Evaluation of HIV-1 p24 Antigenemia and Level of CD8 + CD38 + T cells as Surrogate Markers of HIV-1 RNA Viral Load in HIV-1YInfected Patients in Dakar, Senegal Pascale Ondoa, MD, PhD,* Tandakha Ndiaye Dieye,† Chris Vereecken,* Makhtar Camara,† Abdoul Aziz Diallo,† Katrien Fransen,* Amber Litzroth,* Souleymane Mboup,† and Luc Kestens* Summary: Alternative, affordable, and simple assays to monitor antiretroviral therapy (ART) in resource-poor settings are needed. We have evaluated and compared a heat-denatured (HD) HIV p24 amplified enzyme-linked immunosorbent assay from Perkin-Elmer and CD38 + CD8 + T-cell levels, determined by flow cytometry, for their capacity to predict viral load (VL) in HIV-1Yinfected patients from Senegal. Median fluorescence intensity (MFI) of CD38 expression on memory (CD45RO + ) CD8 + T cells correlated better with RNA VL than HD p24 antigenemia (R = 0.576, P G 0.0001 vs R = 0.548, P G 0.0001). MFI of CD38 expression on memory CD8 + T cells could predict detectable RNA VL (VL = 2.6 log 10 ) with a sensitivity of 87% and a specificity of 74%. A comparable sensitivity (89%) could be reached for HD p24 assay, but only to predict RNA VL of more than 5 logs, which might lead to unacceptable delays in clinical decision making. The clinical use of the HD p24 assay to monitor ART in Senegal would require more comparative data about the kinetics of p24 antigen and HIV RNA in peripheral blood as well as further evaluation regarding its sensitivity toward subtype A and CRF02. MFI of CD38 expression on memory CD8 + T cells appeared to be a better alternative to monitor ART in HIV-infected patients from Senegal. Key Words: HIV, alternative viral load, resource-poor settings, antiretroviral therapy, immune activation, p24 assay (J Acquir Immune Defic Syndr 2006;41:416 Y 424) E ffective highly active antiretroviral therapy (HAART) decreases the level of viral replication, improves the CD4 count, and reduces the mortality rate associated with HIV infection. 1 Recent accessibility of HAART in developing countries has focused the attention on the discrepancy between the availability of antiviral drugs and the lack of infrastructure for monitoring HIV infection in resource-poor settings. In standard practice, CD4 count and HIV RNA viral load (VL) are considered as the end-point parameters to monitor the efficacy of antiviral treatment. 2,3 The measure- ment of both CD4 count and RNA VL can cost up to US$100, although there is a trend toward significant price reductions. Because of resource constraints in developing countries, the World Health Organization recommends an absolute minimal monitoring of therapy on the basis of clinical evaluation of the patient. 4 Despite being the best indicator of the immu- nologic reconstitution, CD4 count is only considered as a desirable test, whereas HIV RNA VL was categorized as optional. Although these recommendations aim to facilitate the scaling up of antiretroviral treatment (ART) in resource- limited settings, they may actually result in a dramatic delay in the diagnosis of immunologic and/or virological drug failure. Such a situation could contribute to a global dimi- nution of ART efficiency as well as the emergence of a high level of drug resistance. Affordable alternative methods and instruments for the determination of CD4 count and VL have thus recently been developed but have not yet become largely implemented in resource-poor settings. Probably because VL is considered as a second priority for the follow-up of ART, alternative VL assays are presumably in an earlier stage of development than are the low-cost CD4 cell assays. As well, virological and immunologic parameters have been proposed as potential surrogate markers for monitoring of virus replication. The heat-denatured (HD) HIV p24 antigen (Ag) enzyme- linked immunosorbent assay (ELISA) is considered a good candidate as an alternative VL load test because it is already at an advanced stage of development. An improved amplified version of HD HIV-1 p24 Ag assay from Perkin-Elmer has recently been shown to be capable of predicting disease pro- gression, 5,6 monitoring ART, and detecting early virological drug failure. 7,8 Although the amplified HD p24 ELISA shows excellent clinical performance on subtype B viruses, studies conducted on nonYB subtypes indicated that the assay may underperform as far as the detection and quantification of certain HIV clades, prevalent in some African countries, are concerned. 9,10 It is therefore important that these deficiencies are carefully assessed and improved before the test is trans- ferred to this region. The level of CD38 expression on CD8 + T cells has been shown to have the strongest prognostic significance for dis- ease progression as compared with other markers of T-cell activation such as HLA-DR, CD25, CD69, CD70, and A 2 microglobulin. 11Y13 CD38 is a transmembrane glycol protein with enzymatic, adhesive, and cell signaling activity (reviewed in Ref 14 ). CD38 is expressed during hemopoiesis mostly on immature cells and is reexpressed during the early stage of memory T-cell activation. 14 Elevated levels of CD38 + CD8 + T cells have a negative prognostic value in HIV- infected adults. 15,16 CD38 expression on CD8 + T cells has BASIC SCIENCE 416 J Acquir Immune Defic Syndr & Volume 41, Number 4, April 1, 2006 Received for publication September 2005; accepted December 2005. From the *Institute of Tropical Medicine, Department of Microbiology, Antwerp-Belgium; †Immunology Unit, Laboratory of Bacteriology-Virology, Le Dantec Hospital, Cheikh Anta Diop University, Dakar, Senegal. Reprints: Dr Pascale Ondoa, MD, PhD, Institute of Tropical Medicine Nationalestraat 155-2000B Antwerp, Belgium (e-mail: Pondoa@itg.be). Copyright * 2006 by Lippincott Williams & Wilkins Copyr ight © Lippincott Williams & Wilkins. Unauthor iz ed reproduction of this article is prohibited.