(359 days)
HemosILTM AcuStar Anti-ß2 Glycoprotein-I IgG: Fully automated chemiluminescent immunoassay for . the semi-quantitative measurement of anti-ß, Glycoprotein-I (anti-B>GPI) IgG antibodies in human citrated plasma and serum on the ACL AcuStar, as an aid in the diagnosis of thrombotic disorders related to primary and secondary Antiphospholipid Syndrome (APS) when used in conjunction with other laboratory and clinical findings.
HemosILTM AcuStar Anti-B2 Glycoprotein-I IgM: Fully automated chemiluminescent immunoassay for . the semi-quantitative measurement of anti-B2 Glycoprotein-I (anti-B>GPI) IgM antibodies in human citrated plasma and serum on the ACL AcuStar as an aid in the diagnosis of thrombotic disorders related to primary and secondary Antiphospholipid Syndrome (APS) when used in conjunction with other laboratory and clinical findings.
. HemosIL AcuStar Anti-B2 Glycoprotein-I IgG Controls: For the quality control of the Anti-B2 Glycoprotein-I IgG assay performed on the ACL AcuStar.
. HemosIL AcuStar Anti-B2 Glycoprotein-I IgM Controls: For the quality control of the Anti-B2 Glycoprotein-I IgM assay performed on the ACL AcuStar.
HemosIL AcuStar Anti-B2 Glycoprotein-I IgG is a chemiluminescent two-step immunoassay . consisting of magnetic particles coated with human purified ByGPI which capture, if present, the anti-ByGPI antiphospholipid antibodies from the sample. After incubation, magnetic separation, and a wash step, a tracer consisting of an isoluminol-labeled anti-human IgG antibody is added and may bind with the captured anti-B2GPI IgG on the particles. After a second incubation, magnetic separation, and wash step, reagents that trigger the luminescent reaction are added, and the emitted light is measured as relative light units (RLUs) by the ACL AcuStar optical system. The RLUs are directly proportional to the anti-ß2GPI IgG concentration in the sample.
The ACL AcuStar anti-B2GPI IgG assay utilizes a 4 Parameter Logistic Curve (4PLC) fit data reduction method to generate a Master Curve. The Master Curve is predefined and lot dependent and it is stored in the instrument through the cartridge barcode. With the measurement of calibrators, the predefined Master Curve is transformed to a new, instrument specific 4PLC Working Curve. The concentration values of the calibrators are included in the calibrator tube barcodes.
. HemosIL AcuStar Anti-B2 Glycoprotein-I IgM is a chemiluminescent two-step immunoassay consisting of magnetic particles coated with human purified ByGPI which capture, if present, the anti-B2GPI antiphospholipid antibodies from the sample. After incubation, magnetic separation, and a wash step, a tracer consisting of an isoluminol-labeled anti-human IgM antibody is added and may bind with the captured anti-ß2GPI IgM on the particles. After a second incubation, magnetic separation, and wash step, reagents that trigger the luminescent reaction are added, and the emitted light is measured as relative light units (RLUs) by the ACL AcuStar optical system. The RLUs are directly proportional to the anti-B2GPI IgM concentration in the sample.
The ACL AcuStar anti-B3GPI IgM assay utilizes a 4 Parameter Logistic Curve (4PLC) fit data reduction method to generate a Master Curve. The Master Curve is predefined and lot dependent and it is stored in the instrument through the cartridge barcode. With the measurement of calibrators, the predefined Master Curve is transformed to a new, instrument specific 4PLC Working Curve. The concentration values of the calibrators are included in the calibrator tube barcodes.
HemosIL AcuStar Anti-B2 Glycoprotein-I IgG Controls: The Low and High Anti-S, Glycoprotein-I . IgG Controls are prepared by means of a dedicated process and contain different concentrations of human anti-ß2 Glycoprotein-I IgG antibodies.
. Low Anti-B2 Glycoprotein-I IgG Control: Control intended for the assessment of precision and accuracy of the assay at the normal or around cut-off anti-B2 Glycoprotein-I IgG levels.
. High Anti-B2 Glycoprotein-I IgG Control: Control intended for the assessment of precision and accuracy of the assay at the abnormal anti-B2 Glycoprotein-I IgG levels.
. HemosIL AcuStar Anti-B2 Glycoprotein-I IgM Controls: The Low and High Anti-B2 Glycoprotein-I IgM Controls are prepared by means of a dedicated process and contain different concentrations of human anti-B2 Glycoprotein-I IgM antibodies.
