McCormick Magazine

McCormick pioneers low-cost HIV test


hiv testIn developing countries, where 1.5 million HIV-positive women become pregnant each year, residents face a unique problem: babies born to HIV-infected mothers carry the virus's antibodies — the usual indicator of the virus in the body and what most HIV tests look for — whether they are infected or not.

Early detection in infants is important. While adults can manage the disease for decades, an infant who isn't treated will likely die within a year or two. But tests available today that search for other indicators of the virus, such as DNA or proteins, use complex instrumentation, require specimens to be transported long distances, and can be cost prohibitive.

David Kelso, professor of biomedical engineering and director of the Center for Innovation in Global Health Technologies, and his research group now offer an answer: two new low-cost, portable HIV tests that use DNA or proteins to test for HIV. Though the tests — which use different technologies and have different accuracies and costs — still need to go through clinical trials, researchers are optimistic that they hold the solution to the problem.

Taking on the challenge
McCormick researchers began working on the problem in 2006 after receiving a grant from the Bill & Melinda Gates Foundation as part of a new initiative called Grand Challenges in Global Health. The program — which calls on scientists to undertake 14 challenges, including creating new vaccines, treating infections, and improving nutrition — gave out grants totaling $436 million.

Kelso, along with colleagues at Northwestern's Kellogg School of Management and Feinberg School of Medicine, decided to take on the challenge of improving diagnostics for resource-limited settings.

"The idea was that if we did the product development here with the Gates funding, then most of the risk would be taken out of it," Kelso says. "We could transfer the working devices to a diagnostic company who could then manufacture and distribute them." Those companies are Abbott and Inverness Medical Innovations.

To jumpstart the process, Abbott provided Kelso's lab with a working HIV test that uses polymerase chain reaction (PCR) technology that extracts RNA from blood to detect the virus. "Our job was to replace the existing large, expensive instrument with something that was small, cheap, and battery operated," Kelso said.

Inverness gave Kelso's lab an HIV test that works like a home pregnancy test — the patient's finger is pricked and a drop of blood is put on a small strip. Chemicals within the strip perform the test, and the results come in the form of a pink line on the end of the strip.

This strip test is portable, simple, and extremely low cost — but it tests for antibodies, which, in the case of infants with HIV-positive mothers, is not reliable. So Kelso's goal was to create a test that used the same strip but detected the presence of the p24 protein of HIV. "That required a much more sensitive assay," Kelso says, "and a way to free the p24 protein from the infants' own antibodies."

In addition to the two infant tests, Kelso's lab also worked on a similar low-cost, quick test that could measure the viral load of HIV in a patient. The test shows how much of the virus is still active in a patient's body, which in turn indicates whether a drug is working or whether the patient has a strain of HIV that doesn't respond to that medication.

"It also gives patients positive feedback," Kelso says. "Often there are side effects to these drugs, and it's a way to show them they are getting better even though they may be feeling worse." The test also indicates if the virus mutates and becomes resistant to the drug — since the last thing anyone wants is a drug-resistant strain of HIV being transmitted.

Economies of scale
kelso and jangamIn 2006 Kelso and his lab — which includes a dozen research professors, postdoctoral researchers, and graduate students — went to work trying to find novel ways to perform these tests. "We broke the problem down into smaller pieces, and we let the scientific method tell us which one is the best," Kelso says.

For the strip test that detects the p24 protein, researchers made two breakthroughs: they developed a way to separate the protein from the infant's antibodies, and they changed the indicator line from pink to black, making it easier to see.

"We pretty much had to start from scratch," says Zaheer Parpia, a graduate student working on the project. "We did a lot of chemistry, and then we developed a way to heat the test" — a process that breaks the protein free from antibodies. What resulted was an HIV test that is just a tiny strip of paper and may cost as little as 50¢ to manufacture.

While the test is inexpensive, it sacrifices a measure of accuracy. At stake in these tests are two performance indicators: sensitivity and specificity. Sensitivity is the percentage of the results that will be positive when HIV is present — a test that is 100 percent sensitive would yield no false negatives. Specificity is the percentage of the results that will be negative when HIV is not present — a test that is 100 percent specific would yield no false positives. In other words, if a test's specificity is 90 percent, 10 percent of people who take the test will be incorrectly told they have HIV. The strip test has a sensitivity of 90 and a specificity of 98 percent.

"In most situations, it's better to err on the side of specificity," Kelso says. "You don't want to tell somebody they have HIV when they don't." So why use an HIV test if it's not 100 percent accurate? Accessibility and cost. These tests are easy to use, require no training, and are ideal for rural areas with few or no health-care options. And while 50¢ to $1 may not seem like a large amount to pay for an HIV test, in many countries it's a half day's wage and a significant part of what residents spend on health care in a year.

Faster, better, cheaper
The PCR test, which amplifies DNA, required researchers to find a way to extract nucleic acids in order to replicate them to detect the disease. "You need to first purify the DNA, and that process usually takes a half-dozen steps and several reagents and is done by complex, expensive robots in laboratories," Kelso says.

