Detection of C. difficile toxin as a model assay for performing fully automated high-throughput RT-PCR on clinical stool samples

Ulrich Eigner; Dominik Nörz; Svenja Reucher; Jan Furrer; Jingtao Sun; Kristina Chu; Melissa Kolb; Nadine Hefner; Susanne Pfefferle; Marc Lütgehetmann

doi:10.1016/j.mimet.2020.105882

Detection of C. difficile toxin as a model assay for performing fully automated high-throughput RT-PCR on clinical stool samples

J Microbiol Methods. 2020 May:172:105882. doi: 10.1016/j.mimet.2020.105882. Epub 2020 Feb 28.

Authors

Affiliations

¹ MVZ Laboratory Dr. Limbach, Heidelberg, Germany.
² Center for Experimental Medicine, Institute of Medical Microbiology, Virology and Hygiene, Hamburg, Germany.
³ Roche Diagnostics International AG, Rotkreuz, Switzerland.
⁴ Roche Molecular Systems Inc, Pleasanton, CA, USA.
⁵ Center for Experimental Medicine, Institute of Medical Microbiology, Virology and Hygiene, Hamburg, Germany. Electronic address: mluetgeh@uke.de.

PMID: 32119956
DOI: 10.1016/j.mimet.2020.105882

Abstract

Background: The cobas® omni Utility Channel enables users to integrate lab-developed tests (LDTs) on the cobas® 6800 System to perform molecular diagnostics with high-throughput capacity and full automation. At present, there are no CE- or FDA-approved tests for stool pathogens on this system. To assess the performance of stool as a matrix, we evaluated the analytical and clinical performance of an LDT for detection of Clostridioides difficile (C. difficile) toxin B using the Utility Channel (C.diff_UTC).

Methods: A 10% stool suspension prepared from liquid stool samples diluted in phosphate buffered saline was used for analysis. Limit of detection (LoD) was determined in six dilutions with 126 replicates/dilution. Clinical evaluation was performed using 514 predetermined patient stool samples from two study sites in Germany. The C.diff_UTC was compared with LC 480 amplification and an LDT or the R-BioPharm C. difficile assay. Discrepant results were further analyzed using the GeneXpert C. difficile assay.

Results: Limit of detection was 23.48 cfu/mL (95% Confidence Interval [CI]: 19.14-31.01) with inter-run variation of <2 cycle thresholds at 3 × and 10 × LoD. No cross-reactivity was observed with a panel of fecal organisms and pathogens. Bioinformatic analysis showed coverage of the major C. difficile toxinotypes by the primer/probe set. Clinical evaluation revealed sensitivity of 96.7% (95% CI: 88.7-99.6) and specificity of 99.3% (95% CI: 98.0-99.9) compared with the reference method; inhibition rate was 3.5% (18/514).

Conclusion: Using a predesigned primer/probe set, the C.diff_UTC assay features analytical performance and clinical sensitivity and specificity comparable to currently available nucleic acid amplification tests (NAATs) and is suitable for high-throughput testing. This was a proof-of-concept study, indicating the cobas Utility Channel could likely be adapted for other clinically relevant stool pathogens in outbreak scenarios.

Keywords: Clostridioides difficile; Clostridium difficile; Cobas 6800; High-throughput diagnostic; Hospital-acquired gastroenteritis.

Publication types

Research Support, Non-U.S. Gov't

MeSH terms

Bacterial Proteins / genetics
Bacterial Proteins / isolation & purification*
Bacterial Toxins / genetics
Bacterial Toxins / isolation & purification*
Clostridioides difficile / genetics*
Clostridioides difficile / isolation & purification*
Clostridium Infections / diagnosis*
Clostridium Infections / microbiology
Feces / chemistry*
Germany
Humans
Nucleic Acid Amplification Techniques
Reverse Transcriptase Polymerase Chain Reaction / methods*
Sensitivity and Specificity

Substances

Bacterial Proteins
Bacterial Toxins
toxB protein, Clostridium difficile