Choosing the proper cancer treatment can be a shot in the dark. That’s because the best treatment can depend on specific aspects of a tumor, which doctors can’t always identify. Anirudh Joshi (with cofounders Damir Vrabac, 28, and Viswesh Krishna, 23) started Palo Alto, California–based Valar Labs hoping to deliver that crucial information. Valar, which has raised $26 million in VC funding, feeds images of patients’ tumors (limited for now to bladder cancer) into its AI models to suggest the best treatment. “We weren’t physicians, we weren’t oncologists coming into the field,” says Joshi, an engineer by training. “I realized where medical devices could be transformative was on the software side.” A study of more than a thousand bladder cancer patients found that Valar’s first AItest, called Vesta, was able to accurately determine whether a typical treatment would be effective. Patients can now access the test with their physicians at 20 hospitals around the nation and get reimbursed by their insurance companies. Joshi says Valar could be “the next billion-dollar company in cancer,” and he’s working on developing similar tests for more cancer types, including pancreatic, ovarian and lymphoma. Because they are software, Vesta’s products do not currently require FDA oversight.
Subscribe to FORBES: https://www.youtube.com/user/Forbes?sub_confirmation=1
Fuel your success with Forbes. Gain unlimited access to premium journalism, including breaking news, groundbreaking in-depth reported stories, daily digests and more. Plus, members get a front-row seat at members-only events with leading thinkers and doers, access to premium video that can help you get ahead, an ad-light experience, early access to select products including NFT drops and more:
https://account.forbes.com/membership/?utm_source=youtube&utm_medium=display&utm_campaign=growth_non-sub_paid_subscribe_ytdescript
Stay Connected
Forbes newsletters: https://newsletters.editorial.forbes.com
Forbes on Facebook: http://fb.com/forbes
Forbes Video on Twitter: http://www.twitter.com/forbes
Forbes Video on Instagram: http://instagram.com/forbes
More From Forbes: http://forbes.com
Forbes covers the intersection of entrepreneurship, wealth, technology, business and lifestyle with a focus on people and success.
Subscribe to FORBES: https://www.youtube.com/user/Forbes?sub_confirmation=1
Fuel your success with Forbes. Gain unlimited access to premium journalism, including breaking news, groundbreaking in-depth reported stories, daily digests and more. Plus, members get a front-row seat at members-only events with leading thinkers and doers, access to premium video that can help you get ahead, an ad-light experience, early access to select products including NFT drops and more:
https://account.forbes.com/membership/?utm_source=youtube&utm_medium=display&utm_campaign=growth_non-sub_paid_subscribe_ytdescript
Stay Connected
Forbes newsletters: https://newsletters.editorial.forbes.com
Forbes on Facebook: http://fb.com/forbes
Forbes Video on Twitter: http://www.twitter.com/forbes
Forbes Video on Instagram: http://instagram.com/forbes
More From Forbes: http://forbes.com
Forbes covers the intersection of entrepreneurship, wealth, technology, business and lifestyle with a focus on people and success.
Category
🛠️
LifestyleTranscript
00:00When you're starting to build a company from scratch, there's a lot of hesitancy or fear
00:04to put yourself out there.
00:05When you remove that fear, and when you remove kind of any threshold to try to make something
00:10happen, that's when things happen.
00:13Anirudh, thank you so much for joining us.
00:18To kick us off, can you give me a 30-second pitch about who you are and what you do?
00:22Yeah, no, thank you so much for having me.
00:25I'm Anirudh Joshi.
00:26I'm the co-founder and CEO of Vela Labs.
00:28Vela Labs is an AI diagnostic company that was founded to help cancer patients navigate
00:34what is the right treatment pathway for them.
00:36If you had to explain what you're doing to a 10-year-old, what would you say?
00:40Yeah.
00:41I mean, the most base explanation of this is, today, different drug options don't have
00:48a way to know what's more effective versus another.
00:52And what we do at Vela is really simplify that for the physician and make them have
00:55information in their hands as to this treatment is going to be better than this treatment.
00:59So you should pick this treatment for this patient and this other treatment for this
01:02other patient.
01:03Because at the end, cancer therapy is very personal.
01:06And each patient has a different tumor.
