2886. Retrospective Case–Control Study of the Performance of the Karius Test, a Plasma Microbial Cell-free DNA Next-generation Sequencing test, to Detect Invasive Mold Infections in Hematopoietic Cell Transplant Recipients with Pneumonia

BACKGROUND: Pulmonary invasive mold infections (IMI) cause significant morbidity and mortality after hematopoietic cell transplant (HCT). Noninvasive diagnostic options are limited, particularly for non-Aspergillus (non-Asp) molds. Given differences in activity and toxicities of antifungals, early diagnosis and targeting of specific IMI is important. We evaluated the performance of the Karius Test, a plasma microbial cell-free DNA next-generation sequencing (NGS) test, for detecting IMI in HCT recipients. METHODS: We conducted a retrospective case–control study of 24 HCT recipients with proven non-Asp pulmonary IMI, 51 probable/proven Aspergillus pulmonary IMI, and 20 controls with nonfungal pulmonary infections. All subjects had plasma obtained within 14 days of diagnosis. Workup included bronchoalveolar lavage (BAL) and/or biopsy, with fungal stains/culture and galactomannan testing of BAL and serum. Plasma cell-free DNA was extracted and NGS performed (Karius, Redwood City, CA). Human reads were removed and remaining sequences aligned to a curated database including over 300 fungi. Organisms present above a predefined significance threshold were reported. A higher sensitivity research-use only pipeline was also used. Analysis of sequencing data was blinded to all clinical data. RESULTS: We identified pathogenic molds in 19/24 (79%) of subjects with proven non-Asp IMI, including Mucor, Rhizomucor, Scedosporium, Rhizopus, and Cunninghamella spp. In Aspergillus proven/probable IMI, A. fumigatus was identified in 13.7% (7/51) of subjects. In 3 other subjects with proven/probable aspergillosis, we also identified Rhizomucor miehei and R.pusillus, and Cunninghamella, consistent with clinical findings. The use of an optimal-sensitivity pipeline identified an additional 9 subjects with Aspergillus spp. and other pathogenic molds, increasing detection of molds to 37.3% (19/51). Specificity for molds in negative samples was 100% (20/20). CONCLUSION: The Karius plasma NGS test is a noninvasive means of detecting IMI with high sensitivity for non-Aspmolds in HCT recipients with pulmonary disease. Further assay optimization may increase sensitivity for Aspergillus. This may be a useful adjunctive test for diagnosing IMI, and larger studies should be conducted. DISCLOSURES: All Authors: No reported Disclosures.