IAS: Two studies suggest that it may be possible to screen for acute primary HIV infections in sub-Saharan Africa

Theo Smart
Published: 28 July 2005

Two studies presented on Monday at the Third IAS Conference on HIV Pathogenesis and Treatment in Rio de Janeiro, Brazil reported that a significant number of acute HIV infections, which are undetectable using standard HIV antibody-based tests, can be recognized using other screening methods - even in resource-limited settings in sub-Saharan Africa.

Both teams suggest that early detection could be particularly important for prevention interventions, although further study needs to be done to demonstrate this idea, and to find ways of scaling up (and reducing the cost) of screening for acute infections.


Even though many Africans ill with acute HIV infection undoubtedly go to clinics looking for care to relieve their symptoms, the condition is rarely diagnosed, largely because no standardized simple testing methodology exists. During acute HIV infection, the patient has no detectable antibodies to the virus, (which are what the ELISA HIV tests - used for standard HIV testing - look for); and the symptoms of acute HIV infection are generally hard to distinguish from other acute infections such as the flu or other infections that may be prevalent in the community.

However, there are a variety of reasons why it might be useful to be able to detect these early infection sooner than the ELISA antibody tests permit:

  • During acute infection, viral load is elevated and neutralizing antibodies are absent, which greatly increases the risk of HIV transmission. In fact, this phase may be the most infectious stage of the disease.
  • Identifying primary HIV infection thus has profound implications for both prevention and clinical management.
  • The early infection period is of interest for studies of HIV pathogenesis (understanding how the infection causes disease). Early detection is also critical as an endpoint in clinical trials of prevention interventions.
  • Accurate acute infection diagnosis could also be used to more reliably estimate trends HIV incidence from cross-sectional seroprevalence surveys.

Two studies presented at the Rio conference tacked the problem of diagnosing acute HIV infections in different ways.

Pooled HIV RNA testing in South Africa

Dr. Wendy Stevens and colleagues from the University of Witwatersrand (Wits) conducted a cross sectional study assessing the prevalence of antibody-negative acute HIV infection among 1906 South African men and women visiting a primary health care clinic (the Esselen Street Clinic in Johannesburg) for treatment of their sexually transmitted infections or to learn their HIV status.

All the samples were first screened for HIV with the ELISA test - a total of 672 or 35.2% of tests came back positive.

The study then tested the antibody-negative or indeterminate samples with the HIV RNA test (the Roche MONITOR HIV-1 version 1.5 assay) that can detect HIV genetic material in a sample of blood a couple weeks before the ELISA tests can detect HIV antibodies. But rather than run separate HIV RNA tests on each of the patients - which would have been prohibitively expensive - the Wits team ran tests on pooled samples of blood.

In other words, samples were stored until there were at least one hundred specimens to pool. This pool was then amplified for evidence of HIV RNA. If the pool tested negative, then all the specimens were deemed to be negative. However, if the pool tested positive, progressively smaller pools (of 50, then of 10) were tested until each of the HIV RNA positive specimens were isolated.

In this case, 1200 HIV negative samples were pooled, three of the pools of 100 specimens were positive for HIV-1 RNA, then three were positive in the pools of 50, and seven pools of ten were positive. Individual sample testing revealed eight HIV RNA samples positive. Four of eleven indeterminate HIV ELISA samples were also tests and found to be positive for HIV RNA. Thus, the total number of acute HIV infections was twelve.

The prevalence of acute infections was 0.6% of all the patients in the study, and 0.99% in the antibody-negative patients screened.

If this figure is used to calculate the incidence rate per year (using a 28 day window between when PCR testing and the ELISA can detect HIV) the incidence rate per year of acute HIV was 12.9%. However, researchers have revised that calculation because the window between PCR and ELISA HIV detection is much shorter with the current generation of ELISA than with previous tests (around twelve days). If that shorter window period is used, the incidence rate per year in this cohort was 30.1%.

