Novel diagnostic algorithms to improve antimicrobial management for Ugandan patients with chemotherapy-associated febrile illness

University Of Minnesota, Minneapolis MN

Investigators

Abstract

ABSTRACT: Over 1 million people with cancer live in African countries, 30-40% of whom are people with HIV (PWH). Chemotherapy-associated febrile illness is a frequent complication that occurs during cancer treatment and is associated with morbidity and mortality. Small-scale studies also suggest that the microbiology of chemotherapy-associated febrile illness differs significantly for those living in Africa. For example, our previous pilot study showed that 19% of Ugandans with solid tumors who developed chemotherapy-associated febrile illness tested positive for tuberculosis (TB). Furthermore, PWH have higher rates of chemotherapy-associated bone marrow suppression and are at increased risk for developing opportunistic infections when compared with their HIV-negative counterparts. Yet, international guidelines for diagnosis and treatment of chemotherapy- associated febrile illness do not reference any studies conducted in Africa and do not account for the patient’s HIV status. Point-of-care (POC) tests are a potentially cost-effective way to identify the microbiologic causes of febrile illness for patients living in low- and middle-income countries. Urinary lipoarabinomannan (LAM) is a first-line POC test for TB diagnosis among PWH. However, the sensitivity and specificity of urinary LAM has not been evaluated among patients with cancer. Given the high prevalence of HIV and TB throughout Africa, it is critical to develop guidelines that incorporate HIV and TB diagnostics among patients with cancer. In this study, we will use POC tests to develop a cost-effective diagnostic algorithm to guide antimicrobial management for patients with solid tumors receiving chemotherapy in Uganda. We proposed to use clinical microbiology diagnostics and next generation metagenomic sequencing to understand the association between HIV status and the microbiology of chemotherapy-associated febrile illness among Ugandan inpatients with solid tumors (Aim 1). We will assess the diagnostic accuracy of urinary LAM to detect TB among febrile patients with cancer (Aim 2). Finally, we will create a diagnostic algorithm that incorporates the patient’s HIV status and uses POC tests to guide antimicrobial management for patients with chemotherapy-associated febrile illness among Ugandan patients with solid tumors to assess an optimal, cost- effective diagnostic strategy (Aim 3). The proposed research aims will support Dr. Gulleen in meeting her career development training goals of 1) gaining proficiency in using metagenomic sequencing, 2) developing expertise in assessing the diagnostic accuracy of POC tests among novel populations, and 3) learning principles of cost-effectiveness analysis. The K23 award will facilitate her transition to an independent physician-scientist who evaluates and implements novel diagnostic algorithms for managing chemotherapy-related infections among PWH. Dr. Gulleen will conduct these activities with outstanding mentorship from experts in global health, metagenomic sequencing, clinical laboratory diagnostics, and cost-effectiveness analysis.