Many HIV-infected individuals on effective antiretroviral therapy with suppressed HIV RNA levels in blood as well as in cerebrospinal fluid have signs of low-level intrathecal inflammation. Our aim is to study if changing the nucleoside backbone in an antiretroviral regimen can decrease the residual intrathecal immune activation.

HIV infects the central nervous system (CNS) very early in the infectious course and establishes a chronic infection with markedly increased intrathecal immune activation. Modern antiretroviral treatment (ART) is effective not only to supress HIV in the blood and lymph system, but also in the CNS. After commencement of ART, markers of intrathecal immune activation such as cerebrospinal fluid (CSF) neopterin decreases markedly, but remains slightly increased in a substantial number of patients despite several years of receiving ART. CSF markers of immune activation correlates to markers of neuronal injury and the axonal marker neurofilament light protein (NFL) has been found to be slightly increased in asymptomatic patients compared to HIV-negative controls despite effective ART. This indicates a linkage between inflammation and neuronal injury, however, the long-term effect of low-grade immune activation within the CNS is not known.

CSF neopterin correlates also to levels of CSF HIV RNA in patients on ART when measured with ultrasensitive methods in the range 0.3-20 copies/mL highlighting the potential for the CNS to serve as a viral reservoir and for persistent infection to cause subclinical CNS injury.

There are several indications that abacavir could act pro-inflammatory, a factor that might influence the proposed increased risk for coronary heart disease during abacavir treatment.

Tenofovir alafenamide fumarate (TAF) is a new prodrug for tenofovir and have some interesting differences compared to the currently used prodrug, tenofovir disoproxil fumarate (TDF). Both TDF and TAF require conversion to the active drug tenofovir diphosphate (DP). TDF is converted initially to tenofovir in the blood, then tenofovir is taken up into cells. In lymphocytes, macrophages, and some other cells, it is phosphorylated to the DP form. TAF, however, is largely delivered as TAF to lymphocytes and macrophages, then metabolized intracellularly to tenofovir and phosphorylated to the active tenofovir DP. Thus, plasma levels of tenofovir are much lower with TAF than with TDF, and tenofovir DP levels are much higher within lymphocytes and macrophages. This is of large interest since macrophages and microglia (a tissue macrophage) is the main target cells for HIV in the CNS.

A longitudinal study on HIV in CNS has been ongoing since more than 30 years in Gothenburg, Sweden. Subjects are followed with clinical follow up and annual lumbar punctures (LP), both before and during ART. Blood sampling is performed in parallel. The study protocol is approved by the Regional Ethical Review Board in Gothenburg and patients are included after written informed consent.

This is a pilot study and if we find interesting changes after treatment switch the idea is to continue with a larger study with design dependent on results and with enough power to demonstrate differences. However, the results from this study will be presented in a scientific publication irrespective of the outcome.

1. No change in CSF viral load or CSF immune activation after switch.

   a. This outcome reassure a high effect in CNS also by TAF

2. Decrease in CSF immune activation after switch from abacavir (ABC)/lamivudine (3TC)

   a. This would be a very interesting finding that in such case needs to be tested in a larger setting.

3. Decrease in CSF viral load and/or CSF immune activation after switch in both groups

   a. This is an unlikely outcome but would imply that increased intracellular concentrations of tenofovir DP in macrophages and microglia could decrease an ongoing low-grade replication in the brain

4. Increase in CSF immune activation after switch in one or two groups a. This is a highly unlikely outcome

Inclusion Criteria:

• Stable ART with TDF/FTC or ABC/3TC + 3rd agent (PI/r, NNRTI, or INI) >18 months
• Plasma HIV RNA < 50 copies/mL >12 months
• Lumbar puncture performed in the longitudinal CSF study (University of Gothenburg) between 6 and 18 months before study entry

Exclusion Criteria:

• Neurological symptoms or severe neurocognitive impairment
• Pre-treatment CNS opportunistic infection/ tumour or HIV-associated dementia
• Pregnancy