B-cell immune responses in HIV positive and HIV negative patients with tuberculosis evaluated with an ELISA using a glycolipid antigen

Summary

The diagnostic value of the PGL-Tb1 enzyme-linked immunosorbent assays (ELISA) was established following a survey study using sera from 220 Tuberculosis patients (including 69 HIV coinfected) and 324 controls. A higher percentage (76.8%) of the HIV-seropositive compared to the HIV-seronegative (58.9%) TB patients were ELISA positive (p=0.02) with a specificity of 94%. In HIV-positive TB patients, ELISA sensitivity was identical for all sites of disease and antibody levels were not affected by the CD4+ counts, PPD results, age or bacterial yield. Combining data for both the smear microscopy and ELISA maximized sensitivity. The kinetics of anti-PGL-Tb1 antibody was evaluated in cohort studies using sera collected before, during and after treatment for clinical TB for 79 TB patients (including 39 HIV coinfected). Statistically significant ELISA signals were observed in 51.3% of HIV-seropositive TB patients prior to the diagnosis of clinical TB and elevated antibody levels persisting 18 months after the end of antituberculous chemotherapy. Asymptomatic development of antibody also occurred in 22.7% of a cohort of 44 HIV-positive patients with a high risk of tuberculosis, but no correlation was found between persisting elevated antibody levels and progression to active disease. This antibody response in absence of disease, might reflect the control of an incipient tuberculosis infection by antituberculous prophylaxis or through an improved protective immune response associated with antiretroviral therapy.

Introduction

The interaction between tuberculosis (TB) and the human immunodeficiency virus (HIV) infection represents one of the most devastating associations, responsible for many deaths each year, mainly in developing countries.¹ As much as 60% of the TB caseload in Africa is attributable to HIV infection.² Rapid diagnosis of TB still relies on the detection of Acid Fast Bacilli (AFB) by smear microscopy, especially outside industrialized nations. However, this method possesses a highly variable sensitivity (22–78%).³ Its predictive values are low in TB patients coinfected with HIV, because of atypical clinical presentations in such patients,⁴ its poor specificity due to frequent infection with mycobacteria other than M. tuberculosis (MOTT) and a higher proportion of smear-negative HIV-seropositive patients among cases diagnosed with pulmonary TB.2, 5 Failure to make a rapid diagnosis of TB, during the AIDS phase of HIV infection, increases the mortality rate (45% versus 19%).6, 7 Despite potential new recommendations for mycobacterial culture in the HIV-seropositive patient with suspected TB, as the diagnostic gold standard,⁸ culture is not available in all countries and can require up to 8 weeks to become positive. Finally, despite several advantages, molecular amplification methodology is expensive and requires sophisticated laboratory facilities and optimal internal quality control.⁹

New diagnostic tests for TB are thus needed to replace or facilitate smear microscopy for the identification of smear-positive cases and to improve the diagnosis of smear-negative cases. In addition to screening suspected TB cases, smear microscopy is also used to measure the progress of treatment of smear-positive cases.¹⁰ Thus, a new test designed to perform both functions (diagnosis and follow-up) would be highly desirable. We therefore tested an alternative approach, namely serology, for rapid TB diagnosis and potential treatment follow-up. This strategy is known to be inexpensive, simple and rapid. The use of serology in the diagnosis of TB has a long record in the TB literature, but has never been well developed due to its low diagnostic values with poor specificity and sensitivity.¹¹ Since the 1990s, newer approaches have been chosen using enzyme-linked immunosorbent assays (ELISAs) and highly purified antigens or recombinant proteins.12, 13 Improvement in these assays has been obtained by using several different antigens simultaneously.14, 15 However, several evaluations of these tests in HIV-positive patients with TB have been at best inconclusive16, 17 mainly because HIV coinfected TB patients have been shown to be poor serological responders to protein antigens.18, 19, 20, 21, 22

ELISA tests using a panel of non-protein antigens, such as glycolipids specific for M. tuberculosis, have been developed23, 24 and evaluated.13, 25, 26, 27, 28, 29, 30, 31 About 65–70% of TB–HIV coinfected patients had serum reactivity to at least one glycolipid antigen and maintained the diverse antibody repertoire previously observed in HIV-negative TB patients.³² The reason why the antibody response to these glycolipid antigens is preserved in HIV-seropositive TB patients, despite declining CD4 T lymphocyte counts has been reported to be due to the novel CD1-restricted lipid antigen presentation pathway.33, 34 This is in sharp contrast to the classical response to T-cell-dependant peptide antigens being MHC restricted.³⁵

The underlying cause of the apparent lack of antibodies in 30–35% of the TB patients was mainly due to the presence of antibodies in circulating immune complexes and their detection improved the overall sensitivity without changing specificity.36, 37, 38

This current study was performed to evaluate the clinical utility of our in-house ELISA assay that uses one specific glycolipid antigen (PGL-Tb1) for the diagnosis and treatment monitoring of TB in HIV-positive patients compared to HIV-negative patients. In addition, we determined the performance of this assay to predict the likelihood of PPD-positive HIV coinfected patients to develop active TB.