01:08So we help guide and personalize treatment pathways for them.
01:12Amazing.
01:13And Vela Labs is on a big mission, and I want to get to that.
01:16But before we get there, I want to hear your story.
01:20How did you get into the health care industry?
01:23What are your passions otherwise?
01:25And what is kind of the founding story of Vela?
01:28Totally.
01:29Health care has always been a passion of mine.
01:31And this was dating all the way back to high school, where I started reading about retinal
01:36implants and cochlear implants and how we could use technology to actually bring back
01:40the sense of sight and the sense of hearing to both deaf and blind people.
01:43And that was mind-blowing to me.
01:45Because we had amazing technology, and we had progressed in so many different fields.
01:51But health care was just beginning to reap the benefits of all of those technological
01:55advances.
01:56So for me, from the beginning, it was always the most exciting part was, how can we combine
02:01different advances and new technologies to actually solve problems that we were seeing
02:05in medicine and health care?
02:07So I actually got started as a biomedical engineer in my undergraduate degree at Georgia
02:12Tech and worked at a few medical device companies through that period, leveraging mechanical
02:17engineering, electrical engineering, and computer science and different facets of the medical
02:21device process.
02:22And increasingly, as I was going through that, one thing that I was beginning to realize
02:27more and more is, we had reached some of the physical limitations almost from a basic sciences
02:34perspective when it came to the mechanical and electrical engineering side of things.
02:39But really, where medical devices could be transformative was on the software side and
02:44artificial intelligence.
02:45And this was in some of the early days when machine learning was starting to be applied
02:49to health care data.
02:50So I actually got more and more into that pathway of leveraging AI in health care.
02:56And after my undergrad, I went to Microsoft for a couple of years, really, to learn how
03:00to bring AI systems to production and make sure that people and users could access low-latency,
03:07high-performance models.
03:09From there, I came to Stanford, where I wanted to continue my journey in being able to bridge
03:15medicine with AI.
03:16And Stanford was a very unique environment because you had the medical school and the
03:21computer science department right across from each other.
03:23And there was a lot of crosstalk between the two groups, where engineers and computer scientists
03:28were always with doctors building systems to help improve health care in various facets.
03:34So that was a really exciting time.
03:35And that's when we got more and more deeper into this space.
03:39And I met my co-founders there.
03:41So Vishwesh Krishnan and Damir Rabak, met them while I was at Stanford.
03:45We were working in the same research group, building a lot of these systems.
03:49And the genesis of Velar was really, we spent a year just talking to hundreds and hundreds
03:55and hundreds of oncologists and just trying to understand, the technology was clearly
04:01there, from a computer science and AI perspective.
04:05But where were the clinical unmet needs where physicians could actually leverage those advances
04:10to improve patient care?
04:12And what we realized was, post the diagnosis of cancer, for a majority of cancer patients
04:19today, there's a lot of uncertainty when it comes to what treatment is the best for them.
04:23And this uncertainty is, the medical oncologists feel it all the time.
04:28And the patients feel it too.
04:29So I have chemotherapy A, chemotherapy B, maybe immunotherapy, maybe surgery.
04:35But which one is really going to give me the best shot at a cure?
04:38So genetic testing and genomics sort of made a first pass at this and was helpful in really
04:46starting the revolution of precision medicine.
04:48But our realization was, just 15 to 20% of patients have mutations in the genome that
04:53can actually guide treatment.
04:55So we took an approach of using AI and images of the patient's tumor samples to actually
05:00guide and predict what is the right treatment option for each patient.
05:04What was kind of the process of getting there and making sure that your AI is accurate and
05:10what kind of went into it?
05:12Talking to oncologists, the research, perfecting the artificial intelligence.
05:18Tell me a bit about the process.
05:19Yeah.
05:20So it's a great question.
05:22And AI in medicine should be and is now being held to the same standard that new treatments
05:30or new genetic testing has been held to historically.
05:33And what that implies is large scale studies, which use hundreds or maybe thousands of patients
05:41across the country, maybe across the world, to be able to validate whether these systems
05:46work as they say they should.
05:48So we actually, thanks to all our partners, our physician partners across the nation and
05:54across the world, were able to validate our algorithms on over 1,000 patients on over
05:5930 different hospital systems across the world.
