A New Beginning for Semiconductor Sequencing
By Aaron Krol
February 2, 2015 | In December 2010, Ion Torrent of Connecticut launched the first commercial product based on a new technology called semiconductor sequencing. The Ion PGM, sold at half the price of its nearest competitors, was taking a bold step into reading DNA electronically. Unlike other gene sequencers whose bulky optical systems looked at fluorescent dyes fixed to DNA molecules, the PGM used an array of microchips that could read electrical signals given off when DNA binds to a nucleotide. Ion Torrent, which had been acquired by the larger Life Technologies just months before and had the resources to move aggressively into the genomics market, hoped to seize the momentum in a rapidly shifting field, and become the sequencer of choice for genetics labs around the world.
Things have not quite panned out for semiconductor sequencing in the years since. Ion Torrent followed the Ion PGM with the more advanced Ion Proton: both have mostly fizzled under heavy competition. After a second acquisition by Thermo Fisher Scientific, the company is now racing to keep its customers, increase the speed and throughput of its instruments, and get the nod from the FDA to sell its technology for clinical use, but is lagging behind market leader Illumina on every count.
Chris Toumazou, who pioneered the field of semiconductor sequencing at his company DNA Electronics (DNAe), says he’s not surprised Ion Torrent has struggled. “Ion Torrent are competing with Illumina, who have got full genomes right up to a high spec,” he says. Ion Torrent uses the same basic platforms as Toumazou’s company — Thermo Fisher has a non-exclusive license to use DNAe’s patents for genome-wide applications — but the two companies have very different visions for the future of the technology.
In October 2011, Bio-IT World ran a cover feature on Toumazou, a microelectronic engineer who had made an unlikely turn into biotechnology in the middle of his career. At the time that story was published, Toumazou had spent nearly ten years working on semiconductor sequencing, first in his academic lab at Imperial College London and later at DNAe. After a decade perfecting a handheld microchip that could read short genetic sequences, Toumazou seemed about to strike it big. While Ion Torrent provided a major payoff by licensing his patents, DNAe itself was toying with a miniaturized device that could be used inside a doctor’s office or hospital; Toumazou was gearing up to release a series of pharmacogenetic tests on the platform, to help physicians tailor drug prescriptions to their patients’ genetic profiles.
By going after the most high-throughput uses of DNA technology, Toumazou believes, Ion Torrent has missed out on a key advantage of semiconductor sequencing: its ability to scale down. Because DNAe’s “Genalysis” platform has no optical sensors, there’s no bulky equipment to hold back miniaturization. A Genalysis device can be held between two fingers and used for fast, highly targeted sequencing.
“My heart was always in the point of care: the speed, cost, scalability,” says Toumazou. “That was where I saw the real value of the semiconductor technology.”
And how about those pharmacogenetic tests?
“We had a couple of interesting programs, with Glaxo and with Pfizer,” he says. But gradually he came round to the idea that these partners were going to slow walk the technology, that industry wasn’t ready to limit its patient pool by targeting mass market drugs like blood thinners to specific subpopulations.
“Does big pharma really want this?” Toumazou asks rhetorically. “The number of times I’ve stood at personalized medicine conferences, and you can hear how much big pharma wants it — but do they really want it?”
Skin Deep
As a matter of fact, there is one place you can use a DNAe product for targeted genotyping today. It’s a shop on Bond Street, the nexus of London’s high-end fashion world and, not coincidentally, the second most expensive real estate location in Europe.
The shop is called GENEU (pronounced gene-you). It looks a bit like an Apple store, and it sells skin cream. If you book an appointment, you’ll be greeted by a personal care assistant with a PhD, who collects your saliva using a cheek swab. Your DNA is then extracted from that swab and fed into a Genalysis device, which half an hour later will have concocted your personal U+ profile: a set of genetic variants that GENEU believes affect your skin health.
You leave the store with a skin cream in which ingredients have been blended according to your unique DNA code. The business is apparently quite profitable.
GENEU’s service is not exactly an ennobling vision of the future of medicine, but, says Toumazou, he’s got his foot in the door. “You can take the stigma away from medical technology by presenting it in a luxury environment,” he says. “What you’re doing is getting the consumer to appreciate the value of personalization in an industry that doesn’t have the big pharma regulatory barriers.”
The trick for DNAe is to find a medical niche for the Genalysis platform where the need for a fast, highly portable device is obvious, and a larger whole genome sequencer would not get the job done. Pharmacogenetics made sense because doctors might want to run tests in their offices before prescribing a new drug, but the interests of big pharma aside, that field has had a hard time convincing the medical community that DNA is a better indicator of drug response than age, BMI, or traditional blood tests.
Toumazou is also interested in oncology, which is leading the way on connecting therapies to personal genetic information. Many next-generation cancer medicines are only effective against tumors with certain types of mutations, making it well worth physicians’ while to dip into their patients’ genomes. But here, DNAe might have trouble competing with whole genome sequencing. Cancer tests are rarely urgent, and the range of genes involved in choosing a therapy can be very wide, making it more practical to send specimens to a lab with an Illumina or Ion Torrent machine to sequence hundreds of genes at once. Plus, until someone figures out how to reliably extract tumor DNA from the bloodstream, getting tissue samples from patients will remain an invasive and time-consuming process, limiting the benefits of a fast turnaround for the test itself.
