Mucositis in Immunocompromised Hosts: No Association Between Mucositis and Mucosal Barrier Integrity Organism Detection on Plasma Next Generation Sequencing

Background Plasma next generation sequencing (pNGS) for microbial organisms is useful in the workup of immunocompromised hosts with suspected infection. However, the interpretation of pNGS is still an area of investigation. PNGS results often include bacteria/fungi know to colonize the gastrointestinal (GI) tract. The clinical relevance of these results, especially in immunocompromised patients, is unclear. We hypothesized that patients with mucositis would be more likely to have increased detection rates of microorganisms that colonize the GI tract due to barrier disruption leading to transient bacteremia. Methods Inpatient pNGS (KariusTM) results over four-years (3/2020-10/2024) were reviewed retrospectively. Patients with malignancy or receiving hematopoietic stem cell transplant (HSCT) were included. Patient data was extracted from electronic charts into REDCapTM. Statistical analyses (descriptive, Fisher t-test, Mann-Whitney U) were performed with GraphPad PRISM. A Pediatric Oncologist specified criteria for mucositis and adjudication complex or unclear cases. Results Eighty-three pNGS tests met inclusion criteria. Nineteen tests (22.9%) detected organisms classified as Mucosal-Barrier-Injury-Associated (MBI) by the National Healthcare Safety Network (Table 1). There was no association between MBI detection and presence of mucositis, neutropenia, cancer diagnosis, ICU admission, or symptoms. Patients with MBI were older than non-MBI patients (median age 16 vs 10.5 years, p=0.0142). MBI organism detections were more likely to lead to a diagnosis (p=0.044) and prompt antibiotic changes (p=0.0025). MBI detection was less common in HSCT patients (10.5%, p=0.0488). Eighteen tests occurred in patients with documented mucositis (Table 2). Although not associated with MBI detection, mucositis was correlated with GI symptoms, skin findings, and febrile neutropenia. Other examined patient factors were not different between patients with and without mucositis. Conclusion Contrary to our hypothesis, mucositis was not associated with increased detection of MBI organisms by pNGS. Despite prolonged neutropenia and presumed higher risk of mucosal disruption and infections, HSCT and AML patients had very low rates of MBI detection (3/19, 15.7%), likely related to antibiotic prophylaxis and close inpatient monitoring during ANC nadirs. At this institution, detection of MBI organisms was often interpreted as clinically relevant and leds to antibiotic changes. These conclusions are limited by small sample size and single center design. Future studies could examine the role of mucositis severity, duration of neutropenia, infection-related mortality. Antibiotic exposure as a potential confounder should be further investigated. This study raises important questions on the relationship between clinical symptoms and MBI detections among the immunocompromised for which further study is needed.