Secondary Logo

Journal Logo

Basic Science

Immune Activation by Mycobacteriumtuberculosis in HIV-Infected and -Uninfected Subjects

Wyndham-Thomas, Chloé MD*,†; Corbière, Véronique PhD*; Selis, Elodie MD*; Payen, Marie-Christine MD; Goffard, Jean-Christophe MD, PhD; Van Vooren, Jean-Paul MD, PhD; Mascart, Françoise MD, PhD*,§; Dirix, Violette PhD*

Author Information
JAIDS Journal of Acquired Immune Deficiency Syndromes: January 1, 2017 - Volume 74 - Issue 1 - p 103-111
doi: 10.1097/QAI.0000000000001157



HIV infection is associated with a generalized immune activation, involving both the innate and the adaptive immune response. This systemic immune activation causes a sustained T cell turnover with increased apoptosis and senescence, a progressive decline in thymic and secondary lymphoid organ output because of tissue fibrosis, and a persistent circulation of pro-inflammatory cytokines. These interrelated pathogenic mechanisms lead to the exhaustion of immune resources and thus contribute to the progressive damage of the immune system and progression to AIDS.1 Even after achieving long-term viral suppression with combined antiretroviral therapy (cART) and hampering the progression to AIDS, people living with HIV (PLHIV) have immune activation biomarker levels that remain higher than in the general population. This residual immune activation is believed to be involved in the early onset of age-related non–AIDS-defining diseases, such as cardiovascular disease, non–AIDS-defining malignancies, bone disease, renal disease, liver disease, and neurocognitive impairment.2,3

Immune activation in HIV infection is driven by various mechanisms, including gastrointestinal mucosal dysfunction with consequent microbial translocation, modification in the host's immune modulation mechanisms, and ongoing antigenic stimulation by HIV or HIV gene products.2,4 Chronic coinfections with herpes family viruses and hepatotropic viruses are also believed to be implicated. Cytomegalovirus coinfection has been associated with an increased risk of severe non–AIDS-defining events and deaths in PLHIV,5 and an 8-week trial of valganciclovir in HIV/cytomegalovirus coinfected patients has resulted in a decrease in markers of CD8+ T cell activation.6 Similarly, PLHIV with hepatitis C virus (HCV) coinfection present a higher systemic immune activation than HIV- and HCV-monoinfected subjects,7,8 and these levels of immune activation can be reduced by HCV-specific treatment using ribavirin/interferon-alpha.8

Mycobacterium tuberculosis (Mtb) infection is one of the most frequent coinfections of PLHIV, and an estimated 30% are concerned.9 Infection by Mtb can be grossly divided into 2 main clinical entities, active disease (active tuberculosis, TB) and asymptomatic infection (latent tuberculosis infection, LTBI). If TB has been shown to induce important immune activation and inflammation in both HIV-infected10–15 and HIV-uninfected subjects,11–13,16 the impact of LTBI is less well described. Recently, Sullivan et al14 found in a study set in a high TB incidence setting that the percent of CD8+ T cells expressing the surface activation marker CD38 was increased in HIV-infected subjects with LTBI when compared to HIV-infected subjects with no evidence of Mtb infection.

To further investigate the potential role of Mtb coinfection, and in particular LTBI, to the systemic immune activation in HIV pathology, this work compares the levels of immune activation according to Mtb infection status (no evidence of infection, LTBI, or TB) among both HIV-infected and -uninfected subjects living in a low TB incidence country. Immune activation is measured by various biomarkers selected to reflect the activation of the different components of the immune system: markers of inflammation (interleukin-6), coagulopathy (fibrin degradation product D-Dimer), monocyte activation (soluble protein sCD14), and lymphocyte activation (expression of CD38 and HLA-DR on CD8+ T cells). These biomarkers have all been associated to progression to AIDS and/or mortality in HIV-infected persons.17,18


Ethics Statement

The protocols for this study (P2011/311 and P2011/113) received approval from the leading ethics committee ULB-Hôpital Erasme (aggregation number OMO21), and all participants signed an informed consent.

Study Groups

Both HIV-uninfected and -infected adults were prospectively recruited from 5 Brussels-based hospitals. The participants were divided into groups according to their Mtb infection status: TB, LTBI, or no evidence of infection. TB diagnosis was based either on microbiological proof (culture-positive or polymerase chain reaction–positive specimens) or on a high clinical suspicion with favorable response to anti-TB treatment. In HIV-uninfected subjects, LTBI diagnosis was based on a history of Mtb exposure and either a tuberculin skin test (TST) induration >15 mm or a recorded TST conversion (increase in 10 mm in 1 year). In contrast, for the HIV-infected patients, LTBI was defined by a TST induration >5 mm and/or a positive QuantiFERON-Gold In tube (QFT-GIT; Qiagen, Venlo, the Netherlands) scored according to the manufacturer's instructions. TB was excluded in all the patients with LTBI by clinical assessment and chest x-ray. Patients with no history of TB exposure and a negative LTBI screen were defined as having no evidence of Mtb.