“DNAe [invented] a solution that was looking for a problem,” Toumazou admits. “We’re pretty open to all options. This is blood-to-result. It’s a diagnostic, really, so it’s a very different sell to a sequencing machine.”
The Germ of an Idea
When would you need to sequence a small amount of DNA in a hurry? For clinicians dealing with infectious disease, the answer is all the time.
Bacterial infections can be tricky to diagnose, because their symptoms are often non-specific and it takes time to grow cultures in the lab or run a series of one-off tests to confirm suspected diagnoses. Meanwhile, an infection that progresses to sepsis can become life-threatening in a matter of hours. With a platform like Genalysis, even a small amount of genetic information could be used to confirm the species and strain of an infectious agent, as well as flag mutations that give rise to antibiotic resistance or virulence factors.
The DNAe team has always known that bacteria — and viruses and fungi — were promising targets for their technology. Splitting samples into different chambers on the Genalysis chip, they could run multiplexed genotyping tests for a variety of infectious agents at once. Alternatively, they could perform targeted sequencing of gene regions broadly shared across different species. “We’ll be looking at very wide-ranging primers that allow us to sequence within a particular targeted range,” says Toumazou. “It’s so versatile, because you can amplify, you can genotype, you can hybridize, and you can sequence all on the same chip.”
Chris Toumazou, CEO of DNA Electronics, holding up a Genalysis chip. Image credit: DNA Electronics
The obstacle has been bringing samples to the platform. There’s only so much advantage to be had in quickly sequencing a pathogen’s DNA on a microchip if users still need a full suite of lab equipment to isolate the DNA in the first place. But with infectious disease looking more and more like DNAe’s best point of entry to the clinic, the company is now pressing ahead with pathogens. Toumazou successfully raised enough funding in a venture round last April to scale up operations, and earlier this year, DNAe paid around $24 million to acquire nanoMR, a small company based in New Mexico with a product that quickly extracts bacterial and fungal cells from blood.
“One of the USPs [unique selling points] of our technology has always been its miniaturization capability and its speed,” says Toumazou. “So we were looking for a front end that could be scaled in a very similar way.” nanoMR’s Pathogen Capture System is a portable platform that works by attaching tiny magnetic beads to antigens, which bind with antibodies on the surface of their target cells. The beads can be used to grab pathogens from whole blood, and then magnetically separated from the blood sample.
Toumazou plans to combine the Pathogen Capture System with the Genalysis platform to build a single instrument that goes straight from sample to answer. “It will be a cartridge-based system,” he says, “a bit like a USB stick, a portable unit that you plug in.” DNAe’s software will also cover interpretation of the results, so physicians who run tests on the instrument can get back a diagnosis and recommended drug courses.
Chips on the Table
Bringing a product like this to the clinic, where it would be used to make life-or-death decisions for vulnerable patients, is a harder proposition than bespoke skin care. But Toumazou is not the only one who sees targeted sequencing as the future of infectious disease diagnostics. Illumina and Ion Torrent sequencers are already being used to monitor bacterial outbreaks in real time, and it would be a short (though sensitive) leap from there to diagnostic tests. A newer handheld sequencer made by Oxford Nanopore in Toumazou’s native England is also showing some early promise in this area. And regulators are slowly warming up to the idea of using new DNA technologies in the clinic.
“The FDA are getting a lot more relaxed on this,” says Toumazou, citing companies like BioFire and T2 Biosystems, who both have FDA-cleared platforms that use DNA signatures to identify pathogens. (For a profile of an instrument along these lines, see Bio-IT World’s feature on QuantuMDx.) These companies’ tests are more specific than the targeted sequencing Toumazou wants to do at DNAe, but they do suggest that genotyping, at least, is within the FDA’s comfort zone.
“We know that we’ve got this wonderful ability to scale semiconductor chips, so you can highly target,” he says. “I think that semiconductors will play a very important role in this next decade, where you’ve got optics that don’t scale down to the molecular level or the chip level, and you’ve got nanopore that doesn’t scale up to the level where you can get any accuracy.”
Toumazou’s excitement for semiconductor sequencing has not diminished in the years since Ion Torrent first brought his technology to light. If anything, he’s ready to raise the stakes.
“This has to be global,” he says, noting that point-of-care tests for infectious disease are most urgently needed in low-income countries with little access to hospitals. “Southeast Asia, Africa, these are the places we really need to target. This will really test the technology to its extremities.”
For a platform currently serving the luxury fashion market, it’s a tall order to develop a product that could be deployed in India while clearing regulators in the U.S. and Europe. But Toumazou’s unwavering optimism is not out of place for a sequencing technologist, a breed of inventors that has gotten used to shattering new barriers year after year. And even as mainstream genomics has continued to trend toward bigger, more elaborate sequencing projects, the thrill of reading short strings of DNA with a chip the size of a postage stamp has not worn off. As DNAe turns once more to new horizons, there’s still reason to think Genalysis has not yet had its chance to shine.