(117 days)
Immunoturbidometric assay for the in vitro quantitative determination of CRP in human serum and plasma on Roche automated clinical chemistry analyzers.
Measurement of c-reactive protein aids in the evaluation of the amount of injury to body tissues.
The C-Reactive Protein Gen 3 assay is a particle enhanced turbidimetric assay. Human CRP agglutinates with latex particles coated with monoclonal anti-CRP antibodies. The precipitate is determined turbidimetrically at 570 nm.
Here's a breakdown of the acceptance criteria and study information for the Tina-Quant C-Reactive Protein Gen. 3 device, based on the provided text:
Acceptance Criteria and Device Performance
The provided document describes modifications to an existing device (Tina-Quant C-Reactive Protein Gen 3) and claims substantial equivalence to its predicate device (Tina-Quant C-Reactive Protein (Latex), K032336). Therefore, the acceptance criteria are implicitly defined by demonstrating comparable core performance characteristics to the predicate device. The information below presents the modified device's performance alongside the predicate's performance for comparison, which serves as the "acceptance criteria" through the lens of substantial equivalence.
Table of Acceptance Criteria (Predicate Performance) and Reported Device Performance (Modified Device)
| Feature | Acceptance Criteria (Predicate Device Performance - K032336) | Reported Device Performance (Modified Device - K082444) |
|---|---|---|
| Measuring Range | Roche/Hitachi 902: 1-265 mg/L | |
| Roche/Hitachi 717/Modular D: 1-265 mg/L, 1-398 mg/L with rerun | ||
| Roche/Hitachi 904/911/912: 1-260 mg/L, 1-520 mg/L with rerun | ||
| Roche/Hitachi 917/Modular P: 1-280 mg/L, 1-560 mg/L with rerun | Roche/Hitachi 901/912/917/Modular P/Modular D analyzers: 0.3-350 mg/L. Dilution of samples via the rerun function is a 1:2 dilution. | |
| Precision (Within Run) | Control 1: Mean (mg/L) 3.36, SD (mg/L) 0.09, %CV 2.76 | |
| Control 2: Mean (mg/L) 22.17, SD (mg/L) 0.44, %CV 1.96 | ||
| Control 3: Mean (mg/L) 51.12, SD (mg/L) 0.90, %CV 1.77 | ||
| H Pool 1: Mean (mg/L) 5.76, SD (mg/L) 0.14, %CV 2.50 | ||
| H Pool 2: Mean (mg/L) 150.15, SD (mg/L) 1.14, %CV 0.76 | Control 1: Mean (mg/L) 3.6, SD (mg/L) 0.03, %CV 0.85 | |
| Control 2: Mean (mg/L) 42.2, SD (mg/L) 0.26, %CV 0.61 | ||
| H Pool 1: Mean (mg/L) 0.9, SD (mg/L) 0.03, %CV 4.00 | ||
| H Pool 2: Mean (mg/L) 1.6, SD (mg/L) 0.02, %CV 1.02 | ||
| H Pool 3: Mean (mg/L) 18.4, SD (mg/L) 0.09, %CV 0.48 | ||
| Precision (Between Run) | Control 1: Mean (mg/L) 3.51, SD (mg/L) 0.16, %CV 4.61 | |
| Control 2: Mean (mg/L) 22.01, SD (mg/L) 0.62, %CV 2.81 | ||
| Control 3: Mean (mg/L) 50.41, SD (mg/L) 0.94, %CV 1.86 | ||
| H Pool 1: Mean (mg/L) 5.99, SD (mg/L) 0.15, %CV 2.53 | ||
| H Pool 2: Mean (mg/L) 146.31, SD (mg/L) 2.63, %CV 1.80 | Control 1: Mean (mg/L) 3.1, SD (mg/L) 0.08, %CV 2.7 | |
| Control 2: Mean (mg/L) 41.4, SD (mg/L) 0.86, %CV 2.1 | ||
| H Pool 1: Mean (mg/L) 0.5, SD (mg/L) 0.03, %CV 6.2 | ||
| H Pool 2: Mean (mg/L) 1.5, SD (mg/L) 0.05, %CV 3.3 | ||
| H Pool 3: Mean (mg/L) 39.1, SD (mg/L) 0.73, %CV 1.9 | ||
| Analytical Sensitivity | Functional Sensitivity: 0.88 mg/L | |
| Lower Detection Limit: 0.425 mg/L | Limit of Quantitation (Functional Sensitivity): 0.6 mg/L | |
| LoB: 0.2 mg/L | ||
| LoD: 0.3 mg/L | ||
| Interferences | Icterus: No significant interference up to 60 mg/dL | |
| Hemolysis: No significant interference up to 950 mg/dL | ||
| Lipemia: No significant interference up to L index of 1700 | ||
| Rheumatoid Factor: No interference up to 1200 IU/mL | ||
| High dose hook effect: No false results up to CRP concentrations of 1200 mg/L | Icterus: same (implicitly up to 60 mg/dL) | |
| Hemolysis: No significant interference up to 1000 mg/dL | ||
| Lipemia: No significant interference up to L index of 1000 | ||
| Rheumatoid Factor: same (implicitly up to 1200 IU/mL) | ||
| High dose hook effect: same (implicitly up to 1200 mg/L) | ||
| Method Comparison | Slope (Passing Bablok): 1.020 | |
| Intercept: 0.000 | ||
| Coefficients of correlation (r): 1.000 | Comparison was performed against Tina-Quant C-Reactive Protein (latex) on Hitachi 917. Specific numerical results (slope, intercept, correlation coefficient) for the modified device against the predicate are not provided in this section, but the predicate's values are listed as the reference, implying the modified device aims to match these. |
Study Information
The provided document describes a Special 510(k): Device Modification submission for the Tina-Quant C-Reactive Protein Gen 3 assay. The core of the "study" is a comparison of the modified device's performance characteristics against its predicate device (Tina-Quant C-Reactive Protein (Latex), K032336), to demonstrate substantial equivalence.