For the HIV-uninfected participants, we enrolled 27 patients with TB (HIV−/TB group), 21 subjects with LTBI (HIV−/LTBI group), and 26 with no evidence of Mtb infection (HIV−/no.Mtb group). HIV negativity was based on self-disclosure except for the members of the TB group that were all screened with a fourth-generation enzyme-linked immunosorbent assay (ELISA) HIV antibody assay. Among the HIV−/TB group, only 3 subjects had significant comorbidities (diabetes, sickle cell anemia, and chronic obstructive pulmonary disease, respectively). Based on self-disclosure, none of the HIV−/LTBI or the HIV−/no.Mtb subjects had significant comorbidities. Most of these subjects were health care workers invited to participate in the study upon routine TST at the occupational medicine clinic.

Considering the HIV-infected participants, 11 subjects with TB (HIV+/TB group), 8 subjects with LTBI (HIV+/LTBI group), and 29 with no evidence of Mtb infection (HIV+/no.Mtb group) were enrolled. All but one of the HIV+/TB subjects had relevant comorbidities at the time of enrolment, namely chronic hepatitis B (n = 2), chronic hepatitis C (n = 1), lymphoma (n = 2), Kaposi sarcoma (n = 1), bacterial infection (n = 3), and chronic kidney disease (n = 1). Five of them (45%) had already initiated cART. In contrast, all the HIV+/LTBI and HIV+/no.Mtb were cART naive and had no significant comorbidities. These subjects were part of a previous study,19 and those with underlying hepatitis B, hepatitis C, active neoplasia, acute infection, or recent vaccination (within the past month) were excluded for the present work. Prospective clinical follow-up was organized for HIV+/LTBI and HIV+/no.Mtb. The initially HIV+/no.Mtb subjects who were prescribed cART for ≥3 months and who reached undetectable viral loads made up the HIV+.cART/no.Mtb subgroup (n = 9). The HIV+/LTBI group achieving the same criteria and that had never received LTBI preventive therapy made up the HIV+.cART/LTBI group. However, numbers being small for this group (n = 5), we recruited an additional 4 HIV-infected patients to increase its size. These 4 patients were under stable cART with undetectable viral loads and, on retrospective analysis, were found to have had a positive QFT-GIT within the past 2 years but no history of prophylactic anti-TB treatment.

The demographic and biological characteristics of the participants are summarized in Table 1. The viral load, CD4+ T cell count, CD4+ percentage, and CD4+/CD8+ ratio of the HIV-infected participants were obtained from routine analysis performed at the same time or at maximum within 1 month of biomarker measurement.

Demographic and Biological Characteristics of the Study Groups

Biomarker Analysis

For the TB patients, blood sampling for biomarker analysis was made before or within the first 5 days of anti-TB treatment. For the subjects presenting LTBI or without evidence of Mtb infection, blood sampling was made either the same day or at least 3 months after TST, and before any isoniazid preventive therapy.

Flow Cytometry Analysis of CD38 and HLA-DR Expression on CD8+ T Lymphocytes

Fresh whole blood (100 μL) was stained with CD45-PerCP (clone 2D1), HLA-DR-FITC (clone L243), CD4-APC-H7 (clone SK3), CD8-APC (clone SK1), CD3-V450 (clone UCHT1), and CD38-PE (clone HB7). All antibodies were obtained from BD Biosciences (Erembodegem, Belgium). After 30 minutes incubation at room temperature, red blood cell lysis was achieved using 2 mL of BD-lysing solution. Flow cytometric analysis was performed using a FACS Canto II (Becton Dickinson) and FlowJo software (Tree Star, Ashland, OR). Samples with <10,000 CD45+CD3+ cells were excluded from analysis. The gating strategy applied to define CD8+CD38+, CD8+CD38high, CD8+CD38+HLA-DR+, and CD8+CD38highHLA-DRhigh cell population is available as supplementary material (see Supplemental Digital Content 1,, and an illustrative example of the flow cytometry plots obtained is shown in Figure 1. A CD8+CD38highHLA-DRhigh gate is not usually reported in articles. It was applied here to count a subset of cells that formed a cluster among the CD38 and HLA-DR double positives on the flow cytometry images (Fig. 1).

