Search Results
Found 1 results
510(k) Data Aggregation
(149 days)
Tosoh AIA-PACK RBC FOLATE is designed for in vitro diagnostic use only for the quantitative measurement of red blood cell folate (RBC FOLATE) in whole blood (Heparin or EDTA) samples on TOSOH AIA system analyzer. Measurement of RBC Folate is used in the diagnosis and treatment of anemia.
The Tosoh AIA-PACK RBC FOLATE is a competitive enzyme immunoassay which, after sample hemolysis and sample pretreatment, is performed entirely within the AIA-PACK. First, sample hemolysis is performed to achieve complete hemolysis of erythrocytes and deconjugation to monoglutamate in the whole blood sample using sample hemolyzing reagents (containing ascorbic acid). After sample hemolysis, sample pretreatment is performed to release folate from endogenous binding proteins in the hemolysate sample using sample pretreatment reagents (containing sodium hydroxide and dithiothreitol). Folate present in the pretreated test sample competes with enzyme-labeled folate for a limited number of binding sites on a fluorescein labeled bovine folate binding protein which then binds to anti-FITC (fluorescein isothiocyanate) antibody immobilized on magnetic beads. The beads are washed to remove the unbound enzyme-labeled folate and are then incubated with a fluorogenic substrate, 4-methylumbelifery| phosphate (4MUP). The amount of enzyme labeled folate that binds to the beads is inversely proportional to the folate concentration in the test sample. A standard curve using a range of known standard concentrations is constructed and unknown folate concentrations are calculated using this curve.
The Tosoh AIA-PACK RBC FOLATE is a device designed for the quantitative measurement of red blood cell folate (RBC FOLATE) in whole blood samples, used in the diagnosis and treatment of anemia. The device's performance was evaluated through various studies to demonstrate substantial equivalence to a predicate device (Bayer K010050, ADVIA Centaur and ACS: 180 Folate Immunoassay).
The document provides performance data for several criteria: Recovery, Dilution, Linearity, Precision, Correlation, Specificity, and Limit of Detection (LoD)/Limit of Quantitation (LoQ). Acceptance criteria are not explicitly stated as distinct numerical targets but are implied by the reported performance and the overall conclusion of substantial equivalence based on established CLSI protocols.
Here's a breakdown of the acceptance criteria and the study results:
1. Table of Acceptance Criteria and Reported Device Performance
| Acceptance Criteria Category | Specific Metric | Acceptance Criteria (Implied by CLSI Protocols and Performance) | Reported Device Performance |
|---|---|---|---|
| Recovery | Percent Recovery | Generally 90-110% (standard for spiked recovery studies) | Ranged from 94.2% to 102.7% (demonstrated good recovery) |
| Dilution | Percent Recovery (after serial dilution) | Generally 90-110% (standard for dilution linearity studies) | Ranged from 100.0% to 114.7% (demonstrated good linearity upon dilution) |
| Linearity | Linear range with ±10% difference | Linear from 0.5 to 40.0 ng/mL, within ±10% | Demonstrated linearity from 0.5 to 40.0 ng/mL, within ±10% difference. The claimed linearity was 0.62 to 24.0 ng/mL. |
| Precision | Within-run CV (%) | Typically < 10% for low concentrations, < 5% for higher. | Control 1 (low): 5.9%, Control 2: 3.4%, Control 3 (high): 2.2%, Sample A3: 2.2% |
| Total CV (%) | Typically < 15% for low concentrations, < 10% for higher. | Control 1 (low): 8.2%, Control 2: 4.6%, Control 3 (high): 3.3%, Sample A3: 4.2% | |
| Correlation (EDTA vs. Heparin) | Correlation Coefficient (r) | Typically > 0.95 for strong correlation | 0.99 |
| Slope (Deming Regression) | Close to 1 (e.g., 0.9-1.1) | 0.959 (0.922 to 0.996) | |
| Y-intercept (Deming Regression) | Close to 0 | - 3.890 (-13.512 to 5.732) | |
| Correlation (vs. Alternate Method) | Correlation Coefficient (r) | Typically > 0.90 for good correlation | 0.87 (Deming Regression), 0.86 (Linear Regression with hematocrit correction) |
| Slope (Deming Regression) | Close to 1 (e.g., 0.9-1.1) | 1.06 (0.950 to 1.168), 1.00 (0.897 to 1.107 with hematocrit correction) | |
| Y-intercept (Deming Regression) | Close to 0 | -19.554 (-44.640 to 5.532), -26.60 (with hematocrit correction) | |
| Specificity | Cross-reactivity (%) for common interferences | Low percentage (e.g., < 1%) | Amethopterin: 0.66%, Leucovorin: 0.34% |
| LoD/LoQ | Limit of Detection (LoD) | As determined by CLSI EP17-A protocols | 0.43 ng folate/mL |
| Limit of Quantitation (LoQ) at 20% CV | As determined by CLSI EP17-A protocols | 0.62 ng folate/mL |
2. Sample Size and Data Provenance for Test Set:
- Recovery Study: 3 whole blood (EDTA) samples, each spiked at 3 different levels.
