Founder Stories: Chad Robins, Co-Founder and CEO of Adaptive Biotechnologies

This video made possible by Killer Visual Strategies. Visit KillerVisualStrategies.com.

Pleasure to be here. Thanks, Tim, for the introduction and GeekWire for hosting this session. I'm Luke, uh, and, uh, some of you, uh, may notice a little something different about me. I've got this beard going, and it's for a reason. Uh, GeekWire has covered my Climb to Fight Cancer campaigns for the Fred Hutch here in Seattle, and I just announced today that, uh, We're doing another million-dollar campaign next year, a trek to Everest base camp, which will bring together 20 biotech executives. And we're going to raise $1 million for Fred Hutch. I'm also going to go to Antarctica before I get there. That's the beard reason. But it's been a great experience to kind of leverage my biotech experience and network toward the great scientific research center that we have here in our community. So I just want to let anybody who might want to contribute or check that out, you can go to timbermanreport.com later today. So now to get started with our conversation today, really happy to have Chad here. Chad, as many of you know, led the biggest deal of the year in Seattle biotech IPO on the NASDAQ over the summer, Adaptive Biotechnologies. What's your stock trading at this morning? Chad?

Probably around $31.

Yeah, not too bad. Several billion dollar market valuation. Immune sequencing. Sequencing of the adaptive immune system. Now, this is not a biology conference, and I don't want to put everyone here to sleep first thing in the morning, but I think it does help to just walk through some of the basics here with what you do. And once we know that, we can have a better sense of where you might go with this. So I covered Chad and his company with the very first story 10 years ago when he spun out of Fred Hutch. And the thing that captivated me then and still does today is this idea that you can sequence the immune system cells. So if you were to scrape off one of my skin cells today, it's got the same code in it that when I was a baby or when I was a teenager or today. And that skin cell will die and it will regenerate and, you know, there's not a lot, you don't need to do a lot more further analysis once you know that code to begin with. But with the adaptive immune system, the B cells and the T cells that we have in our blood, they have this ability to adapt to all the pathogens that we encounter. That doorknob you touched, you know, you've got genes that are reshuffling and rearranging to adapt to all of these different environmental stimuli we encounter. And it can change our health in a big way. It can lead to— so we have an ability— Adaptive started out with this idea that they could sequence those specific immune cells. And that can tell us a lot of useful information about cancer, perhaps, about autoimmune disease, when your immune system goes awry. They've really got the secret sauce here. The software to analyze that data, multiple partnerships they put together over the years. This is an exciting idea. This is what captivated Wall Street, that we can get a better sense at a fine-grained molecular level what's going on in your immune system, whether it's healthy or tilting toward disease or, you know, in a disease state. So with that, just kind of like basics.

I'm not sure I need to do anything anymore. That's beautiful. That's why we have you here.

Thank you. So I've been doing this for 20 years. So, so Chad, you, you've built this company in a very stepwise fashion. I would say you didn't go public until you were 10 years old. As a company. You started with offering immune sequencing as a service to fellow researchers. Your brother Harlan, co-founder of the company, presented this stuff at scientific conferences. A lot of people said, hey, this is interesting, I'd like to do that too. You built your business slowly, stably, got some revenue in, and then you branched out. How have you thought about seizing this really big platform opportunity with immune sequencing as a business. Sure.

