(134 days)
The OSOM C. difficile Toxin A/B Test is an immunochromatographic assay intended for the qualitative detection of Clostridium difficile toxins A and/or B in human stool samples. This test is intended as an aid in the diagnosis of C. difficile-associated disease (CDAD) in patients with symptoms of CDAD.
The OSOM C. difficile Toxin A/B Test is a rapid test which can detect the presence of Clostridium difficile toxins A and B in human stool samples. A test kit contains 25 OSOM test stick devices and 25 disposable pipettes. The OSOM C. difficile Toxin A/B Test is a qualitative assay that employs immunochromatographic, dipstick technology. The test format is a sandwich immunoassay, with a single test zone on the nitrocellulose dipstick to detect Toxin A and/or Toxin B ("blue/gray" line) and a single control line zone to indicate proper sample flow ("red" line). The test procedure involves binding of C. difficile Toxin A and/or Toxin B from a patient stool sample to blue colored latex particles conjugated to a monoclonal antibody against Toxin B or a polyclonal antibody against Toxin A. When Toxin A and/or B is present in the sample, it will form a partial immune complex with the antibody-conjugated colored particles. The OSOM C. difficile Toxin A/B Test stick, when placed in the sample mixture, initiates sample migration along the nitrocellulose membrane. If C. difficile toxin A or toxin B is present, a blue/gray line will appear in the test line region indicating a positive result. A red control line must appear for the results to be valid. If C. difficile toxins are not present, only the red control line will appear. An invalid test occurs when no control line appears.
Here's a breakdown of the acceptance criteria and the study that proves the device meets them, based on the provided text:
Acceptance Criteria and Reported Device Performance
| Performance Metric | Acceptance Criteria (Implicit) | Reported Device Performance |
|---|---|---|
| Clinical Study 1 (Overall) | Adequate sensitivity, specificity, and agreement compared to cytotoxicity assay for diagnostic aid. | |
| Sensitivity | Sufficiently high to detect positive cases. | 88.2% (95% CI, 80.5 - 93.1%) |
| Specificity | Sufficiently high to correctly identify negative cases. | 96.8% (95% CI, 95.7 - 97.7%) |
| Agreement | Overall high agreement with the reference method. | 96.2% (95% CI, 95.1 - 96.9%) |
| Clinical Study 2 (Comparison to Predicate) | Performance of the new device should be comparable to or better than the predicate device. | |
| OSOM Sensitivity | Comparable to predicate. | 87.5% (95% CI, 75.3 - 94.1%) |
| OSOM Specificity | Comparable to predicate. | 90.1% (95% CI, 85.2 - 93.5%) |
| Predicate Sensitivity | Baseline for comparison. | 70.8% (95% CI, 56.8 - 81.8%) |
| Predicate Specificity | Baseline for comparison. | 97.5% (95% Cl, 94.3 - 98.9%) |
| Analytical Sensitivity | Ability to detect C. difficile toxins at specified concentrations. | Detected 15 ng/mL for Toxin A and 40 ng/mL for Toxin B. |
| Cross-Reactivity | No false positives with common bacteria and viruses typically found in clinical samples. | All listed bacteria and viruses gave negative responses. |
| Interfering Substances | No effect on performance from common substances found in stool or used by patients. | No effect from the 11 tested substances. |
| Reproducibility | Consistent results across different operators, sites, and testing days. | Expected result 99.2% (357/360) of the time. |
Study Details
-
Sample sizes used for the test set and the data provenance:
- Clinical Study 1 (Overall performance): 1274 de-identified excess loose or watery stool specimens.
- Provenanc: Obtained from patients at five sites in the United States.
- Nature: Retrospective (specimens submitted for CDAD testing).
- Clinical Study 2 (Comparison to Predicate): 250 paired sample results (OSOM and Predicate Device).
- Provenance: Not explicitly stated, but implies from a clinical trial site, likely within the US given the overall study context.
- Nature: Retrospective (using existing clinical trial samples).
- Analytical Sensitivity: Not reported as a sample size, but involved serial dilutions prepared from purified toxins.
- Cross-Reactivity: Not reported as a numerical sample size, but tested against a panel of 28 bacterial isolates and 6 viral isolates.
- Interfering Substances: Not reported as a numerical sample size, but tested against 11 potential interferents.
- Reproducibility: 360 total test results (12 samples x 2 operators/day x 5 days x 3 sites = 360)
- Clinical Study 1 (Overall performance): 1274 de-identified excess loose or watery stool specimens.
-
Number of experts used to establish the ground truth for the test set and the qualifications of those experts:
- The ground truth for the clinical studies was established using a cytotoxicity assay. The text does not specify the number of experts or their qualifications for performing this assay, beyond stating it was "performed at Genzyme using a standard C. difficile toxin cytotoxicity assay." It notes that "Those performing the cytotoxicity assay testing were blinded to the OSOM test results," which indicates an effort to reduce bias.
-
Adjudication method for the test set:
- No explicit adjudication method is described for the primary clinical study. The cytotoxicity assay served as the direct reference standard.
- Discrepant testing was performed for some cases:
- 12 cytotoxin positive, OSOM negative cases were tested with PCR for tcdB gene.
- 37 cytotoxin negative, OSOM positive cases were tested with PCR for tcdB gene (one sample not available).
- This suggests a form of post-hoc analysis for discordant results, but not a formal adjudication process to establish the initial ground truth for all samples.
-
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:
- No. This study is not an MRMC comparative effectiveness study involving human readers and AI. It is a study comparing an immunoassay device (OSOM C. difficile Toxin A/B Test) to a laboratory reference method (cytotoxicity assay) and a predicate device.
-
If a standalone (i.e. algorithm only without human-in-the loop performance) was done:
- Yes, for the OSOM C. difficile Toxin A/B Test. The study evaluates the performance of the OSOM C. difficile Toxin A/B Test, which is a rapid immunochromatographic assay device, as a standalone diagnostic tool. Its performance is assessed directly against the ground truth (cytotoxicity assay) without human interpretation in a human-in-the-loop scenario. The device itself produces the qualitative "positive" or "negative" result.
-
The type of ground truth used:
- The primary ground truth used for the clinical studies was a cytotoxicity assay, which detects biologically active C. difficile toxins.
- For discrepant analysis, a PCR-based molecular method (tcdB gene detection) was used as a secondary reference.
-
The sample size for the training set:
- The document does not report a separate training set size. This is common for traditional in vitro diagnostic (IVD) devices like this immunochromatographic assay, which are based on biochemical reactions and antibody binding, rather than machine learning algorithms that typically require extensive training data. The "training" in this context would be the development and optimization of the assay itself.
-
How the ground truth for the training set was established:
- As there is no explicit "training set" in the context of machine learning, there's no ground truth established for it. The assay's performance characteristics (e.g., analytical sensitivity, cross-reactivity) are established using purified toxins or cultured microorganisms, which serve as known positive and negative controls during the assay development and validation phases.
§ 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.