Search Results

VTEC-RPLA "SEIKEN" is an in vitro diagnostic device for the detection and identification of verotoxic device for in culture isolates of E. coli and is intended as an aid in the diagnosis of Enterohemorrhagic E. coli (EHEC) infections.

VTEC-RPLA "SEIKEN" is a reagent system which can be used for the detection of verotoxins in culture isolates of E. coli to aid in the diagnosis of EHEC infections. It employs antibody-antigen reactions as part of the assay and is based on reversed passive latex agglutination (RPLA). VTEC-RPLA utilizes polyclonal antibodies specific for each verotoxin type, VT1 and VT2, and thus may be used to identify toxins.

This document describes the VTEC-RPLA "SEIKEN" device, an in vitro diagnostic procedure for detecting and identifying verotoxins in E. coli culture isolates to aid in diagnosing Enterohemorrhagic E. coli (EHEC) infections.

Here's an analysis of the acceptance criteria and the study that proves the device meets them:

1. Table of Acceptance Criteria and Reported Device Performance

Note on Sensitivity: While the VTEC-RPLA "SEIKEN" has a higher minimum detectable concentration (25 pg/well for both VT1 and VT2) compared to the predicate device Premier EHEC (7 pg/well for VT1 and 15 pg/well for VT2), the correlation study still showed 100% sensitivity and specificity when directly comparing both devices with clinical isolates. The FDA's substantial equivalence determination implies that this difference in analytical sensitivity does not compromise clinical performance for the intended use.

2. Sample Size Used for the Test Set and Data Provenance

• Study 1 (in-house, Correlation):
  • Sample Size: 100 Escherichia coli strains (80 verotoxin-producing, 20 non-verotoxin producing).
  • Data Provenance: Obtained from the Japanese National Institute of Infectious Disease. Retrospective (as strains were obtained and then tested).
• Study 2 (external, Correlation):
  • Sample Size: 178 Escherichia coli strains (147 verotoxin-producing, 31 non-verotoxin producing) isolated from humans.
  • Data Provenance: Tested by the Tokyo Metropolitan Research Laboratory of Public Health. Retrospective.
• Sensitivity: 2 verotoxin-producing strains (DK-EC-VT1 and DK-EC-VT2).
• Specificity: 8 verotoxin-producing and 12 non-producing strains of E. coli.
• Reproducibility: Same 2 verotoxin-producing strains as Sensitivity study.
• Dilution: 3 verotoxin-producing strains (DK-EC-PS 1 (VT1), DK-EC- PS 4 (VT2), and DK-EC-PS 7 (VT1 and VT2)).
• Stability: Same 3 reagent lots as Sensitivity study, tested over time.

3. Number of Experts Used to Establish the Ground Truth for the Test Set and Qualifications of Those Experts

The document does not specify the number or qualifications of experts used to establish the ground truth for the test set.

However, for the external correlation study (Study 2), the ground truth for the 178 E. coli strains was established by parallel testing using:

• VTEC-RPLA "SEIKEN"
• PCR (Polymerase Chain Reaction)
• Vero cell assay

The Vero cell assay is a widely accepted method for detecting verotoxicity. PCR is also a molecular method often considered a gold standard for pathogen detection and characterization. The document implies these methods collectively served as the reference for determining whether a strain was "verotoxin-producing" or "non-verotoxin producing."

4. Adjudication Method for the Test Set

The document does not explicitly describe an adjudication method for conflicting results if they occurred. However, given that both correlation studies reported 100% sensitivity and 100% specificity for VTEC-RPLA "SEIKEN" compared to the predicate (Study 1) and compared to PCR and Vero cell assay (Study 2), it suggests there were no conflicting results or that discrepancies were resolved in a manner not detailed here. The ground truth seems to be established by the consensus of the reference methods (for Study 2) or the predicate device (for Study 1).

5. If a Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study Was Done

No, a Multi-Reader Multi-Case (MRMC) comparative effectiveness study was not performed. This device is an in-vitro diagnostic (IVD) reagent system, not an imaging device or algorithm requiring human interpretation. Therefore, the concept of "human readers improving with AI vs. without AI assistance" is not applicable here.

6. If a Standalone (i.e., algorithm only without human-in-the-loop performance) Was Done

Yes, the studies conducted are essentially standalone performance evaluations of the VTEC-RPLA "SEIKEN" device. It is a laboratory-based assay where the "performance" is the device's ability to detect and identify verotoxins. There is no "human-in-the-loop" component in the sense of an algorithm providing assistance for human interpretation; the device generates a result directly.

7. The Type of Ground Truth Used

• Study 1 (Correlation): The ground truth for the 100 E. coli strains was established by comparison with the Premier EHEC Reagent System, a legally marketed predicate device.
• Study 2 (Correlation): The ground truth for the 178 E. coli strains was established using a combination of PCR and the Vero cell assay. The Vero cell assay can be considered a form of biological activity assay or functional assay, which approximates outcomes data in terms of cellular toxicity, and PCR provides definitive genetic identification of toxin genes.
• Sensitivity, Specificity, Reproducibility, Dilution: The ground truth involved using known verotoxin-producing and non-producing strains, and purified verotoxin preparations with defined concentrations. This is based on established expert classification through prior characterization of these strains and preparations.

8. The Sample Size for the Training Set

The document does not mention a "training set" in the context of machine learning or AI development. This is an in-vitro diagnostic test, not an AI/ML-based device. The studies described are validation and verification studies for its performance.

9. How the Ground Truth for the Training Set Was Established

Since there is no "training set" for an AI/ML model, this question is not applicable. The document describes how the reference materials (strains, purified toxins) used in the verification studies were characterized, as outlined in point 7.

Ask a specific question about this device