June 9, 2003 — It’s not the inmates taking over the asylum, but similar: mathematicians taking over medicine.
AgaMatrix Inc., a nine-person startup based in Cambridge, Mass., has developed a new technology to measure blood sugar levels in diabetics. Rather than measure glucose through a series of biochemical tests, AgaMatrix has put a digital signal processor in its biosensor to filter out extraneous blood chemicals using mathematical algorithms.
As a result, AgaMatrix says, it can get a more accurate reading with less blood — as little as 100 nanoliters. That would let diabetics take a blood sample by pricking various parts of their body with a microneedle, rather than jab a regular needle in their finger — a huge selling point for the diabetes market, where a chief obstacle to treatment is patient reluctance to put pin under skin.
“It doesn’t sound like much unless it’s your finger you have to jab the needle into,” said Paul Kelly, an angel investor and biotech entrepreneur who put startup cash into AgaMatrix. “It’s one of the real barriers.”
AgaMatrix was started by two men, Sonny Vu and Sridhar G. Iyengar, who met at the University of Illinois as undergraduates in the 1990s. Iyengar went on to Cambridge University in England and spent seven years developing the algorithms now at the heart of AgaMatrix.
Iyengar founded the company in 2000; Vu joined the next year. The two landed several hundred thousand dollars in seed money in 2002, and hope to close their first round of venture capital for several million dollars in the next few months.
AgaMatrix essentially takes a computational approach to a biological task: detecting glucose levels in a person’s blood. Current technology uses a biosensor to take the blood sample, and then runs a series of biochemical tests to filter out other material like Vitamin C that might give a false reading.
AgaMatrix replaces those biochemical processes with a digital signal processor (DSP) that manipulates the biosensor directly, to discard automatically any pollutants and focus solely on the glucose. “We say give us the wet chemistry with all its defects. … We make the signal cleaner,” Vu said.
Vu admits that filtering biological processes through mathematical formulas is a novel approach to life sciences. To his knowledge, AgaMatrix is the only business pursing this idea. “It’s kind of like a new kid on the block doing something out of his field,” he said.
Thomas Davison, president of Sontra Medical Corp. in Franklin, Mass., said AgaMatrix is simply another step on the evolutionary ladder of glucose sensors as they move from traditional pin-prick approaches to noninvasive, constant monitoring. Sontra is developing a glucose sensor of its own, a patch that rests on the skin.
“Accuracy is not the issue,” Davison said. “The big issue is how to get diabetic patients to be more compliant.”
AgaMatrix’s product consists of two parts: the algorithms, which can be etched onto a DSP chip, and the electrochemical sensing strip that holds the actual blood sample. The strips are the profit center; they cost a few cents each to make, and 12 million diabetics use the strips eight or more times per day in the United States alone.
Two other elements will be critical to AgaMatrix’s success: approval by the U.S. Food and Drug Administration and partnerships with large medical equipment makers. Vu expects FDA approval to take only a few months, since glucose sensors are well-known devices; if all goes according to plan, AgaMatrix equipment would be on the market by late 2004.
Vu said the company is talking to six major manufacturers of glucose monitors, such as Johnson & Johnson, Roche and Bayer. They would do all the packaging and marketing of the technology.
411 Massachusetts Ave.
Cambridge, Mass., 02139
AgaMatrix was founded in 2000 by Sridhar Iyengar and incorporated in July 2001 with business partner Sonny Vu.
Small tech-related products and services
AgaMatrix is developing electrochemical and DSP (digital signal processing) technology to enhance the performance of medical device biosensors. Improvement in biosensors can help reduce the invasiveness and increase the accuracy of diagnostic devices. Applications for AgaMatrix’s technology range from home glucose measurement kits to point-of-care hospital tests. In the company’s business model, electrochemical testing strips and DSP chips will be marketed to medical device manufacturers. The technologies may also prove to be relevant to nonmedical industries, including environmental monitoring, industrial gas sensing and the military.
In June 2002, AgaMatrix received $500,000 in seed financing from incTANK Inc. and individual investors (including biotech supporter Paul Kelly). The company is in the middle of a $2-4 million funding round expected to close in spring/summer 2003.
Barriers to market
As new, less-invasive testing technologies are introduced, patient compliance and willingness to experiment with new techniques will be key to product success. AgaMatrix will also need to clear the FDA approval hurdle, as well as cementing key partnerships with medical device manufacturers.
Though no other company appears to be using mathematical algorithms as a component of diagnostic testing, many firms are in the blood glucose or point-of-care testing niche, serving as both competitors and potential AgaMatrix partners. They include:
What keeps them up at night
“I think the top concern for us is the same as it is with many other medical device companies — marketing and distribution,” said co-founder Sonny Vu. “I don’t think getting a partner is the problem; rather, the question is, can we get one (soon enough) that is visionary enough to be excited not just about our glucose application but also about our future applications.”
— Research by Gretchen McNeely