(363 days)
ACE Hemoglobin A lc (HbA lc) Reagent is intended for the quantitative determination of percent hemoglobin A lc in venous whole blood collected in K2-EDTA tubes using the ACE Axcel® Clinical Chemistry Systems. This test is intended for use in clinical laboratories and physician office laboratories to monitor long term blood glucose control in individuals with diabetes mellitus. For in vitro diagnostic use only.
The ACE Hemoglobin A1c (HbA1c) Reagent assay requires a pretreatment step of denaturation of the whole blood samples, which is performed off-line. The red blood cells in the sample are lysed by the Hemoglobin Denaturant and the hemoglobin chains are hydrolyzed. For determination of HbA1c, a latex agglutination inhibition assay is used. In the absence of HbA1c in the sample, the synthetic polymer containing the immunoreactive portion of HbA 1c Agglutinator Reagent will agglutinate with the antibody-coated microparticles in the HbA1c Antibody Reagent. The presence of HbA1c in the blood sample competes for the antibody binding sites and inhibits agglutination. The increase in absorbance, monochromatically at 692 mm, is inversely proportional to the HbA1c present in the sample. For the determination of total hemoglobin, all hemoglobin derivatives in the sample are converted to alkaline hematin. The reaction produces a green colored solution. which is measured bichromatically at 573 nm/692 nm. The intensity of color produced is directly proportional to the total hemoglobin concentration in the sample. The concentrations of both HbA 1 c and total hemoglobin are measured, the ratio is calculated, and the result reported as percent HbA1c.
1. Acceptance Criteria and Reported Device Performance
The acceptance criteria for the ACE Hemoglobin A1c (HbA1c) Reagent device are implicitly established by demonstrating comparable performance to the predicate device, the DCA 2000+ System for Hemoglobin A1c (K951361), across a range of analytical performance characteristics. While explicit numerical acceptance criteria are not always stated, the study aims to show that the new device's performance aligns with acceptable standards for HbA1c measurement in clinical diagnostics.
Here's a table summarizing the reported device performance:
| Performance Characteristic | Acceptance Criteria (Implied by Predicate & Clinical Relevance) | Reported Device Performance (ACE Alera & ACE Axcel) |
|---|---|---|
| Limit of Quantitation (LoQ) | Clinically relevant lower limit for HbA1c measurement. | ACE Alera: 2.5% HbA1c |
| ACE Axcel: 2.5% HbA1c | ||
| Linearity (HbA1c) | Strong correlation (r² close to 1) and a regression equation with a slope near 1 and y-intercept near 0 across the measuring range, indicating accurate and proportional measurement of HbA1c. | ACE Alera (Range 2.7%-13.0% HbA1c): y = 0.987x + 0.3, r² = 0.9948 |
| ACE Axcel (Range 2.4%-13.1% HbA1c): y = 0.954x + 0.3, r² = 0.9936 | ||
| Linearity (Total Hemoglobin) | Strong correlation (r² close to 1) and a regression equation with a slope near 1 and y-intercept near 0 across the measuring range of total hemoglobin. | ACE Alera (Range 1.4-22.2 g/dL): y = 1.006x + 0.10, r² = 0.9978 |
| ACE Axcel (Range 1.2-21.8 g/dL): y = 0.997x + 0.20, r² = 0.9964 | ||
| Precision (Within-Run %CV) | Low %CV for different HbA1c levels, indicating consistent results within a single analytical run. Typically, 0.97) and regression parameters (slope near 1, intercept near 0) indicating agreement with the predicate device. Confidence intervals for slope should include 1 and for intercept should include 0. | ACE Alera (n=101, Range 3.2-12.8% HbA1c): y = 0.979x + 0.05, Correlation = 0.9839, SE = 0.32, CI slope (0.944-1.015), CI intercept (-0.21-0.31) |
| ACE Axcel (n=102, Range 2.5-12.8% HbA1c): y = 0.983x - 0.03, Correlation = 0.9832, SE = 0.34, CI slope (0.948-1.019), CI intercept (-0.29-0.24) | ||
| Comparative Analysis (POLs vs. Predicate) | Similar strong correlation and regression parameters to in-house comparative analysis, demonstrating robust performance in typical clinical laboratory settings. | ACE Alera (POLs): Correlation range 0.9892 to 0.9945. Slopes generally close to 1 (e.g., 0.967, 0.984, 0.981). Intercepts generally close to 0 (e.g., 0.34, -0.02, -0.09). |
| ACE Axcel (POLs): Correlation range 0.9885 to 0.9960. Slopes generally close to 1 (e.g., 1.000, 0.993, 0.980). Intercepts generally close to 0 (e.g., -0.28, -0.12, 0.02). | ||
| Analytical Specificity | No significant interference from common endogenous substances or therapeutic compounds within specified concentrations. | Interferents: Bilirubin (≤ 53 mg/dL), Triglycerides (≤ 1100 mg/dL), Ascorbic Acid (≤ 6 mg/dL), Sodium Fluoride (≤ 1200 mg/dL), Acetaldehyde (≤ 100 mg/dL) showed no significant interference. |
| Cross-Reactivity | No significant interference from common hemoglobin variants or modified hemoglobins. | Non-Interfering: Acetylated Hb (2000 mg/dL), Carbamylated Hb (2000 mg/dL), Labile A1c (1440 mg/dL), Non-glycated Hb (HbA0) (1725 mg/dL), HbA1a+b fraction (100 mg/dL) showed no significant interference. |
| Known Interferences (within certain concentrations): HbD (≤ 36.3%), HbE (≤ 22.5%) showed no significant interference. High HbF (> 10.1%), High HbC (> 14.0%), and High HbS (> 17.1%) will result in inaccurate HbA1c results. These interferences are acknowledged and will be included in labeling. | ||
| Measuring Range | Consistent with or broader than the predicate device to coverclinically relevant HbA1c values. | Candidate Device: 2.7 – 13.0% HbA1c |
| Predicate Device: 2.5 – 14.0% HbA1c (The candidate device's range is slightly narrower at the upper end but still covers the critical clinical range). |
Study Details:
2. Sample Size and Data Provenance for Test Set:
- Linearity: 11 samples were used for both HbA1c and Total Hemoglobin linearity studies, run in 4 replicates each, for both the ACE Alera and ACE Axcel systems.