Thanks, Luke, and thanks, PeakWire, for having me here. First of all, I want to go back to this concept of the adaptive immune system and then talk about our immune medicine platform, and then we'll talk about how the products overlay onto the platform. So first, your adaptive immune system is just that. It's a system. It actually detects and treats almost all diseases in the exact same way. So whether you have cancer, an infectious disease, or autoimmune disorder, you have these receptors, these defender cells that are floating around in your body, and their job as a diagnostic, as a natural diagnostic or detector of disease, is to go in and find this disease. Once it finds the disease, your immune system springs into action and winds up going and killing that disease. And so this natural diagnostic, which is your receptor, then becomes a natural therapeutic, which goes and kills the disease. So if we keep that basic concept in mind, then we can start to understand as we kind of resolve these different components of your immune receptors and how they find and bind to what we call clinically relevant antigens or signals of disease, then we can talk about the immune platform and then how we're attacking it from a product perspective. So first, our platform is comprised of 4 different components. The first, as Luke mentioned, is just the ability to apply next-generation sequencing technology to be able to decode the adaptive immune receptors. So as what Luke was saying is that every other cell in your body, what we call germline DNA, the code, Your DNA is actually the same. But our immune system, because we need to be able to respond to this vast array of any potential diseases or kind of what we call pathogens that we might encounter, your DNA system actually, the code, the DNA has to rearrange. And so what we're able to do, the first step, is to be able to resolve that genetic rearrangement. And so this led to our first set of products. The first is in, life sciences research called Immunoseek. And what that product is, is no matter what research you're doing on the immune system, whether it be autoimmune disorders, infectious disease, or whether you're looking at cancer and the dynamics of the immune system, now we have a much better tool. Think about like the pickaxe for the gold rush or the blue jeans. We're an enabler of that research at a much more accurate, specific, and quantitative level. If you think about just macro perspective in terms of kind of the hardware software industry, you know, started off with hardware and kind of the IBMs of the world, but when, fitting since Balmer just talked about it, when Microsoft came along, you made these machines then, there were two components of these machines that made them clinically relevant. One is how can a doctor make a decision on a patient to impact care, and how can you use this technology to impact how drugs are discovered. So the second technology that we developed was a first clinical application, and that's called Clonoseq, and that's for measuring what we call minimal residual disease in certain blood cancers. And so if you have a certain type of blood cancer, it's actually a cancer of the immune system. And at diagnosis, you have this one immune system cell that becomes cancerous and has gone awry. Or out of control, so it metastasizes and it basically crowds out all your other immune receptors.

That's the clone.

That's the clone. That's what the clone and Clonoseq, right? And so think about it as a fingerprint of your cancer. So we get that at diagnosis and then the goal of therapy is to myeloblade or kill, knock down that cancer clone. And then simply we can watch for this clone to grow back over time and we can be very, very sensitive to see this clone, this immune receptor sequence, in relation to all your other cells. So what does that allow a doctor to do?

Let's just stop here for people who don't, who aren't into this. So the traditionally, you're with me, right? So if you've got a leukemia or lymphoma, you go into the doctor, you get chemo or radiation, maybe there's some surgery that can be done, Immediate question is, on your follow-up visit, did we get all the cancer? Yep. And until adaptive came along, we had ways of doing that, you know, CT scans, MRIs, imaging. And then there was something called flow cytometry where you could count the cells. You could take a blood draw and say, okay, how many of those cancerous clones are lurking there in the blood? Is it darn near undetectable or is there a lot there? Is it coming back in a relapse? But what Adaptive does is it actually has 1,000 times more sensitive. So you could detect one cancer cell out of a million, right? In that blood sample. That's the clone. It's coming back. So that's the evidence of what he said, minimal residual disease. That is completely enabled by this technology platform we did not have before, and suddenly now minimal residual disease is a term. It's a thing. You go to scientific conferences, everybody's talking about what's your MRD, you know, and if those patients have it, well, then what do you do?

