to live.
This is a way to have very shortclinical trials
that actually are representativeof a longer period.
(32:48):
I mean, imagine a depression trial.
I remember there wasthere was one trial on depression.
It was something like six weeks. Right.
If you're talking about major depressivedisorder, a six-week
window is a hardone to draw conclusions from.
And we do the best we can.
But having something like thisthat is a reflection in the future
of what this is doing to your proteins,I think is very exciting.
(33:16):
I mean,
yeah, I'm thinking about data that's new.
Let's say there's an era of proteins,versus big data generation
for biomarker discovery,then what is coming next?
The in vitro diagnostics era is booming,for example? Then some of these biomarkers
be like customized andused for clinical diagnosis.
So how do you see this road map?I know this is difficult to predict, but
(33:39):
what do you see this comingactually from your perspective?
I think so. I think so.
And Cindy's point, I think is really- so to sort of touch on that real quick
and then and I'll touch on that, Sarantis,because they're certainly connected.
But they're slightly different in my view.
So this idea of having a risk scoreto help, you know, shorten a trial, right.
(33:59):
Give you some sort of a surrogateend point or some sort of early read.
I mean, I remember,you know, Kari [Stefansson] in a
presentationhe gave mentioning that, you know,
if you could apply this,you know, risk score,
you could
cut the time of the cardiovascularoutcomes trial, you know, significantly,
I think by more than halfand save hundreds of millions of dollars.
(34:23):
Right.
And I think broadly thatwould help everybody
because the
the companies developing therapeuticswould not have to spend so much money,
it would be less expensiveand the right patients would get,
you know, the right drug
because they're at higher riskif you use an enrichment strategy.
So I think that's absolutely coming.
There's no doubt about it. But then,
(34:43):
you know the
real end game, I think, Sarantis, was
what you've referred to in terms of thisin vitro diagnostics piece.
You know, so I was recently visiting RocheDiagnostics, you know, in Basel.
And they're, you know, worldleaders in diagnostic tests
and by and large, today it's a single-plexassay.
(35:04):
Yeah.
You measure one thing.and there's a lot of reasons for that
you know it's challenging to havemultiplexed assays validated to the level
that today we're used tobeing required from the FDA and others.
But biologically and just
you know, if you just think about
(35:25):
the complexity of disease, single marketis probably not the best thing to do.
So I do think that's comingand I hope in the rest of my career
I have a role to play in that
because if we can have very predictive
multi marker tests to be usedin the diagnostics space,
(35:46):
that to mewill be the biggest societal benefit
that can come from all of the amazing workthat's happening right now.
I think that that's where this all goes.
And you can just imagine a futurewhere there's much more resolution
to your personal risk for disease,your personal response
to therapies that we just don't see today.
(36:07):
So yeah,I think that's where it goes.
It's a hard road.
Well, but we're already seeing multi-gene testing in cancer
and stratifying and diagnosingto help better serve cancer patients.
So I think there's still a lotto be done there and I think you know that
pan-cancer study that
came out of Mathias Uhlén's team,which we've talked about on the podcast
(36:32):
before, is a great placewhere proteomics is making inroads.
So, yeah, fantastic.
Also,
to add onto that, as we said,one biomarker is not enough.
We have a lot of examples in papers
where you see the
additive value of having morethan one biomarker
that are really great.Erik Michaëlsson, Mathias Uhlén,
(36:54):
and you know, there are plenty of papersand so there is a divide.
And I think that if onlythe community start realizing that
having more than one biomarker willincrease the value of their work
Yeah yeah.
And it really just depends on
who you're talking to in termsof what they think the next big thing is.
Right? You asked for my opinion,I gave you my opinion.
You know, someone else could say, "Hey,I just want to measure
(37:18):
ten, you know, million samplesand then we're going to get much
richer insightsinto the next best drug targets.
And then that's going to createmore efficient pipelines and a better,
you know, drug development universein the next 50 years."
And yes, I think that'll happen, too.
(37:38):
But but yeah, there's just on allends of the drug development spectrum,
these innovations,you know, that Olink and others have made
I think are really,really going to be transformative.
And they already are.
They already are. But it's so early.
I'm sorry, not to go on a tangent,but it really is early.
Yeah.