Representative flow cytometry plots of CD38 and HLA-DR expression on CD8+ T cells. Flow cytometry plots for the expression of CD38 and HLA-DR by CD8+ T cells in an HIV-uninfected subjects with no evidence of M. tuberculosis infection (A) and in an HIV-infected subject with active TB (B). These plots are representative of the 2 extremes of immune activation levels found in this study. i, CD8+CD38+HLA-DR+; ii, CD8+CD38highHLA-DRhigh percentage.

Soluble Biomarker Analysis

Interleukin-6 (IL-6) and sCD14 levels were measured in cryopreserved plasma samples using classical sandwich ELISA (Human IL-6 CytoSet; Life technologies, Gent, Belgium and Human sCD14 Immunoassay; R&D systems Europe, Abingdon, United Kingdom). Plasma samples were tested, in duplicate, following the manufacturer's instructions. Detection thresholds were 20 and 320 pg/mL for IL-6 and sCD14, respectively. Prior plasma dilution was required for sCD14 measurement (1/400).

The plasma samples were thawed a second time for the quantitative determination of D-Dimers. This measurement was done by the Hemobiology Clinic of the Erasme Hospital, using an immunoturbidimetric assay (INNOVANCE D-Dimer; Siemens Healthcare Diagnostics Products GmbH, Marburg, Germany).

Statistical Analysis

Results were analyzed using GraphPad Prism version 5.04 (GraphPad Software, La Jolla, CA; A Mann–Whitney U test or a Kruskall–Wallis test was used to compare continuous variables. The relationship between categorical variables was assessed using a Fisher exact test or χ2 test, whereas correlations were analyzed by Spearman rho test. A P-value <0.05 was considered significant.


Expression of CD38 and Co-expression of CD38/HLA-DR on CD8+ T Cells

As shown in Figure 2, in HIV-uninfected subjects, the median percentages of CD8+CD38+, CD8+CD38+HLA-DR+, and CD8+CD38highHLA-DRhigh were significantly higher in patients with TB than in subjects presenting LTBI or with no evidence of infection by Mtb (CD8+CD38+: 73.2%, 51.3%, and 56.7%; CD8+CD38+HLA-DR+: 14.7%, 8.5%, and 6.6%; CD8+CD38highHLA-DRhigh: 2.5%, 1.5%, and 0.9% for TB, LTBI, and no Mtb, respectively). Corresponding P-values, ranging from <0.0001 to 0.0041, are shown in Figures 2A–D. For the HIV-infected patients, only the median expression of CD8+CD38+HLA-DR+ was significantly higher in patients with TB than in those with LTBI or those with no evidence of Mtb (75.5%, 55.0%, and 54.2%; P = 0.0303 and P = 0.0327, respectively, Fig. 2C). In HIV-infected or -uninfected patients, no significant difference in surface marker expression was found between those with LTBI and those with no evidence of Mtb infection. Interestingly, the levels of CD8+CD38+, CD8+CD38high, CD8+CD38+HLA-DR+, and CD8+CD38highHLA-DRhigh were higher in the HIV+/no.Mtb group than in the HIV−/TB group (P = 0.0256, <0.0001, <0.0001, and <0.0001, respectively, Figs. 2A–D).

CD8+ T cell surface activation markers according to HIV and M. tuberculosis status. The expression levels of (A) CD8+CD38+, (B) CD8+CD38high, (C) CD8+CD38+HLA-DR+, and (D) CD8+CD38highHLA-DRhigh are shown for the different study groups: HIV-uninfected subjects (white boxes) with no evidence of M. tuberculosis infection (HIV−/no.Mtb; n = 24), with LTBI (HIV−/LTBI; n = 21), or with active TB (HIV−/TB; n = 25) and HIV-infected subjects (gray boxes) with no evidence of M. tuberculosis infection (HIV+/no.Mtb; n = 26), with LTBI (HIV+/LTBI; n = 5), or with active TB (HIV+/TB; n = 7). For each group, the top and bottom of the box represents the 25th and 75th percentile, the band inside the box is the median, the whiskers extend from the 10th to the 90th percentile, and the dots represent the outliers from this range. Groups are compared 2-by-2 using Mann–Whitney U test. Significant P-values are indicated.

cART induced a progressive decline in the levels of CD8+ activation surface markers, but only the percentage of CD8+ T cells co-expressing CD38 and HLA-DR remained higher in virally suppressed HIV-infected patients when compared to HIV-uninfected patients (CD8+CD38+HLA-DR+: P < 0.0001; CD8+CD38highHLA-DRhigh; P = 0.0076, Fig. 3A). When comparing the HIV+.cART/no.Mtb and the HIV+.cART/LTBI, no significant difference in the expression of any of the CD8+ activation markers was found (data shown for CD8+CD38highHLA-DRhigh, Fig. 3B).

