FDA 510(k) Search - By Innolitics (SaMD and AI Experts)

The iDart™ Lyme IgM ImmunoBlot Kit is an immunoblot assay intended for the in vitro qualitative detection of IgM antibodies to Borrelia burgdorferi in human serum. The iDart™ Lyme IgM ImmunoBlot Kit is intended to detect antibodies to Lyme Screen Antigen (LSA) and multiple other B. burgdorferi antigens following a modified two-tier test methodology. Positive results from the iDart™ Lyme IgM ImmunoBlot Kit are supportive evidence for the presence of antibodies and exposure to B. burgdorferi. Negative results do not preclude infection with B. burgdorferi. iDart™ Lyme IgM ImmunoBlot Kit is intended to aid in the diagnosis of Lyme disease and the test kit should only be used on samples from patients with clinical history, signs and symptoms consistent with Lyme disease. The iDart™ Lyme IgM Immunoblot Kit is not intended as a screen for asymptomatic patients.

Test results are to be used in conjunction with information obtained from the patient's clinical evaluation and other diagnostic procedures.

Device Description

The iDart™ Lyme IgM ImmunoBlot tests are line immunoblot assays. Antigenic proteins specific for Borrelia species that cause Lyme Disease are produced by recombinant DNA technology in Escherichia coli. The purified proteins are then applied as discrete lines on a nitrocellulose membrane along with two control proteins. The iDart™ Lyme IgM ImmunoBlot Kit contains IgM ImmunoBlot strips and the proteins are applied in the following order: C1 (IgG/IgM – conjugate control), C2 (Protein L – calibrator/serum control), P93, P41 (2 antigen bands), P39 (2 antigen bands), P23 (9 antigen bands), P31 (9 antigen bands), P34, C10 and LSA (a chimeric VlsE peptide termed the Lyme Screen Antigen).

AI/ML Overview

Here's an analysis of the acceptance criteria and study detailed in the provided FDA 510(k) clearance letter for the iDart™ Lyme IgM ImmunoBlot Kit.

Note: The provided text details the performance characteristics but does not explicitly state "acceptance criteria" in a separate section with numerical thresholds. Therefore, the "Acceptance Criteria" column in the table below is inferred from the achieved "Reported Device Performance" and common expectations for such devices (e.g., high agreement, specificity, and sensitivity). The "study that proves the device meets the acceptance criteria" refers to the clinical and analytical performance studies conducted.

1. Table of Acceptance Criteria and Reported Device Performance

Criteria Category	Acceptance Criteria (Inferred)	Reported Device Performance
Reproducibility	High agreement across sites, operators, and runs (e.g., ≥95% agreement)	100% agreement on all test results across runs, days, and sites for all panel samples (negative, high negative, low positive, moderate positives, high positive). (N=90 replicates per sample across 3 sites, 2 operators, 5 days, 2 runs/day).
Analytical Specificity (Endemic)	High Agreement (e.g., ≥95%) in endemic healthy population	99.44% specificity (1 false positive out of 177 samples) on samples from apparently healthy individuals in endemic areas (US, CDC, Bay Area Foundation NY, MA, WI).
Analytical Specificity (Non-Endemic)	High Agreement (e.g., ≥95%) in non-endemic healthy population	99.21% specificity (1 false positive out of 127 samples) on samples from apparently healthy individuals in non-endemic areas (US, CDC, IGeneX, Inc.).
Cross-Reactivity	High specificity (e.g., ≥95%) in samples with potentially interfering conditions	97.94% specificity (5 false positives out of 243 samples) in cross-reactivity study (various bacterial/viral infections, autoimmune diseases). Note: 2 Leptospira samples and 1 Mononucleosis, 1 H. pylori, 1 Parvovirus-19 sample were positive with iDart Lyme IgM. The Leptospira samples were also positive by STTT.
Interference (Endogenous Analytes)	No interference from common endogenous substances at specified concentrations	No interference observed from Bilirubin (1, 15mg/dL), Albumin (3.5, 5g/dL), Cholesterol (150, 250mg/dL), Triglycerides (150, 500mg/dL), Hemoglobin (10, 20g/dL) in positive, low positive, and negative Borrelia IgM samples.
Method Comparison (PPA)	Acceptable Positive Percent Agreement (PPA) with STTT (e.g., lower bound of 95% CI > 80-85%)	90.91% PPA (95% CI: 81.55%– 95.77%) against STTT in a cohort of 997 serum samples.
Method Comparison (NPA)	Acceptable Negative Percent Agreement (NPA) with STTT (e.g., lower bound of 95% CI > 95%)	98.07% NPA (95% CI: 96.96%– 98.77%) against STTT in a cohort of 997 serum samples.
Clinical Sensitivity (CDC Panel)	Acceptable sensitivity across disease stages compared to STTT (e.g., comparable or better performance)	Overall Sensitivity (CDC Panel): 65.8% compared to STTT's 55.7%. Stage I: 66.0% (STTT 56.0%), Stage II: 88.9% (STTT 77.8%), Stage III: 55.0% (STTT 45.0%). The iDart kit showed higher sensitivity than STTT across all stages in this particular panel.
Clinical Specificity (CDC Healthy Controls)	High specificity (e.g., ≥95%) in healthy controls for CDC panel	97.9% agreement (3.1% false positive rate) for healthy controls within the CDC panel (N=95). Note that the table reports "Agreement," which implies specificity here.
Clinical Specificity (CDC Disease Controls)	High specificity (e.g., ≥95%) in disease controls for CDC panel	98.9% agreement (1.1% false positive rate) for disease controls within the CDC panel (N=84). Note that the table reports "Agreement," which implies specificity here.
Fresh vs. Frozen Samples	Equivalent performance between fresh and frozen samples	Performance is equivalent between fresh and frozen samples. (N=63 fresh, N=63 frozen; same positive/negative counts).
Antibody Class Specificity	Specific detection of IgM antibodies (i.e., minimal cross-reactivity with IgG)	The kit specifically detected IgM; treatment with human IgM blocked reactivity, while treatment with human IgG did not affect reactivity in positive samples. Negative samples remained negative under all conditions.

