Pediatric Fungal Network's (PFN) STudy of Rare Invasive Fungal DisEases in Immunocompromised Pediatric Patients (STRIDE) Project 3

Children'S Hosp Of Philadelphia, Philadelphia PA

Investigators

Abstract

PROJECT 3 SUMMARY/ABSTRACT Invasive fungal disease (IFD) is a significant cause of morbidity and mortality in immunocompromised pediatric patients. Early and accurate diagnosis is crucial for improving outcomes, but current diagnostic methods have limitations. Project 3 of this this Pediatric Fungal Network STudy of Rare Invasive Fungal DisEases in Immunocompromised Pediatric Patients (PFN-STRIDE) Rare Diseases Clinical Research Consortium (RDCRC) aims to develop and validate novel, non-invasive approaches for diagnosing IFDs in pediatric patients by utilizing host response signatures and artificial intelligence (AI)-driven imaging analysis. Both aims will be accomplished using cohorts assembled from a prior NIH funded PFN study called DOMINIC (Non-Invasive Diagnosis Of Pulmonary Mold INfections in Immunocompromised Children) and from Project 1 of PFN-STRIDE. The first aim of Project 3 will develop and validate host transcriptomic signatures of IFD from banked and prospectively collected whole blood samples of at-risk immunocompromised children enrolled at participating PFN sites. Using bulk RNA sequencing and machine learning approaches, we will generate signatures that can accurately distinguish IFD from other clinical syndromes. The second aim will leverage AI-driven imaging analysis to automate detection of radiographic features consistent with pulmonary IFD in immunocompromised pediatric patients. We will develop prediction models using chest CT scans from the NIH funded PFN study DOMINIC, aiming to distinguish radiographic IFD features from other infectious or non-infectious processes. These models will be validated using images from prospectively enrolled patients on PFN-STRIDE. This project offers several innovative aspects, including: 1) the first large-scale study of host transcriptional biomarkers for IFD diagnosis in pediatric patients; 2) the development of AI tools for interpreting CT imaging in pediatric pulmonary IFD; and 3) the creation of a unique IFD biobank and imaging repository linked to detailed microbiologic and clinical data. Successful completion of this project will result in development of novel IFD diagnostic tools that can be assessed in future studies of immunocompromised pediatric patients, potentially improving time to diagnosis. The established biobank and imaging repository will serve as valuable resources for future IFD research. This work aligns with strategic priorities determined by international groups and consortia, addressing critical knowledge gaps in pediatric IFD diagnosis.