As I’ve just been catching up on my vacation reading with Marya Zilberberg’s helpful new book, “Between the Lines,” the first thing that caught my eye were the statistics on the test’s accuracy. The test had a sensitivity of ~92% compared to in-office testing results of 99%. So one out of every 12 test results would be falsely negative.
Next was the not entirely reassuring plea of one of the FDA advisory committee members, Dr. Brooks Jackson, professor and chair of pathology at Johns Hopkins University Medical School, “We shouldn't let the perfect be the enemy of the good.”
So, practically speaking, what do these results mean?
Perspective on the problem of HIV
According to the CDC, about 1.2 million people in the U.S. have HIV, the virus that causes AIDS; there are approximately 50,000 new HIV infections each year. Worryingly, 20 percent of HIV-infected people are not even aware that they are infected (26% in Canada). And one-third of those with a new HIV diagnoses progressed to AIDS within 12 months. So there is a strong public health incentive to identifying patients earlier so that they are less likely to transmit the infection to others, and so that they receive earlier treatment.
The long path of HIV testing
Testing has been limited to medical settings—first to hospitals (1980s), where a requirement was pre- and post- test counseling of the patient.
In 2002, FDA approved the first CLIA-waived rapid HIV test, which allowed for testing to be done in mobile outreach centers. The FDA looked at a staged approaches to achieving the goal of a home-use test, looking not only at test accuracy, but at test interpretation, and willingness of test subjects to medical follow-up. The first phase looked at test sensitivity and specificity when the test was run by trained personnel. This culminated in the 2004 approval of the OraQuick ADVANCE Rapid HIV-1/2 antibody test on oral fluid. This test showed a 99.3 percent sensitivity and 99.8 percent specificity.
The second phase looked at sensitivity and specificity of untrained users under observation, in 2008. In the Phase 2B Observed use trial, there were 2% of operational errors observed. These errors prevented users from obtaining an HIV test result; revisions were then made to the packaging and labeling instructions. When those results looked promising, with a sensitivity of 97.9% and specificity of 99.79%, the third phase extended testing to the intended-use setting.
Surprisingly, in this final Phase 3 testing, in a more real-world scenario, Specificity was 99.98%, but the sensitivity dropped to 92.9%. OraSure’s briefing document says “Subjects with false negative results were not clustered in any particular site and were not asymmetrically distributed according to race, gender, sexual orientation or measured literacy score…The reasons for these false negative results based on self-reported outcomes of an individual’s self-testing are unknown.”Bottom line: one out of every 12 test results would be falsely negative.
Reviewing the Orasure background materials to the FDA and the staged testing details, I’m impressed with the care and level of detail that went into this product’s development—including not only the actual testing, but tests of label understanding, revisions, and creation of the call center.
Risks and down-sides of HIV home testing
The FDA itself notes a number of negatives: “The potential risks of an HIV home-use test include not fully comprehending the limitations of a screening test, such as misunderstanding a negative result when testing during the window period or the incomplete understanding of a false positive result. Home-use tests lack live counseling, resulting in the potential for adverse outcomes if the user obtains an unexpected result without concurrent live counseling.
Additional risks include the inability to reach individuals for medical follow-up, partner notification, and public health reporting. There is also the potential for coercive testing and testing by unprepared minors."
One critical decision will be the price point that OraSure sets for their test. Currently priced at $17.50 for clinics, estimates are in the $40-60 range. While that will be good for investors, it’s not so good for patients. In surveys, 46% said they would pay no more than $20 for a test; 13% said they would only take the test if it were free. And undoubtedly many who would benefit from an OTC test will be unable to pay that much several times a year, which is the recommended interval for some high-risk groups. So from a public health perspective, the pricing will be critical as to the test’s impact.
A simple home HIV test might pose an unexpected problem for HIV vaccine trials. In a 2010 trial, 42% of vaccine trial participants developed antibodies to the vaccine, and these antibodies cross-reacted with the test, causing them to be falsely positive. Clinical trial participants could readily detect if they received the vaccine or not by these results.
Benefits of OraQuick
Because this new test can be done in private, at home, it is hoped that more people will be tested. Importantly, 41% of Phase 3 study participants found to be HIV positive reported they had never before been tested; many indicated they would not have been tested without the OTC self-testing option. Projections are that in a million tests, 9,087 additional HIV positive individuals would be identified, who would have been missed by traditional testing methods.
I’m concerned about the relatively low sensitivity, but looking at projections, the approval now makes more sense from a public health perspective. It comes down to the risk/benefit for the individual vs. the community as a whole. For the individual who receives the 1/12 false negative result, the test has the potential to be quite harmful. They might indulge in riskier behaviors and put their partners at additional risk. But for the ~9,000/million who are identified as HIV positive, and for their partners, this test is extraordinarily beneficial, enabling earlier treatment. The test is also projected to prevent 700 additional HIV transmissions per million people. So the FDA concluded that if just 1% of Americans uses the test, OraQuick could identify up to 45,000 previously undetected HIV-positive people and prevent around 4,000 new infections each year. I wonder if anyone has calculated yet the breakpoint where providing the test for free (or perhaps on a sliding scale) would be a cost-effective strategy. I hope OraQuick will be priced to have the greatest public health impact.
Home testing kits have long been available for pregnancy tests. OraQuick is exciting as the the first OTC, in home test for an infectious disease. Its impact? That will depend on the cost, public outreach and education, and how well can directions be followed by individuals who are poorly literate or have impaired ability to understand.
Photo of OraQuick testing: Marcello Casal JR/ABr; Agência Brasil
"Molecules to Medicine" banner © Michelle Banks