(77 days)
The Roche ONLINE Phenytoin assay contains an in vitro diagnostic reagent system indicated for the quantitative determination of phenytoin, an anti-convulsant drug, in human serum or plasma on automated clinical chemistry analyzers. Measurements obtained by the device are used in the diagnosis and treatment of phenytoin overdose and in monitoring levels of phenytoin to ensure appropriate therapy.
The Roche ONLINE Phenytoin assay contains an in vitro diagnostic reagent system indicated for the quantitative determination of phenytoin, an anti-convulsant drug, in human serum or plasma on automated clinical chemistry analyzers. Measurements obtained by the device are used in the diagnosis and treatment of phenytoin overdose and in monitoring levels of phenytoin to ensure appropriate therapy. The proposed labeling indicates the Roche/Hitachi 911, 912, 917, and Modular P analyzers can be used with the Roche ONLINE Phenytoin reagent kits.
Acceptance Criteria and Study for Roche ONLINE Phenytoin Assay
1. Table of Acceptance Criteria and Reported Device Performance
The provided document describes a comparison to a predicate device (Roche CEDIA Phenytoin II Assay), not a study against pre-defined acceptance criteria for a novel device. The "acceptable results compared to the predicate device" serves as the implicit acceptance criterion. The key performance metrics reported are for precision and method comparison.
| Performance Metric | Acceptance Criteria (Implicit - Comparison to Predicate) | Roche ONLINE Phenytoin Assay Performance | Roche CEDIA Phenytoin II Assay Performance (Predicate) |
|---|---|---|---|
| Method Comparison | To be substantially equivalent to the predicate device when compared against the Abbott TDx Phenytoin Assay. | Y = 0.99X - 0.865 (vs Abbott TDx) | Y = 0.99X - 1.43 (vs Abbott TDx) |
| Correlation Coefficient (R) | To be substantially equivalent to the predicate device. | R = 0.992 (vs Abbott TDx) | R = 0.993 (vs Abbott TDx) |
| Measurement Range | To be substantially equivalent to the predicate device. | 1.68 to 40.0 µg/mL | 1.1 to 40.0 µg/mL |
| Precision (Level 1: Mean) | To be substantially equivalent to the predicate device. | 6.49 µg/mL | 6.3 µg/mL |
| Precision (Level 1: CV% within run) | To be substantially equivalent to the predicate device. | 2.7% | 3.2% |
| Precision (Level 1: CV% total) | To be substantially equivalent to the predicate device. | 6.1% | 5.1% |
| Precision (Level 2: Mean) | To be substantially equivalent to the predicate device. | 13.39 µg/mL | 14.8 µg/mL |
| Precision (Level 2: CV% within run) | To be substantially equivalent to the predicate device. | 1.4% | 2.0% |
| Precision (Level 2: CV% total) | To be substantially equivalent to the predicate device. | 4.5% | 3.1% |
| Precision (Level 3: Mean) | To be substantially equivalent to the predicate device. | 22.87 µg/mL | 26.8 µg/mL |
| Precision (Level 3: CV% within run) | To be substantially equivalent to the predicate device. | 1.5% | 1.3% |
| Precision (Level 3: CV% total) | To be substantially equivalent to the predicate device. | 4.3% | 2.3% |
2. Sample Size Used for the Test Set and Data Provenance
- Sample Size for Test Set (Method Comparison):
- Roche ONLINE Phenytoin vs. Abbott TDx Phenytoin Assay: N = 106
- Roche CEDIA Phenytoin II vs. Abbott TDx Phenytoin Assay: N = 108
- Data Provenance: The document does not specify the country of origin of the data. It also does not explicitly state whether the data was retrospective or prospective, though clinical diagnostic assay comparisons are typically prospective clinical performance evaluations or studies on patient samples.
3. Number of Experts Used to Establish the Ground Truth for the Test Set and Their Qualifications
This is a clinical chemistry assay for quantitative determination of phenytoin. Ground truth for such assays is typically established by comparing the device's results to a recognized, accurate reference method or a legally marketed predicate device. In this case, the Abbott TDx Phenytoin Assay served as the reference method against which both the new device and the predicate device were compared. No "experts" in the sense of physicians reviewing images or cases are usually involved in establishing ground truth for quantitative laboratory assays; the ground truth is the result from the reference method.
4. Adjudication Method for the Test Set
Not applicable. This is a quantitative laboratory assay comparison, not a diagnostic imaging or qualitative assessment requiring adjudication of results from multiple observers. The comparison is against a reference assay.
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
Not applicable. This is a clinical chemistry assay for quantitative determination of a drug in human samples, not an AI-assisted diagnostic imaging device or a device involving human readers.
6. If a Standalone (i.e. algorithm only without human-in-the-loop performance) was done
Yes, the study describes the standalone performance of the Roche ONLINE Phenytoin assay. It quantifies the level of phenytoin directly from human serum or plasma samples on automated clinical chemistry analyzers. There is no human-in-the-loop performance described for the assay itself; the human involvement is in interpreting the quantitative results for clinical decision-making.
7. The type of Ground Truth Used
The ground truth used for the method comparison study was the results obtained from the Abbott TDx Phenytoin Assay, which is a recognized method for phenytoin quantification.
8. The Sample Size for the Training Set
Not applicable. This document describes a traditional in-vitro diagnostic device (IVD) based on reagent systems, not an AI/Machine Learning algorithm that requires a training set. The "training" for such devices involves optimizing the chemical reactions and analytical parameters during development.
9. How the Ground Truth for the Training Set Was Established
Not applicable. As noted above, this is not an AI/Machine Learning device that utilizes a training set with established ground truth in the typical computational sense. The performance characteristics of the assay are established through analytical verification and validation studies.
§ 862.3350 Diphenylhydantoin test system.
(a)
Identification. A diphenylhydantoin test system is a device intended to measure diphenylhydantoin, an antiepileptic drug, in human specimens. Measurements obtained by this device are used in the diagnosis and treatment of diphenylhydantoin overdose and in monitoring levels of diphenylhydantoin to ensure appropriate therapy.(b)
Classification. Class II.