CD8+ T cell surface activation markers in HIV-treated patients: comparison between subjects with no evidence of M. tuberculosis infection and LTBI. A, The percentage of CD8+CD38highHLA-DRhigh among CD8+ T lymphocytes found in HIV-infected patients (subjects with no evidence of M. tuberculosis and LTBI confounded, black dots) before the instauration of combined antiretroviral therapy (HIV+/no.cART), after 3–6 months of treatment (HIV+/3–6 mo cART), and after at least 6 months of therapy (HIV+/>6 mo) are shown. These groups are compared to healthy controls (HIV−/no.Mtb, represented as black triangles). Groups are compared 2-by-2 using Mann–Whitney U test. Significant P-values are indicated. B, Among the HIV-infected subjects with undetectable viral loads and at least 3 months of combined antiretroviral therapy, the CD8+CD38highHLA-DRhigh percentage is compared between subjects with no evidence of M. tuberculosis infection (HIV.cART/no.Mtb) and subjects with LTBI (HIV.cART/LTBI). No significant difference is found using a nonparametric Mann–Whitney U test (P > 0.05).

Soluble Markers IL-6, sCD14, and D-Dimers

Levels of sCD14 and D-Dimer, as shown in Figure 4, were significantly higher among the HIV−/TB group than for the HIV−/LTBI and HIV−/no.Mtb groups with respective medians of 2.18 × 106, 1.5 × 106, and 1.25 × 106 pg/mL for sCD14 and of 2000, 380, and 170 ng/mL for D-Dimers. Likewise, the median levels were significantly higher for the HIV+/TB group when compared with the HIV+/LTBI and the HIV+/no.Mtb groups (sCD14: 2.82 × 106, 1.81 × 106, and 1.84 × 106 pg/mL; D-Dimers 6089, 724, and 489 ng/mL). Corresponding P-values are shown in Figures 4A, B. Among the HIV-uninfected subjects, the D-Dimer levels were higher in those with LTBI than for those with no evidence of Mtb infection (380 versus 170 ng/mL, P = 0.0447). However, when applying a Bonferroni correction for multiple comparisons, the P-value obtained here is not statistically significant (P ≤ 0.017 required). The median levels of sCD14 and D-Dimer in the HIV+/no.Mtb group were lower than in the HIV−/TB group (sCD14: 1.84 × 106 pg/mL versus 2.18 × 106 pg/mL, P = 0.0183; D-Dimers: 2000 versus 489 ng/mL, P < 0.0001).

Soluble biomarker levels according to HIV and M. tuberculosis status. Plasmatic concentrations of (A) sCD14 and of (B) D-Dimer are shown for the different study groups: HIV-uninfected subjects (white boxes) with no evidence of M. tuberculosis infection (HIV−/no.Mtb), with LTBI (HIV−/LTBI), or with active TB (HIV−/TB), and HIV-infected subjects (gray boxes) with no evidence of M. tuberculosis infection (HIV+/no.Mtb), with LTBI (HIV+/LTBI), or with active TB (HIV+/TB). For each group, the top and bottom of the box represents the 25th and 75th percentiles, the band inside the box is the median, the whiskers extend from the 10th to the 90th percentile, and the dots represent the outliers from this range. Groups are compared 2-by-2 using Mann–Whitney U test. sCD14 measurement: n = 22, 19, and 25 for HIV−/no.Mtb group, HIV−/LTBI group, HIV−/TB group, respectively, and n = 28, 8, and 12 for HIV+/no.Mtb group, HIV−/LTBI group, and HIV+/TB group; D-Dimer measurement: n = 22, 19, and 24 for HIV−/no.Mtb group, HIV−/LTBI group, HIV−/TB group, respectively, and n = 28, 8, and 11 for HIV+/no.Mtb group, HIV−/LTBI group, and HIV+/TB group, respectively. Significant P-values are indicated.

Both sCD14 and D-Dimer levels were significantly decreased with cART and, after 6 months of treatment, these were no longer different from those found in HIV−/no.Mtb controls (results not shown). No significant difference was found between the HIV+.cART/no.Mtb and the HIV+.cART/LTBI (results not shown).

