(399 days)
Immunoassay for the in vitro quantitative determination of N terminal pro Brain natriuretic peptide in human serum and plasma. This assay is used as an aid in the diagnosis of individuals suspected of having heart failure. The test is further indicated for the risk stratification of patients with acute coronary syndrome and heart failure. The test may also serve as an aid in the assessment of increased risk of cardiovascular events and mortality in patients at risk for heart failure who have stable coronary artery disease.
The electrochemiluminescence immunoassay "ECLIA" is intended for use on cobas e immunoassay analyzers.
Elecsys proBNP II (updated assay) is a second-generation assay by Roche Diagnostics for the in vitro quantitative determination of N-terminal pro-Brain natriuretic peptide (NT-proBNP) in human serum and plasma with increased biotin tolerance. The electrochemiluminescence immunoassay "ECLIA" is intended for use on cobas e immunoassay analyzers.
The cobas e family of analyzers employs the electrochemiluminescence immunoassay "ECLIA" technology. The assays are an 18-minute (Elecsys proBNP II) and 9 minute (Elecsys proBNP II STAT) application following a sandwich principle using two monoclonal antibodies which are specifically directed against NT-proBNP.
Acceptance Criteria and Device Performance Study for Elecsys proBNP II and Elecsys proBNP II STAT Assays
The document describes the analytical performance studies conducted for the Elecsys proBNP II and Elecsys proBNP II STAT assays, which are in vitro diagnostic devices. These assays are intended for the quantitative determination of N-terminal pro-Brain natriuretic peptide (NT-proBNP) in human serum and plasma, aiding in the diagnosis of heart failure, risk stratification, and assessment of cardiovascular event risk. The assays have been modified to include increased biotin tolerance.
The studies aim to demonstrate that the updated assays meet predetermined acceptance criteria for various analytical parameters, ensuring their safety and effectiveness and substantial equivalence to their predicate devices.
1. Table of Acceptance Criteria and Reported Device Performance
The provided document doesn't explicitly list "acceptance criteria" alongside "reported device performance" in a single, dedicated table for all parameters. However, the "Conclusion" sections for each analytical study implicitly state whether the acceptance criteria were met. Based on the provided text, a table can be constructed as follows:
| Acceptance Criteria Category | Specific Metric | Predetermined Acceptance Criterion (Implicitly Met) | Reported Device Performance (Elecsys proBNP II) | Reported Device Performance (Elecsys proBNP II STAT) |
|---|---|---|---|---|
| Analytical Sensitivity | Limit of Blank (LoB) | ≤ 8 pg/mL | All lots met ≤ 8 pg/mL | All lots met ≤ 8 pg/mL |
| Limit of Detection (LoD) | ≤ 10 pg/mL | All lots met ≤ 10 pg/mL | All lots met ≤ 10 pg/mL | |
| Limit of Quantitation (LoQ) (Intermediate precision 20% CV) | ≤ 36 pg/mL (Inferred from reported LoQ values) | 32.7 - 35.7 pg/mL | 7.28 - 13.6 pg/mL | |
| Precision | Repeatability (Within-run precision) | Low CV% (Specific numerical thresholds not stated) | See tables on pages 8-9 for detailed CV | See tables on pages 9-10 for detailed CV |
| Intermediate Precision (Within-laboratory precision) | Low CV% (Specific numerical thresholds not stated) | See tables on pages 8-9 for detailed CV | See tables on pages 9-10 for detailed CV | |
| Inter-Instrument Variability (Inter-laboratory precision) | Low CV% (Specific numerical thresholds not stated) | See table on page 10 for detailed CV | See table on page 11 for detailed CV | |
| Linearity/Reportable Range | Measurements across claimed measuring range are linear | Not explicitly quantified, but demonstrated | Linearity data on page 15 | Linearity data on page 15 |
| Interference | Bilirubin (Conjugated & Unconjugated) | No interference up to 25.0 mg/dL | No interference up to 25.0 mg/dL | No interference up to 25.0 mg/dL |
| Hemoglobin | No interference up to 1000 mg/dL | No interference up to 1000 mg/dL | No interference up to 1000 mg/dL | |
| Lipemia | No interference up to 1500 mg/dL | No interference up to 1500 mg/dL | No interference up to 1500 mg/dL | |
| Biotin | No interference up to 3500 ng/mL | No interference up to 5000 ng/mL | No interference up to 5000 ng/mL | |
| Rheumatoid Factor | No interference up to 1500 IU/mL | No interference demonstrated | No interference demonstrated | |
| Albumin | No interference up to 7 g/dL | No interference demonstrated | No interference demonstrated | |
| Cross-Reactivity | Absence of significant cross-reactivity with various substances | Recovery % close to 100% (Implied) | See tables on pages 19-20 | See tables on pages 19-20 |
| Exogenous Interference | No interference with listed common and special therapeutic drugs | No significant interference (Implied) | No interference seen with tested drugs | No interference seen with tested drugs |
| Matrix Comparisons | Equivalence between Serum, Li-Heparin, and K2-EDTA plasma | Slope close to 1, Intercept close to 0, High 'r' | Slope 0.990-1.01, Intercept -0.985-0.898, r≥0.998 | Not explicitly detailed for STAT |
| Method Comparison | Substantial equivalence to predicate device | High Pearson's r, Passing-Bablok Slope close to 1, Intercept close to 0 | Pearson's r ≥ 0.999, Slope 0.98, Intercept -2.88 | Pearson's r ≥ 0.999, Slope 1.01, Intercept -1.60 |
2. Sample Sizes and Data Provenance
-
Test Set Sample Sizes:
- Precision (Repeatability & Intermediate): 8 human serum samples and 2 controls (n=10 total) for each assay, measured in 84 determinations over 21 operating days. The exact number of individual patient samples contained within the "8 human serum samples" is not specified but appears to be 8 distinct pools/matrices.
