Search Results

The BioArray Solutions ENA IgG BeadChip™ Test System is intended for use in testing human serum for the presence of human IgG class antibodies to six extractable nuclear antigens, SSA, SSB, Sm, Sm/RNP, Scl-70, and Jo-1. The presence of these autoantibodies can be used in conjunction with clinical findings and other laboratory tests to aid in the diagnosis of systemic lupus erythematosus, Sjögren's syndrome, scleroderma and myositis. This ENA IgG BeadChip is for use with Array Imaging System 400 (AIS400). This test is for in vitro diagnostic use.

The ENA IqG BeadChip™ Test System is a randomly encoded microarray-based immunoassay for antibodies to extractable nuclear antigens. The device is designed to detect IgG class antibodies in human serum to six extractable nuclear antigens: SSA, SSB, Sm, RNP/Sm, Jo-1, and SCL-70. Each antigen is covalently immobilized to a spectrally distinguishable bead type. A and of bead types is constructed by mixing all of the bead types of interest including antigen beads, positive control beads, negative control beads, and system control beads. The bead mixture is immobilized as a BeadChip™ microarray on a silicon chip allowing for the simultaneous detection of the auto-antibodies of interest by the AIS 400.

Here's an analysis of the acceptance criteria and study details from the provided 510(k) summary:

Acceptance Criteria and Device Performance

The device's performance is primarily assessed through a comparison with a predicate device (ENA IgG ImmuStrip™ Test System). The acceptance criteria are implicitly defined by the reported "Percentage Positive Agreement" and "Percentage Negative Agreement" with the predicate device for each antigen.

Table of Acceptance Criteria and Reported Device Performance

Note: The specific numerical acceptance criteria (e.g., "must be >90%") are not explicitly stated in the document. They are inferred from the reported high agreement percentages, indicating that the device aims to closely match the predicate's results.

Study Details

1. Sample size used for the test set and the data provenance:

   • Sample Size (Method Comparison Study): 229 samples were used for the method comparison study against the predicate device for each antigen. However, the agreement tables show varying "sub-total" numbers for analysis depending on the antigen due to "EQU" (equivocal) or "ND" (not determined) results. For example, for SSA, the total included in the agreement calculation was 218.
   • Sample Size (Site-to-Site Reproducibility): 176 samples (52 normal, ~25 positive for each antigen) were provided to three testing sites. A total of 192 samples were run at each site, with 181 being included in the final analysis after exclusions.
   • Data Provenance: Not explicitly stated but inferred to be clinical human serum samples. The document does not specify countries of origin. The studies appear to be prospective, as they involve testing samples for performance evaluation.

2. Number of experts used to establish the ground truth for the test set and the qualifications of those experts:

   • The primary "ground truth" used for the method comparison study is the result obtained from the predicate device, the ENA IgG ImmuStrip™ Test System. The predicate device itself would have undergone its own validation studies.
   • For the reproducibility and cutoff determination studies, clinical samples (pooled positive, pooled negative, normal samples, and disease positive samples) were used. These samples would have had a pre-established status (e.g., "positive for anti-SSA") which would serve as the ground truth. The section does not specify the number or qualifications of experts involved in establishing the initial status of these samples.

3. Adjudication method (e.g., 2+1, 3+1, none) for the test set:

   • The document describes a comparison against a predicate device rather than expert consensus adjudication for establishing ground truth for the method comparison study. The predicate device's result is taken as the reference.
   • For the site-to-site reproducibility, "randomly encoded" samples were distributed, and agreement between the sites was assessed. No explicit adjudication method for discrepant results between sites is mentioned, beyond excluding samples due to "specimen control failure" or "sample preparation error."

4. 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 device is an in vitro diagnostic immunoassay system, not an AI-assisted diagnostic tool requiring human reader interpretation in the context of MRMC studies. The detection method is fluorometric and analyzed by a program, not directly by human readers.

5. If a standalone (i.e. algorithm only without human-in-the-loop performance) was done:

   • Yes, a standalone performance study was done. The device itself, consisting of the ENA IgG BeadChip™ and the Array Imaging System 400 (AIS 400) with its ENA Analysis program, generates the test results. The method comparison study and reproducibility studies assess the performance of this system independently in generating results comparable to the predicate device and in achieving consistent results.

6. The type of ground truth used (expert consensus, pathology, outcomes data, etc.):

   • The primary ground truth for the method comparison study is the results from the legally marketed predicate device (ENA IgG ImmuStrip™ Test System).
   • For reproducibility and cutoff determination, the ground truth for "positive" and "negative" samples (e.g., pooled positive sample, pooled normal serum sample, disease positive samples, normal samples) would likely be based on clinical diagnosis and/or established laboratory characterization of these samples.

7. The sample size for the training set:

   • The submission does not explicitly describe a "training set" in the context of a machine learning or AI model development. This device is an immunoassay system.
   • However, for cutoff determination, 130 normal samples were used to establish the normal range and positive cutoff values for each antigen. This could be considered analogous to a reference or normalization set rather than a training set for an algorithm in the AI sense.

8. How the ground truth for the training set was established:

   • As there isn't a "training set" in the AI sense, the establishment of ground truth for method comparison relies on the predicate device's results.
   • For the cutoff determination, the ground truth for the 130 normal samples was that they were from "healthy, asymptomatic human subjects." This is established through clinical criteria and screening for apparent health.

Ask a specific question about this device