Detectable IL-6 levels were only recorded for 6 of the 66 HIV-uninfected patients tested (5/25 TB, 0/19 LTBI, and 1/22 with no evidence of Mtb) and 5 of the 48 HIV-infected patients tested (3/11 TB, 2/8 LTBI, and 1/29 with no evidence of Mtb). Comparisons between groups could therefore not be made.

Relationship Between Activation Markers and the Biological/Demographic Characteristics

The relationship between the activation markers and the biological/demographic characteristics of the cART-naive HIV-infected subjects are shown in Table 2. An association between higher CD8+CD38high, CD8+CD38+HLA-DR+CD8+CD38highHLA-DRhigh percentages was found with younger age. Gender was associated with D-Dimer levels and CD8+CD38highHLA-DRhigh percentages. D-Dimer levels were also associated with ethnic origin, being higher for subjects of sub-Saharan origin. Higher levels of HIV viral load and severity of immune deficiency, whether measured by CD4+ T cell count, CD4+ percentage, or CD4+/CD8+ ratio, correlated with the surface markers, but no association was found with soluble marker levels.

Relationship Between Markers of Immune Activation and the Biological/Demographic Characteristics of HIV-Infected Patients


Overall, TB was associated with an increase in biomarkers of immune activation, in both HIV-infected and -uninfected subjects.10–12 The greatest levels of immune activation were found in dually infected patients (HIV and active TB), although this group of subjects had other potentially immune-activating comorbidities. As discussed here, the role of LTBI on immune activation biomarkers remained inconclusive and warrants further investigation.

Previous studies have reported, both in HIV-infected and in HIV-uninfected subjects, that TB is associated with an increase in both soluble and cellular immune activation biomarkers that will progressively decline on TB treatment,13,15,20 suggesting an association with the level of bacterial burden. In the present study, CD8+CD38+ percentage, D-Dimer and sCD14 levels were all significantly increased in the course of TB in the HIV-uninfected subjects. In addition, the CD8+ co-expression of CD38/HLA-DR was significantly increased, an observation not consistently reported in literature as some have described a decrease in HLA-DR expression with TB.21 Globally, comparable results were found for the HIV-infected patients although these must be interpreted with caution as comorbidities that may increase immune activation were present in the HIV+/TB group: despite multicentric enrolment, finding HIV-positive patients with TB but no previous cART, and no other comorbidities (the strict criteria applied to all the other groups) proved virtually impossible. This is inherent to the greater immune deficiency that characterizes these patients.

The reported effect of LTBI on immune activation biomarkers is not as clear, and studies to date have focused on HIV-uninfected individuals. Rodrigues et al13 found no statistically significant difference between health care workers with recent TST conversion and healthy volunteers in CD8+ T cell expression of CD38. Wergeland et al16 found the same result for the co-expression of CD38 and HLA-DR on CD8+ T cells, although a trend towards more immune activation was noted in LTBI subjects that had persisting QFT-GIT positivity despite preventive therapy. Shitrit et al22 did not find a significant difference in D-Dimer level when comparing subjects with LTBI to a control group, whereas Lopes et al23 showed an increase in plasmatic IL-6 in TB contact patients with positive TST when compared with healthy controls. As LTBI is a heterogeneous entity made up of a broad spectrum of disease that extends from controlled nonreplicating infection to subclinical disease,24–26 and as Mtb-associated immune activation seems to relate with the level of bacterial burden, it is possible that noteworthy immune activation may only arise in LTBI with residual bacterial replication. This may account in part to the discordant results found in the literature. Concerning HIV-infected patients specifically, to our knowledge, only one other study has recently published on the effect of LTBI on immune activation biomarkers.14 In this study by Sullivan et al14 and set in South Africa, LTBI was defined by a TST induration >5 mm and/or a positive RD-1–specific interferon-gamma ELISpot. The investigators found an increase in the CD8+ T cell expression of CD38 in HIV-infected subjects with LTBI compared with subjects with no evidence of Mtb infection. This result was not confirmed in the work presented here, although to conclude on the absence of a significant difference with an 80% power, groups of 195 persons would be required. Notably, Sullivan et al14 found no differences in CD38/HLA-DR co-expression on CD8+ T cells, sCD14, or IL-6 between the 2 groups.