. Low Anti-B2 Glycoprotein-I IgM Control intended for the assessment of precision and accuracy of the assay at the normal or around cut-off anti-B2 Glycoprotein-I IgM levels.
. High Anti-B2 Glycoprotein-I IgM Control: Control intended for the assessment of precision and accuracy of the assay at the abnormal anti-B2 Glycoprotein-I IgM levels.
The provided text describes performance data for the HemosIL AcuStar Anti-B2 Glycoprotein-I IgG and IgM assays, which are in vitro diagnostic (IVD) devices. The acceptance criteria and performance studies for IVDs differ significantly from those for algorithm-based medical devices or AI. Therefore, much of the requested information (like sample size for test set, data provenance, number of experts for ground truth, adjudication method, MRMC studies, standalone performance, and training set details) is not applicable or cannot be directly extracted from this document in the context of an AI/algorithm-based device.
However, I can extract information related to the performance characteristics provided for these IVDs, which serve as their "acceptance criteria" and "proof of meeting acceptance criteria" in the context of a 510(k) submission for an IVD.
Here's the information extracted and adapted where possible:
Acceptance Criteria and Reported Device Performance
For IVD devices like those described, performance is typically assessed through precision, sensitivity, specificity, and agreement with predicate devices or clinical outcomes. The "acceptance criteria" are implicitly met if the reported performance characteristics are deemed sufficient for the device's intended use and demonstrate substantial equivalence to predicate devices during the 510(k) review process.
1. Table of Acceptance Criteria and Reported Device Performance:
| Performance Metric | Acceptance Criteria (Implicitly met by 510(k) clearance) | Reported Device Performance (HemosIL AcuStar Anti-β2 Glycoprotein-I IgG) | Reported Device Performance (HemosIL AcuStar Anti-β2 Glycoprotein-I IgM) |
|---|---|---|---|
| Precision | Adequate for clinical use and comparable to predicate | - Low Control: 19.1 U/mL (Mean), 7.8% CV (Within run), 11.2% CV (Total) | - Low Control: 4.32 U/mL (Mean), 3.4% CV (Within run), 6.4% CV (Total) |
| (% CV Within Run) | - High Control: 429 U/mL (Mean), 3.0% CV (Within run), 3.8% CV (Total) | - High Control: 63.0 U/mL (Mean), 2.4% CV (Within run), 4.3% CV (Total) | |
| (% CV Total) | - Plasma Samples A-F: CVs ranging from 2.5% to 6.9% (Within run) and 3.3% to 10.9% (Total) | - Plasma Samples A-F: CVs ranging from 2.4% to 4.5% (Within run) and 5.2% to 8.3% (Total) | |
| Clinical Sensitivity | Adequate for diagnosis (compared to clinical groups) | 64.1% (95% CI: 53.5%-73.9%) - (Vs. PAPS & SAPS combined) | 29.3% (95% CI: 20.3%-39.8%) - (Vs. PAPS & SAPS combined) |
| Clinical Specificity | Adequate for diagnosis (compared to clinical groups) | 90.8% (95% CI: 86.3%-94.2%) - (Vs. PAPS & SAPS combined) | 95.2% (95% CI: 91.6%-97.6%) - (Vs. PAPS & SAPS combined) |
| Overall % Agreement | Adequate for diagnosis (compared to clinical groups) | 83.2% (95% CI: 78.6%-87.1%) - (Vs. PAPS & SAPS combined) | 76.3% (95% CI: 71.3%-80.9%) - (Vs. PAPS & SAPS combined) |
| Positive Agreement (Vs. Predicate) | Comparable to predicate device | 100.0% (95% CI: 93.0%-100.0%) vs. REAADS IgG Anti-β2GPI | 63.8% (95% CI: 48.5%-77.3%) vs. REAADS IgM Anti-β2GPI |
| Negative Agreement (Vs. Predicate) | Comparable to predicate device | 80.8% (95% CI: 71.7%-88.0%) vs. REAADS IgG Anti-β2GPI | 96.8% (95% CI: 92.8%-99.0%) vs. REAADS IgM Anti-β2GPI |
| Overall Agreement (Vs. Predicate) | Comparable to predicate device | 87.3% (95% CI: 80.9%-92.2%) vs. REAADS IgG Anti-β2GPI | 89.3% (95% CI: 84.2%-93.2%) vs. REAADS IgM Anti-β2GPI |
2. Sample size used for the test set and the data provenance:
- Clinical Outcome Studies Test Set: 321 frozen citrated plasma samples.
