➤ Adaptive Biotechnologies Corp. is collaborating with Microsoft Corp. to collect and analyze data from thousands of patients who have been exposed to, recovered from or have COVID-19.
➤The companies are making the data from this project, dubbed ImmuneRace, freely available for researchers, as well as using it themselves to try to better understand the body's immune response to the virus.
➤ Seattle-based Adaptive, which filed for an IPO last year, has also partnered with Amgen Inc. to find powerful antibodies that the biotech giant can harness to create a new therapy for the coronavirus.
Harlan Robins, Adaptive's chief scientific officer, co-founded the company in 2009 with his brother and current CEO Chad Robins. Its Immune Medicine Platform uses a combination of bioinformatics, software and machine learning to read the genetic code of an individual's immune system. Having first partnered with Microsoft in 2018 to map how T cells — which play a vital role in the body's immune response — bind to antigens, the two companies expanded their arrangement by launching ImmuneRace in May 2020 and then ImmuneCode in June. The result is a database the company has described as a "treasure trove of information that could provide one consistent and trackable measure" of immune responses to COVID-19.
Robins spoke to S&P Global Market Intelligence about Adaptive's coronavirus partnerships. The following interview has been condensed and edited for clarity.
Adaptive Chief Scientific Officer Harlan Robins
Source: Adaptive Biotechnologies
S&P Global Market Intelligence: What made you want to focus your company on the immune system in particular?
Harlan Robins: Basically, once you start looking into it, it's incredible; it's got this amazingly difficult job of trying to protect you against all sorts of things that you've never seen before.
COVID is a great example, right? There's a new virus that just showed up in our system that we've never seen before. And the death rate seems really high, but as a population, it's not that high — it's a few percent. Even if no one did anything and this was 1,000 years ago, our immune systems would have done a pretty good job protecting us against this pathogen.
Speaking of COVID-19, can you talk about your partnership with Microsoft on the ImmuneRace and ImmuneCode projects?
Normally, we obviously wouldn't do this, but because of the global crisis, we said, "Okay, why don't we just go as fast as we can and give as much information as we can to the world?" As of right now, we don't make vaccines, but knowing what part of the virus your immune system sees is vital to making a vaccine, for example.
So we thought if we could make this data available at scale to as many researchers as possible, we could help the global effort in fighting this. So that's what we set out to do with Microsoft. Adaptive maps the T cells to what parts of the virus they are actually seeing, but there's a lot of different ways the immune system can see those parts of the virus. And we generate a data set with a lot of that information, but Microsoft provides machine learning so we can learn from the connections that we've already made and expand them computationally.
How would a diagnostic created from the ImmuneCode database be different from others?
Antibodies or the serology — the B cell — are primarily designed to prevent you from getting reinfected at some point in the future. And because of that, the T cells come up much faster, and they also have immunological memory, so they persist long after.
There's that gap between when you might not even have [polymerase chain reaction] results anymore and might not have the virus, but you haven't yet created enough antibodies to test positive for serology. So basically, the testing has a time gap in the middle there, and we're hoping that we can fill that time gap, because the T cells do exactly that.
In your collaboration with Amgen, you've said that you're looking for "the Michael Jordan of antibodies." What do you mean by that?
We're trying to find the ones that actually have the function of being able to knock out the virus on its own without the requirement that there's hundreds of them to coat the virus. We just want one or two that [are] powerful enough to destroy the virus or to knock out the virus' life cycle. And it has to have a bunch of other properties too — viruses can mutate, so you have to make sure you hit a spot that the virus can't mutate from. You're literally looking for some superstar antibodies that can do miracle work.
How will the therapy created with Amgen be different from other antibody therapies in development now?
Adaptive's strength is really our scale — not scale in terms of the size of our company, but our ability to probe the immune system more broadly.
If you can throw in two or three antibodies that all hit totally different parts of the virus, but are also all independently able to neutralize the virus, then even if the virus mutates and one of those antibodies is no longer functional, the other two should be able to knock [it] out. So that's really what our strength is — the ability to search a much broader space than some of the other groups looking.