The long term goal of this project is to provide information critical to the development and use of a low-cost, point-of-care, rapid, simple, and highly accurate diagnostic tool that can help clinicians make a microbiological diagnosis in children who present with signs and symptoms suggestive of tuberculosis disease (TB). At least two major factors complicate the diagnosis of pediatric TB. First, conventional tests used to diagnosis TB in adults (i.e., sputum smear microscopy and culture) are not widely applicable to children who are often unable to produce sputum or have paucibacillary TB that cannot be detected with these methods. Second, among children with TB symptoms, a sizeable proportion ofthe disease burden may be due to potentially fatal pathogens other than Mycobacterium tuberculosis (Mtb). One strategy that may improve TB diagnosis in children is to use a multiplexed approach that includes both screening for other possible etiologic agents of illness and rigorous attempts to isolate Mtb from more accessible biologic sources. This study has three specific objectives: 1. We will identify the spectrum of respiratory pathogens that occur in children who present with a clinical picture consistent with TB in three different settings with high TB burdens: Peru, Haiti, and Cambodia. The results of this aim will help inform the selection of pathogens that could eventually be included in a multipathogen diagnostic for children with symptoms of TB. 2. We will use sensitive PCR techniques to identify the most promising easily-accessible sources (i.e., stool, blood, nasopharyngeal swabs, gastric secretions from entero-strings) for TB diagnosis in children so that future diagnostic tools may be optimized for the detection of Mtb in these samples. 3. We will determine whether cell free Mtb DNA can be detected in serial plasma and urine samples of children and adults with TB and describe the host-related factors that alter the sensitivity. These results will help identify optimal samples and strategies for the non-invasive diagnosis of Mtb using urine.

The lack of an accurate diagnostic tool to confirm tuberculosis disease (TB) in children is a major barrier to case detection and treatment initiation. Children often cannot produce sputum for testing or have TB that is not detectable using conventional tools. This study will provide critical information to inform the development of an accurate TB diagnostic that uses specimens that are easily obtained among children of all ages.