Over the last four years our studies have shown that cycles of brief interruptions of Highly Active Anti-Retroviral Therapy (HAART) are not associated with adverse events, lack of resuppression upon reinitiation of therapy, loss of recall responses or lack of restoration of CD4 levels to pre-interruption levels upon achieving viral suppression following reinitiation of therapy. The long-range goal of this proposal is to determine whether, in suppressed patients, intermittent interruptions of HAART result in maintenance of immune parameters (e.g. CD4 counts, recall responses etc.) comparable to continuous HAART, while reducing overall long-term toxicity and cost. Specifically, we propose to test the hypothesis that repeated cycles of 2-8 weeks off HAART followed by 16 weeks on therapy (leading to a maintenance strategy decreasing drug exposure by 33%) is not inferior to continuous therapy over the same period, with non-inferiority defined by the sustained cellular and humoral immune response to a de novo antigen. Functional end-point of retained immune reconstitution will be evaluated in conjunction with viral suppression to <400 copies/ml and retention of CD4 cell count above baseline at the final observation when both intermittent and continuous study arms are on therapy. Additionally, we hypothesize that the cyclic therapy intervention will result in a significantly lower therapy-related toxicity while maintaining CD4 Tcell counts at levels significantly higher than pre-therapy levels during period off and on treatment. We will test these hypotheses in treatment-naive subjects with 200-350 CD4 cells/mm3 who successfully achieve viral suppression to <50 copies/ml during a 24 week """"""""run-in"""""""" period on lopinavir/ritonavir, lamivudine, stavudine to include a complete vaccination series against rabies from week 16 to 22 (de novo antigen) before randomization into study arms in a single-center, randomized, two-arm non-blinded study (n=74). We will monitor immune reconstitution and adherence to therapy and determine changes in the immune status of patients following HAART interruption. Therapy-induced toxicities will be monitored by assessing fat distribution, glucose/insulin metabolism, mineral bone density and liver, kidney and pancreatic function tests. We will also monitor viral resistance outcomes by determining genotypic changes in the HIV-1 protease and reverse transcriptase regions over time. In addition to addressing the needs of South Africa in relation to development of sustainable and affordable treatment strategies, this study advances our understanding of immune reconstitution in clade C HIV-1 infected subjects and of treatment interruption as a strategy to decrease drug toxicity in therapy-naive chronically infected persons. This hypothesis-driven proposal represents an international multidisciplinary research effort by the Wistar Institute, the Clinical HIV Research Unit and Departments of Haematology, Chemical Pathology, Medicine (Endocrinology Division) at the University of the Witwatersrand, the AIDS Virus Research Unit at the National Institute for Communicable Diseases in Johannesburg, and the University of Pennsylvania's Center for Clinical Epidemiology and Biostatistics.