Contact details

Professor Robert J Wilkinson

Professor in Infectious Diseases
Department of Medicine

Tel: +44 (0)20 7594 3891
Email:

Professor Robert J Wilkinson

University of Cape Town:

 Mycobacterial Immunology Group

Understanding Tuberculosis in Cape Town, South Africa

HIV and TB are the most pressing public health problems in Africa: in South Africa up to 1000 people die prematurely per day because of HIV, many due to co-existent TB. The epidemiology and clinical features of HIV associated TB are well researched, but few studies have addressed in detail the cellular mechanisms. This programme aims to answer questions arising from the clinical care of patients with these infections. Projects underway at present include the following
 
How can tuberculosis infection in HIV infected people be more effectively diagnosed and prevented?

This project is led by Molebogeng Rangaka, who is an EDCTP Fellow who has recently been awarded a Research Training Fellowship by the Wellcome Trust. An HIV driven epidemic of tuberculosis is underway in Khayelitsha township where the annual risk of TB amongst people infected by HIV can be as high as 1 in 3. The limited TB control measures presently available need urgently to be augmented. Several commercial tests for tuberculosis infection that may have greater sensitivity and specificity than the standard skin test are now in use in Europe and North America. However, neither has been tested rigorously in the context of HIV infection or in children. We have completed and brought to publication three studies in Khayelitsha township in which we demonstrate that there is a very high prevalence of latent tuberculosis infection, and that at least one form of the Interferon gamma release assay is less impaired in HIV infected people that the skin test. However, poor agreement between the tests means that prospective studies to determine which best predicts the subsequent risk of tuberculosis and benefit from isoniazid preventive therapy (IPT) are therefore necessary. The opportunity to perform such evaluation has been taken in the context of a large (n=1200) clinical trial of IPT plus cART versus cART alone for the prevention of tuberculosis in HIV infected people. This trial is funded by the Medecins Sans Frontieres and by the Department of Health of South Africa. Partial funding for the adjacent study of the predictive value of the QuantiFERON-Gold in tube test has been provided by the Foundation for Innovative Novel Diagnostics.

Two separate collaborative studies are examining the hypothesis that tuberculosis may be associated with a distinct transcriptomic and/or proteomic signature. The first of these conducted in collaboration with Dr Anne O’Garra at the MRC National Institute for Medical Research will especially focus on immunopathogenesis and immune regulation and the effect of treatment of tuberculosis on these potential biomarkers. The other project is a multicentre study administered by Brighton and Sussex Medical School and funded by the European Union (EU) that aims to evaluate the diagnostic potential of biosignature. The novelty of the EU study is that the Translational approach is patient- rather than technology-driven, with emphasis on early iterative testing and evaluation.

By what immune mechanisms do preventive therapies decrease susceptibility to TB?

The three preventive therapies under investigation are combined antiretroviral treatment (cART), IPT and vitamin D. The project on cART led by Katalin A Wilkinson, who is an MRC Senior Investigator Scientist. It is now clear that cART reduces the risk of TB. However the drugs do not affect TB directly: they act by improving the immune response. We have completed and are analysing a longitudinal study that detailed the degree and nature of restitution of the immune response to mycobacteria during cART. Dr Beate Kampmann, a Wellcome Trust Intermediate Fellow at Imperial College London, is extending these studies to HIV infected children in a project that aims to determine whether restitution is sufficient to contemplate boosting immunity by novel subunit vaccination.
 
The work on vitamin D completes a PhD conducted by Adrian Martineau. The American Journal of Respiratory and Critical Care Medicine, Journal of Immunology and Journal of Clinical Investigation published three high impact publications in 2007. Dr Martineau is now investigating the possible treatment shortening and tissue protective effects of vitamin D in the context of a randomised controlled trial of vitamin D or placebo as adjunctive therapy for pulmonary tuberculosis. We have noted that, in addition to its effect on the induction of antimicrobial peptides, vitamin D also modulates Matrix metalloproteinase (MMP) expression. These enzymes play a key role in tissue degradation and, as such, may contribute to the tissue damage associated with tuberculosis.

What is the cause and best management of the HIV-Tuberculosis-associated Immune Reconstitution Inflammatory Syndrome?

