HIV-positive patients with chronic active or reactivating viral coinfections (eg, cytomegalovirus or herpes simplex virus coinfections) have higher plasma HIV RNA levels, lower T cell counts, and greater progression to AIDS compared with HIV-positive patients without evidence of infection/reactivation [31
]. For HCV coinfection, however, there have been conflicting reports [10
]. Consequently, our findings from this large cohort study, which included 1307 women, are notable. First, women with high baseline HCV RNA levels have an increased risk of AIDS, independent of CD4 cell count and HIV RNA level. Second, there was an almost 2-fold increased probability of developing AIDS for HCV-positive women compared with HCV-negative women who never had a CD4 cell count of <200 cells/μ
L and for women who remained ART naive. Third, HCV coinfection was associated with increased CD8 activation. Finally, our most important finding was the statistically significant association between the level of activated CD8 T cells and incident AIDS among HCV-positive viremic women. HCV-positive viremic women with >43% activated CD8 T cells had an almost 3-fold increased risk of AIDS-defining conditions and/or AIDS-related deaths compared with HCV-positive viremic women with <26% activated CD8 T cells. This was not found for HCV-negative women. In contrast, high levels of CD4 activation similarly predicted AIDS in both groups of women. These data suggest that the increased risk of HIV disease progression among HCV-coinfected women with high levels of CD8 activation may be due to immune dysfunction. Importantly, we also showed that women with high percentages of CD8+
T cells had a 50%–70% decreased AIDS risk, which suggests better immunity among this group of women.
To our knowledge, the Swiss HIV Cohort Study [17
] was the first large study to demonstrate that HCV accelerated HIV disease independent of injection drug use. Other studies found no effect of HCV on HIV disease progression, although recent studies have reported increased risk [9
]. Our study, in which only 10%–20% of the participants were active injection drug users during follow-up, demonstrates an increased risk of an AIDS-defining condition with increasing HCV RNA level. Furthermore, our study assessed outcome from the time of study entry to the time of an AIDS-defining condition adjusting for ART in time-dependent Cox models, whereas some studies limited outcome from initiation of highly active ART or excluded women with CD4 cell counts of <200 cells/μ
L at baseline, neither of which was done in our study. Although the effect of HCV genotype on immune activation is unknown, most women in our study were infected with genotype 1, which is associated with higher HCV RNA levels [33
]. In our study, higher HCV RNA levels correlate with higher levels of immune activation. Differences in genotype or in rates of active injection drug use may explain differences between the results of our study and those in previous reports [19
Immune activation has been closely linked to HIV disease progression [1
], but to our knowledge this association has not previously been reported in the setting of HCV coinfection. Our finding of increased incidence of AIDS-defining conditions in relation to high levels of immune activation suggests that there is impaired T cell function in HCV-positive viremic women that may potentially put them at higher risk of HIV disease progression compared with HCV-negative women.
There are a number of factors that could potentially influence immune activation and in turn pathogenesis of HIV disease, including age, race, gender, injection drugs, use of highly active ART and antibiotic prophylaxis, smoking, alcohol, HCV genotype, and extent of liver damage. We evaluated for each of these in univariate models and then in multivariate models when appropriate.
