In-depth: Should SA invest in the latest HIV tests?
South Africa’s mass HIV testing programme has been a huge success over the last decade, so much so that the country has almost hit the global target of 95% of people with HIV knowing their status. This success has largely been built on the use of many millions of rapid HIV tests.
The UNAIDS targets for 2030 are that 95% of people living with the virus know their status, 95% of those who are diagnosed are on treatment, and 95% of those on treatment are virally suppressed. According to Minister of Health Dr Joe Phaahla, South Africa has reached 94, 78, 89.
Now, new fourth-generation rapid HIV tests and a new testing algorithm may be able to push the country even closer to the 100% mark – where effectively all people with HIV have been identified and diagnosed.
But there are questions about the performance and high cost of the new tests that have so far prevented their wider adoption in South Africa. Some experts nevertheless believe there may be a place for them in certain contexts and in particular groups. Newer tests could, for example, help diagnose recently infected people in who HIV is not yet detectable using the currently available tests. They might also help diagnose people who become infected while taking antiretroviral medicines to prevent HIV infection (called pre-exposure prophylaxis or PrEP).
“Rapid tests are the main tool used to identify people with HIV in South Africa,” says Professor Adrian Puren, who is the Executive Director of the National Institute of Communicable Diseases.
He says this is based on the fact that roughly 80% of the South African population are uninsured and therefore rely on the public health sector. Rapid HIV tests can be done in any location and do not require sophisticated laboratory infrastructure or the high-level expertise of a lab technician or pathologist.
They are designed to be used by most people, yet not just anyone can perform them. There is a degree of training involved but, once trained, most lay people can conduct rapid HIV tests accurately and safely. They are, therefore, cheaper than lab-based tests because they don’t require high-level equipment or expertise which is necessary in our public sector where, according to the National Department of Health’s Dr Thato Chidarikire, between 15 and 18 million screening tests are conducted each year. Chidarikire is the Director of HIV Prevention Programmes at the department.
Additionally, rural and outlying areas that are far from established laboratories are well-served by the rapid testing programme.
Rapid tests also have the advantage of delivering a quick result, usually within 30 minutes. This is an important factor in the public sector as people do not need to return to a health facility at a later date to receive their results and can be initiated on treatment immediately. Many people do not have the means for multiple trips to a health facility that may be far away from where they live.
A universal-test-and-treat strategy built around the use of rapid tests was implemented in South Africa in September 2016 “in hopes of attaining the UNAIDS 90-90-90 targets by 2020”, according to a 2021 paper.
“In the public sector, if a patient ends up in a hospital or if a patient has a suspected HIV infection that was not identified with rapid tests, they may be tested using the laboratory system,” says Puren.
Public sector laboratory testing is run by the National Health Laboratory Service.
Laboratory-based HIV tests used for diagnostic purposes are usually enzyme-linked immunosorbent assays (ELISA).
The private sector does not make routine use of rapid tests, according to Professor Eftyhia Vardas, who is a professor of medical virology at the University of Stellenbosch and the head of virology at Lancet Laboratories in Johannesburg.
While she mentions the advantages of rapid tests for the public sector context, she said that these are not used routinely in the private sector where the drawbacks are considered significant (and avoidable via lab-based testing) and where there aren’t the same cost constraints.
According to Chidarikire, her department pays between R7 and R8 for a rapid test. At the Lancet laboratories, the cost of two ELISA tests is between R280 and R300.
“There are variables with rapid point-of-care tests such as how frequently are the staff at clinics doing competency tests? We check competency every month to make sure our lab technicians haven’t forgotten anything. The competency of the person reading the test can have an impact on the end result,” she says.
Additionally, she says rapid tests have a very high sensitivity rate which can result in more false positives than lab-based tests.
HIV tests have to conform to guidelines on sensitivity and specificity. The sensitivity of a test refers to the percentage of results that will be accurately positive when HIV is present. High sensitivity means that the test is geared to never miss a positive case and is therefore more likely to cause false positives. Specificity refers to the percentage of results that will be accurately negative when HIV is not present. High specificity is more likely to cause false negatives.
