August 2010
Feature Story

Josh P. Roberts

“It’s about time” that combination antigen/antibody HIV diagnostic assays were introduced into the United States, says the CDC’s Bernard Branson, MD, about the FDA’s June appro­val of Abbott Diagnostics’ Architect HIV Ag/Ab Combo, the first combination assay in the country. Such so-called fourth-generation tests have been in use overseas for about 10 years and are now considered standard of care in France, Germany, the United Kingdom, and elsewhere. “This will allow us to catch up with Botswana,” says Dr. Branson, associate director for laboratory diagnostics, Division of HIV/AIDS Prevention, Centers for Disease Control and Prevention.

Abbott Diagnostics will begin shipping the assay to customers this fall, according to Abbott spokesperson Darcy Ross. The Combo test is licensed to run on Abbott’s Architect i1000SR and i2000SR platforms.

“The CDC has been actively working, certainly since 2007, to bring fourth-generation tests to the U.S. market,” Dr. Branson says. “That has involved efforts both with manufacturers and with the FDA, and we’re glad to see those efforts paying off.”

Until now, first-line HIV diagnostics involved looking for antibodies produced in response to HIV 1 groups M and O, as well as HIV-2. “The third-gen tests were tests that had enhanced antibody detection and used an antigen sandwich configuration with highly selected immunodominant recombinant or peptide antigen,” explains Michael Busch, MD, PhD, professor of laboratory medicine at the University of California, San Francisco.

The problem is the serological window during which test results are negative despite the presence of infection. According to data cited by Patel, et al. (Arch Intern Med. 2010;170:66–74), it takes about 22 days for detectable HIV antibodies to appear.

The fourth-generation advance is to retain the third-generation assay’s high level of antibody sensitivity and add the ability to detect antigen—using monoclonal antibodies created against a highly conserved determinant of the gag p24 protein, cross-reactive with all known strains of HIV, says John Hackett, PhD, manager of emerging pathogens and virus discovery programs at Abbott. “It’s a way of directly detecting the presence of the virus rather than looking for antibodies.”

Because virus has to be made before the immune system can react to it, antigen is produced (and should be detectable) before antibody. Data presented at the 2010 HIV Diagnostics Conference by three manufacturers of fourth-generation assays “showed that, with seroconversion panels, such assays could yield a 10–12-day reduction in the window period relative to third-generation (IgM-sensitive) antibody assays, hence allowing for the detection of nearly 80% of acute HIV infections otherwise detectable only by RNA assays” (Expert Rev Anti Infect Ther. 2010;8:631–633).

Among high-risk populations, Dr. Branson says, those who are antibody-negative but have acute infection represent 10 percent of all the positives. “So the current antibody tests are missing 10 percent of the infected people, [and the] fourth-generation should pick up at least 80 percent of them.”

And that’s important, says Peter Leone, MD, medical director of the North Carolina HIV/STD Prevention and Control Branch, University of North Carolina, “because we think those first eight weeks after acquiring infection may account for anywhere from 15 to 50 percent of transmission events. So it’s a short period of time, but it’s a significant contributor to ongoing HIV transmission.”

“The problem you get into is not only that they’re more infectious—and they are—they’re anywhere from 10 to 20 times more infectious during this early phase, and that’s due to the fact that the virus is at incredibly high levels during that period of time,” Dr. Leone says. “But also behaviors during that period tend to be behaviors that actually facilitate HIV transmission as well. So when you add those two things together, you contribute significantly to transmission within the general population.”

Simply informing people that they are HIV-positive leads them to reduce their risk behavior by 68 percent, Dr. Branson says. “So that a person who knows they’re infected is 3.5 times less likely to transmit HIV than a person who’s infected and doesn’t know it.”

If HIV p24 antigen appears in the blood so much earlier in an infection than does anti-HIV antibody, why not just test for p24? Once the body starts making antibody, it tends to bind up the virus—and thus the antigen. “And so it turns out that after the acute phase, the antigen is not present anymore at a detectable level through these tests,” Dr. Leone explains. “Which is why we haven’t been able to use just the p24 antigen as the sole HIV test.”

Fourth-generation tests are designed to detect the spectrum from acute to chronic infection, says Abbott’s Dr. Hackett—to give a yes/no answer. Most are not designed to discriminate between acute and chronic infection because they do not in­dicate whether it was antibody or antigen that was detected. None differentiate which strain of the virus the detected antigen/ antibody represents.