06:02And that allowed us to really demonstrate scientifically and medically to the physician
06:07community, but also from an AI perspective that when this algorithm makes a certain prediction
06:14that it's accurate and it's better than anything that's come before it.
06:18How does a patient get access to Raytheon Labs tests?
06:21Fantastic question.
06:22Yeah.
06:23So it's, you know, one of the biggest mistakes I feel like many new technology companies
06:28make, especially when they try health care, is they try not only disrupting technology,
06:33but they try disrupting workflows.
06:35And that is a recipe for disaster because, you know, at the end, the medical workflows
06:39and medical and systems that hospitals are used to have been built in a certain way over
06:44decades.
06:45And if you're trying to change something, it's better to start with changing one thing
06:48versus changing everything all at the same time.
06:50So for us, the approach that we took is let's make this so simple and fit workflows so perfectly
06:59that any hospital in the US, whether it's, you know, the community hospital down the
07:03street or, you know, one of the best academic hospitals north of here, Cleveland Clinic,
07:07like all of those folks could use the same technology.
07:10And the way we did that is, luckily, we had the good fortune of coming after the genetic
07:15testing revolution, which had a proper workflow for all hospitals to leverage diagnostic testing.
07:23So we slotted into that same workflow that hospitals use today for patients for genetic
07:27testing for our diagnostic testing.
07:30So from a hospital system and a physician perspective, it's exactly the same workflow
07:36they've been used to for the last 10 years.
07:38But behind that is really revolutionary new technology that'll bring biomarker testing
07:43and precision medicine to way, way more patients.
07:46Does it work with insurance companies and are they reimbursing patients for the tests?
07:50Yeah.
07:51So diagnostic testing typically goes through the payers, at least in the United States,
07:55where payers like Medicare and commercial payers will reimburse for the use of these
08:00tests by patients.
08:02And then globally, it's a mix.
08:04Some countries are more cash based, some countries are payers.
08:06But yes, in the US, majority payers.
08:08Is there sort of a case where you would work with pharma companies?
08:12There is.
08:13Yeah.
08:14So in a number of...
08:15It's a great question.
08:16In a number of ways, pharma companies are interested in what we're building, and for
08:20a number of reasons.
08:21So one, pharma companies are always trying to push the needle in terms of improvements
08:27and treatments, giving more treatment options to patients.
08:30Now, in order for them to do that, they'll need to know who is the right patient population
08:35who will benefit most from a new therapy that they're bringing to market.
08:39And the way they can find that out is using tests like ours, because we've developed our
08:43tests to understand who are the responders and who are the non-responders to standard
08:47of care today.
08:49And the folks who are non-responders to standard of care today are really the folks who need
08:53new therapeutic advancements and who will benefit the most from it.
08:57So that's where, actually, there's a lot of interest, even from pharma companies, to use
09:00this.
09:01And right now, if I'm correct, you're focused on bladder cancer.
09:04We are.
09:05Are you going to start looking at other areas of cancer as well?
09:09Yeah.
09:10I mean, it's very appreciating.
09:11So bladder cancer was where we got our start.
09:14We put out the first AI test in bladder cancer to predict response to an age-old therapy
09:20that's been used for the last 20 years in the space.
09:23And the idea was, can we identify those who are non-responders to this?
09:26Because there are so many new therapies and so many new alternatives who are coming to
09:29market where patients can benefit from that.
09:32But this is a platform that fundamentally is pan-tumor and pan-cancer.
09:37And we have already published a number of other cancer types as well, like pancreatic,
09:42ovarian, et cetera.
09:44And we are looking at the whole spectrum of solid tumors that exist.
09:47Amazing.
09:48So there's a lot of really smart people in science, in health care, who work behind the
09:52scenes.
09:53Not everybody wants to become an entrepreneur, because that's its own journey.
09:58That's running a business.
10:00What made you take the leap?
10:02Yeah.
10:03It's a great question.
10:04I think when I first got started in high school and my undergrad, I actually wasn't focused
10:11on doing this in an entrepreneurship sense.
10:14It was more like, through industry or through academia, can I build these systems?
10:18And then what I increasingly realized was, neither academia or big companies, big medical
10:24device companies or big tech companies, would be able to actually bring these technologies
10:28to market and scale it in a way that patients will get access to.
