PROJECT SUMMARY/ABSTRACT Mycobacterium tuberculosis is a leading cause of sepsis and the leading killer among people living with HIV (PLWH) in sub-Saharan Africa. The diagnosis of tuberculosis (TB) is challenging to confirm in this population due to the paucibacillary and disseminated disease state, coupled with a frequent inability to produce sputum even among those with a pulmonary focus of TB. There is a pressing need for improved TB diagnostics, particularly those that rely on readily-available specimens such as urine. One example of this is the AlereLAM assay, which detects a TB cell-wall fragment (lipoarabinomannan, or “LAM”) from urine. The AlereLAM has been endorsed by the World Health Organization (WHO) since 2015 and performs relatively well among people with advanced HIV/AIDS, however the sensitivity is still modest at 56%. The Fujifilm SILVAMP is a newer urinary LAM assay that uses high affinity monoclonal antibodies and a silver amplification step to aid interpretation. Early studies have demonstrated improved sensitivity at 71% among PLWH, but the assay has not yet been tested among a subset of PLWH who present with undifferentiated sepsis. There is room for improvement in the accuracy of these assays. Therefore, we will evaluate the ability for a ~30 minute urine processing step, using the Ceres Nanotrap, to augment the yield of these rapid urinary LAM assays. This project serves as a substudy of the ATLAS Trial, a randomized controlled trial using a 2x2 factorial design of immediate empiric anti-TB therapy initiation and sepsis-specific doses of anti-TB therapy along with standard of care among people living with HIV who present with sepsis to our collaborative sites in Tanzania and Uganda. We will compare the performance between the AlereLAM and SILVAMP assays before and after processing the urine with the Ceres Nanotrap which concentrates urinary LAM and removes interfering substances within the urine. We will also evaluate the diagnostic accuracy and incremental yield of rapid assays to detect urinary LAM as compared to robust diagnostic reference standards (including blood/sputum culture for TB, sputum smear and TB-PCR testing, bacterial culture of blood/sputum/urine, as well as a novel multiplex PCR assay to identify the etiology of sepsis from blood and TB-PCR testing of urine) that are designed to identify TB and non-TB etiologies of sepsis, as well as a clinical reference standard that is partly informed by the ATLAS Trial interventions and outcomes. We will leverage the interventional ATLAS trial design and pursue exploratory analyses to understand the epidemiologic impact of a more accurate, rapid, TB diagnostic as well as the clinical impact of timelier initiation of anti-TB treatment on mortality. Our work within this unique trial infrastructure can inherently hasten the implementation of improved urinary TB diagnostics into clinical practice in settings with high TB/HIV burden for critically ill people.