Currently in the United States, a positive result on any HIV test has to be followed up with a confirmatory test before the results are reported. This generally means a positive antibody test is followed by a Western blot or an antigen test, or perhaps a transcription-mediated amplification assay for HIV-1 RNA (generally referred to as NAT or NAAT, for nucleic acid amplification testing).

In light of the sensitivity of the third-generation assays compared with the others (with the exception of NAT), and the cost and skill involved in NAT, members of the Association of Public Health Laboratories and the CDC convened workgroups to propose updates to the two-decades-old algorithms for point-of-care and laboratory HIV testing. The workgroups’ proposals were released before the FDA approved the fourth-generation assay in June. But the groups are now interested in feedback on the potential uses of fourth-generation assays and ask that comments and suggestions be sent to hiv.algorithm@aphl.org.

Among the goals of the confirmatory algorithms should be to answer the questions that remain after a positive third- or fourth-generation test, says Dr. Branson, one of the authors of the joint APHL/ CDC proposals. This can be accomplished in a variety of ways, including through the use of NAT, immunoassays, or HIV-1 versus HIV-2 antibody differentiation tests.

At present, some public health organizations are using a NAT pooling strategy, pioneered by Dr. Leone and his colleagues at UNC, as the primary screening test (Pilcher, et al. N Engl J Med. 2005;352:1873–1883). Batches of up to 90 samples are tested together. If the resulting pool tests positive, then samples are tested in smaller batches until the one infected with HIV is discovered. The process will detect virus in the blood about four days earlier than will a fourth-generation antigen/antibody screen. But there are tradeoffs: Pooled NAT is expensive and complicated to perform and has a turnaround time of about 10 days for results. That “might not sound significant, but if you’re looking at someone who’s highly infectious [and] is continuing to have unprotected sex, we want to be able to identify them as soon as possible,” Dr. Leone says.

Pooled RNA testing hasn’t caught on in the United Kingdom, says Martin Fisher, MD, a physician at Brighton and Sussex University Hospitals in the UK, where fourth-generation testing has been standard of care for several years. Perhaps this is because of the availability of easy-to-use, less expensive fourth-generation assays, he says. In fact, Drs. Leone and Branson predict the demise of pooled RNA testing as a diagnostic tool in the United States for precisely that reason.

Blood banks, though, will continue to screen using NAT. “Since we have the systems in place and licensed excellent tests for the HIV nucleic acids, we’re not going to take a step back and go to fourth-generation testing,” explains Dr. Busch, who is also senior vice president for research and scientific affairs at Blood Systems Inc., Scottsdale, Ariz. “It’s a similar situation in Europe and Japan and Australia.”

Another major advantage of the Architect HIV Ag/Ab Combo is its ability to be run on a random-access platform. The platform does not require patient specimens to be bracketed with controls; instead, controls must be run once every 24 hours of testing time. The assay can’t be run on any other platforms, which means that, at least for now, to do a fourth-generation HIV test in the United States, a lab will have to have an Abbott system.

Bio-Rad’s blood virus division manager, Patrice Deletoille, PhD, says the company has completed clinical trials and will submit a fourth-generation assay to the FDA within the coming weeks. Unlike the Abbott assay, Bio-Rad’s assay can be run manually or on a fully automated enzyme immunoassay testing platform.

“We need multiple vendors,” Dr. Busch says. “We want a number of choices both to overcome [issues] about instrument costs and exclusivity, but also to keep pricing competitive.” If the cost of adoption is too high, Dr. Branson cautions, fourth-generation tests could succumb to the same fate as Abbott’s third-generation test when it came to market in 1992: “It really was not adopted by almost any public health labs through 2006 because it was fivefold more expensive” than other available HIV diagnostic assays.

Abbott’s Ross says the price of the new Ag/Ab Combo assay will be “in line with other tests that are currently available. And it’s less expensive than the [less sensitive point-of-care] rapid tests.”

Dr. Hackett cautions that price shouldn’t be a laboratory’s only consideration in deciding which assay/platform combination to use. Most important, he says, is the assay itself. “Don’t assume that all HIV combo assays are created equal,” he says, explaining that Abbott has taken great care to make sure its assay recognizes the less common and emerging strains of the virus.

But if Europe is any indication, there’s room here for multiple players in the HIV combination antigen/ antibody assay market. “My expectation,” Dr. Busch says, “is that we’ll have a second fourth-generation licensed within four to six months and a third within a year.”