- Precision (Inter-Instrument): 8 native human serum sample pools and 2 quality control levels (n=10 total) for each assay, with 75 determinations per sample/QC level (due to 5 days, 3 laboratories, 25 determinations/site, 5x5=25, 3 sites x 25 = 75 total reported).
- Analytical Sensitivity (LoB): 60 determinations of an analyte-free sample.
- Analytical Sensitivity (LoD): 5 low-level human serum samples, with 60 determinations per reagent lot.
- Analytical Sensitivity (LoQ): 9 native, unaltered serum samples for Elecsys proBNP II and 10 native, unaltered serum samples for Elecsys proBNP II STAT, each tested with 25 measured values.
- Linearity/Assay Reportable Range: One high analyte human, native serum sample diluted to 11 concentrations.
- Endogenous Interference (Bilirubin, Hemoglobin, Lipemia, Biotin, Rheumatoid Factor, Albumin): Three different analyte concentration levels (low, medium, high) in human native serum samples. Specific number of samples at each level not explicitly stated but implied to be several.
- Cross-Reactivity: Two human native serum samples (low and high analyte levels) spiked with potential cross-reactants.
- Exogenous Interference (Drugs): Two human native serum samples (low and high analyte concentrations).
- Matrix Comparisons: Single donor samples drawn into Serum (reference), Li-Heparin (98 pairs), and K2-EDTA plasma (111 pairs).
- Method Comparison: 1928 subjects for Elecsys proBNP II STAT and 1940 subjects for Elecsys proBNP II.
-
Data Provenance: The document generally refers to "human serum samples" and "native human serum samples" without specifying the country of origin. The studies are described as internal (e.g., "one internal site" for precision). There is no explicit mention of the data being retrospective or prospective, but the nature of in vitro diagnostic device performance studies (analytical validation) typically involves prospective testing of samples under controlled laboratory conditions, simulating diagnostic use.
3. Number of Experts and Qualifications for Ground Truth
The document does not describe the use of human experts to establish "ground truth" for the test set in the context of diagnostic interpretation (e.g., radiologists, cardiologists). This document details the analytical performance of an in vitro diagnostic assay, not an AI/imaging diagnostic device that requires expert adjudication of images. The "ground truth" in this context refers to the true analytical concentration of NT-proBNP in the samples, established through well-defined laboratory methodologies and traceable standards, or comparison to a previously cleared predicate device.
4. Adjudication Method for the Test Set
Not applicable. As this is an analytical validation of an in vitro diagnostic assay, there is no "adjudication method" in the sense of multiple human readers or experts resolving discrepancies in diagnostic interpretation. The methods involve quantitative analytical measurements of biochemical markers.
5. Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study
Not applicable. This is not an imaging or AI-assisted diagnostic device that would typically undergo an MRMC study. The study focuses on the analytical performance of a laboratory immunoassay.
6. Standalone (Algorithm Only) Performance
Not applicable in the typical sense for medical imaging AI. The "algorithm" here is the chemical reaction and measurement process of the immunoassay itself. The analytical performance metrics (precision, sensitivity, linearity, interference, matrix comparison) are effectively the "standalone performance" of the device. The method comparison study directly compares the performance of the updated device against its predicate (older version) without human intervention in the result generation.
7. Type of Ground Truth Used
The ground truth used for these analytical studies is primarily measured analytical concentration, established through:
- Reference materials: Calibrators and controls with known concentrations.
- Spiking experiments: Adding known amounts of analyte or interfering substances to samples.
- Dilutions: Creating samples with predictable concentrations.
- Comparison to predicate device: The method comparison studies compare the new device's results against the results from the previously cleared Elecsys proBNP II and Elecsys proBNP II STAT (older versions) as the reference.
- Consensus laboratory methods/standards: Studies like LoB, LoD, LoQ, and precision follow CLSI (Clinical and Laboratory Standards Institute) guidelines, which are established consensus standards for analytical validation.
There is no mention of pathology, clinical outcomes data, or expert consensus interpretation of results as a primary ground truth in this analytical performance section.
8. Sample Size for the Training Set
Not applicable. This document describes the analytical validation of laboratory assays, not a machine learning model that requires a "training set" in the same sense. The assay works based on established biochemical principles (electrochemiluminescence immunoassay, ECLIA, utilizing specific antibodies) and wet-lab procedures, not on learned patterns from a large dataset. Reagent formulation and process optimization might involve internal development data, but it's not a "training set" in the AI/ML context.
9. How the Ground Truth for the Training Set was Established
Not applicable, as there is no "training set" for an AI/ML model in this context. The "ground truth" for the development of the assay itself would be based on fundamental analytical chemistry principles, extensive laboratory testing, control materials, and established reference methods for quantifying NT-proBNP.
§ 862.1117 B-type natriuretic peptide test system.
(a)
Identification. The B-type natriuretic peptide (BNP) test system is an in vitro diagnostic device intended to measure BNP in whole blood and plasma. Measurements of BNP are used as an aid in the diagnosis of patients with congestive heart failure.(b)
Classification. Class II (special controls). The special control is “Class II Special Control Guidance Document for B-Type Natriuretic Peptide Premarket Notifications; Final Guidance for Industry and FDA Reviewers.”