But graduate student Sujit Jangam said the team set to work on finding better solutions. "We just tried to think outside the box to miniaturize the process," he says. The result was a battery-operated machine that is about the size of a toaster. This test costs more — $1 to $2 — and requires more power than the strip test, but initial results show the test to be 100 percent sensitive and specific.

For the viral load test — which uses some of the same technology as the PCR test — researchers took a test that used a device the size of a table and found a way to use a microfluidic platform that would simplify the process.

Besides having to make the tests smaller and more portable, researchers faced another parameter: time. To best serve the clients — who may come in for one checkup and never return — the devices had to find a result within an hour. Thanks to the work of Kelso's lab, both infant tests easily meet that standard and can provide patients with results in just one visit.

"It's not our place to make the choice who uses which test," Kelso says. "That choice has to be made by the people in the country who are allocating limited funds and can best assess the trade-offs."

But how to find out who wants what? That's where the Kellogg School of Management comes in.

Designing for developing countries
kelso and grad studentsWhile Kelso and his lab worked on technology, faculty and students at Kellogg went about studying the markets.

"We take a broader view on product development," says Kara Palamountain, research associate professor at Kellogg and the executive director of Northwestern's Global Health Initiative. "While engineers often focus on the end user, we talk not only with end users but also with purchasers, distributors of the equipment, those who influence whether the equipment should be purchased at all. We bring the strategic perspective."

Palamountain teaches students in the Global Initiative in Management program at Kellogg, which has sent students to 17 developing countries to interview government officials, nongovernmental organizations, and health-care providers to figure out which product attributes are important.

"Each country has its own system," Palamountain says. "In some, the government officials favor accuracy, while the health-care provider favors portability. If we make something acceptable to the provider, it doesn't matter if the purchaser doesn't approve it. We have
to decide what the trade-off will be."

Palamountain and her students have also studied health-care settings, measuring labs for dust (a concern with the optics in the tests) and alerting Kelso's group to the time requirement. "One of the things we discovered early on is that time is critical — to give the patients results in the same visit and because HIV progresses more rapidly in infants," she says. "So we made that choice over other directions we could have taken."

The group traveled to Uganda over spring break to show stakeholders prototypes — developed by design firm IDEO — in order to get their reactions. "Having a prototype really helps bring out the discussion," Palamountain says.

A group effort
Researchers plan to take the two infant tests to Africa for testing in late summer. Since mother-to-child HIV transmission in the United States is so rare, researchers have only been able to test on samples of HIV positive infants that were collected 20 years ago. The strip test will be evaluated in a lab near the University of Cape Town in South Africa — an institution Northwestern collaborates with. Researchers there collect 30 to 40 infant blood samples a day, and up to 10 percent are HIV positive. The DNA PCR system will be tested at Mulago Hospital in Kampala, Uganda. Testing of the viral load test won't start for another year.

"We decided to concentrate on the infant tests first because there is a much more compelling need to have a rapid test in remote settings," Kelso says. "There are more than a million infants born to HIV-positive mothers every year, and right now we think the best place to intervene is when infants get their shots. Most mothers will bring their babies to a clinic or a mobile van in order to get their diphtheria and tetanus shots, so while they're getting their shots, we can run this test and advise them to get antiretroviral treatment for their babies if necessary."

For Kelso, the project has been a chance to bring real-world product development into a research lab. "There are three PhD degrees that will result from this research, and in the course of doing their thesis work these students will have developed a product," Kelso says. "It has turned out to be a very rich area for PhD research."

Those PhD students — Parpia, Jangam, and Kunal Sur — say the project has inspired them to work on diagnostics in their careers. "We've come from places that are resource-limited settings," says Sur, who, along with Jangam, hails from India. "So it's especially interesting and scientifically challenging to work on a test that could have an impact like this."

Though they faced a few roadblocks along the way ("And we saw a little magic," Sur says), the students say one of the best experiences was working with researchers from outside the lab — hearing from experts at Abbott and Inverness and finding new ways to look at the technology in light of the Kellogg market research. "It's great to have that interaction — like you're working in a small company," Sur says.

In a field where research can confound laymen and even scientists outside the field, the students say it's nice to work on a project that everyone can understand. "I like the fact that I can explain this to anyone really easily," Jangam says. "By working on this, I've been enlightened as to how much goes into a product. You don't think about it until you're sitting there, examining every piece that goes into the test."

The students already have several provisional patents under their belts — the new platforms they developed for the tests could be used in other areas, like STD testing — and Jangam and Parpia will travel to Africa in late summer to assist in field tests.

All cite Kelso as an inspiration: "Our motivation is Professor Kelso," Parpia says. "He's the driving force."

"He comes into the lab before any of us and leaves after all of us," Sur says.

"We can't match him," Parpia says. "And we're supposed to be in our prime."

But Kelso is quick to praise the efforts of his group. "This is real-world product development," he says. "The end goal is not a paper in Science. It's a humanitarian product on the market. And it's different than your perception of university labs where a professor does the research. These problems are too complicated for any one person to solve. You have to bring together a group of people with different specialties and perspectives and figure out how to work together to solve these problems."

—Emily Ayshford