To detect signs of disease in the blood, clinicians look for molecules that signal the presence or risk of an illness. These blood “biomarkers” can be toxins, proteins, DNA, RNA, or other molecules.
For a large number of diagnostic tests, DNA is the marker of choice, and for one simple reason: Molecules in the blood are very small and hard to measure at low concentrations. But scientists are able to detect DNA thanks to a quick, inexpensive technique invented in the 1980s called polymerase chain reaction (PCR), which repeatedly copies a piece of DNA until there are thousands to millions of replicas of the same strand.
Proteins, on the other hand, are not easily copied because the molecules are far more complex—strings of organic compounds folded into elaborate 3D shapes.
Cambridge, Massachusetts-based Quanterix has a new approach to detecting proteins—picking them out of the blood one at a time. “We’ve invented a way to make proteins as sensitive as DNA,” says Quanterix CEO Kevin Hrusovsky. “It’s rocket science in the blood.”
To get a peek at this “rocket science,” we visited Quanterix’s headquarters in a suburban town northwest of Boston—a nondescript gray building tucked in the rear of a business park.
The Simoa 2.0 allows single molecules to be counted in human samples of blood, spit, urine and more. Quanterix
The company’s technology, called Simoa for Single Molecule Array, is a white and blue box the size of an extra-large refrigerator. Inside is a system sensitive enough to detect even a single protein in a vial of blood, says Hrusovsky. That’s like locating one grain of sand in 2,000 Olympic-sized swimming pools, he adds.
Simoa has already been used to detect a protein in the blood that might predict when a healthy person is going to have a heart attack, and another that signals if someone with a traumatic brain injury is at risk for brain degeneration.
Most recently, independent researchers described a new application for Simoa in a series of three papers. Their aim: to detect a protein called neurofilament light (NFL), a biomarker for concussions that had never before been measured in blood. The concentration of NFL in college football players’ blood, it turns out, increases over the course of a season due to head impacts. This marker might be used to detect the onset of brain injury before symptoms are present.
In a tour of the company’s “Accelerator Lab,” where they process 500 to 800 client samples per day (the Simoa units are also available for sale), vice president and general manager Mark Roskey opens up the machine to give us a look inside.
The Simoa 2.0 platform is fully automated: Insert several test tubes of blood (or saliva, urine, feces, even breath condensate), enter a few instructions on the touchscreen, then receive your results in about an hour. Roskey describes the process as “bleed-to-read.”
Inside, the system is a bit more complicated. Luckily, there’s a robot in charge.
A robotic arm, lit from above by blue LEDs, zips in, sucks up a bit of the sample, and whisks it away for three rounds of washes and incubation.
First, the sample is mixed with magnetic beads coated in antibodies to bind the protein one is searching for. Antibodies are sticky, monogamous molecules—they latch onto one molecule and hold tight. Say, for example, we wanted to find a protein called tau in a blood sample. Tau is produced by neurons and implicated in Alzheimer’s and other severe brain diseases, so it is an enticing biomarker but only present in vanishingly small quantities in the blood.
To find tau, we’d mix our blood with magnetic beads coated in an antibody that binds tau. Once coated, the beads look like miniature Koosh balls, and any tau present in the sample gloms onto the end of a spiny antibody.
Next, the robot washes the sample to get rid of everything that isn’t tau—other proteins, blood cells, etc. Two more rounds of mixing and washing attach an additional antibody and an enzyme to the tau protein. These additions make the bead glow if a tau protein is attached—we’ll see why in a minute.
Finally, a second robotic arm hauls the prepared sample to a thin, plastic CD-shaped disc etched by a laser with 24 rectangles each the size of a baby’s fingernail. Each rectangle contains 216,000 femtoliter-size wells—that’s 10-15 liters per well, enough space for one magnetic bead.
Now the fun begins: The robot loads the disc on a stand as if to play music. But instead, it pours a bit of the sample over each well, and the beads settle into them. Then the wells are sealed with oil, and a light turns on.
If a desired molecule is present, the well glows. Quanterix
If a bead has an attached tau protein, the well glows. A computer takes a picture of the disc and analyzes how many wells are lit up and therefore how many tau proteins were present in the blood sample.
A researcher can look for up to 10 proteins at once by making the beads in different colors. The company hopes to soon expand that to 35, says Roskey. The technology can also detect DNA, RNA, and microRNA, but at a price. Currently, PCR can detect DNA and RNA for between $1 and $2 per assay, while tests with Simoa cost about $5 per assay. Hrusovsky hopes the Simoa process will eventually drop to less than $1 per test.
In addition to the work in concussions, the Simoa system has been used to detect biomarkers of infection, cancer recurrence, and inflammation. But to be useful outside the lab, the system will need to be accessible to hospitals and doctors, not just large pharmaceutical companies and research institutions that can afford the machine, which retails for $150,000 to $200,000 per unit.
That could soon be a possibility: Last week, the company announced that it had raised an additional $46 million from investors, some of which is earmarked for developing a desktop version of the machine. “We want to get the test closer to the patient,” says Hrusovsky. “Getting a point-of-care, low-cost device is key.”
Megan is an award-winning freelance journalist based in Boston, Massachusetts, specializing in the life sciences and biotechnology. She was previously a health columnist for the Boston Globe and has contributed to Newsweek, Scientific American, and Nature, among others. She is the co-author of a college biology textbook, “Biology Now,” published by W.W. Norton. Megan received an M.S. from the Graduate Program in Science Writing at the Massachusetts Institute of Technology, a B.A. at Boston College, and worked as an educator at the Museum of Science, Boston.