Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2024 Oct 10;13(20):6044.
doi: 10.3390/jcm13206044.

Rapid and Robust Identification of Sepsis Using SeptiCyte RAPID in a Heterogeneous Patient Population

Affiliations

Rapid and Robust Identification of Sepsis Using SeptiCyte RAPID in a Heterogeneous Patient Population

Robert Balk et al. J Clin Med. .

Abstract

Background/Objective: SeptiCyte RAPID is a transcriptional host response assay that discriminates between sepsis and non-infectious systemic inflammation (SIRS) with a one-hour turnaround time. The overall performance of this test in a cohort of 419 patients has recently been described [Balk et al., J Clin Med 2024, 13, 1194]. In this study, we present the results from a detailed stratification analysis in which SeptiCyte RAPID performance was evaluated in the same cohort across patient groups and subgroups encompassing different demographics, comorbidities and disease, sources and types of pathogens, interventional treatments, and clinically defined phenotypes. The aims were to identify variables that might affect the ability of SeptiCyte RAPID to discriminate between sepsis and SIRS and to determine if any patient subgroups appeared to present a diagnostic challenge for the test. Methods: (1) Subgroup analysis, with subgroups defined by individual demographic or clinical variables, using conventional statistical comparison tests. (2) Principal component analysis and k-means clustering analysis to investigate phenotypic subgroups defined by unique combinations of demographic and clinical variables. Results: No significant differences in SeptiCyte RAPID performance were observed between most groups and subgroups. One notable exception involved an enhanced SeptiCyte RAPID performance for a phenotypic subgroup defined by a combination of clinical variables suggesting a septic shock response. Conclusions: We conclude that for this patient cohort, SeptiCyte RAPID performance was largely unaffected by key variables associated with heterogeneity in patients suspected of sepsis.

Keywords: SIRS; SeptiCyte; host immune response; phenotype; sepsis; sepsis likelihood; stratification.

PubMed Disclaimer

Conflict of interest statement

K.N., T.D.Y., D.S., J.T.K., S.C., R.F.D. and R.B.B. declare they are present or past employees or shareholders of Immunexpress, Inc. X.W.M. declares that she is a present employee of Princeton Pharmatech, Inc. R.R.M. III discloses that, over the time period of relevance to this work he was paid (as a consultant) an honorarium by Immunexpress for conducting a critical review of a Clinical Evaluation Report of SeptiCyte technology, submitted by Immunexpress to the Therapeutic Goods Administration (TGA) of Australia. N.R.A. declares that he has received grants from the NIH and Department of Defense for non-related work and that payments from these grants are made to his institution, the University of Colorado. N.R.A. also declares that he is a committee member of the Pfizer Paxlovid U.S. Medical Advisory Committee formed to broadly discuss COVID-19-related therapeutic priorities and populations of interest. B.K.L. declares that he is a consultant for Karius Inc., is a member of the Scientific Advisory Board for Seegene Inc., and has received honoraria for speaking for Qiagen Inc. No other competing interests are declared.

Figures

Figure 1
Figure 1
Box and whisker plots of SeptiCyte RAPID performance in patients with (A) hyperglycemia, (B) impaired immunity, (C) hypertension, (D) cardiovascular disease, (E) kidney disease, and (F) obesity. The dotted horizontal lines indicate the SeptiScore band boundaries as defined by Balk et al. [16].
Figure 2
Figure 2
SeptiCyte RAPID performance stratified by infection source. The “other” group included the following (N per group): cellulitis (3), Fournier’s gangrene (1), hip arthroplasty (1), osteomyelitis (1), toe infection (1), sacral wound (1), post-surgical sternal wound (1), bladder/prostate abscess/peritonitis from cecum microperforation (1), influenza (1), tracheitis (1), erysipelas (1), skin or soft tissue necrotizing fasciitis (1). The dotted horizontal lines indicate the SeptiScore band boundaries as defined by Balk et al. [16]. Significance: p <= 0.01 **, p <= 0.001 ***.
Figure 3
Figure 3
Box and whisker plot of SeptiCyte RAPID performance for patients treated vs. not treated with immunosuppressants. The dotted horizontal lines indicate the SeptiScore band boundaries as defined by Balk et al. [16].
Figure 4
Figure 4
Box and whisker plots of the influence of antibiotic treatment (Abx) initiation time on SeptiCyte RAPID performance. The day of blood draw is defined as day 0. (left) treatment initiated −1 day to 0 days relative to blood draw; (middle) antibiotic treatment initiated on the same day as blood draw. (right) antibiotic treatment initiated +1 day after blood draw. The dotted horizontal lines indicate the SeptiScore band boundaries as defined by Balk et al. [16].
Figure 5
Figure 5
PCA plot of the sepsis group (N = 176) with superimposed HC using 16 phenotypic variables. In the plot, the peripheral points in each subgroup were used to define the cluster boundaries for that subgroup. Sepsis subgroup 1 (black) N = 110. Sepsis subgroup 2 (red) N = 50. Sepsis subgroup 3 (green) N = 15. Sepsis subgroup 4 (blue) N = 1.
Figure 6
Figure 6
Performance for discriminating sepsis vs. SIRS in different phenotypic subgroups defined by individual driving variables in PCA/HC and other biomarkers used for sepsis adjudication. Red: SeptiScore. Black: driving variable. Additional details corresponding to this figure are presented in Supplement Table S5.
Figure 7
Figure 7
Sepsis subgroups 1 (red; N = 96) and 2 (blue; N = 80) identified by k-means clustering. The small, most seriously ill subgroup 3 from the PCA/HC analysis (black; N = 15) is contained entirely within k-means group 2. The large colored symbols indicate the centroids of the subgroups.

References

    1. Liu V.X., Bhimarao M., Greene J.D.M., Manickam R.N., Martinez A., Schuler A., Barreda F.M., Escobar G.J. The Presentation, Pace, and Profile of Infection and Sepsis Patients Hospitalized Through the Emergency Department: An Exploratory Analysis. Crit. Care Explor. 2021;3:e0344. doi: 10.1097/CCE.0000000000000344. - DOI - PMC - PubMed
    1. Lengquist M., Varadarajan A., Alestam S., Friberg H., Frigyesi A., Mellhammar L. Sepsis mimics among presumed sepsis patients at intensive care admission: A retrospective observational study. Infection. 2024;52:1041–1053. doi: 10.1007/s15010-023-02158-w. - DOI - PMC - PubMed
    1. Seymour C.W., Kennedy J.N., Wang S., Chang C.-C.H., Elliott C.F., Xu Z., Berry S., Clermont G., Cooper G., Gomez H., et al. Derivation, Validation, and Potential Treatment Implications of Novel Clinical Phenotypes for Sepsis. JAMA. 2019;321:2003–2017. doi: 10.1001/jama.2019.5791. - DOI - PMC - PubMed
    1. Fohner A.E., Greene J.D., Lawson B.L., Chen J.H., Kipnis P., Escobar G.J., Liu V.X. Assessing clinical heterogeneity in sepsis through treatment patterns and machine learning. J. Am. Med. Inform. Assoc. 2019;26:1466–1477. doi: 10.1093/jamia/ocz106. - DOI - PMC - PubMed
    1. Reddy K., Sinha P., O’Kane C.M., Gordon A.C., Calfee C.S., McAuley D.F. Subphenotypes in critical care: Translation into clinical practice. Lancet Respir. Med. 2020;8:631–643. doi: 10.1016/S2213-2600(20)30124-7. - DOI - PubMed

LinkOut - more resources