10:32So it was almost a force of necessity that if we needed to bring this technology to market,
10:35we had to start our company to do it.
10:38And that was sort of the driving motivation to start a company.
10:45We strongly believed this was something that was needed, and physicians would benefit from
10:49it.
10:50And the only pathway to doing that was starting a company.
10:51Now one of the things that I think is the beautiful part of Silicon Valley and even
10:58the Stanford ecosystem, is it lowers that barrier to entrepreneurship, which is, there's
11:04always a mental barrier for a lot of folks of starting a new company, venturing out into
11:08the unknown.
11:09And I think what is beautiful about that environment, that ecosystem, is you see others do it and
11:15that kind of mentally lowers the barrier for a lot of people saying, hey, this isn't as
11:20scary as folks may actually think, and it makes more people take the leap.
11:25And I think that's a good thing, especially, and it keeps the US in one of the leading
11:28powers here.
11:29You know, the other piece that is needed when you start a company is you need great co-founders,
11:34and you also need great investors.
11:35And that's something that we got very lucky on on both fronts.
11:39I didn't want to start a company until I met the right group of people.
11:42And I did.
11:43I had a good fortune meeting my co-founders in grad school.
11:46And our investors as well have always been very long-sighted investors, PRVC, Andreessen
11:51Horowitz, and DCVC.
11:53They've always had a long-term vision of where this is going, and that's also highly beneficial
11:58to companies like ours and really makes this thing possible.
12:00What are some of the lessons learned so far in running a business?
12:05So I think, yeah, I mean, one of the, you know, this may sound, you know, tongue-in-cheek,
12:11but it's not, is, you know, a lot of people initially, when you go zero to one, when it's
12:15the first, when you're starting to build a company from scratch, there's a lot of hesitancy
12:20or fear to put yourself out there.
12:22But that is the, you know, putting yourself out there is what actually ends up driving
12:28things and making things happen.
12:30We weren't physicians.
12:31We weren't oncologists coming into the field.
12:33But now over the last three years, we've been able to develop connections and collaborations
12:37with over 200-plus physicians, and part of the reason for that is initially even just
12:41cold emailing, like things that would feel like, oh, would they really respond to me?
12:47Would they know who I am?
12:48This, that.
12:49At the end, like when you remove that fear, and when you remove kind of any threshold
12:54to trying to make something happen, that's when things happen.
12:57And I think like a lot of people, and myself included, I think when I look back, that's
13:02some of the things that, you know, helped us from those very early days.
13:05Now, as you start scaling, one of the things that we began to realize and began to, you
13:10know, understand is like, you know, building the team is the biggest part here, is like,
13:14you know, you can have amazing technology, you can have the right business model, and
13:18everything could be perfect, but at the end, you need to build a self-sustaining engine
13:21to function, especially, you know, at an executive level.
13:25And you know, one thing that we want to be able to do is really build Valar into an engine,
13:30both from the technology side, but then also from the, you know, commercial side and the
13:34company side to be able to run and continue to generate new and new diagnostics for every
13:40part of cancer.
13:41And I imagine, you know, letting go of that fear, cold emailing people, putting yourself
13:44out there, that's how you raised $26 million in funding today.
13:48But tell me a little bit more about that.
13:50Tell me about the very first round you raised, and how you went about pitching yourself,
13:54and were there yeses and noes?
13:57I want to hear all about it.
13:58Yeah.
13:59So it's a great question.
14:00So, you know, we went through, so the first check-in to Valar was from ParaVC, and that
14:05was a pre-seed investment.
14:06We were part of their accelerator program over the summer, and then we went on to raise
14:10our seed round with Andreessen Horowitz, and Vinita Agarwala, and Jay Rogani a few
14:14months after that.
14:16And in that phase, I would say, you know, one of the lessons that Page Money, he's one
14:20of the founders of ParaVC, told us is fundraising is like a soccer game.
14:25It's, you know, all 90 minutes count.
14:29And even if it's 0-0 or 1-1 till the 90th minute, and you get a goal in extra time,
14:34it's still a win, right?
14:35And that's all that matters.
14:37And I will say that that was very true.