In vitro studies have shown that CD8 T cells have a lower threshold for activation and proliferation compared with CD4 T cells [37
]. Ongoing antigen-driven activation of CD8 T cells ultimately leads to CD8 T cell exhaustion and replicative senescence, which lead to inability to fight opportunistic pathogens [4
]. The factors that influence T cell activation among the women included in our study are most likely multifactorial and may be a direct consequence of activation in the liver or may be related to extrahepatic replication of HCV [38
]. Alternatively, host-specific factors (eg, cytokines) may drive persistent T cell activation. Furthermore, factors that drive immune activation may increase the available targets for further viral replication. Finally, HCV infection may impair T cell maturation more globally to a more immature primed activated phenotype and also may impair responses to Toll-like receptors, suggesting that both the innate and adaptive arms are affected [24
Our study systematically evaluated the association between HCV viremia, immune activation, and AIDS outcome. We determined that (1) HCV viremia is associated with AIDS outcome, independent of injection drug use, HIV RNA level, CD4 cell count, and ART (); (2) HCV viremia is associated with CD4 and CD8 activation, independent of HIV RNA level (); and (3) high levels of CD8 activation are associated with AIDS in HCV-positive viremic women but not in HCV-negative women. Our multivariate models showed that the causal pathway between T cell activation and HIV disease progression cannot be completely explained by HIV or HCV viral load, and other mediators of CD8 activation may be involved in increasing AIDS risk. This may be because substantial immune activation occurs in tissues such as the liver, and viral load may not reflect the extent of the activation. For instance, recently there has been evidence that the gastrointestinal tract is a site for activation of T cells as a result of microbial translocation. Gut-associated T cell depletion may also have a significant effect on HIV pathogenesis [8
]. From the intestines, T cells and microbial antigens circulate directly through the liver, where activation may continue. Whether the liver plays a role in amplifying or modulating this activation, especially with prior HCV infection, is unknown, but a recent study found evidence of increased microbial translocation among HCV-positive individuals with cirrhosis [43
]. In our study, liver biopsy data were unavailable. However, we did evaluate for liver disease by use of the aspartate to platelet ratio index [30
] and found no statistically significant correlation with AIDS risk. Furthermore, we found no association between alcohol use and AIDS outcome. On the other hand, although at the time of AIDS diagnosis there were no statistically significant differences in individual AIDS-defining conditions noted between HCV-negative women and HCV-positive women, HCV viremic women were more likely to report wasting syndrome, bacterial pneumonia, and encephalopathy at the end of follow-up. This finding suggests that HCV-positive viremic women may be at continued risk for certain AIDS-defining conditions and further supports the importance of treatment in this group. The pathogenesis of these findings may be HCV-related infection of the central nervous system and/or progressive liver disease including cirrhosis. Bacterial pneumonia is increased among HIV-infected individuals, especially injection drug users [44
], and cirrhosis may play a role in some who are coinfected with HCV. However, continued immune activation is also associated with encephalopathy, and some studies have suggested that microbial translocation related to gut-associated immunodeficiency may be a contributing factor [43
]. Furthermore, wasting syndrome may be a manifestation of chronic and severe gut-associated immunodeficiency that may be exacerbated by progressive HCV disease and/or alcohol use.
Our study results confirm those of previous studies that the presence of HLA-DR on CD8 T cells without CD38 expression appears to be protective [36
]. This is consistent with reports showing that elite controllers have higher percentages of HIV-specific and global CD8+
T cells [49
T cells appear to have high proliferative capacity and cytotoxic activity upon antigenic stimulation. Our study results suggest further research is needed to better understand the path of CD8 activation, especially the protective effect of HLA-DR expression.
Although our study is unique, it has limitations. We did not have information on the timing of HIV or HCV infection. Furthermore, HCV-positive women and HCV-negative women may be different because of differing routes of HIV infection (injection drug use vs sexual). Immune activation markers were measured for only a subset of the study population [7
], and not all phenotypic markers were measured at the same time. Nevertheless, our results confirmed previous observations by other investigators [1
]. Finally, we evaluated HCV RNA levels only at baseline because not all women underwent multiple measures. We assumed that HCV RNA levels would not change substantially in women who were not treated for HCV infection, on the basis of published studies that found that even with initiation and discontinuation of highly active ART there is only a 0.43–0.59 log change in HCV viral load [50
]; such small changes would not affect our findings.
In conclusion, our study demonstrates that HIV-coinfected HCV-positive viremic women are at increased risk for AIDS-defining conditions compared with HCV-negative women, possibly because of high levels of activation of T cells, especially CD8 T cells, which indicates increased immune dysregulation in this population of women. Lower levels of activation of both CD8 and CD4 T cells and activation of CD8 T cells expressing only HLA-DR is protective against AIDS. Further study is needed to understand better the pathogenesis of T cell activation, especially of CD8 T cells in relation to HIV disease. HCV-positive viremic women may benefit from treatment of HIV and HCV infection to prevent significant immunologic changes and improve long-term outcome. Assessing CD4 and CD8 T cell activation could help clinicians evaluate their patients’ risk of developing AIDS.