Study Proving Device Meets Acceptance Criteria

The studies detailed in the "PERFORMANCE CHARACTERISTICS" section of the 510(k) summary are designed to prove that the iDart™ Lyme IgM ImmunoBlot Kit meets the necessary performance standards for its intended use. These include:

Reproducibility Study: Tested consistency across multiple sites, operators, and runs.
Analytical Specificity Studies: Evaluated performance in healthy endemic and non-endemic populations, and cross-reactivity with various non-Borrelia pathogens and autoantibodies.
Interference Study: Assessed impact of common endogenous substances.
Method Comparison with STTT: Compared the device's accuracy (PPA and NPA) against the Standard Two-Tier Test methodology, which is a recognized approach for Lyme disease diagnosis.
CDC Serum Panel Study: Evaluated sensitivity and specificity across different disease stages and in various control groups using a well-characterized reference panel.
Fresh and Frozen Samples Comparison Study: Demonstrated stability of samples under different storage conditions.
Antibody Class Specificity Study: Confirmed the specific detection of IgM antibodies.

Additional Information:

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

Reproducibility: A panel of coded samples with different levels of anti-B. burgdorferi IgM (negative, high negative, low positive, moderate positives, and high positive samples). 90 replicates per sample were generated. The specific number of distinct samples in this panel is not explicitly stated beyond "a panel of coded samples".
Analytical Specificity (Endemic): 177 samples from apparently healthy individuals in the US from endemic areas (CDC and Bay Area Foundation NY, MA, WI).
Analytical Specificity (Non-Endemic): 127 samples from apparently healthy individuals in the US from non-endemic areas (CDC and IGeneX, Inc.).
Cross-Reactivity Study: 243 serum samples from patients with bacterial or viral infections, as well as sera from patients with diagnoses that could be confused with Lyme disease (CDC, IGeneX, Inc., New York Biologics (NY), Kamineni Life Sciences Pvt. Ltd, Hydrabad (India), Warde Medical Laboratory (MI)).
Interference Study: One positive, one low positive, and one negative Borrelia IgM samples (tested in singlicate with various spiked interfering agents).
Method Comparison with STTT: 997 prospectively collected serum samples procured from IGeneX, Inc.
CDC Serum Panel: 258 serum samples from the CDC panel. These included patients with Lyme disease at different stages and Lyme disease look-alike infections, and healthy controls.
Fresh and Frozen Samples Comparison Study: 63 fresh samples and 63 frozen samples.
Antibody Class Specificity: 8 previously tested patient samples (4 negatives, 4 positives).

