These results, in a small group of scIL-2 treated, virally suppressed HIV-infected patients suggest that the amount of CD8 cellular death at baseline may correlate with the ability to restore CD4 cell counts and CD4:CD8 ratios after immune reconstitution involving ART with or without additional scIL-2 therapy. These differences were detectable after two years of study enrollment. Increased CD8+ T cell apoptosis was a characteristic of about 70% of previously studied patients where we found a negative correlation between high CD8+ T cell apoptosis and low CD4+ T cell percentages . Others have also suggested that the activation level of CD8+ T cells in HIV-infected patients is an independent predictor of immune status, separable from viral load and CD4 cell count [24,25]. Our data in virally suppressed patients suggest that levels of CD8+ T cell apoptosis also may independently predict response to therapies, including scIL-2, by not only increasing the absolute number CD4 cells, but perhaps by regulating the levels of CD8+ T cell expansion.
The role of lymphocytic apoptosis in the response to ART with or without IL-2 has been studied by several groups. Most suggest apoptosis levels of both CD4+ and CD8+ T cells tend to be reduced after ART (14–16). However, in multiple studies, the levels of apoptosis still remains higher than in controls, which likely indicates the presence of continued immune activation which was previously demonstrated in the lymph nodes . IL-2 therapy seems to have little effect on apoptosis levels in the long-term. Preferential apoptosis of activated CD8+ T cells was also seen by another group at 5 days post scIL-2 therapy similar to our observations . However, there was little difference in spontaneous CD8+ T cell apoptosis levels with time, suggesting that IL-2 did not change the inherent level of CD8+ T cell apoptosis.
The work that most closely approximates the work reported here followed 17 patients during the first 2–3 years of ART and showed that lymphocyte apoptosis was reduced most in those who had the best response to therapy . Although the beginning levels of apoptosis appear different between good responders and poor responders in that study, the authors did not note that this might be an important variable for the outcome. They do show a negative correlation between the percentage of Annexin V+ CD8+ T Cells and CD4+ T cells over two years of ART therapy. Another difference in our work is that the patients were already virally suppressed when baseline apoptosis was measured. Our data, with a small group of patients indicates that the starting level of apoptosis reflects long term outcome to therapy and that the level of CD8 T cell death is independent of viral load.
IL-2 is known as the consummate T cell growth factor; however, IL-2 not only enhances proliferation and survival of T cells, it can also signal death in susceptible cell populations. We suggest that the activated CD38+ DR+ population, which dies after acute administration of IL-2, is indicative of this type of response to IL-2 in vivo. Since CD8+ T cells do not upregulate CD25, it is likely they respond to IL-2 via binding to CD122, (B chain IL-2R), which at the time of these studies was not commercially available. Of importance is that we saw no increase in CD4+T cell death, even though there was CD38+ DR+ activation marker upregulation in the treated people at Day 5 of the scIL-2 cycle. This is in contrast to the conclusions of another report; however, the data in that manuscript also clearly show more apoptosis in the activated CD8+ T cells than the CD4+ T cells after scIL-2 administration at Day 5 . We think that activation of T cells by IL-2 likely leads to differential outcomes in CD4+ vs. CD8+ T cell subpopulations. Recent data, which examined DNA turnover in CD4+ and CD8+ T cell subsets of HIV-infected people after long-term scIL-2 therapy, are supportive of this idea because IL-2 therapy enhanced CD4+ T cell survival, but not CD8+ T cell survival .
The mechanism responsible for enhanced normalization of CD4 cell counts in those with less CD8+ T cell death is unclear. It may be associated with a better overall CD4 cell count at the beginning point in the low CD8+ T cell apoptosis group. The CD4 cell count nadir in this group was about 100 cells/mm3 higher and the median was higher than in the high CD8+ T cell apoptosis group, which is consistent with our previous observations that a high level of CD8+ T cell apoptosis was correlated with fewer CD4+ T cells .
Our interpretation of this data is that the amount of CD8+ T cell apoptosis at baseline indicates existing immune activation to HIV antigens, which can continue in some patients, even without detectable viremia [14-16]. CD8+ T cells, therefore, get activated and die at an increased level in some individuals. Even though CD4 cell count increases in IL-2 treated patients, the amount of proliferation in the long term actually becomes reduced as shown in two studies [3,4]. It has been suggested that proliferation, as measured by Ki67 staining, found in cells in the SG2M phase of the cell cycle, is similar to measurements of activation using CD38 and DR expression . Hence, the better long-term outcome in those with less apoptosis might be due to less inherent activation of cells from these patients [28,29]. In the one IL-2 long-term non-responder patient studied by Kovacs, et al , the DNA turnover of both CD4+ and CD8+ T cells remained high, even after 22 cycles of scIL-2, indicating that chronic activation in some people is not relieved by IL-2 therapy. Of interest is that this patient also had about a log10 more HIV RNA (32,873 copies/mL) than other patients. However, we saw no correlation in our small study with viral load, which remained undetectable or barely detectable throughout the study . We did not have access to information on viral set point in most of the patients. Non-responders to IL-2 also have more T cell activation at baseline as shown by others [3,4], which could also be reflected by higher levels of CD8+ T cell death. The underlying reason for this is not clear, but could be related to inherent host differences or a "set point" of CD4 cell count, below which an enhanced CD8+ T cell apoptosis is triggered.
Another possible mechanism to account for the role of CD8+ T cell apoptosis relates to what drives the proliferation of T cells. That is, homeostatic mechanisms, like that driven by IL-7, are one way that T cell proliferation is driven. By contrast, activation-induced proliferation is another mechanism whereby T cells divide. We think it likely that the predominant form of CD8 proliferation occurring in those with high CD8+ T cell apoptosis is activation-induced, rather than homeostatic-driven proliferation, because death occurs mainly with activation induced proliferation, not homeostatic driven proliferation . This may account for the slightly higher CD8 cell counts in those with high apoptosis both before and after IL-2 therapy. These individuals fail to normalize CD4 T cell numbers because CD8 activation and death continues in spite of attempts at CD4 cellular reconstitution.