(253 days)
Diazyme 1,5-anhydroglucitol (1,5-AG) Assay is an enzymatic method intended for the quantitative determination of 1,5anhydroglucitol (1,5-AG) in serum or plasma. The 1,5-AG Assay is for the intermediate term (preceding 1-2 weeks) monitoring of glycemic control in people with diabetes. For in vitro diagnostic use only.
Diazyme's 1.5-AG assay is an enzymatic method intended for the quantitative determination of 1,5-anhydroglucitol (1,5-AG) in serum or plasma. The assay uses the enzyme pyranose oxidase (PROD) to oxidize the 2nd position hydroxyl group of 1,5-AG and to detect the generated hydrogen peroxide by colorimetry using peroxidase (POD). To eliminate reactive glucose in sample, it is pretreated by enzymatic reactions using hexokinase and pyruvate kinase (PK). Hexokinase uses adenosine triphosphate (ATP) to convert glucose into non-reactive glucose-6-phosphate (G-6-P), generating adenosine diphosphate (ADP). The reaction is driven to completion with PK, as ADP is phosphoralated to ATP during the conversion of phosphoenolpyruvate (PEP) into pyruvate.
The provided document is a 510(k) premarket notification for the Diazyme 1,5-AG Assay, a device for in vitro quantitative determination of 1,5-anhydroglucitol (1,5-AG) in serum or plasma to monitor glycemic control in people with diabetes.
Here's an analysis of the acceptance criteria and the study that proves the device meets them, based on the provided text:
Key Takeaways:
- This submission focuses on substantial equivalence to a predicate device (GlycoMark™ 1,5-anhydroglucitol (1,5-AG) K031604) rather than a novel, de novo AI/ML device. Therefore, many standard AI/ML specific criteria like MRMC studies, human-in-the-loop performance, and expert consensus for ground truth are not applicable or detailed in this context.
- The studies presented are primarily analytical performance studies to demonstrate that the new device performs comparably to the predicate device and meets established laboratory testing standards.
1. Table of Acceptance Criteria and Reported Device Performance:
The document implicitly defines acceptance by comparing the Diazyme 1,5-AG Assay's performance to the predicate device and established CLSI (Clinical and Laboratory Standards Institute) guidelines for analytical validation.
| Acceptance Criteria (Implied) | Predicate Device Performance (GlycoMark™ 1,5-AG) | Reported Device Performance (Diazyme 1,5-AG Assay) |
|---|---|---|
| Indications for Use Equivalence | Same as Diazyme 1,5-AG Assay | Same as GlycoMark™ 1,5-AG Assay: Quantitative determination of 1,5-anhydroglucitol (1,5-AG) in serum or plasma for intermediate term (preceding 1-2 weeks) monitoring of glycemic control in people with diabetes. For in vitro diagnostic use only. |
| Assay Principle Equivalence | Pyranose oxidase (PROD) enzyme, colorimetry | Same: Uses pyranose oxidase (PROD) to oxidize 1,5-AG and detect generated hydrogen peroxide by colorimetry using peroxidase (POD). Pretreatment with hexokinase and pyruvate kinase (PK) to eliminate reactive glucose. |
| Test Objective Equivalence | Quantitative determination of 1,5-AG | Same: For the in vitro quantitative determination of 1,5-AG in human serum or plasma. |
| Type of Test Equivalence | Quantitative | Same: Quantitative |
| Specimen Type Equivalence | Human serum or plasma | Same: Human serum or plasma. |
| Product Type Equivalence | Assay reagent kit: liquid stable two reagent system | Same: Assay reagent kit: liquid stable two reagent system |
| Analytical Linearity/Reportable Range | Linear up to 110 µg/mL | Linear up to 116.7 µg/mL. Also linear in the clinically relevant range of 0.6-10.0 µg/mL. (Based on CLSI EP6-A). The recovered values showed good agreement with expected values, with errors generally between 0% and -8%. |
| Analytical Precision (CV%) | 0.8 – 3.8% (2 controls and 2 serum samples) | CV% less than 5% across various samples and runs. (Based on CLSI EP5-A2). Within-run CV% ranged from 0.5% to 2.3%. Total CV% ranged from 0.9% to 4.8%. |
| Accuracy (Method Comparison) | 89.6% concordance (vs. Roche Tina-Quant A1c) | Strong correlation ($R^2$ = 0.9995) with the predicate GlycoMark™ 1,5-AG assay. Slope = 1.0164, Y-intercept = -0.2042. (Based on CLSI EP9-A2). Correlation Coefficient (R) = 0.9997. Standard Error of Estimate = 0.51. The rationale states "excellent correlation." |
| Limit of Blank (LoB) | Not specified | 0.3 µg/mL (Based on CLSI EP17-A2). |
| Limit of Detection (LoD) | Not specified | 0.5 µg/mL (Based on CLSI EP17-A2). |
| Limit of Quantitation (LoQ) | Not specified | 0.6 µg/mL (Based on CLSI EP17-A2). |
| Analytical Specificity (Interference) | Not specified | Less than 10% deviation caused by: Free Bilirubin (5 mg/dL), Bilirubin Conjugated (5 mg/dL), Hemoglobin (125 mg/dL), Ascorbic Acid (37.5 mg/dL), Triglyceride (1000 mg/dL), Glucose (1000 mg/dL), Maltose (500 mg/dL), Uric Acid (20 mg/dL), Creatinine (10 mg/dL), Urea (20 mg/dL). (Based on CLSI EP7-A2). |
| Reference Range (Normal Population) | Not specified | Males: 8.19 to 32.19 µg/mL (Non-parametric 5th-95th percentiles). Females: 6.00 to 29.10 µg/mL (Non-parametric 5th-95th percentiles). (Based on CLSI C28-A3 guideline). Note: This is a study result not a performance claim compared to predicate. |
2. Sample Size Used for the Test Set and Data Provenance:
- Precision Study: Six serum samples (tested repeatedly over 240 measurements for each sample to calculate different CV% components).
