(259 days)
The Cystatin C assay is an in vitro diagnostic test used in the quantitative immunoturbidimetric determination of cystatin C in human serum and plasma on the Alinity c system.
Measurement of cystatin C aids in the diagnosis and treatment of renal diseases.
For laboratory professional use only.
The Cystatin C assay is an automated clinical chemistry assay.
Cystatin C is a particle-enhanced turbidimetric immunoassay (PETIA) developed to accurately and reproducibly measure cystatin C levels in serum and plasma. Latex particles coated with anti-human cystatin C antibody agglutinate when mixed with sample containing human cystatin C. The change in absorbance due to agglutination of the reaction mixture is proportional to the quantity of human cystatin C in the sample.
The provided text is a 510(k) clearance letter for an in vitro diagnostic test (Cystatin C assay), not an AI/ML-driven medical device. Therefore, the information requested for AI/ML device studies (such as MRMC studies, expert adjudication, training set ground truth, etc.) is not applicable to this document.
However, I can extract and present the performance acceptance criteria and reported device performance from the provided document, even though they relate to a traditional laboratory assay rather than an AI/ML model.
Here's an interpretation based on the provided text for a traditional in vitro diagnostic device:
Based on the provided FDA 510(k) clearance letter for the Cystatin C Test System, the device is a quantitative immunoturbidimetric assay for the determination of cystatin C in human serum and plasma. The performance of this device was demonstrated through various nonclinical studies.
1. Table of Acceptance Criteria and Reported Device Performance
For in vitro diagnostic tests like the Cystatin C assay, "acceptance criteria" are typically defined by ranges or thresholds for performance characteristics such as precision, accuracy (bias), linearity, and limits of detection. The document describes the reported performance measured during the studies, and implicitly, these figures met the internal acceptance criteria set by the manufacturer (and approved by the FDA for clearance).
Here's a summary of the reported device performance:
| Performance Characteristic | Acceptance Criteria (Implicit from Industry Standards/Predicate) | Reported Device Performance (Cystatin C Assay) |
|---|---|---|
| Analytical Measuring Interval (AMI) | Defined range of accurate and precise measurement | 0.30 – 10.00 mg/L |
| Extended Measuring Interval (EMI) | Range accessible via dilution/spiking | 10.00 – 40.00 mg/L |
| Reportable Interval | Overall range of reportable results | 0.30 – 40.00 mg/L |
| Within-Laboratory Precision (Repeatability) | Expected low variability | Control Level 1: 0.81 mg/L Mean, 1.0%CV |
| Control Level 2: 4.11 mg/L Mean, 0.6%CV | ||
| Panel A: 0.49 mg/L Mean, 1.7%CV | ||
| Panel B: 0.92 mg/L Mean, 0.8%CV | ||
| Panel C: 5.89 mg/L Mean, 0.5%CV | ||
| Panel D: 8.95 mg/L Mean, 0.8%CV | ||
| Within-Laboratory Precision (Total) | Expected low variability over time | Control Level 1: 1.7%CV |
| Control Level 2: 1.0%CV | ||
| Panel A: 1.8%CV | ||
| Panel B: 0.9%CV | ||
| Panel C: 0.6%CV | ||
| Panel D: 1.0%CV | ||
| Overall Reproducibility | Expected low variability across sites/lots | Control Level 1: 2.0%CV |
| Control Level 2: 1.1%CV | ||
| Panel 1: 5.4%CV | ||
| Panel 2: 1.6%CV | ||
| Panel 3: 1.6%CV | ||
| Accuracy (Bias vs. Reference Material) | Acceptable low bias to a certified reference | Ranged from 1.3% to 1.8% across all reagent and calibrator lots relative to ERM-DA471/IFCC. |
| Linearity | Demonstrated proportional response across AMI | Linear across 0.30 to 10.00 mg/L (AMI). |
| Limit of Blank (LoB) | Lowest detectable signal for a blank sample | 0.03 mg/L |
| Limit of Detection (LoD) | Lowest concentration detectable with 95% probability | 0.05 mg/L |
| Limit of Quantitation (LoQ) | Lowest concentration meeting 25% total allowable error | 0.30 mg/L |
| Hook Effect (Prozone Effect) | No false low results at high concentrations | No prozone effect observed up to 40.00 mg/L. |
| Interference (Endogenous Substances) | No significant interference at specified levels | No significant interference observed for Bilirubin (60 mg/dL), Hemoglobin (1000 mg/dL), Total protein (10.2-11.8 g/dL), Triglycerides (1500 mg/dL). Some interference noted for high Rheumatoid factor and very high Total protein. |
| Interference (Exogenous Substances) | No significant interference at specified levels | No significant interference observed for 16 common drugs (e.g., Acetaminophen, Ibuprofen, Ascorbic acid, Cyclosporin, etc.) at specified therapeutic/toxic levels. |
| Correlation with Comparator Assay | Strong correlation with predicate/comparable method | Correlation Coefficient: 1.00 (Serum, n=161) vs. Comparator Cystatin C assay. |
| Matrix Comparison | Acceptable for various specimen types | Acceptable for Serum, Serum separator, Dipotassium EDTA, Lithium heparin, Lithium heparin separator, Sodium heparin, Tripotassium EDTA plasma. |
| Reference Interval (Expected Values) | Established range for healthy individuals | Adult: 0.59–1.28 mg/L (2.5th to 97.5th percentile) |
Study Details (Relevant to IVD, not AI/ML)
The following points are addressed as much as possible for an in vitro diagnostic device, noting where the requested AI/ML specific information is not applicable.