Data Provenance: The data appears to be a mix of retrospective and prospective. Samples from "apparently healthy individuals" and "patients with bacterial or viral infections" can be either. The "Method Comparison with STTT" explicitly states "Prospectively collected serum samples" from IGeneX, Inc. The CDC Serum Panel is a reference panel, usually well-characterized retrospectively. Locations include the US (CDC, Bay Area Foundation NY, MA, WI, IGeneX, Inc., New York Biologics (NY), Warde Medical Laboratory (MI)) and India (Kamineni Life Sciences Pvt. Ltd, Hydrabad).

3. Number of experts used to establish the ground truth for the test set and the qualifications of those experts:
The document does not explicitly state the number of experts or their qualifications for establishing ground truth for the clinical study samples. For the "Method Comparison with comparators (STTT)", the comparison is against "FDA-cleared EIA and immunoblot as part of the standard two-tier test methodology (STTT)". This implies the ground truth for these samples was established by the results of the STTT, which is a standardized and accepted diagnostic algorithm, rather than by individual expert interpretation. For the CDC Panel, the samples are "from patients diagnosed with Lyme Disease at different stages" or "Lyme disease look-like infections", suggesting that the CDC provided established diagnoses based on their protocols, which would implicitly involve expert clinical and laboratory evaluation.

4. Adjudication method (e.g. 2+1, 3+1, none) for the test set:
The document does not describe an explicit adjudication method (like 2+1 or 3+1) for establishing ground truth for the clinical test sets. The ground truth for the method comparison study was the result of the "FDA-cleared EIA and immunoblot as part of the standard two-tier test methodology (STTT)". For the CDC panel, it was based on the provided "diagnosis".

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 MRMC comparative effectiveness study was done. This device is an in vitro diagnostic (IVD) kit for laboratory use (an ImmunoBlot assay, specifically evaluating IgM antibodies) rather than an AI-powered image analysis system or a device requiring human-in-the-loop interpretation that would typically necessitate such a study. The "reading" of the immunoblot strips is performed visually but the interpretation criteria are clearly defined, not requiring subjective interpretive discretion usually addressed by MRMC studies.

6. If a standalone (i.e. algorithm only without human-in-the-loop performance) was done:
This question is not directly applicable in the terms typically used for AI/software devices. The iDart™ Lyme IgM ImmunoBlot Kit is a lab-based immunoassay. Its performance characteristics (sensitivity, specificity, reproducibility) were evaluated as a standalone device (without external assistance like AI or human readers for interpretation, other than the standard visual reading of the bands per the kit's instructions), as demonstrated by the analytical and clinical studies. The results are generated by the chemical reaction on the strip and interpreted based on predefined criteria. It's an "algorithm only" in the sense of the assay's biochemical process and visual interpretation rules, operating without additional "human-in-the-loop" modifications to its fundamental mechanism or output.

7. The type of ground truth used (expert consensus, pathology, outcomes data, etc):
The ground truth used primarily appears to be:

Standard Two-Tier Test (STTT) results: For the majority of the clinical method comparison study (N=997), the device was compared against the results from "FDA-cleared EIA and immunoblot as part of the standard two-tier test methodology".
Clinical Diagnosis from CDC: For the CDC reference panel, samples were from patients "diagnosed with Lyme Disease at different stages" or "Lyme disease look-like infections". This implies a ground truth established through a combination of clinical evaluation, laboratory tests, and expert judgment according to CDC protocols.

8. The sample size for the training set:
Not applicable/Not specified. This is an in vitro diagnostic kit (ImmunoBlot assay), not a machine learning or AI-based device that typically requires a distinct "training set" for model development. The antigens used are recombinant proteins, and the assay's interpretation criteria are rule-based, not learned from a dataset.

9. How the ground truth for the training set was established:
Not applicable. As stated above, this device does not utilize a training set in the context of machine learning. The design and validation of immunoblots are based on known antigen-antibody interactions and assay development methodologies.

Summary

Regulation Number and Section

§ 866.3830

Treponema pallidum treponemal test reagents.(a)
Identification. Treponema pallidum treponemal test reagents are devices that consist of the antigens, antisera and all control reagents (standardized reagents with which test results are compared) which are derived from treponemal sources and that are used in the fluorescent treponemal antibody absorption test (FTA-ABS), theTreponema pallidum immobilization test (T.P.I.), and other treponemal tests used to identify antibodies toTreponema pallidum directly from infecting treponemal organisms in serum. The identification aids in the diagnosis of syphilis caused by bacteria belonging to the genusTreponema and provides epidemiological information on syphilis.(b)
Classification. Class II (performance standards).