Improving treatment outcomes for patients with pulmonary tuberculosis in Tanzania : host and pathogen factors

Abstract

Tuberculosis (TB) causes more than 1.2 million deaths each year globally. Yet, survivors of tuberculosis are left with long term sequelae of chronic inflammatory responses with consequent reduced quality of life. Biomarkers of monitoring tuberculosis disease activity, clinical response and mortality might assist in reducing mortality and ultimately in improving quality of life post tuberculosis disease.

The general aim of the thesis was to describe mortality among patients treated for a first episode of tuberculosis in Tanzania in relation to selected known pathogen related predictors of poor tuberculosis outcomes, namely HIV and anti-tuberculosis drug resistance. We also explored the role of cytokines as part of the host immune response, in relation to mortality and in modulating lung damage.

In Paper I: We show an overall mortality of 3.4 – 9.3% among patients treated for tuberculosis in Tanzania, and that TB/HIV patients were at a higher risk of death compared to TB mono-infected patients. The best benefits of antiretroviral therapy (ART) in reducing mortality for TB/HIV co-infected patients occurred when ART was initiated after 14 days of anti-tuberculosis therapy (ATT).

We also observed (Paper II) that among 861 patients with tuberculosis, those with isoniazid resistance without concomitant rifampicin resistance had an increased risk of unsuccessful treatment outcome (death or treatment failure or loss from follow up - combined).

In Paper III, we observed that patients who exhibited, in cell-based assays, higher interferon gamma (IFN-γ) responses against cytomegalovirus (CMV), Epstein Barr virus (EBV) or Mycobacterium tuberculosis ESAT-6 antigens at the time of TB diagnosis had a survival benefit following treatment for TB. However, IFN-γ responses to some viral antigens (H5N1 and HSV-1) as well as other mycobacterial antigens (Ag85A, Rv2958c, Rv0447c) were not significantly different between patients who survived and those who died.

In Paper IV, we observed very high levels of interleukin 6 (IL-6) in serum from 234 patients with pulmonary tuberculosis, as compared to levels of IL-6 in serum from seven healthy controls. Other cytokines (IFN-γ, TNF-α, IL-2, IL-10, IL-17A and IL-21) were also analyzed at the time of TB diagnosis. Unlike with the other cytokines which returned to pretreatment levels or below following ATT, IL-6 levels at end of ATT were significantly higher than their corresponding pre-treatment levels. In addition, we also found that higher IL-6 levels at TB diagnosis correlated with survival of patients with pulmonary tuberculosis, as well as with severe lung injury, defined by chest x-ray score more than 80 at diagnosis.

In conclusion, between 3 and 9 out of 100 Tanzanian patients treated for first time tuberculosis will die during treatment. Well planned ART, appropriate clinical monitoring and timely addressing of background isoniazid resistance are essential to improve treatment outcomes. While anti CMV and EBV immune responses may serve to stratify mortality risk, adjunct therapy with IL-6 may serve as a possible target in reducing lung damage and consequently aid to improve quality of life following Mycobacterium tuberculosis disease in the future.