College of American Pathologists
Printable Version

  Feature Story


cap today

Rapid HIV test whips up waiver debate

February 2003
Anne Paxton

If you don’t care whether the test is endorsed by the Food and Drug Administration, there are Web sites eager to sell you a rapid test for HIV—one that will give results at home in only three minutes.

But for those who know that’s too good to be true—or who prefer vetted tests—the FDA has provided another option: On Nov. 7, 2002, the agency approved the first rapid point-of-care HIV test. The OraQuick Rapid HIV-1 Antibody test, manufactured by OraSure Technologies Inc., Bethlehem, Pa., is authorized for use in the U.S. by trained personnel. A whole-blood point-of-care test using lateral flow immunoassay technology, it requires only a fingerstick specimen and promises a 20-minute turnaround time.

It has been long awaited. "This is a major development," says James Versalovic, MD, PhD, assistant professor of pathology and microbiology at Baylor College of Medicine and director of microbiology at Texas Children’s Hospital, Houston. "That’s why there’s so much attention. Until OraSure came out with this next-generation rapid HIV, we simply did not have a test good enough to meet standards for FDA approval and potential CLIA waiver."

But are point-of-care HIV tests simple enough to be waived under the Clinical Laboratory Improvement Amendments? Or should they be performed only by laboratories certified to perform moderately complex tests? The answer is still under debate.

Serological testing for HIV has been available for nearly 20 years, but turnaround time for the standard ELISA test is a minimum of five hours, and—because it is usually performed in batches—can be up to two weeks.

Abbott-Murex Diagnostics, Norcross, Ga., has had FDA approval since 1992 for a rapid HIV-1 test called Single Use Diagnostic System for HIV-1, or SUDS. A microfiltration enzyme immunoassay, SUDS is faster than ELISA but requires centrifugation and must be performed in a laboratory. The Centers for Disease Control and Prevention estimates the entire test procedure, from obtaining the specimen until results are available, takes between 48 and 107 minutes, depending on the proximity of the laboratory.

The OraQuick test device is placed in a vial containing a small drop of blood in developer solution. It includes an internal control, so if the device displays two reddish bands, the result is reactive, while one band indicates a negative result. In practice, says the CDC, results are available in 30 to 45 minutes when the test is performed at the point of care.

Don Braakman, a public affairs spokesman for Abbott Diagnostics, OraSure’s marketing and distribution partner, says, "While SUDS has been an effective, reliable test for many years, OraQuick offers an important advance in HIV testing in the U.S." Abbott announced in January that it will discontinue the SUDS test in mid-2003.

Rapid HIV testing is well established in Europe. Since 2000, a rapid HIV test called MiraTes has been available over the counter at pharmacies in the Netherlands, Greece, and Hong Kong, and for €19.50 can be ordered online at Dutch Web sites for delivery by mail (though, again, it is not authorized for distribution here).

In the U.S., however, rapid HIV tests have been stalled by a patent problem. Urged by the CDC, most manufacturers initially sought to develop a test that would detect both HIV-1 and the less common HIV-2. The patent for HIV-1 presented no obstacle because it is controlled by the National Institutes of Health, which licenses it freely. But BioRad Laboratories Inc. owns the patent for HIV-2—and has refused to license HIV-2 to companies developing a point-of-care HIV test. Abbott’s Determine test recognizes both HIV antibodies, but is available only outside the U.S.

A few years ago, the FDA and CDC told manufacturers that a test for HIV-1 only would be acceptable, and while OraSure is the first to win approval, other companies’ applications are pending.

Experts believe a startling number of HIV cases await detection.
The CDC estimates there are 900,000 cases in the U.S., increasing by some 40,000 per year. But despite the 20 million HIV tests conducted each year, only two-thirds of HIV cases have been diagnosed.

The Bush administration has pushed strongly to have the test reclassified from "moderately complex" to "waived" under CLIA. HHS secretary Tommy Thompson, who maintains that Bush has made AIDS a higher priority than any previous president, actually requested when the approval of OraSure’s test was announced that the company file a waiver application for it.