World Health Organization guidelines require rapid tests to have minimum sensitivity rates of 99% and minimum specificity rates of 98%, but this does not eliminate all inaccuracies.
“Rapid tests have built-in high sensitivity because they are designed to not miss a case but this does, unfortunately, result in some false positive cases,” says Vardas.
Instead, at the Lancet laboratories, two ELISA tests are used when testing for HIV – a screening test that is followed by a confirmatory test in the case of a positive result.
However, Puren tells Spotlight that it is not necessarily the higher sensitivity of the rapid tests themselves that result in higher false positive rates, but rather that there are uncontrollable factors around rapid testing in the field when compared to the controlled environment of a laboratory.
According to Puren, there are many layers of protection in the current system ensuring that the tests are giving as accurate results as possible.
A two-test strategy, which is in line with WHO guidelines, is followed by both the public and the private sector, with the public using rapid tests most of the time.
The first test is called a screening test and, according to Puren, a rapid test with a high sensitivity rate is usually used with the aim of not missing a positive case. The second test is called a confirmatory test and is usually a product with higher specificity.
Despite this, false negatives and false positives do occur and are more likely with rapid tests. According to Vardas, there are cases of people being initiated onto HIV treatment who do not actually have HIV. She said this can take a “psychological toll” on affected individuals above the obvious impact of taking medication without needing to.
The country is preparing to move to a three-test strategy which is recommended by the WHO in areas with test positivity rates below 5%. Positivity rates are the percentage of tests done that result in a positive diagnosis. While Puren says we have high rates in some contexts (as high as 30%), there are places where positivity rates are closer to this 5% mark. This is a result of our maturing HIV programme – where most people who are currently getting tested are not infected with HIV.
Puren explains that the rate of false positives increases when the positivity rate falls, so a third confirmatory test is needed to ensure that misdiagnosed cases are kept down to a minimum.
Chidarikire tells Spotlight that the National Department of Health will implement a three-test strategy in the upcoming national tender for rapid tests. The current tender expires in June 2023.
Vardas says that about 10% of all HIV tests in the private sector are positive so there is no need for them to move to a three-test strategy.
But there are other factors in place to mitigate against the likelihood of misdiagnosis and improve test accuracy.
All HIV tests used in South Africa have WHO prequalification status meaning they have undergone a rigorous approval process by the organisation.
Furthermore, tests used in the country have to be licenced by the South African Health Products Regulatory Authority (SAHPRA). According to head of medical devices at SAHPRA Dr Dimakatso Mathibe, SAHPRA does not evaluate diagnostic tests themselves but uses a reliance pathway to award licences. This means they evaluate (rely on) the approval the device has received in six other jurisdictions including the United States Food and Drug Administration and the European Medicines Agency, “recognising the work they have done” vetting these products.
When awarding the national tenders for HIV tests, she says, Treasury and the National Department of Health do confirm with SAHPRA whether the product is fully licenced.
SAHPRA also has the power to recall any products that are problematic and individuals or groups are able to report possible issues with any locally-available HIV tests to the authority.
Training of testers
Research shows that while false positives and false negatives “should be anticipated” with rapid diagnostic tests, the risk is very low when using the correct testing strategies (such as the two-test strategy in high-prevalence areas like South Africa).
A 2017 systematic review of the literature published in the Journal of the International AIDS Society noted that most misdiagnoses with rapid tests occurred due to suboptimal testing strategies, poor management of supplies (the physical tests weren’t managed correctly), user errors – including difficulty interpreting weak reactive lines (faint lines appearing on test strips), and, finally, retesting people on HIV treatment. This means most misdiagnosed cases aren’t due to the actual quality of the tests themselves.
To prevent misdiagnosis due to issues with how tests are conducted, the department has an ongoing training programme. Chidarikire says that all nine provinces have Regional Training Centres and “implementers should be trained every two years”.
Third-generation versus fourth-generation tests
A big hurdle in all HIV programmes is the fact that many tests aren’t able to detect HIV immediately after infection. There is what is called a ‘window period’ from infection until a test can diagnose an HIV infection.