- Dilution Study: 3 whole blood (EDTA) samples.
- Precision Study: 3 commercially available controls and 1 whole blood (EDTA) control.
- Correlation (EDTA vs. Heparin) Study: 50 patient samples.
- Correlation (vs. Alternate Method) Study: 101 patient specimens.
- LoD/LoQ Study: Blank sample with 60 replicates; 7 low-level samples with 10 replicates each.
The document does not explicitly state the country of origin for the patient samples used in the correlation studies, nor does it specify whether the data was retrospective or prospective. Given the nature of an in vitro diagnostic device regulatory submission, it is typically presumed to be clinical samples, often from a diverse patient population, collected prospectively or obtained retrospectively under controlled conditions for the purpose of the study.
3. Number of Experts and Qualifications for Ground Truth:
The studies described are primarily analytical performance studies for an in vitro diagnostic (IVD) device, not directly involving human interpretation of results in a diagnostic setting (i.e., not an image-based AI device relying on expert radiologists for ground truth). Therefore, the concept of "experts establishing ground truth" in the manner requested is not applicable here.
- Ground truth for these studies is established through reference methods and known concentrations of analytes in controls and spiked samples, measured by established laboratory practices and instrumentation.
- The "experts" involved would be qualified laboratory personnel (e.g., clinical laboratory scientists, chemists) conducting the assays and analyzing data according to standard operating procedures and CLSI guidelines. Their qualifications are inherent in their ability to perform and interpret such analytical studies.
4. Adjudication Method for the Test Set:
Not applicable. As described above, the ground truth is based on quantitative analytical measurements, not on subjective expert consensus requiring adjudication.
5. Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study:
No, an MRMC comparative effectiveness study was not done. This type of study is relevant for diagnostic devices that involve human interpretation (e.g., reading medical images) where the AI aids human performance. The AIA-PACK RBC FOLATE is an automated analytical assay, and its performance is evaluated directly against known analytical values or comparative analytical methods, not human readers.
6. Standalone (Algorithm Only) Performance:
Yes, the studies effectively demonstrate standalone performance. The "device" in this context is the AIA-PACK RBC FOLATE assay run on a Tosoh AIA system analyzer. All the reported performance data (Recovery, Dilution, Linearity, Precision, LoD/LoQ, Specificity) demonstrate the intrinsic analytical capabilities of the assay without direct human intervention in the interpretive phase. Human operators perform the sample preparation and run the analyzer, but the measurement and calculation of folate concentration is algorithmic within the system. The correlation studies compare this standalone analytical result to other analytical methods or sample types.
7. Type of Ground Truth Used:
The ground truth used for these studies includes:
- Known concentrations: For recovery and linearity studies, samples were spiked with known amounts of folate to determine expected values. Commercial controls with established target values were also used for precision.
- Reference methods/Comparison methods: For correlation studies, the device's measurements were compared against an "alternate method" (likely another legally marketed Folate immunoassay) and against each other (EDTA vs. Heparin whole blood) to establish agreement.
- Analytical measurement techniques: The core ground truth for an IVD kit is its ability to accurately and precisely measure the analyte (RBC Folate) based on its chemical and immunological principles.
8. Sample Size for the Training Set:
The document does not explicitly mention a "training set" in the context of machine learning. For an immunoassay, the "training" aspect refers to the development and optimization of the assay reagents, protocols, and standard curve generation. This typically involves extensive R&D work using numerous samples and iterations to establish the optimal conditions and calibration. The calibration process for this device uses a 6-point calibration curve. The specific sample size for developing these foundational elements is not detailed but would precede the formal validation studies presented.
9. How the Ground Truth for the Training Set was Established:
As mentioned, "ground truth for a training set" is not directly applicable in the AI/ML sense for a traditional immunoassay. For an immunoassay, the "ground truth" for developing the assay and its calibration would be established through:
- Highly purified folate standards: Used to create the calibration curve and to spike samples for recovery and linearity.
- Reference materials: Certified reference materials or well-characterized patient samples with established folate concentrations from orthogonal or highly accurate methods.
- Methodology optimization: Iterative testing and adjustment of reagent concentrations, incubation times, antibody affinities, and detection limits to achieve the desired analytical performance.
The calibration curve itself (6-point calibration) serves as the "learned" relationship between signal and concentration, established using known folate standards.
Ask a specific question about this device
Page 1 of 1