2084. Performance of the Karius Plasma Next Generation Sequencing Test in Determining the Etiologic Diagnosis of Febrile Neutropenia: Results from a Pilot Study
Session: Poster Abstract Session: Diagnostics - Novel Diagnostics
Saturday, October 7, 2017
Room: Poster Hall CD
Background:  Blood cultures (BC) fail to detect a pathogen in most patients with neutropenic fever (NF). We examined the performance of the Karius next generation sequencing plasma test (plasma NGS) compared to that of BC in chemotherapy-induced NF.

Methods:  Patients 18 years or older with absolute neutrophil count <500 cells/mm3 anticipated for > 7 days were enrolled at the time of BC collection (T0) due to fever. Plasma samples were collected at T0 and twice weekly until neutrophil recovery or discharge. Samples were shipped to the Karius CLIA/CAP laboratory (Redwood City, CA) where cell-free plasma was prepared, DNA extracted, and NGS performed. After removing human reads, remaining sequences were aligned to a curated pathogen database. Organisms present at a significance- level above a predefined threshold were reported. T0 BC and plasma NGS results were compared, excluding organisms (eg, viruses) not recoverable by BC. Positive agreement was defined as plasma NGS identification of at least one isolate seen on BC. Plasma NGS+/BC- results underwent adjudication by 3 infectious diseases specialists. Diagnosis was Definite if microbiology confirmed NGS result within 7 days of enrollment; Probable if clinical, radiologic, and laboratory data were compatible with plasma NGS test result; and otherwise Indeterminate.

Results:  The first 18 patients were evaluated. At the time of enrollment, 6 patients were on antibiotics and 3 were on antifungals. There were 4 positive BC (each monomicrobial) at T0 and all were concordant with plasma NGS. The plasma NGS test identified additional organisms in 2 of the BC+ patients with surgical abdomen. Compared to BC, plasma NGS positive agreement was 100% (4/4) with lack of agreement in 43% (6/14). All 8 NGS+/BC- cases were deemed Probable by adjudication. These included patients with single or mixed organisms in patients with enterocolitis or severe mucositis. In one sample, Aspergillus fumigatus was detected in a patient with new lung nodules.

Conclusion:  The etiologic diagnosis of NF is frequently unknown, leading to broad antibiotics and sometimes delay of targeted treatment. Plasma NGS may provide useful data for managing NF given its ability to detect a breadth of pathogens even when patients are pretreated with antibiotics.

Esther Benamu, MD1, Kiran Gajurel, MD2, Jill N. Anderson, BA3, Hon Seng, BS4, Desiree Hollemon, MSN, MPH4, David Hong, MD4, Timothy A. Blawkamp, PhD5, Mickey Kertesz, PhD4, Heather Ishak, MS3, Paul L. Bollyky, MD, PhD3, Bruno C. Medeiros, MD6, Steven Coutre, MD6, Jose G. Montoya, MD, FIDSA7 and Stan Deresinski, MD, FIDSA7, (1)Division of Infectious Diseases and Geographic Medicine,Department of Medicine, Stanford University School of Medicine, Stanford, CA, (2)Division of Infectious Diseases, Department of Medicine, University of Iowa Carver College of Medicine, Iowa City, IA, (3)Division of Infectious Diseases and Geographic Medicine, Department of Infectious Diseases, Stanford University School of Medicine, Stanford, CA, (4)Karius, Inc., Redwood City, CA, (5)Karius Inc., Redwood City, CA, (6)Division of Hematology, Department of Medicine, Stanford University School of Medicine, Stanford, CA, (7)Division of Infectious Diseases and Geographic Medicine, Department of Medicine, Stanford University School of Medicine, Stanford, CA

Disclosures:

E. Benamu, None

K. Gajurel, None

J. N. Anderson, None

H. Seng, Karius, Inc: Employee , Salary

D. Hollemon, Karius, Inc: Employee , Salary

D. Hong, Karius, Inc.: Collaborator , Research support and Salary

T. A. Blawkamp, Karius Inc.: Board Member and Employee , Salary

M. Kertesz, Karius Inc.: Board Member and Employee , Salary

H. Ishak, None

P. L. Bollyky, None

B. C. Medeiros, None

S. Coutre, None

J. G. Montoya, None

S. Deresinski, None

Findings in the abstracts are embargoed until 12:01 a.m. PDT, Wednesday Oct. 4th with the exception of research findings presented at the IDWeek press conferences.