- Provenance: Samples were "selected" from 6 different groups:
- Primary APS (PAPS)
- Secondary APS (SAPS)
- Systemic Lupus Erythematosus (SLE) but not APS
- SLE-like
- Patients with cardiovascular disorders (not classified in the above four groups)
- Apparently healthy people
- This implies a retrospective collection of samples, likely from multiple sources within a country or countries (not specified beyond "selected").
- Provenance: Samples were "selected" from 6 different groups:
- Method Comparison Studies Test Set:
- IgG: 150 samples (those from the clinical performance study that were within the compared methods' test ranges).
- IgM: 205 samples (those from the clinical performance study that were within the compared methods' test ranges).
- Provenance: Same as the clinical outcome studies (retrospective, unspecified country/countries).
3. Number of experts used to establish the ground truth for the test set and the qualifications of those experts:
- For IVD devices, "ground truth" is typically established by existing diagnostic criteria (e.g., standard objective tests, clinical diagnosis) or comparison to a gold standard method. In this case, the ground truth for the clinical outcome study groups (PAPS, SAPS, SLE, etc.) was established based on "standard objective tests" and clinical findings. It does not involve a panel of human experts reviewing individual cases directly.
- Therefore, this information (number of experts, qualifications, adjudication method) is not applicable in the context of this IVD submission as it would be for an image-based AI device.
4. Adjudication method for the test set:
- Not applicable for this type of IVD study where ground truth is based on established clinical classifications and standard objective tests, rather than subjective expert review needing adjudication.
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:
- No, this was not an MRMC study. This is an in vitro diagnostic assay designed to be read by automated equipment (ACL AcuStar) and generate quantitative results, not an AI device assisting human readers in interpreting images or other complex data. This question is not applicable.
6. If a standalone (i.e. algorithm only without human-in-the-loop performance) was done:
- Yes, this is a standalone performance study. The device itself (HemosIL AcuStar assay on the ACL AcuStar instrument) provides a semi-quantitative measurement of antibodies. The performance metrics presented (precision, sensitivity, specificity, agreement with predicate) are for the device operating independently to produce these measurements. There is no human-in-the-loop component in the direct measurement process. The results are then used by clinicians, but the device's performance is standalone.
7. The type of ground truth used:
- Clinical Diagnosis and Standard Objective Tests: For the clinical outcome studies, the patient groups (PAPS, SAPS, SLE, etc.) were defined based on their clinical diagnosis and results from "standard objective tests." This serves as the clinical ground truth.
- Predicate Device Results: For the method comparison studies, the results of the "REAADS IgG Anti-β2GPI Test Kit" and "REAADS IgM Anti-β2GPI Test Kit" (ELISA assays) served as the comparative ground truth (or reference method).
8. The sample size for the training set:
- Not explicitly stated in this summary. For IVD assays, there isn't a "training set" in the same sense as machine learning. Assay development involves numerous experiments, reagent optimization, and calibration curve generation (which itself uses calibrators and a 4PLC fit model as described). The 4PLC Master Curve is "predefined and lot dependent" and is stored in the instrument, indicating a developmental process to establish this curve. However, a distinct "training set" size for an algorithm is not provided or applicable.
9. How the ground truth for the training set was established:
- Not applicable in the context of an AI training set. For IVD assays, the "ground truth" for developing and calibrating the assay would involve using known-concentration standards (calibrators) and characterized patient samples. The document mentions "calibrators" with "concentration values... included in the calibrator tube barcodes" and a 4PLC data reduction method to generate a "Master Curve," which is "predefined and lot dependent." This process inherently uses reference materials to establish the curve, which is analogous to "ground truth" for the assay's quantitative output.
§ 866.5660 Multiple autoantibodies immunological test system.
(a)
Identification. A multiple autoantibodies immunological test system is a device that consists of the reagents used to measure by immunochemical techniques the autoantibodies (antibodies produced against the body's own tissues) in serum and other body fluids. Measurement of multiple autoantibodies aids in the diagnosis of autoimmune disorders (disease produced when the body's own tissues are injured by autoantibodies).(b)
Classification. Class II (performance standards).