For multiple reasons, research on the influence of LTBI on immune activation is particularly difficult. First, the absence of a gold standard tool for the diagnosis of LTBI invariably results in a debate on the correct categorization of patients within studies. LTBI diagnosis criteria must take into account the local epidemiology of Mtb and the use of BCG vaccination. For this reason, comparison of studies set in different epidemiological backgrounds is not ideal. This study is set in Belgium, a country with a low global TB incidence (8.8/100,000 inhabitants) and an TB epidemiology driven by immigration from TB-endemic regions.27 The restricted use of BCG vaccination in Belgium allows the TST to remain the recommended tool for LTBI screening, as was applied in this study for HIV-uninfected patients.28 However, in accordance with European Centre for Disease Prevention and Control guidelines, QFT-GIT was made simultaneously to the TST in HIV-infected patients to increase the testing sensitivity.29

The other main difficulty when approaching this topic is the lack of specificity of immune activation biomarkers. Indeed, if the biomarkers studied here have all been correlated to a faster progression of HIV-infection to AIDS (independently to CD4+ T cell count and viral load), numerous other medical conditions influence their levels. These include recent vaccination30,31 and inflammatory diseases.32 For this reason, strict inclusion criteria were applied for the subjects with LTBI or with no evidence of Mtb infection (exclusion of all subjects with recent vaccination, infectious or inflammatory conditions). This offered a greater specificity to our results but, as a counterpart, limited the number of subjects included, particularly those with advanced HIV infection. In addition, although the enrolled HIV-uninfected subjects with LTBI or no evidence of Mtb infection were considered healthy, these subjects are probably less well screened for underlying medical conditions than the closely monitored HIV-infected patients, creating a study bias. Other parameters that independently influence biomarker levels include smoking, age, gender, ethnic origin, and body mass index.33–35 Higher IL-6 levels, for example, are independently associated with older age, nonblack race, higher body mass index, lower serum lipid levels, and hyperglycemia.34,36 By this study, it is clear that IL-6 assays must have thresholds below 20 pg/mL to allow proper evaluation, although this increase in sensitivity might be obtained at the expense of specificity.

By and large, the small number of HIV-infected patients in this study is the main limitation of this work as, ideally, numbers need to be sufficient to perform multivariate analysis taking into account all the known confounding factors of immune activation. Nevertheless, this study on the influence of LTBI on biomarkers of immune activation in HIV-infected subjects is the first in a country of low TB incidence, and our results can help design future investigations in the field. The observation that TB-only subjects (HIV−/TB) had significantly higher levels of both sCD14 and D-Dimer than the HIV-only group (HIV+/no.Mtb), whereas the opposite was observed for the surface activation markers, is of particular interest. Notably, Lawn et al12 also showed higher sCD14 levels in TB-only individuals when compared with HIV-only subjects. These results suggest that soluble markers of activation may offer greater sensitivity than cellular markers for identifying Mtb-associated immune activation. Another advantage of soluble markers over cellular markers is that, although both have been correlated with HIV infection progression to AIDS and mortality in untreated patients, soluble markers have been identified as better predictors or correlates of disease (including for non–AIDS-related diseases) in patients under long-term effective cART.1 Finally, although the measurement of the expression of surface markers is affected by the disadvantages of flow cytometry (need for highly trained operators, standardized calibrations required to insure reproducibility, lack of consensus for gating strategy, performed on fresh blood), soluble markers can be measured by standard ELISA technique on stored frozen plasma. Soluble markers would therefore be more suitable for a large European cohort study, as would prove necessary to obtain sufficient numbers of HIV-positive subjects with TB or LTBI to conclude on the present issue.

In low-TB incidence settings, the screening and treatment of LTBI in all HIV-infected patients is being questioned, and there is a trend toward limiting screening to those at greatest risk of Mtb reactivation.37,38 If LTBI is proven to contribute to the residual immune activation found in virally suppressed HIV-infected patients, and therefore indirectly to an increased risk of specific non–AIDS-defining comorbidities, this would have to be taken into account when balancing the pros and cons for LTBI screening and preventive therapy. For this reason, it is important to compare levels of immune activation between LTBI and non-LTBI subjects in HIV-infected individuals virally suppressed by cART. This was made here, and no differences were noted. However, the only biomarker to remain significantly higher in the virally suppressed HIV-infected patients when compared with healthy controls was CD8+CD38highHLA-DRhigh. The small number of subjects tested may have masked the presence of residual immune activation in the treated HIV-infected patients and, in consequence, obscure the potential role of LTBI.

In conclusion, immune activation by TB is well described in HIV-infected and -uninfected subjects. The role of LTBI is, however, not ascertained. Further investigation is needed and warranted, as the participation of LTBI to the residual immune activation of virally suppressed HIV-infected patients would have therapeutic implications. According to our results, soluble markers of inflammation, such as sCD14 and D-Dimer, would offer a better tool than the CD8+ T cell surface activation markers CD38 and HLA-DR for the evaluation of Mtb-associated immune activation.