"More needs to be done to make this OraQuick test available to everyone who needs it," Thompson said, referring to a CLIA waiver. "But the process can’t begin until OraSure applies for the waiver, so I ask them to please apply now."

"OraSure happily obliged" by filing an application, says Dr. Versalovic. "So they’ve gone into this with HHS already having publicly declared its support."

However, the College—along with the Clinical Laboratory Improvement Advisory Committee and the FDA’s Blood Products Advisory Committee—believes waived status for tests like rapid HIV could cause harm.

Nearly half of all unregulated waiver facilities reviewed by the CDC in a recent study failed to follow manufacturers’ instructions for tests, the College pointed out to the Presidential Advisory Council on HIV/AIDS. Waiver under CLIA provides no mechanism to ensure proper oversight, personnel training, quality control, proficiency testing, or quality assurance. And in HIV testing, "a false-positive or false-negative reading could have devastating results for the patient," the CAP wrote.

"It’s a difficult decision," says Angela Caliendo, MD, PhD, director of medical laboratories at Emory University Hospital, Atlanta. "It just has to be a very foolproof test, and it has to be done in an environment where someone who tests positive can receive counseling. HIV is not an easy diagnosis to swallow."

"The misconception is that because a test is simple to perform, that means that anybody can perform it and generate an accurate result," says Jared Schwartz, MD, PhD, director of laboratory medicine and pathology at Presbyterian Health Care Systems, Charlotte, NC. "The other problem is that by definition, a waived test is supposed to be one for which even if the result were erroneous, there would be little risk of harm to the patient. There’s no way I can see how an erroneous HIV result given to a patient could not cause harm."

A false-negative could lead a patient to forfeit the chance for early treatment, he notes. He emphasizes that appropriate counseling must accompany the test: any positive result, false or correct, could cause emotional trauma—and in the past has led to suicide.

"The person telling a patient about a positive HIV test needs to provide emotional support as well as referral to medical care," says Kathleen G. Beavis, MD, vice chair of the CAP Microbiology Resource Committee and section chief of microbiology and virology laboratories at the John H. Stroger Jr. Hospital of Cook County, Chicago. "Just to give someone an HIV diagnosis without any behavioral counseling or medical followup is not a service."

Bernard Branson, MD, chief of the laboratory determinants and diagnostics section in the CDC’s division of HIV/AIDS prevention, believes key studies have already demonstrated that rapid HIV qualifies for waiver. "One of the reasons we conducted an ’untrained user’ study was to see whether the test was simple and reliable enough to get waived," he says.

In that study, people at clinics in Los Angeles and West Hollywood were given written instructions only for the test. They had no prior laboratory training or experience, and they received no coaching. Only five percent of the specimens were identified with incorrect test results. In a second phase of the study, 10 minutes of training, including a demonstration of weak positive and false-positive results, virtually eliminated all invalid test results.

"Our conclusion was, the rapid HIV could qualify as a waived test," Dr. Branson says. "We do anticipate there is a high likelihood it will happen."

OraSure argues that four large populations are likely to benefit the most from a rapid point-of-care HIV test: people tested in public health settings who do not return for their test results, women in labor who do not know their HIV status, health care workers who suffer accidental exposure through needlestick injuries, and military personnel in the field.

At Emory University Hospital, Atlanta, rapid HIV is primarily used for employee exposures to needlesticks. Rapid HIV is useful for needlesticks because the hospital can treat the employee within an hour of exposure, Dr. Caliendo says. "Post-exposure prophylactic therapy is the standard if someone is exposed to HIV-positive blood."

A rapid test will be a particular boon to employees who receive a needlestick injury, Dr. Beavis says. "If they are stuck on a Friday afternoon, they’re typically put on medication until the EIA test results come back on Monday." This wait could be avoided with rapid HIV. But even if all the confirmation and notification questions are resolved, Dr. Beavis would hesitate to substitute rapid HIV for the existing EIA screening test. "From a laboratory perspective, I wouldn’t want this test in our setting to be the routine test; my laboratory receives 70 specimens a day for HIV testing and I can batch them in one EIA run."

