By Elizabeth Gardner
November 15, 2003
>> A POSTAL WORKER finds white powder in a letter, and inserts a sample in an analyzer. Minutes later, good news: The mystery substance is not anthrax or any other bioterror agent.
>> AN OBSTETRICIAN swabs the birth canal of a woman in labor and inserts the sample into an analyzer. Minutes later, bad news: It is a positive result for Group B strep. Prompt treatment with antibiotics saves the baby from possible death or permanent disability.
>> A PARK RANGER collects a water sample from a public beach and inserts it into an analyzer. Minutes later, good news: The Escherichia coli detected is harmless, and the sample is otherwise free of dangerous bacteria. He hoists the "Beach Open" flag....
The "analyzer" in these scenarios may be hypothetical today, but will likely become real, albeit in several guises, in the next five to ten years. (Indeed, the post office is testing an early version now, courtesy of the "war on terror.") Numerous companies are taking advantage of breakthroughs in nanotechnology, microfluidics, and genome research, aided by abundant grant money and venture capital, to develop the next revolutionary diagnostic tool: a DNA analyzer that's fast, portable, affordable, and accurate — even in the hands of an untrained user.
Tests that today require several days in a clinical lab could be slashed to about half an hour and conducted in the field or at the point of care. Tests that are expensive or impractical for routine use — such as screening for single nucleotide polymorphisms (SNPs), the genetic glitches that can signal predisposition to certain illnesses or receptivity to certain drugs — could become affordable and commonplace.
|Five Companies To Watch
|Five of the companies working on DNA analyzers.
Many companies are engaged in this quest, and the product pipeline is filling (see "Five Companies to Watch"
). "There's lots of activity in research and in contact with the agency," says Steve Gutman, director of the FDA's Office of In Vitro Diagnostic Device Evaluation and Safety, which will oversee regulatory approval for medical use of such devices. "This area is a hot one!"
DNA analysis takes time partly because it requires amplifying the target DNA, usually by polymerase chain reaction (PCR), which can take hours or even days. The companies involved in developing portable quick-turnaround devices are exploring ways to speed up PCR, experimenting with other amplification techniques, or dispensing with amplification entirely by employing detection methods that work on extremely tiny samples.
Microfluidics — how liquids behave in channels the size of a human hair or smaller, and the fabrication of those tiny channels — is driving dramatic miniaturization of many lab procedures. For example, PCR used to require three rooms of equipment, each 12 feet square, says John Bishop, CEO of Cepheid. "We've reduced that to a system that's six inches by eight."
The attacks of Sept. 11, and the subsequent anthrax incidents, made it a matter of national security to be able to identify deadly pathogens promptly and easily. The Advanced Technology Program at the National Institute of Standards and Technology (NIST) and the Department of Defense's (DoD) Defense Advanced Research Projects Agency (DARPA) have given grants and contracts for biodetection, in addition to work already funded by the National Institutes of Health (NIH). The Department of Homeland Security recently allocated $350 million to the development of next-generation detection systems for biological and chemical attacks.
The U.S. Postal Service in May awarded a $175-million contract to Northrop Grumman to develop a biohazard detection system to pick up anthrax and other pathogens in post office air samples. One of the partners in the contract is Cepheid, which is working with Applied Biosystems to develop reagents for biothreat detection. Cepheid has also created portable test kits for the DoD for anthrax, plague, and other biological warfare agents.
The U.S. Army has awarded Nanogen two grants totaling $2.6 million to develop miniaturized devices for detecting biological warfare agents in blood samples. And Nanosphere recently scored an undisclosed multimillion-dollar sum from a government interagency task force on terrorism to build a device to detect those same agents in water.
But, fortunately, bioterror prevention is not a high-volume business, and won't in itself drive the DNA analysis market, even if it underwrites research and development. The big money is in common diseases and ailments, from cancer to sore throats.
Frost & Sullivan analyst Isaac Meek pegs the U.S. molecular diagnostics market at about $890 million annually, and predicts that it will grow to $2.4 billion by 2009, propelled by new technologies, expanding knowledge of SNPs, and improvements in cancer detection.
"There are 10 million annual ER visits for respiratory infections," says Graham Lidgard, senior vice president for research and development at Nanogen, where work on a laptop-size detector is being funded by DARPA and the DoD's Dual Use Science and Technology program. "A lot of those viruses are hard to detect, because you need to culture them. If you had a chip that could detect the 10 or 20 most common infections in two hours, you'd know whether to give the patients antibiotics or just send them home to bed."