Older tests had a window period of about six months or more. Third-generation tests, which are the current rapid tests used in the public sector, test only for antibodies that the body produces in response to an infection. One reason for the window period is that it takes time for the body to produce these antibodies after being exposed to HIV.
Fourth-generation tests detect both antibodies and the p24 antigen. The antigen is part of the virus and is what induces the body to manufacture antibodies. The p24 antigen appears in the blood a few days before any antibodies can be detected. Therefore, fourth-generation tests shorten the window period slightly.
Both the private and public sector lab-based programmes already routinely use fourth-generation tests that detect both antibodies and antigens. According to the United States Centres for Disease Control and Prevention (CDC), a lab-based fourth-generation test can detect HIV from between 18 to 45 days after infection.
Vardas says HIV can sometimes be identified earlier – as soon as 14 days after infection.
Professor Francois Venter, Head of Ezintsha at the University of the Witwatersrand, tells Spotlight that the window period can be shortened to as little as ten days in some cases.
According to the CDC, rapid third-generation tests can detect HIV as soon as 23 days after infection but sometimes need as long as 90 days in order to detect the virus. Rapid fourth-generation tests can detect HIV as soon as 18 days after infection but may also take as long as 90 days in some circumstances.
Fourth-generation rapid tests a good option?
This difference may seem small – five days between rapid third and fourth-generation tests, however, as Venter points out, individuals who are acutely infected with HIV are also the most infectious. He says some research shows that up to 50% of all HIV transmission is from people who are acutely infected with HIV – in other words – who have been very recently infected with the virus.
“We need to consider that the benefit of finding these few high spreaders is of high net worth,” he says.
Furthermore, Puren points out that fourth-generation rapid tests could be useful in identifying breakthrough infections for people taking PrEP. PrEP, if taken as prescribed, is incredibly effective at preventing HIV infection but there are some very rare cases of breakthrough infection – infections acquired while on PrEP.
A 2020 case report in BioMed Central described a South African woman in a serodiscordant relationship (a relationship with a person living with HIV) who tested positive for HIV nine months after starting PrEP, despite taking the medication as prescribed. According to the study, retrospective testing found that HIV resistance was confirmed at month three – suggesting that resistance developed due to the continued use of PrEP after being infected.
Retrospective testing with a fourth-generation rapid test detected HIV at month one, suggesting resistance may have been prevented by the use of fourth-generation rapid testing.
Strains of HIV that have resistance to commonly used antiretroviral medicines are more difficult and expensive to treat.
But, according to Chidarikire, the national health department has no plans to introduce fourth-generation rapid tests into the public sector mass testing programme.
“At the moment, the reason why SA is not using fourth-generation tests is because of their stability and performance,” she says.
While third-generation rapid tests have higher than 99% sensitivity and specificity rates, available fourth-generation rapid tests don’t perform as well.
For example, the Alere Determine HIV-1/2 Combo, which is the only licenced fourth-generation rapid test in the country has a sensitivity rate of 95% according to research. Specificity was much higher and comparable to third-generation tests.
Additionally, as Chidarikire points out, “the price of fourth-generation tests can vary from R300 to over R700 or more for a pack of 20 or 25 single tests”. However, she says that the price should improve via bulk procurement if it was put on tender.
Currently, the WHO guidance does not recommend this option for our setting but this may change if the performance of tests is improved.
However, Venter believes these debates equate to “navel-gazing” and if there is a potential benefit to including fourth-generation tests in our programme, it should be done.
In the end, as Vardas said, “No test is an exact duplication of the truth. We strive to move the test result closer to the truth but we can never really guarantee that.”
NOTE: Professor Francois Venter is quoted in this article. Venter is a member of Spotlight’s Editorial Advisory Panel. The panel provides the Spotlight editors with advice and feedback on the quality and relevance of Spotlight’s public interest health journalism. The Spotlight editors, however, remain editorially independent and solely responsible for all editorial decisions. Read more on the role and purpose of the panel here.