Preventing mother-to-child transmission of HIV-1 during labor and delivery is another crucial role for rapid HIV testing. At Stroger Hospital, for example, OraQuick is used as part of a research protocol for women in labor who have not had regular HIV tests done during prenatal visits.

Studies show that starting antiretroviral treatment during labor and delivery and providing it to the newborn can reduce transmission by half. But the HHS Office of the Inspector General found that almost half of obstetricians do not routinely offer HIV testing during labor and delivery. Twenty percent said one barrier to doing so is the inability of available HIV testing technology to produce timely results.

"It’s my feeling the FDA is going to add a few provisos if they decide to waive the test," says Dr. Versalovic. "I’m not categorically opposed to a CLIA waiver as long as other criteria are met—that patients are informed properly, and that package inserts for the test specifically offer access to information or counseling." He believes the OraQuick test meets the criteria for waiver under CLIA. "The FDA needs to evaluate the accuracy data firsthand, but the two criteria for CLIA waiver are that the test be easy to perform and highly accurate. This test so far seems to be living up to that."

Initially, he expects, the test will be used by point-of-service clinics and even some larger laboratories. "But as this unfolds, if the test performs adequately and to a high level, one can envision an over-the-counter test," he says.

Approval of an over-the-counter test is not as far-fetched as one might think, Dr. Versalovic adds. "There was quite a bit of controversy when pregnancy tests came out. Now we forget. But back then, people said, ’How can you offer this outside of a physician’s office, with no access to counseling?’ Now we don’t think about that.

"With the right test, we can begin to talk about it for HIV, but it does have to occur in a stepwise fashion," he continues. "The final judgment will be made out in the field, when we see large numbers of people getting tested. The downside is that people may be ill-prepared for the result."

But he contends that the laboratory community should support a waiver: "Certainly if the FDA moves forward on it, it may lead to overall improvement in diagnostic testing for HIV. The upside is that many more people will get the test and a certain proportion of them are going to seek medical evaluation and treatment."

The fact is, a number of people are not getting the test for HIV, he says. "It may be because of the cost of getting an appointment; even if the care is free there is difficulty in getting to a clinic, plus time and hassle factors. The Centers for Disease Control estimates there are large numbers of people walking around infected with HIV but not aware of it." In his view, "the advantages from a public health point of view outweigh the disadvantages."

AIDS patient advocates have added their voices to those of other waiver proponents. Last year, nearly 100 U.S. AIDS organizations issued a call for broad-based rapid-HIV testing. "Limiting who can administer the tests and where they can be administered perpetuates this epidemic" by disallowing testing at community outreach sites, said Florida AIDS Action in a statement.

The group’s executive director, Gene Copello, MD, said, "We are trying to win a war and we’ve been shown this wonderful new tool, and now we’re being told we may not be able to use it in the most effective manner possible."

Despite the benefits of a rapid HIV test, Dr. Versalovic cautions that “rapid” does not mean an immediate answer in terms of diagnosing HIV. "The standard still holds that anyone who is recently infected may be missed with a test, no matter how you look at it," he says, "because it may take two to three months to develop an antibody response." Even with molecular tests, there is still a one- to two-week window between exposure and early detection.

Nevertheless, looking at point-of-care HIV testing strictly from a business point of view, Dr. Versalovic says, "We’re going to see an increase in laboratory testing for HIV if this is offered. It’s not going to take testing away from laboratories; it’s going to increase the number of patients coming back into clinics for confirmatory and baseline molecular testing."

He points out that "15 or even 10 years ago, when someone came in concerned about HIV, we’d draw blood and do an antibody test, and if it was positive we’d do a confirmatory test, and that was it, end of story.

"Now, we need to measure the level of HIV RNA in the blood if we want treatment. Doctors demand molecular HIV testing, a baseline viral load test, and a followup viral load test four weeks later—assuming the patients come back to the clinic. You can only imagine: if you offer the test at the point of care, the number of people being tested for HIV by molecular methods is going to increase significantly."