Yeah. So first is response to therapy, prognosis, and then, then did the, did the therapy work or not? But then there's a whole set of treatment decisions that we're developing evidence and real-world evidence, real-world data and trial data to show that you can use the test at different points along the patient care continuum. For example, if you're MRD negative, meaning you can't detect the cancer clone, should that patient have a transplant or not? Is it just some invasive procedure that you're going to get no benefit from? Discontinue— well, I'll just say maintenance therapy is a bucket as well. The reason that this test is more important than ever is there's a new set of drugs that are working better than ever. So Take multiple myeloma, which there's Darzalex, which was approved by Janssen. You obviously have Revlimid by Celgene. There's a whole host of therapies that are showing greater efficacy than ever before. However, once a patient has been driven down to MRD negative, meaning we can't detect their cancer clone, these patients are being held on these courses of treatment for a very long time that have some pretty toxic side effects. There's a question, can you take a patient off of maintenance therapy if you're MRD negative, meaning we can't detect the cancer clone? Alternatively, if we see the cancer clone coming back, the whole idea of this is can we determine if a patient's gonna relapse at a molecular level before they present with symptoms in a clinic? So if you see the MRD level, meaning we can count the cancer clones, if they've gone up tenfold in month 3, another tenfold in month 6, you know that By the way, if that's happening, it never goes the opposite way. It's going to always keep climbing. So if we can catch that early at a molecular level, can you intervene with a treatment decision? Now, interestingly, Luke is absolutely right. This concept of MRD, although it's existed a long time in certain what we call disease indications like acute lymphoblastic leukemia, it's new for other diseases like multiple myeloma, and it's new for the non-Hodgkin's lymphomas. However, you think, and this is what kind of the diagnostic odyssey, if you, I talk to my peers at Foundation Medicine or Exact Sciences, both of which I have had either former or past CEOs and chairmen on my board of directors, it is hard to change clinical mindset even if you have a test that is proven by publication that much better. And I'm thinking about that, I've been in the field.

You've gotten your test through the FDA.

Through the FDA.

You got FDA approval.

—Medicare, 165 million covered lives?

Right, right, right. So there you're— now you're thinking about the business, how you're gonna get paid for this, and what kind of questions does this answer that are meaningful for, you know, the people at Regence, for instance. But, you know, first you had to prove that, you know, you could ask and answer these questions. You could get that information about those clones in the blood, and that has now been vetted. You went through, and that was the first of its kind, next-generation DNA sequencing for immune cells. So you're the only game in town for this. You went to Medicare and made your, but then you need to like make a use case. Start with multiple myeloma or start with acute lymphoblastic leukemia. What is this information telling us? What is it good for? What's it worth? Clinical utility.

Yeah.

So what have those dialogues been like? Yeah.

So this is where I was going. And it's interesting because it's not a companion diagnostic, meaning there are certain companion diagnostics which are called biomarkers that says, hey, if you have, for example, this genetic mutation or marker, then you take this drug. This is not that. This is what we call a complementary diagnostic. So let's say there's 5 drugs that treat multiple myeloma. It doesn't matter. This is basically putting a weapon in the arsenal of the clinician to better treat the patient based on clinicians all the time are taking in a variety of different data sources to make the most informed decision on a patient. This is just a much more accurate, more specific, Specific marker that we can count your cancer cells in if it's coming back. Now, as you as a clinician, you're then making a determination, hey, should I intensify? Should I de-escalate therapy? These are all the decisions that are being made. That being said, it's not, your thought-leading clinicians totally get this, but for the vast majority of clinicians, exactly how do I use the test at this point in the patient's care? And so we're stratifying into different buckets I've mentioned some of them. The pre-transplant, post-transplant, confirmation of CR, complete response, by traditional methods, looking at the maintenance therapy regimen, and looking at all these different things and saying, these are the exact use cases, and we're developing both real-world data and evidence, and we're doing trials around these different concepts to show exactly where it uses. And what I was saying is coverage doesn't equal adoption. Even if it removes a barrier to adoption. So we are, Out there in the field, we've converted like we're in almost, I think, every NCCN center but one in the entire country.

National Comprehensive Cancer Network. Yeah. What do you charge for this test?

Right now, Medicare is paying $1,800 a test. Private payers are within kind of 10% of that, 10 to 15% of that. Lower. Lower. But $1,600 to $1,800.

Getting Medicare first is important as a validating step. The private insurers look to Medicare. Presumably they've run the numbers. How do you arrive at $1,800? I mean, that might sound like a lot.