- Linearity Study: Samples encompassing a wide range of 1,5-AG concentrations (Level 0 to Level 10), tested in triplicate for each level. The exact number of distinct patient samples is not specified, but it implies a spiked matrix or dilutions of real samples.
- Method Comparison (Accuracy) Study: 102 patient samples covering the Analytical Measurement Range (AMR).
- Reference Range Study: 140 apparently healthy males and 140 apparently healthy females (Total 280 samples).
- Data Provenance: The document does not specify the country of origin of the data or whether the studies were retrospective or prospective. It refers to "clinical patient samples" for the method comparison study and "apparently healthy males and females" for the reference range study, implying real human samples. Given the nature of in vitro diagnostic device validation, these are typically prospective collection campaigns or use of well-characterized biobank samples.
3. Number of Experts Used to Establish the Ground Truth for the Test Set and the Qualifications of Those Experts:
- This device is an in vitro diagnostic assay for quantitative measurement of a biomarker, not an AI/ML device interpreting medical images or other complex data. Therefore, the concept of "experts establishing ground truth" in the same way as, for example, radiologists marking tumors, is not applicable.
- The "ground truth" for the test set is established by the measurement results obtained from the predicate device (GlycoMark™ 1,5-AG assay) for the method comparison study, and by the inherent concentration values of the samples used in precision, linearity, LoB/LoD/LoQ, and interference studies. These are not dependent on expert visual review or interpretation.
- The predicate device itself was validated and approved by the FDA based on its own clinical and analytical performance.
4. Adjudication Method for the Test Set:
- Not applicable in this context. Adjudication methods like 2+1 or 3+1 are used when human experts disagree on interpretations (e.g., in radiology image reading). For quantitative in vitro diagnostic assays, the "truth" is the measured value, and deviations are assessed statistically (e.g., bias, correlation, CV%).
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, an MRMC comparative effectiveness study was not done. This is an in vitro diagnostic device measuring a chemical biomarker, not an AI system assisting human readers of medical images or other data. The performance is assessed analytically against a predicate device and established laboratory standards.
6. If a Standalone (i.e., algorithm only without human-in-the-loop performance) was done:
- Yes, in essence, the Diazyme 1,5-AG Assay is a standalone device. It performs the quantitative measurement of 1,5-AG in a sample without direct human interpretation of its internal "algorithm" or sensing process for each test. The results are provided as a numeric value. Humans (lab technicians, clinicians) then interpret these numeric results in the clinical context, similar to any lab test. The performance studies (precision, linearity, method comparison, etc.) directly evaluate the performance of this "standalone" assay.
7. The Type of Ground Truth Used (expert consensus, pathology, outcomes data, etc.):
- Analytical Reference Values / Predicate Device Measurements:
- For the method comparison study (accuracy), the "ground truth" was the measurements obtained from the legally marketed predicate device (GlycoMark™ 1,5-AG).
- For linearity, LoB/LoD/LoQ, and precision studies, the "ground truth" refers to the expected or true concentrations of the analytes in the reference materials or spiked samples. These values are typically established through highly accurate reference methods or certified reference materials traceable to international standards (though not explicitly detailed, this is standard practice for IVD validation).
8. The Sample Size for the Training Set:
- Not applicable in the typical AI/ML sense. This is a chemical assay, not a machine learning algorithm that learns from training data. The "training" for such a device involves optimizing reagents and reaction conditions during development, but there isn't a "training set" of patient data in the same way an AI model is trained.
9. How the Ground Truth for the Training Set Was Established:
- Not applicable. As explained above, this device does not have a "training set" in the context of AI/ML. The "ground truth" for the analytical studies are the established reference values or predicate device measurements.
§ 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).