In HIV testing, speed brings more benefits than simply convenience. To encourage people to seek testing, many public health facilities assure anonymity—but they must depend on the testee to return for the result, sometimes weeks later. According to the CDC, in the last few years 9,600 HIV-positive people did not return to the publicly funded site where they were tested to find out their results. Similarly, hundreds of thousands of HIV-negative people failed to return to find out their status.

Dr. Branson predicts that a rapid HIV test will slash these numbers. In a study at Johns Hopkins Medical Institutions’ emergency room, when a specimen was sent to the laboratory, the turnaround time averaged 180 minutes. Sixty percent of the patients left before receiving their test results.

"When they set up a stat laboratory right next to the ER and the average TAT was fewer than 40 minutes, fewer than 10 percent left before getting their results," Dr. Branson says. "So obviously getting a faster turnaround time is very important."

In a survey of persons of color who received counseling plus an OraQuick test at the University of Minnesota, 97 percent agreed with the statement, "I have tested for HIV in the past and I prefer my results the same day," Dr. Branson says. Ninety-four percent said they would recommend rapid testing to a friend, and 99 percent believed they understood the results of their tests.

After rapid HIV testing was introduced in Malawi, the number of people showing up for testing quadrupled in a short period. "We can’t say it will translate everywhere in the United States," Dr. Branson says. "But our experience in Africa, at least, was that people expressed a tremendous degree of confidence in the test. That was unexpected; we had anticipated people would be anxious."

As it turned out, patients liked the rapid test because it assured them the test was actually performed and was done on their sample, and because they were able to observe the test results themselves.

The availability of rapid HIV will probably increase routine screening as well, Dr. Branson believes. "Since 1992, the CDC has recommended routine screening of hospital admissions in areas where there is a high HIV prevalence. But our experience is that very few implement this screening because of the difficulty in getting people test results. Hospitals don’t want to incur liability by finding someone is infected and being unable to inform them. This may allow us to carry out the recommendations we made 10 years ago."

A particular advantage of the point-of-care test, of course, is that it can be performed much more widely. "It takes a very small amount of space to conduct this rapid test," he points out. When OraQuick was piloted at one hospital, a counter in a utility room was used.

But questions have been raised about available rapid HIV tests.
In the United States, Abbott announced in Oct. 2000 that it was experiencing manufacturing problems with the SUDS test. Because certain lots of the test were not meeting internal specifications, Braakman says, Abbott temporarily halted production in order to investigate. "It was not a very convenient test," Dr. Versalovic says, "so only laboratories were using it as a rapid screen for pregnant patients or health care workers with needlestick injuries."

In Canada last April, the national health care agency Health Canada pulled off the market the Fast-Check HIV-1/2 serum assays and whole blood assays manufactured by BioChem Immunosystems Inc., after testing by the British Columbia Centre for Disease Control identified too many false-negative results, Health Canada said. To date the assays have not been restored.

In prelicensing trials, both tests showed greater than 99.9 percent sensitivity and specificity. But after they were on the market for two years, the BCCDC did a prospective evaluation of confirmed HIV-positive specimens and found that seven of 63 positive specimens had nonreactive results by two lots, 11 had indeterminate results by one or both lots, and 14 others showed faint test results.

After the indeterminate results were removed, the sensitivity was 86.5 percent. "This is unacceptable for an HIV screening test," said the BCCDC. Health Canada also issued a nationwide safety advisory for the 14,000 whole blood and 8,425 serum tests that had already been sold.

The loss of the Fast-Check test has left a vacuum at many Canadian hospitals needing a rapid HIV test. "There’s nothing right now," says Daniel Gregson, MD, associate professor of microbiology and infectious disease consultant at the University of Calgary Laboratory Services in Alberta.

Fast-Check had been approved for occupational exposures and was used regularly. "Our protocol was to draw two tubes of blood, one through the routine system in the province for HIV screening, the other for rapid HIV testing. Health Canada removed the assay without a replacement being available. And they don’t realize the cost and clinical implications involved."

Before the Fast-Check test was withdrawn, a rapid HIV test cost $7 with one hour turnaround time. Now, because laboratory testing is centralized, "we are cabbing samples around the city instead of being able to do the tests on site," Dr. Gregson says. "So now the cost is $7 plus $15 for the cab."