Well, there's a whole variety of reasons. One is HECOM, which is health economics. You do a whole series of analysis that says, what does this actually do for the healthcare system overall? And what's the It's kind of value-based pricing. What is the value of a test? If, for example, if you can take patients off of medication, you're not only saving that cost of the medication, but you're also saving treating that patient in the hospital for all the other toxic side effects from that medication. So all that goes into kind of pricing. Also, how many times you're going to use that test over the patient's life cycle, etc. But it's almost like these doctors— I was thinking about this because, you know, not being a medical doctor or scientist, I was just trying to think about about this in layman's terms, is, you know, I'm like, why wouldn't every doctor in the world, like, this is just clearly better. And by the way, we have a lot of tests that are, the data's emerging that I can't say yet are definitively better, I hope to be able to, but this is one where we've looked at the data and said it's almost, in our minds, like malpractice for a doctor not to be using this test. I'm using an extreme statement for a reason. But then I'm thinking about it as like, I went back to myself. I was like, in 2007, I'm going around. I'm using a BlackBerry. And I'm telling everybody, I'm much faster on my BlackBerry than your iPhone. And people are looking at me like, OK, dude. So finally, I start using an iPhone. Within, I think, 6 to 12 hours, I don't know how many apology calls I made saying, hey, I'm an idiot. Old habits die hard. You're just claiming a better technology, but it's like, you're using something as a medical diagnosis, it's just something you've been using in your practice. They're changing, and what changes clinical mindset. So we're in that, by the way, this will happen, it's just a matter, like, it's a matter of time and how you're impacting clinical behaviors and adoption.

But now you're thinking about more than just the initial set of opportunities for clonaSeq. At $1,800, you got a multiple myeloma patient or acute lymphoblastic leukemia, And they've got that question, did we get rid of all the cancer or not? And then the doctor has to figure out, okay, do I wanna get aggressive with that next round of chemotherapy or do I wanna back off? They need to figure that out and run some trials to get them comfortable with the evidence on what the right thing is with that new piece of information that you provide. But you've got other ideas too. You're thinking of like longer term, further out about an immune checkup. On your repertoire, which I sort of alluded to in the beginning, this idea that maybe you come in even once a month to check up and say, "My immune system looks like this. The repertoire, the arsenal of antibodies and T cells that I've got is going to look different than a couple months from now when maybe I've got a touch of the flu or maybe a touch of prostate cancer that's lurking or gosh, maybe the early sign of psoriasis or Crohn's disease or something." You could detect that in its nascent, like in its embryonic kind of stage. And that's like, that's the kind of thing that, you know, you or I could come and get at a primary care physician when we're not a patient with a grave illness. It could really go broad.

So let's, let's continue the extension of the platform and the technology. And then we'll talk about, and what Luke's referring to is a very integrated partnership that we have with Microsoft. And we'll talk about what that means. But first, let's go back to this concept that our immune system detects and treats all diseases in the same way. And this is on the detects all diseases in the same way. So every disease has its own genome. What that means is for each disease— pick Lyme disease, celiac disease, I'm picking those for a reason, or ovarian cancer— each one of those diseases has caught 200 to 500 specific signals of that disease. So those, what we call clinically relevant antigens, those antigens are, think about it as a flare. For each one of these diseases they have a flare. And for each one of these flares, our immune system, our immune system cells or receptors, there's many, many, many that bind to each one of these flares, okay? And what we're doing is, so we've developed both the chemistry and the informatics to start filling out this map disease by disease, okay? And think about it this way, in 2000, 2001, when the human genome was mapped, they didn't map the immune system. The technology, Illumina or the hardware wasn't there, Adaptive wasn't there, the immune system was never mapped. So another analogy is like, and this will be generous, but when Google categorized the world's information on the internet, Essentially what we're trying to do with Microsoft is we're trying to catalog how our bodies, what the interaction is between our immune system and all the diseases that they bind to.

You're defining the disease at the molecular roots. Correct. I mean, we live in the Stone Age on a lot of these diseases that you mentioned, like psoriasis. I mean, how do we diagnose psoriasis today? You walk into a doctor's office, take off your shirt, and they're like, well, you know, 3/4 of your upper body is covered in lesions, and we'll give you some immune suppressant, some blunt thing, and come in a month later, and maybe it's like half, or maybe we've got some biologic. But we don't, we don't actually, you can't take a blood test and get a quantifiable number that says, well, you've got this much psoriasis, and we gave it a drug, and we brought it down by 50%. We just don't have that for huge amounts of autoimmunity. So like the opportunity to come in there and like define the disease like down there and then, you know, and then that opens up the whole world of like how you monitor over time progress. Exactly.

