The origin of the abnormally high levels of immune activation persisting overtime in HIV positive individuals remains unclear. Under the hypothesis of residual low-level viral replication originating from the HIV reservoir, we aimed to investigate the correlation between the levels of immune activation and the size of the reservoir.
In our cohort of patients undergoing long-term effective cART (as determined by the significant recovery of the CD4 + T-cell count, the absence of blips and documented comorbidities) we observed a positive correlation between the CD4 + T-cell activation and the time under cART at 1-year of follow-up, as well as a significant increase of the levels of CD8 + T-cell activation.
In the present study, we found a very weak correlation of the proviral DNA and 2-LTR circles loads with the levels of T-cell activation and proliferation. Studies have described stronger correlations of the proviral DNA during the early stages of infection (21,22). Furthermore, it has been described a progressive reduction of the proviral DNA overtime after the primary infection and thus, it is possible that any previous stronger correlation in our participants could have been lost at some point during their treatment history. Unfortunately, we cannot explore this hypothesis due to the lack of available retrospective samples in our cohort. Still, we believe that this might be the case of our participants, since they have been under a median of 8 years under suppressive cART. In support of this hypothesis, a prospective study has documented the loss of a similar foregoing correlation between the reservoir and immune activation after 4 years of cART in a cohort of HIV positive subjects (23).
On the other hand, the absence of a strong correlation with the proviral DNA in our participants is relevant since we found increasing levels of immune activation (predominantly in CD8 + T-cell), which suggest ongoing indirect or independent mechanisms to the reservoir are influencing immune activation in the long term, despite chronic cART, and the absence of clinical evidence of replication and comorbidities. In this regard, the absence of blips was important for excluding potential viral replication coming from the reservoir under the real-life clinical settings of HIV positive patients. Furthermore, a progressive increase of CD8 + CD38 + T-cell can be concerning and becomes clinically relevant since higher levels have been associated with higher risk of virological failure, increased morbidity and early mortality (11,24–28). It would be noteworthy to further investigate the presence of other markers in subjects with even more years on cART than our participants (with increasing immune activation and even in the presence of associated morbidity), to identify underlying mechanisms driving immune activation other than the HIV reservoir and residual viral replication by themselves.
The stable presence of the reservoir was evidenced by the steady levels of the proviral DNA (in subjects with ≤ 10 versus those with > 10 years of cART) and the lack of correlation with the number of years under cART.
Of note, our findings revealed a consistent positive correlation between the HIV proviral DNA and the levels of pre-treatment viral load, highlighting the influence of the viraemia for the establishment of the reservoir (even after at least 5 years of cART), and the critical role that early initiation of cART will have in future HIV cure trials.
In the present study, we were able to detect 2-LTR circles without overall significant variations over the period of 1 year. Though there was no correlation of the 2-LTR circles with the proviral DNA nor with the immune activation, we consider that the variations of the 2-LTR circles within the participants show that some underlying ongoing viral activity might be present, however, its impact is not yet clear.
The stable levels of T-cells expressing Ki-67 + overtime, and the lack of correlation with the levels of immune activation, indicate that the expression of immune activation markers is not entirely associated with the proliferative activity of T-cells. This suggest that other factors might be stimulating the expression of immune activation markers on T-cells. Of note, we observed that the proportion of T-cells expressing Ki-67 + was higher in the participants in which we observed an increase of the proviral DNA load at follow-up. Although we did not find a correlation between Ki-67 + and the proviral load, we believe that the study of the role of cell to cell HIV transmission and clonal expansion of infected cells could provide a more comprehensive view of the dynamics of the reservoir in the context of chronic effective cART and immune activation.
In HIV positive patients IL-7 can be found in higher levels than in negative subjects, which has been explained as a homeostatic effort to restore the T-cell count. Furthermore, the decrease of the plasmatic levels of IL-7 has been associated with the increase of the T-cell count in HIV positive patients (29). In our cohort, we documented lower IL-7 plasmatic levels at the follow-up observation. Remarkably, we found lower levels in the group of subjects with > 10 years on cART. However, we do not find a correlation with the T-cell count, nor with the levels of immune activation. Interestingly, we found a significant positive correlation at the enrollment point between CD4 + Ki-67 + T-cells and the levels of IL-7, that was lost at the follow-up.
We acknowledge some limitations of the present study. We quantified the proviral DNA and 2-LTR circles from PBMCs as markers of the reservoir, which do not necessarily reflect the number of cells with replication-competent virus that can be found in the actual anatomical reservoirs. However, PBMCs have been effectively used in previous studies to extrapolate the size of the reservoir, given the difficulty to obtain cells directly from the lymph nodes in human subjects. On the other hand, this approach also limited the number of available cells to investigate the impact of the reservoir present in specific T-cell sub-populations.
We are aware that DNA PCR methods detect all HIV genomes irrespectively of whether they are fully infectious or not. Defective proviruses may still be producing viral antigens, stimulating the immune system. However, there is still no consensus about the technique for the exact quantification of the reservoir.
Also, we do not exclude the potential presence of underlying undiagnosed conditions in some of our participants. However, we believe that the characteristics and settings of our participants (suppressive chronic cART, the absence of blips, the significant recovery of the CD4 + T-cell count and the lack of clinical evidence of other diseases) reduce the probability of major bias.