Imagine giving a small blood sample to test for the presence of disease. You're probably saying, "Big deal — Don't we already have that?" and you would be right.
But now imagine giving a blood sample to test for any disease, in one test, all at once. This universal diagnostic is something quite different from today's blood tests and the goal of a collaboration between Adaptive Biotechnologies in Seattle and Microsoft Corporation, announced on January 4, 2018.
Adaptive Biotechnologies develops diagnostics that detect immune system reactions to disease-causing conditions in the body. The company is a spin-off enterprise from the Fred Hutchinson Cancer Research Center, also in Seattle, where scientists focus on the workings of the immune system to learn more about cancer's workings in the human body and find new treatments for it.
On its web site, the company calls the immune system "nature’s most finely-tuned diagnostic," since the system reacts when it senses something in the body isn't right, and accurately interpreting those signals offers a way of detecting disease, even in early stages or tiny amounts. Adaptive Biotechnologies looks primarily at the adaptive immune system — thus its name — which is the immunity that reacts to outside invaders, like flu viruses or E. coli bacteria, or even cancer. The adaptive immune system develops over time to complement our built-in immunity to disease, called innate immunity.
The immune system's work is carried out largely in the blood stream, by certain white blood cells, among them T-cells. Adaptive Biotechnologies zeroes-in on T-cells, particularly proteins on the surface of these cells called receptors that seek out and bind to other proteins called antigens with unique properties for each disease. When this binding takes place, T-cells are activated to fight the invaders, and in some cases develop a staying power, enabling the T-cells to act against pathogens for years at a time. In other cases, these immune activations are not wanted, such as when an organ is transplanted, or in autoimmune disorders, where the immune system mistakenly attacks healthy cells or tissue and results in diseases like lupus and rheumatoid arthritis.
Adaptive Biotech interprets signals from the immune system by genetically analyzing receptors on the surface of T-cells. The company's technology decodes the genetic basis of receptors, returning the sequence of genetic codes that produce the proteins. With this genetic analysis, Adaptive matches the receptors' genetic code to their corresponding antigens that occur when a disease is present in the body. Even a small amount of antigens can trigger T-cell receptors, thus the presence of receptors with a tell-tale genetic code is a sign a particular disease is in the body.
A Map of Receptors to Antigens
The company so far applied this process to diagnostics for specific diseases, including solid tumor cancers, such as lung and breast cancer, and a type of non-Hodgkin lymphoma. The alliance with Microsoft, however, is taking the company in a different direction, toward a diagnostic test that detects a wide range of disorders in one blood sample. The first step in that quest is a mapping of receptors on T-cells to antigens.
While at a high level the task of matching receptors to antigens may seem straightforward, the size of the undertaking is immense. Adaptive says healthy individuals have at least 50 million T-cell receptors flowing through their blood streams, and that's before receptors for diseases join them. A genetic sequence analysis for a single receptor alone would produce a large quantity of data. Multiply that mass of data by tens or hundreds of millions of receptors for a single individual, then matching their sequences to millions of antigens, would require a staggering array of computing power and analytical sophistication, well beyond the capabilities or resources of most enterprises.
That's where Microsoft comes in. In February 2017, Microsoft announced its Healthcare Next initiative to combine artificial intelligence and large-scale cloud computing with companies and organizations in the health and life science industries, making available its computational tools to make an impact on the health of individuals and populations. In this case, the tools are being used to crunch the massive amounts of data generated by sequencing T-cell receptors and matching them to antigens, guided by machine-learning algorithms to make sense of all that data.
Once the map of T-cell receptors to antigens starts taking shape, Adaptive and Microsoft first plan to apply the mapping to diagnostics for individual diseases, but the universal diagnostic is still the eventual goal. "Some conditions like cancer or autoimmune disorders can be difficult to diagnose," says Adaptive's CEO and co-Founder Chad Robins in a company statement, "but this universal map of the immune system will enable earlier and more accurate diagnosis of disease, potentially helping physicians to connect the dots to understand the relationship between disease states and eventually lead to a better understanding of overall human health."
Peter Lee, Microsoft's vice-president for artificial intelligence and research says in a company blog post that "a universal blood test that reads a person’s exposure and response to disease would be, in a word, transformational." And Microsoft apparently isn't just making nice words about the collaboration. The company is taking an equity stake in Adaptive Biotechnologies, although the size of the investment and other financial details were not revealed.
As a commenter pointed out when news of the collaboration broke, early detection of disease in otherwise healthy people may not be a completely good thing, particularly when revealing a disease state could be used as a pre-existing condition to deny insurance coverage -- even if outlawed by the Affordable Care Act. This is yet another case where science and technology are far ahead of law and economics, but it often takes advances like a universal diagnostic to force authorities to take action.