(129 days)
The BioPlex™ 2200 EBV IgG kit is a multiplex flow immunoassay intended for the qualitative detection of IgG antibodies to three (3) separate EBV antigens; Epstein-Barr Virus Nuclear Antigen-1 (EBV NA-1), Viral Capsid Antigen (EBV VCA), and Early Antigen diffuse (EBV EA-D) in human serum. The test system can be used in conjunction with the BioPlex 2200 EBV IgM kit as an aid in the laboratory diagnosis of infectious mononucleosis (IM).
The EBV IgG kit is intended for use with the Bio-Rad BioPlex 2200 System.
Assay performance characteristics have not been established for immunocompromised or immunosuppressed patients, cord blood, neonatal specimens, or infants. Assay performance characteristics have not been established for the diagnosis of nasopharyngeal carcinoma, Burkitt's lymphoma, and other EBV-associated lymphomas.
The EBV IgG kit uses multiplex flow immunoassay, a methodology that greatly resembles traditional EIA, but permits simultaneous detection and identification of many antibodies in a single tube. Three (3) different populations of beads are coated with E. coli derived recombinant proteins, EBV NA-1 (28kD and 45kD), EBV VCA p18 (40kD), and EBV EA-D (28kD) associated with infectious mononucleosis. The BioPlex 2200 System combines an aliquot of patient sample, sample diluent, and bead reagent into a reaction vessel. The mixture is incubated at 37°C. After a wash cycle, antihody, conjugated to phycoerythrin (PE), is added to the dyed beads and this mixture is incubated at 37°C. The excess conjugate is removed in another wash cycle, and the beads are re-suspended in wash buffer. The bead mixture then passes through the detector. The identity of the dyed beads is determined by the fluorescence of the dyes, and the amount of antibody captured by the antigen is determined by the fluorescence of the attached PE. Raw data is calculated in relative fluorescence intensity (RFI).
Three additional dyed beads, an Internal Standard Bead (ISB), a Serum Verification Bead (SVB) and a Reagent Blank Bead (RBB) are present in each reaction mixture to verify detector response, the addition of serum or plasma to the reaction vessel and the absence of significant non-specific binding in serum or plasma. Refer to the BioPlex 2200 System Operation Manual for more information. The instrument is calibrated using a set of seven (7) distinct calibrator vials, supplied separately by Bio-Rad Laboratories. A combination of four (4) vials representing four (4) different antibody concentrations are used for semi-quantitative calibration. The result for each of these antibodies is expressed as an antibody index (AI).
Here's a summary of the acceptance criteria and study details for the BioPlex 2200 EBV IgG Kit, Calibrators, and Controls, based on the provided text:
1. Table of Acceptance Criteria and Reported Device Performance
The document does not explicitly state pre-defined acceptance criteria for the different performance metrics (reproducibility, precision, comparative testing). Instead, it presents the results of these studies as the "Performance Summary." Therefore, the reported performance itself serves as the basis for demonstrating the device's capabilities.
| Acceptance Criteria (Implied) | Stated Device Performance |
|---|---|
| Reproducibility (Consistency across runs, days, sites) | EBV NA-1 IgG: Total CVs for high and low positive panels ranged from 8.5% to 13.8%. Positive Control Total CV was 17.5%. |
| EBV VCA IgG: Total CVs for high and low positive panels ranged from 8.9% to 17.8%. Positive Control Total CV was 6.9%. | |
| EBV EA-D IgG: Total CVs for high and low positive panels ranged from 8.1% to 10.9%. Positive Control Total CV was 19.2%. (Tables 19-21) | |
| Precision (Consistency within Bio-Rad lab) | EBV NA-1 IgG: Total CVs for high and low positive panels ranged from 9.6% to 13.1%. Negative samples had CVs up to 14.5% (High Negative). |
| EBV VCA IgG: Total CVs for high and low positive panels ranged from 8.7% to 13.3%. Negative samples had CVs up to 18.4% (Low Negative). | |
| EBV EA-D IgG: Total CVs for high and low positive panels ranged from 8.3% to 14.9%. Negative samples had CVs up to 30.3% (Low Negative). (Tables 22-24) | |
| Comparative Performance (vs. Predicate EIA) (Agreement with established methods for serological status) | Agreement by Serological Pattern Characterization (BioPlex vs. EIA): |
- EBV NA-1 IgG: Overall Positive Agreement: 95.6% (95% CI: 93.2 - 97.1%). Overall Negative Agreement: 97.4% (95% CI: 94.0 - 98.9%). (Table 27)
- EBV VCA IgG: Overall Positive Agreement: 96.2% (95% CI: 93.9 - 97.6%). Overall Negative Agreement: 99.4% (95% CI: 96.8 - 99.9%). (Table 29)
- EBV EA-D IgG: Overall Positive Agreement: 88.1% (95% CI: 81.1 - 92.8%). Overall Negative Agreement: 84.1% (95% CI: 80.6 - 87.0%). (Table 31) |
| Serological Status Agreement (BioPlex EBV IgG & IgM vs. Predicate Assays) | Comparison of EBV Serological Status: Overall Serological Agreement: 83.3% (95% CI: 80.2 - 86.1%).
