Indirect effects on immune activation may partially account for lingering benefits of failed antiretroviral therapy in the brain

The intriguing finding that anti-HIV therapy continues to provide protective benefit in the brains of HIV-positive people who have failed therapy may be explained by an indirect effect of therapy on immune activation, say researchers in a report in the April 15^th issue of Journal of Acquired Immune Deficiency Syndromes.

What’s more, they suggest that this newly characterised effect on immune activation may help to explain why antiretroviral therapy often performs better than expected in reducing the amount of HIV in the brain.

A 2006 study by Dr Richard Price, University of California, San Francisco, reported that among 123 people with HIV who were failing treatment, the HIV viral load in the cerebrospinal fluid (CSF) was about 10-fold lower than that in plasma. In the current follow-up study, Price and colleagues tested the hypothesis that this difference in therapeutic effect is due to the reduced immune activation often associated with the replication of drug-resistant virus.

In the cross-sectional study, the researchers placed the participants into four groups: 53 people who had been off treatment for at least three months (‘offs’), 30 people who were on treatment but with a viral load above 500 copies/ml (‘failures’), 40 people who were on treatment and with a viral load below 500 copies/ml (‘successes’) and 14 HIV-negative controls.

For each group, the researchers measured viral load in blood plasma and CSF. They also measured the level of activation of CD8 cells and CD4 cells in both compartments, as well as two markers of CSF immune response, white blood cell counts and neopterin levels.

Median blood CD8 cell activation was highest in offs (47.1%) and decreased in failures (34.2%), successes (21.5%) and HIV-negative controls (9.2%).

Overall, CD8 cell activation in CSF was strongly correlated with that in the blood, with median levels of CD8 cell activation being somewhat higher in the CSF samples: 61.3% for offs, 39.1% for failures, 26.7% for successes and 23.9% for controls. CD4 cell activation in the different groups was similar, but not as clearly correlated as CD8 cell activation. CD8 cell activation also correlated with the two other markers of CSF immune response, white blood cell counts and neopterin.

In their previous work, the researchers had found that plasma viral load was not significantly different between failures and offs, but CSF viral load was significantly higher in offs than failures (1.65 log₁₀ difference in medians).

Noting that this difference in CSF viral load between the failures and offs coincided with a difference in CD8 cell activation levels, researchers then evaluated the interaction between three factors: 1) viral load in plasma and CSF, 2) CD8 cell activation in blood and CSF and 3) status as failure or off.

Across the range of plasma viral loads, blood and CSF CD8 cell activation were significantly lower in failures than in offs. This suggests that CD8 cell activation differed depending on the type of virus: wildtype virus (off group) led to high activation while resistant virus (failure group) led to muted activation in the blood and CSF.

Across the range of CSF viral load, blood and CSF CD8 cell activation were not different between the failure group and the off group. From this the authors suggest that CD8 activation in the CSF is the same regardless of whether the virus is wildtype (off group) or resistant (failure group).

Further statistical analysis revealed that plasma viral load, blood CD8 cell activation and neopterin levels predicted CSF viral load across the three groups (off, failure and success). CSF CD8 cell activation, which was significant in univariate analysis, lost significance after blood CD8 cell activation was included.

The researchers assert that these data support a model of HIV replication (and antiretroviral suppression) in the blood and CSF that includes both direct and indirect effects. Briefly, HIV infection in the blood increases plasma viral load and activates blood immune cells. These activated blood immune cells travel through the blood–CSF barrier, populate the CSF and become a primed target for HIV replication in the CSF. In fact, the researchers propose that the bulk of virus detected in the CSF is due to these systemically activated immune cells. (The CSF can also be populated by immune cells derived from long-standing sources in the brain.)

Antiretroviral treatment decreases plasma viral load and CSF viral load, though the direct effect on the latter may be weakened due to the blood–CSF barrier. Treatment also dampens activation of immune cells in the blood. These dampened cells cross into the CSF, but due to their less activated state will lead to lower HIV amplification in the CSF and thus a lower CSF viral load.

In cases of treatment failure due to resistance, plasma viral load will rise due to resumed HIV amplification in the blood. However, due to some unknown mechanism, blood immune cell activation is muted. When these muted immune cells cross the blood–CSF barrier, they provide poor targets for HIV and so lead to reduced amplification in the CSF and a lower than expected CSF viral load.

As a final remark, the researchers write, “these studies suggest that the level of systemic immune activation is an important modulator of this infection and that its downregulation by ART [antiretroviral therapy] may contribute to controlling HIV-1 in this compartment and explain the better-than-predicted responses of CSF HIV-1 to ART.” While not conclusive, these findings by Price and colleagues add to the mounting evidence that that brain-penetrating antiretrovirals are not always necessary for control of HIV in the CSF.