You know, and part of withwhole genome sequencing,
(38:02):
the cost dropping has really enabled things.
And that's an important point.To be honest,
it really is.
You know, if we're just going to befrank and honest about,
you know, the opportunitiesto help as many people as possible,
if a tool is
prohibitively expensive,it's never going to have broad adoption.
(38:22):
And when I joined Olink
things cost a certain amount of moneyand now things cost less.
There's certainly -
We get more from it. Yeah, yeah, yeah.
Yeah, exactly. Yeah.
There's more datacoming out for a lower all over cost.
Right.
And that's what themarket has expected.
That's what peopleare demanding and again
(38:47):
that's very hard. Innovation, ittakes a lot of innovation.
But, you know, that'sI believe I'm excited to be here
because I know that the missionis the democratization of proteomics
to just get it out there,get it in the hands of the best and
brightest analysts out there. Right.
All of the great big data
folks who have developed such great toolsin that genetic space.
(39:09):
And I'll also say, you know,when you were talking to Chris Whelan
well before this whole UK-PPP project
came to fruition, there was
no guarantee that Olink was going to bethe chosen technology.
It's such an honor that the tools
and the priorities that we
thought were important - specificity,all that -
(39:32):
that those were also importantand continue
to be very important to pharma andand then I'm going to also
just point out that we're now at just,as of this year, at about 5400 proteins
and a really increased
streamlined workflow,increased throughput capability,
which is very exciting to see, too. Any
(39:56):
last comments? Yeah.
Please go ahead, Evan.
No, no, I will.
And I hope this I hope I can say thisbecause, you know, I'm an Olink employee
and this is a Proteomicsin Proximity podcast, right?
So I do think that there's going to be
multiple tools eventually that are goingto answer these questions, right?
(40:17):
I mean, I'm not so myopic as to thinkthat Olink is the only tool out there,
I think we
have some really compelling attributes
for the large scale projectsand for these large clinical analyzes.
But I get excited about continuedinnovation across,
you know, the earlier side of the researchspectrum where there could be tools
(40:39):
that can rapidly tell youabout all these different proteoforms
and phosphorylation states.
And yeah, it's a community, right,that's coming together.
And I think that,
there's just somuch has happened in the last
decadethat I've really been focused in the space
and it's goingto continue to evolve. And
(41:02):
I'm grateful thatwe've gotten
13 companies togetherto do something really big.
We continue to be integrally
involved in the strategy ofdrug development
from a large number of the world'sbest companies.
And I just think that it's all leadingto a more efficient process.
(41:25):
I mean,I have X number of years on this planet.
I want my time to be spentmaking a difference
for my kids and their kids.
And I truly believe that this kind of workis going to enable that.
So thank you for having me on.
Yeah, it was great. Sarantis,
any last words from you?
I mean, it was great.It was great to hear
(41:47):
your perspective and I agree with you.
I think that proteomicsis the major research
from now on. And you're going to see
a lot of papers.And it's only the beginning.
And we're looking forward to the upcomingprojects. Fantastic.
Well that's it for us today.
Again, thank you, Evan, for joining us.
(42:08):
Thank you very much. Thank you.
I think there are a couple of authorsthat we may not have said clearly,
and that was Faiez Zannadwho was integral in this.
And Milton Packer,I don't think we mentioned
Milton, who both were integralin really understanding and repurposing,
identifying, repurposing opportunitiesand their
(42:33):
empagliflozin [Jardiance].
And I think there's an "a" in there.
Empagliflozin.
Yeah, so I just wanted to click on thisand we'll put those into the show notes
as well.
Thanks as always to my co-host, Sarantis.
Thank you, of course.If you enjoyed listening
to Proteomics in Proximity, pleaseshare it with a friend or a colleague
(42:56):
who you think might also enjoy it, maybewe'll get more than 11 listeners, we'll see.
And remember, you can reach out to usat Proteomics in Proximity
at PIP@olink.comand you know, anything,
any feedback, positive, negative,
who should we interview?
We would be grateful for the suggestionsand the feedback.
(43:17):
And with that, we'll close.
Thanks, everyone. Thank you.
Thank you for listening to the Proteomicsin Proximity podcast
brought to you by OlinkProteomics. To contact the hosts
or for further information,simply email info@olink.com.
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.