So let's— so first, let's talk about what Microsoft is doing and then talk about how we use this map, which is where you're going. So first of all, after we start building this map out disease by disease, this is, this is actually a web-scale problem. Like this is mostly a tech conference still, but If you think about all the data on the web, like, it's just, it's mind-blowing to think about what kind of data problem it is. So Adaptive went out to the kind of trillion-dollar companies and said, who can partner with us on this, and who can put their massive machine learning capabilities and cloud compute onto this? Microsoft wound up being a great partner for many, many reasons, which we spoke about last year, but just a great partner with us. So what they're doing is essentially using machine learning to look at HEFs and building machine learning models to basically fill in the map disease by disease. So if you have, for example, if you see a new receptor that we've never seen before, they can use their models to say, hey, that receptor binds to that specific antigen for that disease. Now, once you fill that out for each disease, here's the idea. And I'm gonna start, I'm gonna echo what Luke says and give you the long-term vision, but I'm also gonna go back Practically, how are we going to commercialize this in the short term to show that we can actually diagnose disease before we get to this concept of an X-ray for the immune system? So the idea is, once you build this map, we've already built the technology and the test, which I mentioned earlier, called Immunoseek, that can extract from your blood this information on your immune receptors. And then this map that we built, we take these receptors, and use this map as essentially this massive V lookup table. So we can take these receptors and map them to the map. So we know, as long as we've mapped the disease, and we know if you have a certain combination of receptors, we know that those receptors bind to a specific signal disease for that disease, so essentially we can reverse diagnose disease from that receptor. So the idea here, long term, call it, you know, the 6, 7, 8 years from now, You'll be able to walk into a doctor's office, get your blood drawn, and we should be able to diagnose many, many diseases all at the same time through a simple blood test. However, that's not where we're starting because for a lot of reasons. One is the FDA doesn't yet, doesn't think that way, although the FDA's been phenomenal. The private payer and Medicare don't think that way yet. But what we wanna do is go after unmet medical needs where early intervention leads to a better outcome and where we understand, at least initially, kind of what that disease state is. So—

That's something like celiac and Lyme, which you mentioned.

Yes, and which is interesting because there's a lot of efforts for early detection of cancer. And by the way, we are looking at certain cancers. Cancer is a more difficult disease because remember I said that, that space of clinically relevant signals varies disease by disease. It just turns out in cancer that you have many more signals of that disease, so it's a more complex data problem, one of which we do believe will crack. And we also believe we have many advantages over looking at kind of the immune system as opposed to looking at things that are shedding from the tumor. But why are we going like, so example, Lyme disease or see like two different cases, we can kind of walk through both of them. Lyme disease is, there's not a great diagnostic for Lyme disease out there. It's just really, really poor. And so we're, we've got data Now, by linking your receptor to that, those antigens for Lyme disease, that is improving quite rapidly to hopefully being able to have a definitive diagnostic or close to a definitive diagnostic for Lyme disease.

As opposed to, you know, vague flu-like symptoms which people present with at the doctor. It often takes them a long time to figure out, aha, it's Lyme disease. Okay, here's how we treat it. We can get a more fine-grained definition from what you guys can start there. Okay, so a little bit on the operations. You, you know, you're based here in Seattle. You've got a lab in South San Francisco as well. That's where people send— doctors want this information for their own colonoscopy, for their own practice. They send the sample in to you guys. You do the lab work and send them a report back. That's how that works. But you now just recently announced a partnership with Illumina, the DNA sequencing giant, to work on a more distributed model, a kit, where if, you know, if you're the Cleveland Clinic and you've got your own Illumina sequencing machine, you don't really— maybe you don't want to send your samples to Seattle. Maybe you want to do it right there in your own pathology lab. You are going to provide them the chemistry and the kit to get this information. But that's a very different business model. Why did you decide to— do that?