14:40I mean, we obviously, you know, had a great number of yeses from amazing partners who
14:45were very fortunate to be able to work with, but we had a lot of no's.
14:48We had a lot of no's.
14:49And that's very common here in this field, and it's something that I think, you know,
14:53folks just need to recognize and realize is not everyone's going to, you know, resonate
14:58with your idea.
14:59Not everyone's going to be a believer.
15:01But when you find the right person who believes in the technology, at that point, it's magical.
15:07So.
15:08I love that.
15:09Great answer.
15:10But is $26 million enough for the big mission you guys are on, or do you need to raise more money?
15:16Yeah, I mean, so for our current targets and where we're going today, so right now, you
15:21know, Vesta's our first product in bladder cancer, and our goals are scaling that out
15:26while also bringing a few new products to market.
15:29For those aims, it's definitely sufficient.
15:31However, to be, you know, the next, you know, billion-dollar company in cancer, we will
15:36definitely go through more rounds of financing and more fundraising, because at the end,
15:41these things are capital-intensive companies to build, but when they work, they work in
15:46a huge way.
15:47So.
15:48Amazing.
15:49Other than AI, which you are clearly, you know, using for the benefit of healthcare,
15:55what are your predictions for other trends that will be relevant to the healthcare industry
16:01in these next five to 10 years?
16:04Yeah.
16:05It's a good question.
16:06I will say pretty much any, the way to answer that is you look back at the last five years
16:11of technology advancements in any other field, and healthcare is going to benefit from all
16:15of that, because very few of those have even trickled into healthcare yet, which is, you
16:19know, the unfortunate part, but also the opportunity, right?
16:22So as you said, AI is going to completely change and create an abundance of healthcare,
16:28and this is across all domains, not just, you know, for what we do at Veilar, but even
16:32throughout the stack with, you know, summarization of physician notes to reduce burnout, being
16:37able to provide, you know, expert pathology, expert radiology opinions for hospitals that
16:43may not have it.
16:44All of that is going to completely change with AI, and it'll be an incredible opportunity
16:48for both patients and providers.
16:52Beyond that, I think, you know, one of the things that I'm actually quite excited about
16:56is even the, you know, robotics when it comes to surgeries, and we're starting to see more
17:01of that come in, and the combination of robotics with software, where we have now a better
17:07way to visualize what's happening inside the body, a better way to actually guide these
17:13robotics and help them actually influence the rest of the stack.
17:17So for example, things that we do involve collection of biopsy material from patients,
17:22and sometimes it's very hard to collect these biopsies, because they're in very small parts
17:26of the body, which is really tough to get to.
17:29And as we get better and better with, you know, non-invasive surgeries and robotic surgeries,
17:32it's going to fuel the whole ecosystem.
17:35I'm very excited to see what happens next.
17:38I'm curious, is Vedar generating revenue yet, or will it come later down the line?
17:44Yeah, so there are different, like, revenue streams for the company.
17:48One as you, you know, pointed out before is pharmaceutical revenue, which, you know, is
17:53through existing products.
17:55Then there is the clinical revenue, which requires Medicare reimbursement in order to
17:58get to.
17:59We're a few months off that.
18:01Generally, these technologies take about a year or so to get to sort of insurance coverage,
18:06Medicare coverage, et cetera.
18:08But on other facets of the business, we're there.
18:11Amazing.
18:12My last question for you, looking at everything you've achieved so far and how you're helping
18:17the healthcare industry and being an entrepreneur, what would you say to your 15-year-old self?
18:26Yeah, I mean, I think, you know, when I look back, I would say what we're doing today and
18:35what, you know, Vedar is focused on and what I have the incredible opportunity and very
18:43fortunate to have is sort of exactly what I was looking for when I was 15, was an ability
18:49to actually be able to use everything I'd learned to make, bring new technology advancements
18:56to healthcare.
18:57And I think, you know, I'm just incredibly grateful to be able to be sitting here and
19:02being working on exactly that.
19:04So if there's anything I would tell the 15-year-olds, like, it's coming.
19:08That's it?
19:09It's coming?
19:10Yeah.
19:11No.
19:12All right.
19:13Well, thank you so much for joining us.
19:15It's been a pleasure to have you.