An interesting case study was reported as a poster at CROI 2015, of a UK patient whose viral load has remained undetectable for more than ten years since stopping ART in 2004. This was despite viral failure on earlier ART and persistence of a detectable viral reservoir. 
This case is of a 23 year old African woman who was diagnosed in 1997 with subtype C, following three weeks of acute and severe seroconversion symptoms, during which her CD4 count was confirmed <200 cells/mm3 and viral load was >750,000 copies/mL.
Initial treatment was with AZT/3TC/indinavir, switched after two weeks to AZT/3TC/ritonavir (600mg BD) but viral load remained detectable until treatment was changed again (at 94,000 copies/mL) in April 1999. Viral load then became suppressed and remained undetectable until January 2004 when treatment was stopped.
HIV viral load has remained consistently undetectable off-ART since 2004.
In 2014, a detectable viral reservoir was confirmed with total HIV-1 DNA, integrated HIV-1 DNA and 2-LTR circles at 148.93 (95% CI: 76.99 to 229.64), 134.31 (95% CI: 56.47 to 304.39) and 3.89 (95% CI: 0 to 9.15) HIV-1 copies/million PBMCs, respectively.
The extended period of viral control has not been explained by HLA genotype. Moderately potent CD8 T cell responses were similar to clade-matched responses similar to that seen in treatment-naive patients whose viral set-point is <10,000 copies/mL.
However, unusually broad Gag-specific IFN-gamma CD4 responses were detected that targeted multiple regions of Gag that are associated with viral control.
This interesting case study highlights how much we still have to learn about the HIV reservoir and immune control of HIV replication. It is important that these and similar cases are presented and published to connect collaborative research.
Do individuals such as this accumulate functional HIV-specific CD4 memory responses that are able to mediate control (rather than succumb to infection). If so, is this due to a bias to produce chemokines that make the CD4 T cells relatively resistant to HIV?  This phenomenon has been reported in CMV-specific CD4 T cell responses. 
Are these CD4 cells biased toward a cytotoxic phenotype which has been associated with control?  Or are they expressing restriction factors? Or are NK cells involved, with the CD4 responses more of an effect than a cause? Lots of questions…
Finding out whether immunological factors contribute to or determine viral control in such single cases involves detailed, longitudinal assessments. In this case, broadly cross-neutralising antibodies in plasma could be assessed at multiple time points, in addition to the T cell responses already described.
However, the neutralising response against heterologuous virus was not particularly strong when tested a few years ago.
While the authors report the detection of pro-viral DNA, RNA and 2-LTR circles, it would be surprising if at least some of these were not present in a person with a history of high viraemia. The continued existence or not, of a replication-competent reservoir could be demonstrated with viral outgrowth assays. This might explain whether persistent aviraemia is a feature of a defective reservoir or ongoing effector immune responses, or both.
For HIV positive people, these examples are hopeful, although their rarity emphasises that this is not yet something to try at home.
- Kinloch SI et al. Aviremia 10-year post-ART discontinuation initiated at seroconversion. 2015 Conference on Retroviruses and Opportunistic Infections (CROI 2015), 23-26 February 2015, Seattle. Poster abstract 377.
- Walker WE et al. Increased levels of MIPs result in resistance to R5-tropic HIV-1 in a subset of elite controllers. J. Virology, posted online 4 March 2015, doi: 10.1128/JVI.00118-15.
- Casazza JP et al. Autocrine production of Beta-chemokines protects CMV-specific CD4+ T cells from HIV infection. PLoS Pathogens, 30 October 2009. DOI: 10.1371/journal.ppat.1000646.
- Soghoian DZ et al. HIV-specific cytolytic CD4 T cell responses during acute HIV infection predict disease outcome. Sci Transl Med. 2012 Feb 29;4(123):123ra25. doi: 10.1126/scitranslmed.3003165.