(141 days)
VIDAS® C. difficile Toxin A & B (CDAB) assay is an automated test for use on the VIDAS instruments for the qualitative detection of Clostridium difficile toxin A and toxin B in stool specimens using the ELFA technique (Enzyme-Linked Flurorescent Assay).
VIDAS® C. difficile Toxin A & B (CDAB) assay is an automated test for use on the VIDAS instruments for the qualitative detection of Clostridium difficile toxin A and toxin B in stool specimens using the ELFA technique (Enzyme-Linked Fluorescent Assay). The assay principle combines a two-step enzyme immunoassay sandwich method with a final fluorescent detection (ELFA). The Solid Phase Receptacle (SPR), a pipette tip-like device, serves as the solid phase as well as the pipetting device for the assay. The assay reagents are ready-to-use and pre-dispensed in the sealed reagent strips (STRs). Each of the four reaction steps are performed automatically by the VIDAS instrument. The reaction medium (sample/conjugate mixture) is cycled in and out of the SPR several times. Each step is followed by a wash cycle which eliminates unbound components. At the end of the VIDAS CDAB assay, results are automatically calculated by the VIDAS instrument. A test value as well as the qualitative result (positive, negative or equivocal) are provided on the result sheet for each sample.
This document describes the regulatory submission for the VIDAS® CDAB Assay, a device for detecting Clostridium difficile toxins.
1. Table of Acceptance Criteria and Reported Device Performance:
The document doesn't explicitly state "acceptance criteria" as pass/fail thresholds for numerical values. Instead, it presents performance data used to demonstrate substantial equivalence to a predicate device. The performance data is as follows:
| Item | Device [VIDAS CDAB] Performance | Predicate [Premier Toxins A&B] Performance (for comparison) |
|---|---|---|
| Non-clinical (Analytical) Comparison | ||
| Precision/Reproducibility | Total precision: 7.4 – 37.6% CV | |
| Inter-assay precision: 6.8 – 26.8% CV | ||
| Intra-assay precision: 2.9 – 26.3% CV | Within run: 4.1 – 28.9% CV | |
| Between run: 6.2 – 31.7% CV | ||
| C. difficile strain types | A+/B+: 100% (23/23) | |
| A-/B+: 83% (15/18*) |
- 3 of the A-/B+ strains gave equivocal results | A+/B+: 100% (25/25)
A-/B+: 100% (3/3) |
| Limit of Detection (stool) | Toxin A: ≥ 7.73 ng/mL
Toxin B: ≥ 4.55 ng/mL | Toxin A: ≥ 1.4 ng/mL
Toxin B: ≥ 2.4 ng/mL |
| Clinical Studies Comparison | | |
| Positive Agreement (vs Predicate) | 81.3% (95% CI: 73.4–87.6%) | N/A (Predicate is the comparator) |
| Negative Agreement (vs Predicate) | 99.5% (95% CI: 98.8 – 99.9%) | N/A |
| Global Agreement (vs Predicate) | 97.1% (95% CI: 95.9–98.1%) | N/A |
The acceptance criterion, though not quantitatively defined in a table as a "pass/fail" value, is implied by the conclusion of "substantial equivalence" to the predicate device. This suggests that the reported device performance was deemed acceptably similar or competitive to the predicate's performance across these metrics.
2. Sample Size Used for the Test Set and Data Provenance:
- Sample Size for Clinical Studies: 1011 specimens.
- Data Provenance: The clinical studies were conducted in the US and Europe. The document does not explicitly state whether the data was retrospective or prospective. Given the typical nature of device submissions, it is likely a mix or primarily prospective collection for the clinical comparison, but this is not explicitly stated.
3. Number of Experts Used to Establish the Ground Truth for the Test Set and Qualifications of Those Experts:
The document does not mention the use of experts to establish a "ground truth" for the test set in the context of human interpretation of results. The clinical study performance is evaluated as "agreement versus Predicate," meaning the results of the VIDAS CDAB assay are compared directly against the results obtained from the Meridian Premier Toxins A&B assay (the predicate device). Therefore, the predicate device's results implicitly serve as the comparative standard here, rather than a separate expert-adjudicated ground truth.
4. Adjudication Method for the Test Set:
Not applicable. As noted above, the VIDAS CDAB assay's performance was compared against the predicate device (Meridian Premier Toxins A&B assay), rather than an expert-adjudicated ground truth.
5. If a Multi Reader Multi Case (MRMC) Comparative Effectiveness Study was done, If so, what was the effect size of how much human readers improve with AI vs without AI assistance:
This is not applicable. The VIDAS® CDAB Assay is an automated in-vitro diagnostic device for the qualitative detection of Clostridium difficile toxins, not an AI-powered diagnostic tool requiring human interpretation or assistance. Therefore, an MRMC study with human readers and AI assistance would not be relevant.
6. If a Standalone (i.e. algorithm only without human-in-the-loop performance) was done:
Yes, this was a standalone performance study. The VIDAS® CDAB assay is described as an "automated test" that "automatically calculated by the VIDAS instrument." Its performance metrics (precision, strain detection, limit of detection, and agreement with the predicate) are inherently standalone without human intervention in the result determination.
7. The Type of Ground Truth Used:
For the clinical studies, the "ground truth" was effectively the results obtained from the predicate device (Meridian Premier Toxins A&B). The study's aim was to demonstrate substantial equivalence by comparing the VIDAS CDAB assay's output to the predicate's output.
For the analytical studies (precision, strain types, limit of detection), the ground truth would be based on characterized samples and known concentrations/strains used in the laboratory setting.
8. The Sample Size for the Training Set:
The document does not provide information regarding a separate "training set" for the VIDAS® CDAB assay. Given that it's described as an immunoassay (ELFA technique) rather than a machine learning or AI-driven algorithm, a distinct "training set" in the context of algorithm development is not typically applicable in the same way. The assay development would involve extensive R&D and optimization, but not generally a discrete "training set" as understood in AI/ML.
9. How the Ground Truth for the Training Set was Established:
Not applicable, as a discrete training set for an algorithm is not described or implied for this immunoassay device. The "ground truth" during the development and optimization phases would be based on well-characterized laboratory samples with known C. difficile toxin presence and concentrations, similar to how analytical performance metrics are established.
§ 866.2660 Microorganism differentiation and identification device.
(a)
Identification. A microorganism differentiation and identification device is a device intended for medical purposes that consists of one or more components, such as differential culture media, biochemical reagents, and paper discs or paper strips impregnated with test reagents, that are usually contained in individual compartments and used to differentiate and identify selected microorganisms. The device aids in the diagnosis of disease.(b)
Classification. Class I (general controls). The device is exempt from the premarket notification procedures in subpart E of part 807 of this chapter subject to § 866.9.