Well, I would say that for different tests, there's different reasons. So let's start with ClonaSeq, our blood cancer assay, which I described earlier. There are certain institutions, like the MD Andersons of the world, that don't do send-out tests. And political commentary, or human commentary, it's unfortunate because a lot of that is an economic argument because they want to keep— They don't want you to make the money. They want to keep the economics in their own lab.. So we believe that it's incremental revenue to be able to enable everyone who wants to use a test to use it in some way, shape, or fashion. For Immunoseek DX, what we'll call the Microsoft deal, ultimately we believe the dogma in the field is, is that, and we do have a play in therapeutics which we'll hopefully have time to touch on, but we believe contrary to like the dogma that therapeutics are more valuable than diagnostic. We want to kind of flip that equation on its head. So we believe that ultimately, that if this is part of primary care and you're able to truly do an X-ray or screen for the immune system, then that hits everyone in the world, at least first world countries who do kind of primary care checkup every year, every 6 months. A billion people. A billion people, right? And so if you look at the companies that sell a a lot of stuff to, or stuff to everybody as opposed to a small, even the biggest diseases in the world are still a finite, you know, percent of the population. So, that's kind of how we're attacking it. So, we want a distributed kit. Ultimately, if you think about the computer industry, machines got smaller, faster, cheaper, easier to use, Moore's Law, et cetera. We want to provide the equipment closer to the patient. So that this can be run, ultimately we believe the vision is you're gonna get your blood test, kinda pop it in a chip, pop it in some machine that looks nothing like they do today, and you're gonna get your readout of your immune system, and you're gonna basically have all the diseases, there's some ethical question as to what you wanna know, what things you wanna know about and everything else, but to the extent that we've mapped those diseases, we're gonna be able to diagnose those diseases, and you'll just have a checkbox for a doctor, say check, check, check.

This is how you get to higher volumes, lower price product. You're still gonna make some money off of each and every one of these tests that they run. Get it in the hands of people at labs, MD Anderson, Cleveland Clinic, wherever.

And even primary care. And primary care. Hospital systems, any hospital system where a patient goes in to see a doctor, they should be able to, again, this is visionary and again, not what we're starting with, but ultimately what we're building to is the ability to do this, I'll say locally, and when I say locally, sometimes that means kind of in aggregated provider centers, but really closer to the patient. Yep, yep.

Okay, so we started with the research services for the fellow immunologists and researchers out there that just want, if they're asking a question about some samples, they send them into Adaptive, they get an answer back. Same kind of thing at work there with clinical diagnostics. Now you're trying to make the, the, the finances pencil out, getting paid for the information that you provide. That's a second leg of the business. Now you mentioned therapeutics. This is, you know, a lot of people in biotech think, well, that's where all the money is. You know, people don't really pay much money for diagnostic information. They pay big money for, you know, the fancy targeted antibody drug that attacks your cancer. And, you know, you've thought about that. You know, how can we use this immune sequence information to create personalized cell therapies, engineered T cells to hit— to— that can be taken from you, re-engineered, reinfused back to attack your cancer. You have a big partnership with Genentech announced, I think, a year ago.

How's that going? It's going really well. Let me maybe take a step back. I'll describe what it is. And maybe just a commentary that, oh wait, you guys are doing life science research, because we've gotten this before. Wait, you guys are doing life science research and clinical diagnostics and now you've got a therapeutic. Aren't you guys doing a lot of stuff? It seems like you're doing a lot of stuff, right? You might lack focus. Right, right. So the reality is we're doing one thing. It's the immune system. And if we can translate the genetics of the immune system into different applications, That's what we're focused on. And these opportunities arise and we've been intentional about how we built the platform technologies. So let me describe, and I'll make one correction. We actually don't engineer T-cells. We're looking—

You provide the information. So, yes.