He doesn’t see any movement toward a replacement test. "The problem here is, the market is so small that few companies want to go through the expense of applying for licensure with Health Canada for their product, and we are sort of restricted in what we can do if the product is not licensed in Canada."

Many unresolved issues surround the confirmation of results, Dr. Beavis says. Under current procedures, most laboratories do at least two things before releasing a positive EIA result: They repeat that result in duplicate, and if it repeats reactive they send the specimen off for confirmation, usually by Western blot. Then and only then is the patient informed that he or she is HIV-positive.

Dr. Beavis says, "Any laboratory or public health clinic that wants to consider doing a fingerstick has to ask itself: Will we confirm the rapid test result before releasing a positive result? A big question is whether the institution is satisfied to let a result from this test stand by itself, positive or negative. If this result is not confirmed, that would be a change to what we are doing with the EIA test."

The CDC recommends that patients be told their rapid-test results are reactive and counseled about their likelihood of being infected—but they should also receive a simple message such as: "Your preliminary test result was positive, but we won’t know for sure if you are HIV-infected until we get the results from your confirmatory test. In the meantime, you should take precautions to avoid possibly transmitting the virus."

In recent years, rapid HIV has played an increasingly central role in checking the spread of AIDS outside the United States. "From an international point of view in health care, rapid HIV testing is a very important development," Dr. Versalovic says. He notes that World Health Organization published recommendations have called for use of rapid HIV in areas of high prevalence such as pockets of Africa, and have said a single test alone may be sufficient to treat. In areas of lower prevalence, the WHO recommends two or three tests before treatment because of the expense of treatment.

The market for a rapid HIV test in developing countries is far greater than in North America and Europe, reports Bill Whitehead, executive vice president for sales and marketing with Medical Services International, based in Edmonton, Alberta.

His company’s VScan HIV rapid test kit has been initially approved for use in Latvia and is working through regulatory approval in Estonia, Lithuania, Russia, Belarus, and the Ukraine—a market of about 400 million people—and is in the final stages of approval in 15 other countries. More testing is being forced in many countries because legal policies are mandating testing of certain groups such as pregnant women or foodhandlers, Whitehead explains.

The VScan screens for three variations of HIV—type 1, type 2, and the new subtype O. "It’s not a do-it-yourself test, but it’s simple enough that anybody can administer it," Whitehead says. "For developing countries we’ve developed essentially graphical instructions for how to swab, prick the finger, obtain a drop of blood, and so on. It’s very easy to follow."

There is simply less need for a rapid HIV test in North America, Whitehead says. The company has not applied for approval in the U.S. and Canada, though it is now in discussions with the FDA on an accelerated approval of its West Nile test. When Health Canada pulled the Fast-Check rapid HIV test last year, MSI was concerned that it would cast a cloud over all rapid HIV testing. "The first news release was an advisory bulletin out of British Columbia alluding to a rapid HIV test kit. We were one of the companies that got Health Canada to issue a subsequent news release naming the actual company," says Whitehead.

MSI is preparing to conduct a major study in a Caribbean country; Whitehead believes it will demonstrate the cost savings of rapid HIV. "This country right now has a measurable HIV incidence of about two percent. The existing test costs $20 and it takes anywhere from three days to three weeks to get the results, so we’re saying if they use our test for initial screening at a cost of $4 to $8, it’s going to be so much more cost-effective, and our kit gives results in approximately 15 minutes," he says. "They can send the two percent of all positive tests out for the full-blown confirmatory test. But it means their testing budget can be spread a lot further."

Because of the scope of HIV infection, resource issues like these may dominate the discussion in the U.S. over a waiver for point-of-care HIV, but pathologists such as Dr. Schwartz stress that quality must remain a priority.

"Nobody really knows how many people are walking around with HIV that don’t know it, but I’m sure it’s a large number," he says. "And I would still state in my opinion that although it’s important to try to reach as many of these people as possible, I don’t think the risk of giving people erroneous results one way or the other should be taken lightly."

Anne Paxton is a writer in Seattle.