Dr. Stevens concluded that pooled HIV RNA amplification testing strategies was practical for high risk populations in the primary care setting in South Africa. “These results suggest that the impact of HIV testing programs incorporating acute HIV testing could be proportionately greater in the South African primary care setting than in most developed world settings. This maybe a result of acute infections in our population being driven by factors associated with [actual] HIV acquisition [rather than just] symptomatic sexually transmitted, or anxiety over risky sex.”

But while these methods may hold promise for surveillance purposes, they seem unlikely to improve individual patient management or to be of much use in prevention interventions targeting highly infectious patients with acute HIV- infection, because by the time enough samples are pooled to be tested, most patients experiencing acute HIV-infection will have seroconverted (become detectable by ELISA). Furthermore, even with pooled testing strategies, HIV RNA testing is inaccessible or still cost prohibitive in many sub-Saharan African settings.

p24 antigen testing in Rwanda and Zambia

Another approach is p24 antigen testing, which can be performed at more local regional laboratories. p24 antigen (p24A) is an HIV core protein which is detectable early in the course of infection before the development of antibodies to the virus. However, once antibodies do form, they bind to the antigen and make it difficult to detect.

Dr. Susan Allen of the Rwanda/Zambia HIV Research Group and Emory University in Atlanta used p24 (in addition to rapid ELISA tests) to increase the identification of new infections in a cohort of HIV discordant couples (when one partner is HIV and the other is not). The HIV-negative partner in a discordant couple is at significantly higher risk of seroconversion than the general public - particularly if the couple still engages in unprotected sexual intercourse.

Dr. Allen’s study enrolled 2400 partners from Rwanda and Zambia though a couples’ VCT center and performed repeat HIV tests and p24A tests at three-month intervals. Dr. Allen presented data from the Zambian couples.

A total of 115 seroconverters were identified in 30 months of the study (seven to eight per 100 patient years). Of these, 26 (23%) were positive for p24A. Eight of these were antibody positive with two rapid tests at the time p24A was detected, the remaining 18 were antibody negative. All 18 were antibody positive when they returned for repeat testing and sample collection, p24A quickly became undetectable.

Said Dr. Allen, “in a cohort with a seroconversion rate of approximately 2% per three-month interval, one quarter of new infections can be identified during the early, p24A+ phase.”

However, in the context of this study, the samples for p24A screening were also batched for weekly testing (and only evaluated - for the study’s purposes - at three monthly intervals for each couple). To detect acute HIV-infection sooner, she said, it might be necessary to set up “an on call system, 24/7, for home visits, collection and processing of critical samples.” Which she added might be “feasible” - at least for study purposes.

While these screening methods may not be ready for wide-scale implementation, the work provides grounds for furthering research into testing methodologies and prevention strategies aimed at early detection and prevention of HIV transmission.


Stevens W et al. High prevalence of undetected, acute HIV infection in a South African primary care clinic. Third International AIDS Society Conference on HIV Pathogenesis and Treatment, Rio de Janeiro, abstract MoOa0108, 2005.

Allen S. Early detection of HIV infection in discordant heterosexual couples in Africa. Third International AIDS Society Conference on HIV Pathogenesis and Treatment, Rio de Janeiro, abstract MoOa0107, 2005.

Community Consensus Statement on Access to HIV Treatment and its Use for Prevention

Together, we can make it happen

We can end HIV soon if people have equal access to HIV drugs as treatment and as PrEP, and have free choice over whether to take them.

Launched today, the Community Consensus Statement is a basic set of principles aimed at making sure that happens.

The Community Consensus Statement is a joint initiative of AVAC, EATG, MSMGF, GNP+, HIV i-Base, the International HIV/AIDS Alliance, ITPC and NAM/aidsmap

This content was checked for accuracy at the time it was written. It may have been superseded by more recent developments. NAM recommends checking whether this is the most current information when making decisions that may affect your health.

NAM’s information is intended to support, rather than replace, consultation with a healthcare professional. Talk to your doctor or another member of your healthcare team for advice tailored to your situation.