Acute HIV-1 infection in sub-Saharan Africa

HIV-1 transmission rates peak in the first months of infection [1], when viral burden [2] and per-virion infectivity [3] are high. Although the precise contribution of early-stage transmission to overall HIV incidence varies across settings according to local transmission dynamics, numerous modeling [4,5] and phylogenetic [6–10] studies indicate that this early period plays a disproportionate role in the spread of HIV. Detection of acute HIV infection (AHI), comprising the first weeks of infection prior to antibody development [11], has attracted particular attention as an opportunity to avert further transmission and improve both individual and public health [12]. Treatment of AHI with antiretroviral agents can stop the onward spread of infection [12], reduce the viral reservoir [13], potentially restrict viral diversification [14], and increase chances for viral control or even cure of the infection [15]. However, opportunities to address AHI are routinely lost because identification of acute infection is challenging [11]. In almost all settings, HIV testing relies on the detection of antibodies to HIV, which will (by definition) be absent during AHI. Detection of HIV RNA to identify AHI has proven reliable in research studies in both the United States and Africa, with 4–10% of newly diagnosed HIV cases falling within the AHI period [16–20]. Unfortunately, methods based on HIV RNA detection are laboratory-based and are generally too expensive for routine use in all persons being tested for HIV. Fourth-generation, antigen-antibody combination tests represent a promising alternative to RNA-based testing [21]; however, the combination test remains laboratory-based. A reliable point-of-care test for AHI has been difficult to develop [22,23]. These challenges are not insurmountable. Importantly, the clinical manifestations of AHI provide clues that can guide the targeted application of laboratory-based AHI tests. Most newly infected persons will experience an ‘acute retroviral syndrome’ within 2 weeks of HIV acquisition [24,25], and many acutely infected persons will seek care during this period [26–28]. Thus, if acutely infected persons or their clinicians recognize the risk of recent HIV acquisition and/or the presence of symptoms consistent with AHI, targeted AHI testing can be performed. Algorithms based on combinations of symptoms and other factors (such as risk behaviors, antibody test results, or sexually transmitted infection symptoms) have been used in several ‘high-risk’ settings to detect people with AHI [29,30]. Importantly, because different rapid HIV tests have different abilities to detect the first anti-HIV antibodies formed, people with AHI often have discordant results if the appropriate rapid tests are employed [16]. Although the optimal combination of predictors to include in AHI screening algorithms – as well as the prevalence of AHI and the ability of such algorithms to discern it – will vary across settings, targeted approaches can guide the efficient application of AHI screening tests in those most likely to be acutely infected. In this issue of AIDS, Sanders et al.[31] describe the application of such an algorithm among people seeking healthcare at primary care centers in Kilifi, Kenya. Their algorithm, modified from a strategy we developed to detect AHI in an STD clinic in Lilongwe, Malawi [29], assigned points to each of the following AHI indicators: multiple partners in the last 2 months (1 point), generalized body pains (1 point), fever (2 points), diarrhea (2 points), and sexually transmitted infection symptoms (2 points). Confirmatory diagnostic screening, using p24 antigen testing and either pooled HIV RNA testing or follow-up antibody tests, was limited to seronegative or serodiscordant persons under the age of 30 who had a risk score of at least 2. Among persons meeting these AHI screening criteria, the prevalence of AHI was 1.0% (95% confidence interval, CI 0.3–2.2%). When restricted to the febrile subset of patients meeting these criteria, the prevalence of AHI was identical to that of malaria: 1.7% (95% CI 0.5–4.2%). In an earlier study in Uganda, 1.1 and 2.1% of HIV-seronegative individuals with suspected malaria were found to have acute and early HIV infection, respectively [32]. And in Mozambique, 3.3% of HIV-seronegative persons reporting with fever to a district hospital were found to have AHI [33]. Because HIV is pandemic in sub-Saharan Africa, it should be no surprise that AHI is common, and in some places as common as malaria. However, because the symptoms are nonspecific and similar to those of other infections such as malaria and other tropical diseases [34] so prevalent in many HIV-endemic settings, AHI is often not considered. Given the possibility for improving clinical and public health outcomes, it is important that clinicians consider AHI as a potential cause of febrile illnesses. At the very least, detection of AHI can allow linkage to HIV care and trigger important transmission prevention activities, such as risk-reduction counseling and partner notification [35]. Given the high prevalence of late HIV diagnosis in many settings [36], such opportunities may be afforded years earlier than is the case when AHI is overlooked. But detection of AHI will have an even greater impact when firm guidelines for clinical management are put in place worldwide. Immediate antiretroviral treatment is already part of US DHHS and IAS-USA Guidelines [37,38], reflecting the individual benefits of therapy. In a recent study from San Diego, treatment initiation within 4 months of the estimated date of infection was much more likely to result in near-normal CD4+ cell counts than was treatment initiation more than 4 months after infection [39]. From a transmission standpoint, current ‘treatment as prevention’ strategies hold tremendous promise, but they cannot avert transmissions during AHI if AHI is undetected. Indeed, in the absence of diagnosis and treatment strategies for AHI, the percentage of transmissions resulting from this stage of infection will likely grow with increasing treatment coverage of established infection. The addition of strategies that will address transmission during AHI is likely to maximize reductions in HIV incidence [4]. Despite these clinical and preventive benefits of treatment during AHI, the most recent WHO treatment guidelines do not address this topic [40]. AHI is routinely missed in healthcare encounters, despite the presence of clinical, behavioral, and serological clues. The consequences of this problem have been recognized for well over a decade [41]. We believe redoubled efforts to increase awareness of AHI among at-risk populations and their clinicians are warranted. Development of AHI detection strategies, coupled with parallel changes in clinical policy to recommend antiretroviral treatment and robust prevention strategies, offers important medical and public health benefits that should not be overlooked. Acknowledgements The authors acknowledge funding from the National Institutes of Health (KL2 TR001109, UL1 TR001111, R37 DK04938), and the UNC Center for AIDS Research (P30 AI50410). They thank Drs Joseph Eron, Jr. and William Miller for helpful comments. Conflicts of interest M.S.C. has served as a consultant to the Janssen Pharmaceutical Company and Roche Molecular Systems.