BACKGROUNDThe genomic heterogeneity of HIV-1 impedes the ability of consensus sequences in vaccines to elicit effective antiviral immune responses. AGS-004 amplifies translation-competent RNA molecules encoding for Gag, Rev, Vpr, and Nef from the patient's autologous virus and loads them into dendritic cells.METHODSThis phase IIB, multicenter, 2:1 randomized, double-blind, placebo-controlled study enrolled 54 HIV-1-infected patients on antiretroviral therapy with viral loads (VLs) <50 copies per milliliter, current CD4 T-cell counts >450 cells per cubic millimeter, and nadir counts >200 cells per cubic millimeter, to receive intradermal injections of study product into the axillary lymph node region every 4 weeks. At week 16, a 12-week analytical treatment interruption (ATI) was undertaken.RESULTSThere was no difference in the end-of-ATI VL (average of values from weeks 11 and 12) between the 2 arms of the study [4.39 (4.17, 4.69) vs. 4.47 (3.76, 4.64) log10 HIV-1 RNA; P = 0.73]. Between arms, no change between pre-antiretroviral therapy VL and the end-of-ATI VL [-0.06 (0.24, -0.32) vs. -0.17 (0.17, -0.32) log10 HIV-1 RNA; P = 0.43] was observed. When interferon-γ, interleukin-2, tumor necrosis factor α, CD107a, and granzyme b expressions were measured by multicolor flow cytometry, a greater percentage of AGS-004 than of placebo recipients had multifunctional cytotoxic T-lymphocyte responses induced in the CD28+/CD45RA-CD8 effector/memory T-cell population to dendritic cells electroporated with autologous antigens. Adverse events consisted of transient, mild (grade 1) local injection site reactions.CONCLUSIONSDespite the induction of HIV-specific effector/memory CD8 T-cell responses, no antiviral effect was seen after the administration of AGS-004 when compared with placebo.