Incomplete restoration of Mycobacterium tuberculosis-specific-CD4 T cell responses despite antiretroviral therapy

Summary

Objectives

Despite antiretroviral therapy (ART), HIV-infected persons have increased risk of active tuberculosis (TB). PPD and combined ESAT-6 and CFP-10-specific-CD4 (EC-Sp-CD4) responses were examined over 96 weeks.

Methods

HIV-infected, ART-naive Thai adults with CD4 T cell count ≤350 cells/μL starting ART were assessed at baseline, wk4, 8, 12, 24, 48 and 96. PPD and EC-Sp-CD4 T cells were detected by CD25/CD134 co-expression after stimulation with antigens.

Results

Fifty subjects were enrolled, 39 were male, median age 32 yrs, median baseline CD4 T cell count 186 cells/μL and plasma HIV-viral-load 4.9log₁₀ copies/mL. Seventeen were TB-sensitised.

At baseline, 25 had positive PPD and 15 had positive EC-Sp-CD4 response. CD4 T cell count <100 cells/μL was less (P = 0.005) and TB-sensitisation was more likely (P = 0.013) to be associated with positive baseline PPD-Sp-CD4 response.

At wk4, the number of subjects with positive PPD-Sp-CD4 response rose to 35 (P = 0.021). Mean PPD-Sp-CD4 T cells increased at wk4 (P = 0.017) in patients not classified as TB-sensitised.

The number of subjects with positive EC-Sp-CD4 response did not change significantly post ART. In TB-sensitised patients, mean EC-Sp-CD4 T cells declined to below baseline from wk12 (P = 0.010) onwards. EC-Sp-CD4 responses were undetectable in 3 out of 17 TB-sensitised patients.

Conclusions

Restoration of responses to TB-antigens was incomplete and inconsistent under the employed experimental conditions and may account for persistent increased risk of TB despite ART.

Introduction

HIV and TB co-infection is a major global health problem. In 2011, there were 8.7 million incident cases of active tuberculosis (TB) and 0.43 million deaths from HIV-associated TB globally.¹

CD4 T cells play a major role in the control of TB infection.² This is reflected by the 12–20 times increase in TB incidence in HIV-infected persons when compared to HIV uninfected persons.¹ The presence of CD4 T cell interferon-gamma (IFN-γ) response to mycobacterial antigens has been associated with reduced risk of TB in HIV-infected persons.3, 4

Immunodeficiency associated with HIV infection causes a reduction in CD4 T cell response to Mycobacterium tuberculosis (MTB). There is a progressive reduction in the prevalence of positive tuberculin skin test (TST),5, 6 as well as reductions in positive results on Quantiferon TB-Gold-In-Tube assay (QFN-GIT)7, 8, 9 and TSPOT-TB¹⁰ with progressive decline in CD4 T cell count.

Antiretroviral therapy (ART) has reduced the incidence of TB in HIV-infected individuals, by 70–90%.11, 12, 13 ART leads to restoration of cutaneous hypersensitivity¹⁴ and lymphocyte proliferative responses to Tuberculin Purified Protein Derivative (PPD).15, 16 However, this restoration does not occur in all patients and is often incomplete as responses to PPD often remain lower than uninfected individuals.17, 18, 19, 20 Incomplete restoration of CD4 T cell responses to MTB is thought to account for the increased incidence of TB despite long-term ART when compared to HIV uninfected persons.21, 22, 23

Most studies that examined the reconstitution of CD4 T cell response to MTB used PPD as the stimulating antigen. Some antigens found in PPD are also found in Bacille Calmette–Guérin (BCG) and environmental non-tuberculous mycobacteria (NTM) thus reducing its specificity in the detection of recall response to MTB, especially in BCG vaccinated populations.24, 25 Though the rate of false positive reaction on TST to PPD due to exposure to NTM is low, 0.1–2.3% depending on the local prevalence of NTM,²⁶ the effect is nonetheless important in settings where TB prevalence is low. Therefore, the use of PPD alone as a surrogate to detect MTB-Sp-CD4 response is inadequate.27, 28

RD1 (regions of difference) antigens e.g. ESAT-6 and CFP-10 are largely restricted to MTB and are absent in all strains of BCG and NTM except for Mycobacterium kanasii, Mycobacterium marinum and Mycobacterium szulgai.²⁹ Only a few studies have examined reconstitution of CD4 T cell responses using RD1 antigens. Sutherland et al. reported significant increase in ESAT-6 and CFP-10 (EC)-Sp-CD4 cytokine secretion in 16 patients 1 year post ART initiation.³⁰ Wilkinson et al. also reported increases in the number of EC-Sp-CD4 cells by Elispot.²⁰ However, Mendonca's cross sectional study found that ESAT-6-Sp-CD4 responses were lower than HIV uninfected controls even after 3–10 years of ART.²⁸

In this study, PPD and EC-Sp-CD4 responses were evaluated using the CD25/CD134 co-expression assay.³¹ The advantage of this assay is that it requires no a priori assumptions regarding the specific cytokines produced by the antigen specific CD4 T cells. This assay measures the co-expression of CD25 (α chain of IL-2 receptor) and CD134 (a co-stimulatory molecule that is part of the tumour necrosis factor (TNF) receptor superfamily) by CD4 T cells after stimulation with antigens. The levels of co-expression of these molecules are very low on CD4 T cells in the peripheral blood, but are up-regulated upon CD4 T cell activation.³¹ Thus co-expression of these 2 markers after antigen stimulation enables the detection of CD4 T cell recall response. This assay has been used to evaluate memory CD4 T cell response to a wide range of antigens.31, 32, 33, 34 Data suggest that this assay is highly specific and has good sensitivity in immunocompromised persons.31, 35

In this manuscript, we describe PPD and EC-Sp-CD4 responses in 50 Thai patients with advanced HIV infection (unselected for TB infection status) after ART initiation. The duration of follow-up is longer (96 weeks) and the visits more frequent than the previously published studies. We found that the restoration of EC-Sp-CD4 response is incomplete and inconsistent. This may account for the increased incidence of active TB in HIV-infected persons despite ART.