Jeanne Jordan, director of microbiology, molecular diagnostics, and immunology at Magee-Womens Hospital and Research Institute in Pittsburgh, wants her emergency department to be able to quickly distinguish bacterial meningitis (which needs immediate treatment) from aseptic viral meningitis. The current test involves culturing spinal fluid, which takes days, and hospitalizing suspected bacterial meningitis cases until the results come through. With a quick-turnaround test, "they wouldn't have to tie up a hospital bed unnecessarily," she says. "That would have a major impact financially."
Cepheid's Bishop predicts a ready market in more accurate cancer staging, especially during surgery. For example, a surgeon often needs to know whether a patient's breast cancer has spread to a sentinel node. Current techniques involve preparing a slide from a tissue sample and having a pathologist read it, which takes time and isn't always accurate. Says Bishop: "We would be able to detect overexpression of certain genes within 15 to 30 minutes."
As with all new technologies, the portable, foolproof DNA analyzer must navigate a series of hurdles:
Building a complete device. Many companies and research groups are basing their products on new and ingenious ways to detect the presence of a certain DNA fingerprint with just a tiny sample. But the end is just the beginning, says David Burke, associate professor of human genetics at the University of Michigan, whose group creates and studies microfluidic devices. There's also sample preparation — a set of technological challenges in itself. "You can't just make one stage of the process small — you have to make all of them small to capture the value," Burke says. "If someone says, 'The detection piece works, but we have trouble with sample prep,' then it doesn't work." Two of Burke's students founded HandyLab (see "Five Companies to Watch"), which is focused on shrinking complete laboratory procedures.
Regulatory approval. The FDA's Gutman says a device with a completely new way of working would by default be classified as Category III — requiring the most rigorous testing and documentation. Some of the groundwork could potentially be done in veterinary, agricultural, or other nonmedical settings, thereby speeding the process somewhat. "We're very intended-use driven," Gutman says. "The same device for different uses would require different regulatory packages."
Patent issues. Dan Farkas, president of the Association for Molecular Pathology and director of molecular diagnostics at Baylor College of Medicine, Houston, fears what he describes as the "intellectual property quagmire" surrounding the patenting of target gene sequences, which DNA analyzers would need to license. A small company might have to license many sequences in order to offer a viable product, and might get frozen out entirely if a competitor nabs an exclusive license for a sequence physicians are most interested in, such as one for HIV, cystic fibrosis, or hepatitis C. "Imagine how difficult it could be to offer one portfolio in an economically reasonable format," Farkas says.
|"It's a funny thing about physicians — they're very difficult to change. There has got to be a good reason to give up an existing test . . ."
Dan Farkas, Association for Molecular Pathology
But the "gold rush" to patent gene sequences may abate as the U.S. Patent and Trademark Office gets tough. "Patent offices are requiring greater structural detail in [gene] patents," says Stephen Bent, a patent attorney with Foley and Lardner. Applicants used to be able to claim any sequence that fulfilled a given function. "That was OK when biotech was just getting started, but the patent office is less receptive now," he says. On the other hand, different genes with similar functions may be patented separately, potentially expanding the available pool.
Market acceptance. Some vendors are confident that their products will pass a cost-benefit analysis with flying colors, and that a higher cost per test upfront will be offset many times over by the savings from providing more effective care and not providing unnecessary care. "I don't know that it will displace what's being done now, but if it does, it will be because it's faster and better, though not necessarily cheaper," says Vijaya Vasista, chief operating officer of Nanosphere, which is hoping to get a handheld detector into testing by the end of the year.
But Farkas doubts it will be that easy. "It's a funny thing about physicians — they're very difficult to change," he says. "There has got to be a good reason to give up an existing test — it has to be at least as fast, but not more expensive. We'll pay the same $3 for a better test."
And miniaturization can go too far. "You might need to look at 100 microliters in order to detect leukemia early," Farkas says, which is many times larger than the nanoliter-scale samples being contemplated for some microfluidic devices. "If you get the sample too small, I would start to worry about sampling error."
And it has to be simple, says Jordan of Magee-Womens Hospital. "Most point-of-care testing now is very poorly done — people aren't trained or don't care. They mix reagents from different kits, use outdated stuff ... These devices would have to be as easy to use as a home pregnancy test."
Nonetheless, there's no problem thinking up useful things for such devices to do. "Some genes are known to be turned on by toxic substances," says Karen Seta, director of the functional genomics lab at the University of Cincinnati. "The EPA could catch a fish, scrape a scale, and test for toxins without killing the fish."
"The power of these systems is going to be in moving the testing operations to the environment where the samples are," Burke says. "If we can make them small and cheap like computer chips, people will find ways to use them. You'll be able to figure out what kind of fungus is on your roses. The demand is unrecognizably high — it's just a matter of marketing strategy."
Elizabeth Gardner is a Chicago-area writer. She can be e-mailed at firstname.lastname@example.org.
PHOTOGRAPH BY STEPHEN WEBSTER