- Precision (In-house): 4 samples (A, B, C, D) were tested, but the number of runs/replicates to calculate SD and %CV is not explicitly stated in the table. Typically, precision studies involve multiple replicates over several days.
- Precision (Physician Office Labs - POLs): 4 samples were tested across 3 POLs for each instrument (ACE Alera and ACE Axcel). Similar to in-house, the specific number of runs/replicates per POL for SD and %CV calculation is not detailed.
- Comparative Analysis (In-house): 101 samples for ACE Alera and 102 samples for ACE Axcel were compared against the predicate device (DCA 2000+).
- Comparative Analysis (POLs): 50 samples per POL for a total of 150 samples for ACE Alera, and 52 samples for one POL and 50 samples for the other two POLs (total 152 samples) for ACE Axcel were compared against the predicate device (DCA 2000+).
- Analytical Specificity/Cross-Reactivity: The number of samples for these studies is not explicitly stated, but typically involves spiking known concentrations of interferents into samples and measuring the effect.
- Data Provenance: The studies were conducted in-house by Alfa Wassermann Diagnostic Technologies, LLC, and in external Physician Office Laboratories (POLs). Given the nature of performance validation for a diagnostic device, these studies are prospective, as samples are analyzed using the new device and compared against a reference method or predicate. The "country of origin of the data" is implicitly the United States, where the manufacturer and the POLs are located.
3. Number of Experts and their Qualifications for Ground Truth:
The document does not mention the use of "experts" in the traditional sense (e.g., radiologists, pathologists) to establish ground truth for the test set. For an in vitro diagnostic device measuring a quantitative analyte like HbA1c, the ground truth is typically established by:
- Reference Methods: Highly accurate and precise analytical methods, often traceable to international standards (e.g., NGSP, IFCC), which are considered the "gold standard" for measuring the analyte.
- Predicate Devices: Comparison to a legally marketed device that has already established its safety and effectiveness.
In this case, the ground truth for the comparative studies was derived from the predicate device (DCA 2000+ System for Hemoglobin A1c), which is itself NGSP Certified and traceable to IFCC reference materials.
4. Adjudication Method for the Test Set:
Not applicable. Adjudication methods (like 2+1, 3+1) are typically used in clinical studies involving interpretation of medical images or complex diagnostic assessments by human readers, where discrepancies between readers need to be resolved. For a quantitative in vitro diagnostic device, the ground truth is established analytically through reference methods or predicate comparison, not through expert consensus requiring adjudication.
5. Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study:
Not applicable. MRMC studies are generally performed for image-based diagnostic aids or other devices where human interpretation plays a significant role in the diagnostic outcome. This document describes the analytical performance validation of an in vitro diagnostic reagent, which is a quantitative measurement, not an interpretative task for human readers in the context of an MRMC study. Therefore, no effect size of human readers improving with AI vs. without AI assistance is relevant or reported.
6. Standalone (Algorithm Only) Performance Study:
Yes, this entire submission focuses on the standalone performance of the ACE Hemoglobin A1c (HbA1c) Reagent when used with the ACE Alera® and ACE Axcel® Clinical Chemistry Systems. The studies presented (linearity, precision, comparative analysis, specificity, cross-reactivity) all evaluate the direct analytical performance of the device itself, without human intervention for interpretation beyond standard laboratory procedures for operating the instrument and processing samples.
7. Type of Ground Truth Used:
The ground truth used for the comparative analysis studies was the predicate device, the DCA 2000+ System for Hemoglobin A1c. The document explicitly states that the DCA Hemoglobin A1c test method is National Glycohemoglobin Standardization Program (NGSP) Certified and is traceable to International Federation of Clinical Chemistry (IFCC) reference materials and test methods. This indicates that the predicate device serves as a highly standardized and accepted reference for HbA1c measurement.
8. Sample Size for the Training Set:
The document does not explicitly mention a "training set" in the context of machine learning or AI models. This device is a diagnostic reagent kit for a clinical chemistry system, not a software algorithm that requires a separate training phase with a distinct dataset. The performance data presented are for the validation of the finalized device.
9. How the Ground Truth for the Training Set Was Established:
As there is no "training set" in the context of this device being a reagent for a clinical chemistry system, this question is not applicable. The device's design and formulation would have been developed through internal R&D, likely using internal validation and optimization experiments, but these are not typically referred to as a "training set" with established ground truth in the same way as for AI/ML models. The ground truth for the validation of the device's performance (as described above) was established by comparison to the NGSP/IFCC-traceable predicate device.
§ 864.7470 Glycosylated hemoglobin assay.
(a)
Identification. A glycosylated hemoglobin assay is a device used to measure the glycosylated hemoglobins (A1a , A1b , and A1c ) in a patient's blood by a column chromatographic procedure. Measurement of glycosylated hemoglobin is used to assess the level of control of a patient's diabetes and to determine the proper insulin dosage for a patient. Elevated levels of glycosylated hemoglobin indicate uncontrolled diabetes in a patient.(b)
Classification. Class II (performance standards).