In the first phase (phase A), ritonavir-boosted lopinavir (Kaletra®) will be compared with continued HAART. In the second year (phase B) patients on conventional HAART are also offered LPV/r monotherapy to extend the longterm experience of this new strategy.

Only patients willing to give a genital secretion and a spinal fluid sample will be included. All patients must be on a fully suppressive HAART with at least 2 consecutive values of HIV-RNA at the screening visit . After performance of lumbar puncture at baseline, patients will be randomized to continued HAART or LPV/r monotherapy. During the first year of randomized treatment patients will be followed at week 6/ 12 /18 /24 /32 /40 and 48. Lumbar puncture and genital secretion sampling will be repeated at week 48.

Follow up during the second phase (B: W48-96) of the study will be identical to phase A including genital and spinal sampling at week 96. After study termination at week 96, patients may opt to continue monotherapy if results of HIV-RNA in blood and CSF support this decision.

The primary endpoint of the study will be treatment failure in the compartment (CSF and / or genital tract). Since the variability of HIV-RNA determination in CSF and genital secretions is not very well known, a one log increase above the baseline value will be considered as treatment failure in the respective compartment. Only patients who had a complete viral suppression in blood will be considered for compartment evaluation. Patients treated in the monotherapy arm with a CSF HIV-RNA value at week 48 more than 1.0 log10 cp/ml above baseline (= compartment failure) will be switched to a conventional combination treatment. HIV-RNA testing in the genital samples will be performed batchwise at the end of the study.

In addition, patients with a blood treatment failure (two consecutive HIV-RNA detections > 400cp/ml) will be considered as full treatment failures and switched to a rescue regimen at the descretion of the treating physician. For the analysis, these patients will be considered as systemic treatment failure and will not be entered in the analysis of compartmentalized treatment failure. If the rescue strategy was only intensification of adherence and results in full blood viral load re-supression, the patient will still be maintained in the study and compartment evaluations can be performed at w48 and/or 96, respectively.

The secondary aim of the study is the definition of prognostic markers for compartement failures. Potential risk factors associated with mono-maintenance failure are HIV-DNA load at time of treatment simplification, HIV-RNA at the time of first treatment initiation, duration of HIV-RNA suppression before simplification, history of HIV-RNA blips, presence of detectable HIV-RNA in spinal fluid at the time of treatment simplification, changes of level of c-reactive protein (high sensitive methodology, hsCRP) from baseline as a marker of immune-activation during the maintenance therapy.

If funding allows, we will test for the presence of resistant viruses and compare the presence of genetic polymorphism at baseline. We will also measure parameters of immunactivation (hsCRP, CD8+, CD38+).

The study is financed by the Swiss National Science Foundation and the Swiss HIV Cohort Study.