-
Sample size used for the test set and the data provenance:
- Test Set Sample Size: The document does not explicitly state the specific number of samples used for each test (e.g., precision, interference, method comparison). For the "Method Comparison," it refers to a "Scatter plot showing correlation between two methods for measuring C-Reactive Protein," which implies a collection of patient or control samples were run on both methods. However, the exact count is not given.
- Data Provenance: Not explicitly stated. The studies were conducted by Roche Diagnostics, suggesting internal studies. The country of origin of the data is not mentioned, nor is whether the data was retrospective or prospective.
-
Number of experts used to establish the ground truth for the test set and the qualifications of those experts:
- This is an in vitro diagnostic (IVD) device for quantitative determination of CRP. For such devices, "ground truth" is typically established by reference methods or highly accurate analytical techniques, not by expert consensus on interpretations like with imaging. The document traces the device's standardization to CRM 470, which is a certified reference material for proteins. There is no mention of experts establishing a ground truth in the context of clinical interpretation, as this device provides a quantitative measurement.
-
Adjudication method (e.g., 2+1, 3+1, none) for the test set:
- Not applicable. This is an IVD device for quantitative measurement. Adjudication methods like "2+1" typically apply to diagnostic tasks involving human interpretation or subjective assessments, where disagreements between experts need to be resolved.
-
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, an MRMC comparative effectiveness study was not done. This device is an automated in vitro diagnostic assay, not a device that assists human readers (like AI for image analysis). Therefore, comparisons of human reader performance with or without AI assistance are not relevant to this submission.
-
If a standalone (i.e., algorithm only without human-in-the-loop performance) was done:
- Yes, effectively, this is a standalone device performance evaluation. The device is an automated clinical chemistry analyzer that quantitatively measures CRP. The performance metrics reported (precision, analytical sensitivity, interference, method comparison) are intrinsic to the device's analytical function without direct human intervention in the result generation process, beyond sample loading and general operation.
-
The type of ground truth used (expert consensus, pathology, outcomes data, etc.):
- Reference material standardization. The device claims traceability to CRM 470 for standardization. This implies that the "ground truth" for the quantitative CRP measurements is established against this internationally recognized reference material, rather than clinical outcomes, pathology, or expert consensus on a diagnostic interpretation.
-
The sample size for the training set:
- Not applicable. This document is for a device modification of an in vitro diagnostic assay, not an AI/ML algorithm. Therefore, there isn't a "training set" in the sense of data used to train a machine learning model. The assay's parameters would have been developed and optimized through laboratory experiments, but the concept of a "training set" as commonly understood in AI/ML is not directly relevant here.
-
How the ground truth for the training set was established:
- Not applicable. As explained above, there is no "training set" for an AI/ML algorithm. For the initial development and optimization of the assay reagents and methods, ground truth for measuring CRP would typically be established using highly characterized samples with known CRP concentrations (e.g., purified CRP, reference materials like CRM 470, or validated high-accuracy laboratory methods).
§ 866.5270 C-reactive protein immunological test system.
(a)
Identification. A C-reactive protein immunological test system is a device that consists of the reagents used to measure by immunochemical techniques the C-reactive protein in serum and other body fluids. Measurement of C-reactive protein aids in evaluation of the amount of injury to body tissues.(b)
Classification. Class II (performance standards).