Genentech would make the therapy. Yeah, but it's a different strategy than, we're actually looking at naturally occurring. We've got the depth, we can breadth and depth. So that same technology that matches your immune receptors, what we call T-cell receptors, to antigens that we have that underpins the Microsoft deal, this is also underpinning technology for our therapeutic play in cell therapy. So now take a specific cancer. Let's take lung cancer. Lung cancer has certain disease signals or antigens that are associated with it. In this case, like WT1 is a tumor-associated antigen. It's just a disease signal for lung cancer. Now we can, because of our throughput and scale, we can find all of our immune receptors that bind to that specific implicated signal of disease for that cancer. Now what we have done is built a characterization platform to look at those receptors and look at certain properties of those receptors that make them a good targeting molecule for therapeutic use. So more specifically, we look at that receptor and say, how well does it bind? How well does it kill? And what's the safety profile? Does it cross-react with something on your heart tissue that's going to kill you, right? And so we developed this platform called TruTCR where, again, we're not actually taking and screwing with the the T cell receptor, we're using what's naturally occurring in your body, we're just finding those ones that are way out on the tail of your distribution and saying, and then putting them through the screening assay to say, okay, now these ones are really great as targeting molecules for therapy. And what's interesting about this is we're in Seattle, most of you guys have heard of Juno, there's another company called Kite, both of these companies were acquired by big pharma. Novartis had a cell therapy,. And there's emerging technologies and startups that are doing this. But all of these, all of these companies are going after the same markets where the therapies are working on only a small subset of blood cancers. And that's because of the molecule or the targeting molecule itself. It only binds— basically, you can live without your B cells. And what these CAR-Ts do is they basically ablate, basically kill all your B cells. But they're not actually cancer-specific. Whereas if you have a solid tumor, not a blood cancer, if you have a solid tumor, it has specific signals of disease, which I talked about. So we're talking about tumor-specific therapies. And that's the first step of our relationship with Genentech is to help them find the best targeting molecules that can get them potentially into solid tumors.

And then they will take this custom information. It's different for me, different for you, different for someone here in the audience, and they'll make a custom therapeutic.

So the first part I was talking about is what we're calling the shared approach, and then the second, which you're going to, is that second component is for each patient, fishing out of your blood the immune receptors that are binding to your specific tumor and creating a custom therapy, providing that, that's exactly right, providing that information to Genentech where we're actually providing a custom therapeutic for each cancer patient based on that specific patient's immune system and that specific patient's mutations in their specific tumor and being part of the supply chain. It's the first time, by the way, this is in the pipeline. We're working, there's a lot of value engineering, there's a lot of work that needs to be done to get there, but this is exactly kind of where we're going. But the shared approach where we find T cell receptors against these shared neoantigens will be first.

Last thing, I know you're getting this all the time now that you're on Wall Street, publicly traded. People want to know, like, you've got a platform, it's exciting, you can do lots of things with it, but eventually platforms need to give birth to products, products that make money and, you know, make the whole thing sustainable. How do you think about that, that balance and getting that right, keeping people excited about the vast potential that's here But staying focused on a few things and getting great at those and not, not trying to boil the ocean.

That, that's exactly right. I mean, we have to commercialize products. And that's why, for example, the Microsoft deal, we're starting with like celiac disease and Lyme disease where we can hopefully be on market with one. We've committed to having a clinically relevant signal this year, running a trial next year and being on market with our first diagnostic from the Microsoft partnership in 2021. For Clonoseq, that is about getting into the blood. So moving from the bone marrow, which is an invasive thing to do, to take a bone marrow punch to be able to monitor residual disease in the blood. It's about increasing the revenues from our research business over time. And then finally, it's about hitting the milestones in our, in our Genentech deal. But I'll tell you, we just over the last year, we've instituted a PMO, a program management office to exactly help us make these decisions on a more systematic basis, a more thoughtful allocation of capital basis, to say how are we going to allocate capital both for the here and now and for the longer-term vision. Because we need to keep investing in R&D in order to really impact health care and change the world.

Well, you definitely raised a lot of capital. And I suppose you're going to go out and hire a bunch of people that will help you do this. So thank you, Chad. Thanks everybody for listening with us.

Thank you, Luke. I really appreciate it. Thank you, good friends.