-
Sample sizes used for the test set and the data provenance:
- Precision/Reproducibility:
- Within-Laboratory Precision: 80 replicates per sample/control (2 replicates/day for 20 days) for 2 controls and 4 human serum panels.
- Reproducibility: 240 replicates per sample/control (4 replicates, twice/day for 5 days at 3 sites) for 2 controls and 3 human serum panels.
- Accuracy (Bias): Not specified as a separate sample size, but involved testing across 2 reagent lots and 2 calibrator lots.
- Linearity: Not specified as a number of distinct samples, but assessed over a range of concentrations.
- Lower Limits of Measurement (LoB, LoD, LoQ): ≥ 60 replicates of zero-analyte samples for LoB, and ≥ 60 replicates of low-analyte level samples for LoD/LoQ.
- Analytical Specificity (Interference): Each substance tested at 2 analyte levels.
- Method Comparison: 161 serum samples.
- Matrix Comparison: Not specified as a specific number, but evaluated across various tube types.
- High Dose Hook: Not specified as a specific number of samples, but tested up to 40.00 mg/L.
- Expected Values (Reference Interval): 250 apparently healthy individuals (105 females, 145 males) with eGFR > 80, aged 18 to 69 years.
Data Provenance: The document does not explicitly state the country of origin for the clinical samples. The studies are described as "nonclinical performance" studies, meaning they evaluate the analytical performance of the assay itself, rather than diagnostic accuracy in a clinical setting with patient outcomes. These are typically prospective studies conducted in a controlled lab environment.
- Precision/Reproducibility:
-
Number of experts used to establish the ground truth for the test set and the qualifications of those experts:
- Not applicable for this in vitro diagnostic device. Ground truth for analytical performance studies of IVDs is established by the known concentration of analytes in reference materials, calibrators, and characterized control samples, or by comparison to a well-established reference or predicate method. It does not involve expert readers or their consensus.
-
Adjudication method (e.g. 2+1, 3+1, none) for the test set:
- Not applicable for this in vitro diagnostic device. Adjudication methods are relevant for subjective image-based assessments or clinical diagnoses where human interpretation varies. This product is a quantitative chemical assay.
-
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 for this in vitro diagnostic device. MRMC studies are for evaluating the impact of AI on human reader performance, typically in radiology or similar interpretive fields. This device performs automated measurements.
-
If a standalone (i.e. algorithm only without human-in-the-loop performance) was done:
- Analogous concept applies. The performance data presented (precision, accuracy, linearity, limits of detection, etc.) are the standalone performance of the assay on the Alinity c system. It demonstrates the device's ability to measure Cystatin C without human interpretation of the analytical signal.
-
The type of ground truth used (expert consensus, pathology, outcomes data, etc.):
- For the analytical performance studies, the ground truth was established by:
- Certified reference materials: Specifically, ERM-DA471/IFCC for assay standardization and accuracy validation.
- Preparation of known concentrations: Through spiking or dilution of samples to evaluate linearity and limits of measurement.
- Comparison to a validated comparator assay: For method comparison (Cystatin C on Alinity c vs a Comparator Cystatin C assay).
- Characterized control materials: For precision and reproducibility studies.
- Established healthy population studies: For determining the reference interval.
- For the analytical performance studies, the ground truth was established by:
-
The sample size for the training set:
- Not applicable in the AI/ML sense. This is a traditional chemical assay, not an AI/ML model that undergoes a "training" phase with data. The assay's parameters are developed through conventional analytical chemistry principles and validated through the nonclinical performance studies detailed.
- Development would involve formulation and optimization studies, but not "training data" in the AI sense.
-
How the ground truth for the training set was established:
- Not applicable for this in vitro diagnostic device. As noted above, there is no "training set" in the AI/ML context for a traditional immunoturbidimetric assay. The "ground truth" for developing such an assay would be based on fundamental chemical principles, known concentrations of analytes, and established analytical chemistry standards.
§ 862.1225 Creatinine test system.
(a)
Identification. A creatinine test system is a device intended to measure creatinine levels in plasma and urine. Creatinine measurements are used in the diagnosis and treatment of renal diseases, in monitoring renal dialysis, and as a calculation basis for measuring other urine analytes.(b)
Classification. Class II.