Comparison of Acute and Non-acute EBV Serological Status: Overall Serological Agreement: 85.1% (95% CI: 82.1 - 87.7%). (Tables 32-33) |
| Cross-Reactivity (Minimal interference from other conditions) | The majority of samples that elicited a positive result with the BioPlex were also confirmed positive by the corresponding commercially available microplate EIA, indicating reactivity to EBV IgG antibodies rather than cross-reactivity with an interfering factor. (Table 34) |
2. Sample Size Used for the Test Set and Data Provenance
- Sample Size for Comparative Testing: A total of 621 banked serum samples from patients for whom an EBV test was ordered were tested. (This number reduced slightly due to analysis errors, with 618 or fewer samples used in some specific analyses).
- Data Provenance: The samples were "banked serum samples" and tested at 3 U.S. clinical testing sites. The study is retrospective, utilizing existing banked samples.
3. Number of Experts Used to Establish the Ground Truth for the Test Set and Their Qualifications
The document does not explicitly state that experts (e.g., radiologists) were used to establish ground truth for the test set. Instead, the ground truth for comparative performance and serological status was established by "corresponding commercially available microplate EIAs" and "commercially available microplate EIA and agglutination tests." The serological characterization in Table 25 (e.g., Primary Acute, Late Acute) appears to be a predefined algorithm based on results from these predicate assays.
4. Adjudication Method for the Test Set
No adjudication method by human experts is described for the test set. The interpretation of results (positive, equivocal, negative) for the BioPlex 2200 EBV IgG kit was compared directly against the results of predicate EIA assays. For percent agreement calculations, "equivocal results were assigned to the opposite clinical interpretation than that of the corresponding reference assay result" and vice versa for predicate equivocal results (Page 13).
5. Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study
No MRMC comparative effectiveness study was done. This study focuses on the in-vitro diagnostic device's performance compared to predicate devices, not on human reader performance with or without AI assistance.
6. Standalone (Algorithm Only) Performance
Yes, the study primarily evaluates the standalone performance of the BioPlex 2200 EBV IgG Kit (an in-vitro diagnostic device, essentially an algorithm and assay combined) against predicate EIA assays. There is no human-in-the-loop component described for the interpretation of the BioPlex results in these performance studies; the instrument provides an antibody index (AI) which is then categorized as positive, equivocal, or negative based on predefined cutoffs (≤0.8 Al negative, 0.9 and 1.0 Al equivocal, and ≥1.1 Al positive for analytes, Page 6).
7. Type of Ground Truth Used
The ground truth used for the comparative effectiveness study was the results from "corresponding commercially available microplate EIAs" for individual analytes and "commercially available microplate EIA and agglutination tests" for comprehensive EBV serological status. This is a type of reference standard or predicate device comparison.
8. Sample Size for the Training Set
The document does not mention a distinct "training set" for an algorithm or AI model in the conventional sense. This is a 510(k) submission for an in-vitro diagnostic kit, not an AI/ML device where a separate training set would typically be described. The "expected values" section (Tables 13-18) describes the prevalence of EBV IgG antibodies in different patient populations. While these samples contribute to understanding the device's characteristics in various populations, they are presented as "expected values" rather than a formal training set for an algorithm.
9. How the Ground Truth for the Training Set Was Established
As there is no explicitly defined "training set" for an AI/ML algorithm, the concept of establishing ground truth for it does not apply directly in this document. The device's calibration involves a "set of seven (7) distinct calibrator vials" (Page 1), which are used to assign antibody index (AI) values. The specific concentrations or reference values for these calibrators are not detailed, but they would be established by the manufacturer according to internal standards and quality control processes.
§ 866.3235 Epstein-Barr virus serological reagents.
(a)
Identification. Epstein-Barr virus serological reagents are devices that consist of antigens and antisera used in serological tests to identify antibodies to Epstein-Barr virus in serum. The identification aids in the diagnosis of Epstein-Barr virus infections and provides epidemiological information on diseases caused by these viruses. Epstein-Barr viruses are thought to cause infectious mononucleosis and have been associated with Burkitt's lymphoma (a tumor of the jaw in African children and young adults) and postnasal carcinoma (cancer).(b)
Classification. Class I (general controls).