VOLATILE MARKERS OF PULMONARY TUBERCULOSIS IN THE BREATH

背景：肺结核（TB）是一个日益严重的公共卫生问题，但在疑似患病的患者中确诊往往既昂贵又耗时。临床医生迫切需要一种更快、更好、更便宜的活动性肺结核筛查试验。一种理想的肺结核筛查试验应该是高度敏感和特异性高、快速、廉价和无创的。目前使用的检测方法均不符合以下标准：痰液染色检测抗酸杆菌特异性高，但敏感性较差；痰液培养检测分枝杆菌敏感性较高，但成本高，耗时长。然而，一项第一阶段的试点研究提供了证据，表明呼气中挥发性有机化合物（VOCs）的呼吸测试可以提供临床有用的新标志物。呼吸中氧化应激标志物的模式识别分析对活动性肺结核感染患者的敏感性为94.1%，特异性为95.1%。呼吸VOCs模糊逻辑分析预测活动性肺结核感染的敏感性为100%，特异性为83.3%。

DESCRIPTION (provided by applicant): Background: Pulmonary tuberculosis (TB) is a growing public health problem, but confirming the diagnosis in patients with suspected disease is often costly and time-consuming. Clinicians urgently need a screening test for active pulmonary tuberculosis that is faster, better and cheaper. An ideal screening test for pulmonary TB would be highly sensitive and specific as well as rapid, inexpensive and non-invasive. No test in current use fulfills these criteria: sputum staining for acid fast bacilli has high specificity but poor sensitivity, and culturing sputum for Mycobacteria has higher sensitivity, but it is expensive and time consuming. However, a Phase 1 pilot study provided evidence that a breath test for volatile organic compounds (VOCs) in the breath could provide a clinically useful new marker. Pattern recognition analysis of oxidative stress markers in breath identified patients with active pulmonary TB infection with 94.1% sensitivity and 95.1% specificity. Fuzzy logic analysis of breath VOCs predicted active pulmonary tuberculosis infection with sensitivity = 100% and specificity = 83.3%. Hypothesis: A test for breath VOCs combined with pattern recognition analysis or fuzzy logic analysis could provide a new screening test for pulmonary TB that is highly sensitive and specific as well as rapid, inexpensive and non-invasive. Plans to test the hypothesis: We will evaluate the breath test in a subset of patients enrolled in a large multicenter trial designed to evaluate antibody-detection assays in the detection of pulmonary TB. Results of the breath test will be compared to the "gold standard" of sputum culture for M. tuberculosis. Pattern recognition analysis and fuzzy logic analysis of breath VOCs will be employed to develop predictive models in a training set of patients. These models will be evaluated and compared in a test set of patients to determine the sensitivity, specificity and predictive value of the breath test as a marker of active pulmonary TB. Long term aims: a. Scientific: To validate the clinical findings of the Phase 1 study in a larger group of subjects. b. Technical: to develop a comparatively inexpensive miniaturized and rapid breath test suitable for point of care screening. c. Commercial: To obtain Food and Drug Administration (FDA) approval of the device and to market it. Importance of the research: A breath test for pulmonary TB could make an important clinical impact by detecting new infections rapidly at low cost, and also by monitoring the impact of therapy. Also, it could potentially reduce the mortality and morbidity caused by pulmonary TB.