Rapid Evaluation of Biomarkers in Tuberculosis

Tuberculosis is a growing threat to human health throughout the world. If we are to reverse the increasing number of new cases and the rise of cases unable to be treated with the standard drugs we need new drugs and vaccines. The clinical studies in which drugs are tested are time consuming to perform because the treatment of tuberculosis takes a long time and patients may relapse for at least a year after they have finished their drugs. In addition large numbers of patients are required to provide a convincing result.

In the past few years a number of new potential drugs for tuberculosis have been discovered and in order for these to be developed into medicines that can be used in the clinic soon it will be necessary to find new and quicker ways of testing them safely. A biomarker is a measure of a response to treatment which permits a reliable result to be predicted and this reduces the time taken to perform studies and the number of patients required in clinical trials. In this study we will use an international clinical study of tuberculosis treatment that we are performing to test potential biomarkers. These are different ways of measuring the burden of bacteria in the patient?s body and the patient?s immune response to the invading bacteria. Thus, we expect that this study will result in better ways to measure the response to treatment that will be used in all future clinical trials and will prove useful to health care workers treating tuberculosis across the world.

With the proliferation of potential new agents to be evaluated there is a pressing need to develop biomarkers that can shorten the time taken to trial new drugs and regimens. To date, the only accepted end point for tuberculosis clinical trials is bacteriological cure and absence of relapse at one year. This proposal explores clinically relevant and biologically significant biomarkers that predict the ability to render sputum culture negative and prevent relapse.

The international consortium making this proposal is responsible for a pivotal clinical trial evaluating the efficacy of moxifloxacin in tuberculosis treatment (REMoxTB - Rapid Evaluation of Moxifloxacin in Tuberculosis). This study, to be performed to international Good Clinical Practice and monitored by the Federal Drugs Administration will recruit up to 1500 patients and will follow them through treatment and for one year afterwards. Serial measures will be taken to evaluate the efficacy of two experimental regimens and the standard regimen. Thus, REMoxTB will provide an outstanding collection of well characterised clinical samples from a trial setting that will permit the comparison between the conventional measures of efficacy, bacteriological cure and relapse rate as well as several new approaches. We will collect sequential data on the bacterial load as measured by conventional bacteriological means. The delays associated with conventional culture methods will be addressed by using molecular methods; quantitative measurement of mRNA and transrenal DNA will be compared with the conventional measures of efficacy and modelling of bacterial load. We will also collect serial measures of the immune response to M. tuberculosis infection using ELISpot before during and after treatment. The REMoxTB study provides the ideal setting for the qualification of this promising, but yet not fully validated, T cell biomarker. An adaptation of our earlier non-linear model of bacteriological load measurement using a bi-exponential appraoch and the later application of non-linear mixed effects models recently proposed by Davies will be applied to analyse serial sputum culture colony counts. These methods are advantageous as they account for serial correlation and model the two phases of drug action separately.

This proposal evaluates the utility of several biomarkers in a clinical trials setting, providing the opportunity to inform the trials of the novel anti-tuberculous drugs entering the development pipeline.