In view of the prolongation of patients living with HIV's life expectancy, the question of optimization of ART, which is still a life-long treatment, becomes central. While most patients achieve virological success, their treatments often need to be optimized in order to limit adverse events, drugs interactions and to improve adherence. The switch to dual regimen strategies represent one of the approaches for treatment optimization.

Circulating HIV-1 resistant variants can be archived in viral reservoirs, where they can persist for an unknown duration and reemerge in case of therapeutic selective pressure.

There is a need to assess the dynamic of archived Drug resistance associated mutations (DRAMs) clearance in cell-associated HIV DNA after a long period of virological control, in the perspective of ARVs recycling.

The investigators postulate that it could be interesting in the future to recycle ARV drugs (that where classified as "resistant" in the past) in subsequent regimen. The question is particularly important for 3TC/FTC for subsequent new regimen and for the use of dual regimen (disappearance of M184V).

Thus, the investigators propose a retrospective, longitudinal analysis on blood-cell-associated HIV-1 DNA samples in order to investigate by Sanger and Ultra Deep Sequencing the dynamics of decay and persistence of DNA HIV-1 variants harboring key drug resistance-associated mutations to NRTIs, in particular M184V, in patients with sustained virological control for at least 5 years under effective ART.

Inclusion Criteria:

• HIV-1 infected
• Age ≥ 18 years
• Genotypic resistance test performed at time of failure and harboring at least M184V
• Fully suppressed HIV viral load for at least 5 or 10 years.
• Triple therapy or 2 drug regimen during the entire follow-up
• Availability of at least 1 stored whole blood sample /year

Exclusion Criteria:

• No genotypic resistance test available at time of failure.

The studied population will be composed of followed HIV infected patients of the center, of whom approximately 90% are under treatment and virologically suppressed, consecutively selected (in an anti-chronological order on the sample failure date) until gathering the requested number of patients meeting the eligibility criteria.