The authors express their gratitude to all the participants and recruiting physicians. The latter were members of the Infectious Disease Departments of the Hôpital Erasme, CHU Saint Pierre, Hôpitaux Iris Sud, CHU Brugmann, and Universitair Ziekenhuis of Brussels, Belgium. The authors also are very grateful for the help given by the research nurses and data managers of each center, namely, Mrs. Zoe Kipouros, Mrs. Carolien Wylock, Mrs. Virigine Lenoir, Mrs. Laura Riesi, and Mrs. Leslie Andry. The authors thank Mrs. Myriam Libin and Anne Van Praet from the Laboratory of Vaccinology and Mucosal immunity (LoVMI) for their help with the sample analyses, Walter Wijns and Dr. Ingrid Beukinga from the Hemobiology Clinic of the Erasme Hospital for the D-Dimer measurements, and Prof. Christian Melot from the ULB for his help with the statistical analyses.


1. Klatt NR, Chomont N, Douek DC, et al. Immune activation and HIV persistence: implications for curative approaches to HIV infection. Immunol Rev. 2013;254:326–342.
2. Appay V, Sauce D. Immune activation and inflammation in HIV-1 infection: causes and consequences. J Pathol. 2008;214:231–241.
3. Deeks SG, Phillips AN. HIV infection, antiretroviral treatment, ageing, and non-AIDS related morbidity. BMJ. 2009;338:a3172.
4. Brenchley JM, Price DA, Schacker TW, et al. Microbial translocation is a cause of systemic immune activation in chronic HIV infection. Nat Med. 2006;12:1365–1371.
5. Lichtner M, Cicconi P, Vita S, et al. Cytomegalovirus coinfection is associated with an increased risk of severe non-AIDS-defining events in a large cohort of HIV-infected patients. J Infect Dis. 2015;211:178–186.
6. Hunt PW, Martin JN, Sinclair E, et al. Valganciclovir reduces T cell activation in HIV-infected individuals with incomplete CD4+ T cell recovery on antiretroviral therapy. J Infect Dis. 2011;203:1474–1483.
7. Kovacs A, Al-Harthi L, Christensen S, et al. CD8(+) T cell activation in women coinfected with human immunodeficiency virus type 1 and hepatitis C virus. J Infect Dis. 2008;197:1402–1407.
8. Gonzalez VD, Falconer K, Blom KG, et al. High levels of chronic immune activation in the T-cell compartments of patients coinfected with hepatitis C virus and human immunodeficiency virus type 1 and on highly active antiretroviral therapy are reverted by alpha interferon and ribavirin treatment. J Virol. 2009;83:11407–11411.
9. WHO. WHO|TB/HIV Facts 2012–2013. Available at: Accessed August 7, 2013.
10. Vanham G, Edmonds K, Qing L, et al. Generalized immune activation in pulmonary tuberculosis: co-activation with HIV infection. Clin Exp Immunol. 1996;103:30–34.
11. Hertoghe T, Wajja A, Ntambi L, et al. T cell activation, apoptosis and cytokine dysregulation in the (co)pathogenesis of HIV and pulmonary tuberculosis (TB). Clin Exp Immunol. 2000;122:350–357.
12. Lawn SD, Labeta MO, Arias M, et al. Elevated serum concentrations of soluble CD14 in HIV- and HIV+ patients with tuberculosis in Africa: prolonged elevation during anti-tuberculosis treatment. Clin Exp Immunol. 2000;120:483–487.
13. Rodrigues DSS, Medeiros EA, Weckx LY, et al. Immunophenotypic characterization of peripheral T lymphocytes in Mycobacterium tuberculosis infection and disease. Clin Exp Immunol. 2002;128:149–154.
14. Sullivan ZA, Wong EB, Ndung'u T, et al. Latent and active tuberculosis infection increase immune activation in individuals co-infected with HIV. EBioMedicine. 2015;2:334–340.
15. Toossi Z, Funderburg NT, Sirdeshmuk S, et al. Systemic immune activation and microbial translocation in dual HIV/tuberculosis-infected subjects. J Infect Dis. 2013;207:1841–1849.
16. Wergeland I, Assmus J, Dyrhol-Riise AM. T regulatory cells and immune activation in Mycobacterium tuberculosis infection and the effect of preventive therapy. Scand J Immunol. 2011;73:234–242.
17. Nixon DE, Landay AL. Biomarkers of immune dysfunction in HIV. Curr Opin HIV AIDS. 2010;5:498–503.
