Mechanisms for HIV Persistence on ART
This report is part of a series of focused summaries from the “5th International Workshop on HIV Persistence, Reservoirs & Eradication Strategies” held in St Maarten, December 6-9, 2011. Several speakers addressed the mechanisms involved in HIV persistence, which thwart an HIV cure. This report concerns presentations given by Javier Martinez-Picado (Barcelona), Steve Deeks (San Francisco) and Tae-Wook Chun (Bethesda).
Javier Martinez-Picado reviewed evidence for ongoing HIV replication during ART. He first mentioned that we know that low-level plasma viremia (< 50 copies) persists for more than 7-years of therapy and that persistent immune activation parallels residual viremia.
There are different -but non-mutually exclusive- hypotheses to explain why HIV persists despite ART:
- Residual replication in cells located in sanctuary sites where drug levels are suboptimal,
- Long-lived HIV-infected cells that produce virus,
-Proliferation of latently infected cells with regeneration of a stable reservoir of slowly dividing infected cells.
In the last two cases, treatment intensification will be of little help.
He then presented data from his widely cited Nature Medicine paper (Buzon et al.) based on the “IntergRAL” study. A total of 69 patients on suppressive HAART for at least 1 year were randomized to intensify their ART regimen with RAL or continue on their previous ART for 48 weeks.
There were not significant baseline difference between the two groups with a median of 500 CD4 T cells counts, and 5 years of suppressive ART.
Raltegravir did not additionally impacted the ultrasensitive plasma viremia (SCA) at weeks 12 and 48 when compared with the baseline.
They also looked into the different HIV cDNA intermediates. They specifically quantified total HIV-1 DNA in the patients’ PBMCs and found no reduction over the 48 week study period in the group of patients whose treatment had been intensified with RAL, whereas in the control group they observed an slight increase in the total HIV-1 DNA towards the last time points of the study that they could not fully explain.
However, in a minority of patients (13 of 45) they observed an early, transient but significant increase in the quantification of 2LTR circles after the addition of RAL, a difference that was not observed in the control group. They think that the accumulation of these 2LTR circles reflects recent events of infection which integration in the chromosomal DNA has been aborted by the presence of RAL.
The accumulation of 2LTR circles in the 13 individuals followed a similar dynamics, with two or more positive time-points in most of the patients and a decrease in all but one patient towards week 48.
They then attempted to model the accumulation of 2LTR circles. This has been performed by the groups of Ryan Zwrakowsky in the University of Delaware and Hulin Wu at the University of Rochester.
If the level of ongoing active compartment replication were low (R0=0.7), they would expect to see monotonic increase in 2-LTR count following RAL intensification.
Instead, they observed a rapid increase followed by a rapid decrease, consistent with a high level of active compartment replication.
Model fitting to the 13 non-zero patients from their data gives a range of 102-106 for the ratio of active compartment replication to quiescent cell activation.
This is consistent with the hypothesis of ongoing subcritical replication in sanctuary sites in these 13 patients (figure 1).
They also extensively analyzed longitudinal PBMC samples by deep sequencing of episomal and integrated HIV-1 DNA from patients undergoing raltegravir intensification. They used maximum likelihood phylogenies and statistical tests in order to determine molecular compartmentalization.
They found a statistically significant compartmentalization between episomal and proviral DNA samples suggesting that they belong to different viral populations. In addition, longitudinal analysis of episomal and proviral DNA by phylogeny and AMOVA showed signs of non-chronological temporal compartmentalization suggesting that episomal and proviral DNA forms originated from different anatomical compartments.
Collectively, this suggests the presence of a chronic viral reservoir in which there is stochastic release of infectious virus and in which there are limited rounds of de novo infection. This could be explained by the existence of different reservoirs with unique pharmacological accessibility properties, which will require strategies that improve drug penetration/retention within these reservoirs in order to minimize maintenance of the viral reservoir by de novo infection (figure 2).
Steve Deeks reviewed the host responses that can constitute a barrier to an HIV cure.
Despite long-term effective therapy, inflammatory biomarkers remain elevated.
Among patients with undetectable viral load (<400 copies/mL), hsCRP is 40% higher, IL-6 is 60% higher, and D-dimer is 49% higher, compared with controls (figure 3).
T cell activation also remains elevated during otherwise effective HAART (figure 4).
There is a weak association between cell-based measures of viral persistence and T cell activation in blood, but the association between these factors is much stronger in gut mucosa. Several groups have observed in rectum compared to blood a stronger association between immune activation and HIV persistence.
Although many factors contribute to immune dysfunction, a central mechanism likely involves irreversible lymphoid fibrosis, which predicts poor immune reconstitution, high T cell activation and low HIV-specific T cells.
Inflammation may drive HIV persistence through several non-mutually exclusive mechanisms (up-regulation of “negative regulators”, homeostatic proliferation, increased target cells, lack of effective HIV-specific T cells).
The Deeks described the vicious circle of HIV-associated inflammation and HIV persistence.
Microcrobial translocation and HIV production during ART activates PDCs leading to increased IDO, reduced tryptophan levels, a shift in Th17/Treg ratios and eventually more immune activation. The end result is ongoing T cell activation, lower HIV-specific T cell responses and perhaps HIV persistence.
Inhibitory receptor programmed death 1 (PD-1) is a member of the CD28/CTLA-4 family of T cell regulators (figure 5).
PD-1 (and other negative regulators) may contribute to maintenance of latency. Inhibition of PD-1 ex vivo results in increased HIV production and increased effector functions. Blockade of the PD-1/PD-L1 interaction has been shown to increase HIV-specific CD8+ T cell function in vitro.
Among individuals with durable viral suppression, cell-associated RNA levels are associated with frequency of PD-1 expressing CD4+ T cells.
The ACTG 5301 trial will test the hypothesis that inhibition of PD-1 activates in a relatively specific manner cells containing HIV while simultaneously causing enhancement of HIV-specific immunity.
The study design includes a single arm, as a dose-finding study of anti-PD-1 antibody in a population of patients on long-term HAART, CD4+ T cell counts > 350 cells/mm3. The sample size is 40 (10 subjects/dose) and patients will receive a single IV of open-label MK3475 (Merck) at doses of 0.1, 1, 3, or 10 mg/kg.
In conclusion, although there is only a weak association between plasma HIV RNA levels and immune activation, there are strong associations in GALT, arguing that future studies should focus on tissue. Inflammation may also directly or indirectly contribute to persistence. An ideal intervention that might contribute to HIV eradication would reduce the consequences of immune activation while enhancing the capacity of immune system to clear HIV. Anti-PD1 antibodies could reach these goals.
Tae-Wook Chun reviewed the past and current strategies used to try to get an HIV cure or remission.
Current strategies try to purge the HIV reservoir by activating latently infected cells with drugs such as HDAC inhibitors, with the assumption that these cells with die from infection (figure 6).
Tae-Wook Chun noted that studies done in his laboratory found that HDAC inhibitors induced HIV RNA expression from the cells of patients, but at levels much lower than seen with full mitogen activation (figure 7).
More embarrassing, in such a setting, exposure to HDAC inhibitors and HIV RNA production did not seem to result in cell death.
Key words: HIV PD1, HIV cure, HIV eradication, HIV persistence, PD-1 HIV, PD-1 antibodies HIV