This project is led by Graeme Meintjes who is a Consultant Physician and Wellcome Trust Fellow. Our secondary care facility GF Jooste Hospital serves a population in which over 10000 people are on cART. This rapid roll out in an area where there is a very high incidence of TB has been accompanied by a dramatic increase in numbers of HIV co-infected patients who present with the TB-immune reconstitution inflammatory syndrome (TB-IRIS). This hitherto sporadic and recherché condition has begun to dominate our HIV service. We have assembled the world’s largest prospective case series (over 500 cases) that has been the basis for large immunological studies and a vast specimen bank of serum, culture supernatants, RNA and DNA. We have also completed recruitment of 110 patients to a clinical trial of steroids against placebo for mild-moderate TB-IRIS. We aim to definitively analyse the nature, cause, treatment and diagnosis of this condition.

Why do some patients deteriorate during treatment for tuberculosis?

TB-IRIS is just one of the reasons why people deteriorate during tuberculosis therapy. This is a significant problem for us and so we undertook a pilot study to asses the frequency and reasons for such deterioration. A prospective cohort study was conducted by Dr Dominique Pepper under Graeme Meintjies’ supervision during the first quarter of 2007. Deterioration was defined as clinical worsening or failure to improve after 14 or more days of TB therapy, occasioning referral to hospital. We collected data on TB diagnosis, TB treatment, HIV status and antiretroviral therapy, and investigated reasons for clinical deterioration as well as outcome. During this period, 352 patients met inclusion criteria, accounting for 17% of total medical admissions. More than 70% of patients had an additional illness to TB. Rifampin-resistant TB, TB immune reconstitution inflammatory syndrome (TB-IRIS) and drug resistant bacterial infections (other than TB) were found in 12%, 14% and 3.4% of cases, respectively. Eighty three percent of patients were HIV-infected with a median CD4 count of 89 cells/mm2. Mortality among patients admitted to hospital was 19%. Our findings suggest that co-morbid illnesses, drug resistant TB and other drug resistant bacteria are important reasons for deterioration in HIV infected patients receiving TB therapy. HIV infected patients may be at increased risk of acquiring nosocomial drug resistant pathogens because profound immune suppression results in co-morbid illnesses that require frequent health care facility visits and prolonged inpatient admissions.

What is the pathogenesis of tuberculous pericarditis?

This is a collaborative project with the Department of Cardiology at Groote Schuur Hospital and is led in the laboratory by Kerryn van Veen, a Wellcome Trust Masters Fellow in Public Health and Tropical Medicine. The immunosuppression caused by HIV leads to an increased risk of developing active TB disease with a greater possibility of developing extrapulmonary TB. One of these extrapulmonary manifestations, tuberculous pericarditis, is responsible for more than 50% of pericarditis cases in Africa, compared with the estimated 4% in industrialized countries. In our setting, over 50% of the patients who present with tuberculous pericarditis are co-infected with HIV. Pericardial fluid and whole blood collected from HIV infected and HIV uninfected individuals who present with tuberculous pericarditis are compared in this study. Cytokine concentrations are measured by ELISA and tend to be lower in the pericardial fluid of the HIV infected people than HIV uninfected, although the pericardial fluid had highly elevated concentrations of interferon gamma compared to the blood in both groups. FACS analysis has shown the presence of activated and terminally differentiated effector memory T cells in the pericardial fluid, but with an absence of NK cells. In HIV infected patients, the pericardial fluid had a larger regulatory T cell population than the blood, which could lead to suppressed functioning of the effector T cells. This could explain the decreased concentrations of IFN-, TNF and IL-10 in the pericardial fluid compared to the blood in the HIV infected groups. This is the largest and most comprehensive immunological study of pericardial tuberculosis ever undertaken and is of particular value because of our access to the disease site. We aim to understand immune dysregulation in a highly relevant human model.

Are some antigens of M. tuberculosis preferentially recognised by people with latent tuberculosis infection?

This is partially funded by a component of a Bill and Melinda Gates Foundation/Wellcome Trust Grand Challenges in Global Health programme. The hypothesis is that hypoxia is an important stimulus that maintains a latent infection and that latent bacilli respond with a transcriptomic programme adapted to these conditions. It follows that genes upregulated by hypoxia may encode antigenic proteins that are targets of the immune response. In collaboration we are screening 30 candidate molecules that are expressed by M. tuberculosis under prolonged hypoxic conditions as antigens. The hope is that the immune response to such molecules may serve as a more specific marker for latent tuberculosis and, importantly, that a decline in immune response to such molecules may mark a successfully treated latent infection. This would be important for novel drugs that target latent tuberculosis which would otherwise have to be subject to very large and prolonged clinical studies.

Internal collaborators (Imperial College)
 • Professor Douglas B Young
 • Professor Michael Levin

Internal collaborators (University of Cape Town)
 • Professor Greg Hussey
 • Professor Gary Maartens
 • Professor Bongani Mayosi