18. Duprez DA, Neuhaus J, Kuller LH, et al. Inflammation, coagulation and cardiovascular disease in HIV-infected individuals. PLoS One. 2012;7:e44454.
19. Wyndham-Thomas C, Dirix V, Schepers K, et al. Contribution of a heparin-binding haemagglutinin interferon-gamma release assay to the detection of Mycobacterium tuberculosis infection in HIV-infected patients: comparison with the tuberculin skin test and the QuantiFERON-TB Gold In-tube. BMC Infect Dis. 2015:15. doi:10.1186/s12879-015-0796-0.
20. Mahan CS, Walusimbi M, Johnson DF, et al. Tuberculosis treatment in HIV infected Ugandans with CD4 counts >350 cells/mm3 reduces immune activation with no effect on HIV load or CD4 count. PLoS One. 2010;5. doi:10.1371/journal.pone.0009138.
21. Shankar EM, Velu V, Kamarulzaman A, et al. Mechanistic insights on immunosenescence and chronic immune activation in HIV-tuberculosis co-infection. World J Virol. 2015;4:17–24.
22. Shitrit D, Izbicki G, Shitrit ABG, et al. Normal D-dimer levels in patients with latent tuberculosis infection. Blood Coagul Fibrinolysis. 2005;16:85–87.
23. Lopes FHA, de Assis LC, Neto P, et al. Serum levels of interleukin-6 in contacts of active pulmonary tuberculosis. J Bras Patol Med Lab. 2013;49:410–414.
24. Barry CE, Boshoff HI, Dartois V, et al. The spectrum of latent tuberculosis: rethinking the biology and intervention strategies. Nat Rev Microbiol. 2009;7:845–855.
25. Delogu G, Goletti D. The spectrum of tuberculosis infection: new perspectives in the era of biologics. J Rheumatol Suppl. 2014;91:11–16.
26. O'Garra A, Redford PS, McNab FW, et al. The immune response in tuberculosis. Annu Rev Immunol. 2013;31:475–527.
27. Fonds des Affections Respiratoires–Epidémiologie. Available at: Accessed March 26, 2014.
28. Recommandations Concernant le Dépistage Ciblé et le Traitement de l'infection Tuberculeuse Latente. Available at: Accessed August 12, 2013.
29. ECDC 2011: Use of Interferon-Gamma Release Assays in Support of TB Diagnosis. Available at: Accessed August 7, 2013.
30. Castro P, Plana M, González R, et al. Influence of a vaccination schedule on viral load rebound and immune responses in successfully treated HIV-infected patients. AIDS Res Hum Retroviruses. 2009;25:1249–1259.
31. Lawn SD, Butera ST, Folks TM. Contribution of immune activation to the pathogenesis and transmission of human immunodeficiency virus type 1 infection. Clin Microbiol Rev. 2001;14:753–777. table of contents.
32. Funderburg NT, Stubblefield Park SR, Sung HC, et al. Circulating CD4(+) and CD8(+) T cells are activated in inflammatory bowel disease and are associated with plasma markers of inflammation. Immunology. 2013;140:87–97.
33. Valiathan R, Miguez MJ, Patel B, et al. Tobacco smoking increases immune activation and impairs T-cell function in HIV infected patients on antiretrovirals: a cross-sectional pilot study. PLoS One 2014;9:e97698.
34. Borges ÁH, O'Connor JL, Phillips AN, et al. Factors associated with plasma IL-6 levels during HIV infection. J Infect Dis. 2015;212:585–595.
35. Steel A, John L, Shamji M, et al. CD38 expression on CD8 T cells has a weak association with CD4 T-cell recovery and is a poor marker of viral replication in HIV-1-infected patients on antiretroviral therapy. HIV Med. 2008;9:118–125.
36. Thompson DK, Huffman KM, Kraus WE, et al. Critical appraisal of four IL-6 immunoassays. PLoS One. 2012;7:e30659.
37. Pozniak AL, Coyne KM, Miller RF, et al. British HIV Association guidelines for the treatment of TB/HIV coinfection 2011. HIV Med. 2011;12:517–524.
38. Systematic screening and treatment prophylaxis for tuberculosis in HIV patients in a high income setting in 2013: is it useful? by Ula Maniewski given during the 14th European AIDS Conference powered by–MULTIWEBCAST–State-of-the-art Webcast Services. Available at: Accessed January 3, 2014.

HIV; Mycobacterium tuberculosis; immune activation; CD38; sCD14; D-Dimers; HLA-DR

Supplemental Digital Content

Copyright © 2016 Wolters Kluwer Health, Inc. All rights reserved.