The effect of RLG-containing ART intensification on viral burden, T cell reconstitution, and T cell activation was investigated in 7 chronically-infected HIV+ men who had VL<40 copies/ml for a median of 6.7 years. Despite the limitations of study size and sampling, this study provided several pertinent findings. First, we confirmed the observation by others [1
] that ART intensification does not reduce HIV RNA in the plasma, even though we intensified with one or 2 new agents and applied an assay with LOD of <0.5 copy/ml. This finding, coupled with sequencing studies that show no evidence of drug resistance and little evolution of the plasma RNA over time [2
], suggests that low-level plasma virus is not due to ongoing viral replication in circulating lymphocytes.
Second, we extended these findings by showing that intensification results in no consistent change in cell-associated HIV RNA in either blood (PBMC or CD4+ T cells) or in the duodenum, colon, or rectum. This finding suggests that most of the HIV RNA in the blood and many regions of gut is not the result of ongoing replication that can be impacted by short-term intensification with raltegravir-containing regimens. However, in the ileum, which has the highest baseline HIV Us RNA and RNA/DNA ratio, the Us RNA/106
CD4+ T cells decreased in 5 of 7 participants in conjunction with intensification. Supporting the interpretation that intensification affected viral production, we also observed a trend towards an increase in CD4 content in the ileum and a decrease in CD8+ T cell activation in the ileum and blood. While several recent studies argue against productive infection in the blood of ART-suppressed patients, other findings in ART-suppressed patients would be consistent with ongoing replication. These findings include the presence of 2-LTR circles and functional unintegrated HIV DNA [10
], the fact that there is more HIV DNA in activated CD4+ T cells than resting CD4+ T cells [12
], and the fact that immune activation remains high despite years of ART [13
]. Moreover, ART intensification reduces the half life of the latent reservoir [14
], reduces plasma RNA [15
], increases CD4+ cell counts [15
], and reduces immune activation [15
Studies by ourselves and others have shown that HIV DNA per CD4+ T cell is disproportionately concentrated (up to 10-fold higher) in the gut relative to peripheral blood [35
], which has been attributed to the presence of cryptic low-level ongoing replication in the gut [35
]. If the much higher concentration of HIV DNA in gut CD4+ T cells reflects substantial numbers of newly-infected cells, one would expect a significant proportion of the HIV DNA to be in the labile unintegrated form. If intensification resulted in more complete suppression, one would expect that these labile forms would be cleared [36
]. However, we found that intensification resulted in no consistent decrease in HIV DNA in the gut. The lack of change in gut HIV DNA with intensification suggests that most of this DNA represents latently-infected cells, chronically-infected cells, or defective provirus rather than the result of new infectious events that can be prevented by raltegravir.
In 2 participants (A185 and A186), there was a decrease in HIV DNA in both the PBMC and the peripheral CD4+ T cells, which was reversed at week 16 in the CD4+ T cells but not in the PBMC. This finding suggests that HIV DNA may have a different nature, origin, or stability in different cell types within the blood. Of note, this proportion is similar to the proportion of patients undergoing raltegravir intensification who experienced transient rises in the concentration of 2-LTR circles, a finding which was attributed to recent infections aborted by raltegravir [19
Findings from this study raise new questions. If there is ongoing replication in the ileum, then why is a reduction in RNA not observed in the plasma? The simplest explanation is that the ileum contributes only a small fraction of the total HIV RNA that is measurable in the plasma, perhaps because there is relative compartmentalization (or delayed equilibration) between the two compartments or infection is aborted after local replication in the ileum. Other mechanisms, such as reactivation of latently-infected cells, release of virions captured on the follicular dendritic network, or chronic transcriptionally-active cells, could be responsible for the majority of residual HIV RNA in the blood. It is also unclear why ART intensification had differential effects in the four gut sites. Gut sites could differ in penetration of ART agents, immunologic environments, or the composition of CD4+ T cell populations. The ileum, which is an immune inductive site and is enriched for lymphoid aggregates, seems disproportionately involved in other inflammatory processes such as Crohn's disease and enteric infection with tuberculosis, Salmonella, and Yersinia. Close proximity of CD4+ T cells in the lymphoid aggregates may create an environment that is more conducive to cell-to-cell spread of HIV [37
]. However, one must question whether low-level ongoing replication in the ileum has true consequences for HIV persistence and eradication strategies.
Limitations of our study must be acknowledged. First, we had a relatively small number of participants and were unable to sample extensively at all 4 gut sites. However, 6 to 9 biopsies from each site were used to isolate gut cells, and the relative and absolute levels of HIV and immune markers were remarkably consistent from week 0 to 12. Second, there was some heterogeneity in baseline clinical characteristics, entry regimens, and intensification regimens, which may account for some of the differences in responses to intensification. Third, the period of intensification may have been too short relative to the clearance rate of virions or newly-infected cells. Fourth, inflammation and microbial leakage from the biopsies could have confounded the results and could perhaps account for the observed increase in HIV DNA in the duodenum. Finally, all studies aimed to detect low-level replication suffer from the difficulty that such infectious events could occur in minute, temporally and spatially-discrete foci [47
Despite these limitations, the results add to our understanding of the multiple mechanisms that contribute to the persistence of HIV in patients on ART. The lack of significant intensification-mediated changes in viral RNA in blood and most gut sites implies a process other than the classic model of ongoing replication, such as reactivation from latency or chronic persistence of virions and/or chronically-infected, transcriptionally-active cells. At the same time, the changes in immune activation and CD4 counts would be consistent with ongoing replication in some patients, either in the ileum or at another site. Future studies are needed to confirm the presence and site(s) of ongoing replication and to identify clinical and virological factors that may predict the contribution of ongoing replication to viral persistence. While some patients may potentially derive benefit from intensification of ART, other strategies will be necessary for viral eradication.