LogoFDA 510(k) Search
K Number
K203489
Device Name
SEFRIA PCP Oral Fluid Enzyme Immunoassay
Manufacturer
Immunalysis Corporation
Date Cleared
2021-04-20

(144 days)

Product Code
LCM
Regulation Number
N/A
Type
Traditional
Panel
TX
Reference & Predicate Devices
K183048,K200801,K181135
AI/MLSaMDIVD (In Vitro Diagnostic)TherapeuticDiagnosticis PCCP AuthorizedThirdpartyExpeditedreview
Intended Use

For In Vitro Diagnostic Use.

The Immunalysis SEFRIA PCP Oral Fluid Enzyme Immunoassay is an enzyme immunoassay with a cutoff of 10 ng/mL in neat oral fluid collected by Quantisal II Oral Fluid Collection Device. The assay is intended for the qualitative and semi-quantitative analysis of PCP in human oral fluid with clinical analyzers. This assay is calibrated against PCP.

The semi-quantitative mode is for purposes of enabling laboratories to determine an appropriate dilution of the specimen for confirmation by a confirmatory method such as Gas Chromatography/Mass Spectrometry (GC-MS) or Liquid Chromatography/Tandem Mass Spectrometry (LC-MS/MS) or permitting laboratories to establish quality control procedures.

The Immunalysis SEFRIA PCP Oral Fluid Enzyme Immunoassay provides only a preliminary analytical test result. A more specific alternate chemical must be used in order to obtain a confirmed analytical result. Gas Chromatography/ Mass Spectrometry (GC-MS) or Liquid Chromatography/Tandem Mass Spectrometry (LC-MS/MS) is the preferred confirmatory method. Clinical consideration and professional judgment should be applied to any test result. particularly when preliminary positive results are used.

Device Description

The Immunalysis SEFRIA PCP Oral Fluid Enzyme Immunoassay is an in vitro diagnostic test to detect the presence of PCP in human oral fluid samples collected by Quantisal or Quantisal II Oral Fluid Collection Device.

AI/ML Overview

Here's a breakdown of the acceptance criteria and the study that demonstrates the device meets them, based on the provided text:

Acceptance Criteria and Device Performance

The document does not explicitly state "acceptance criteria" in a separate table or section with numerical targets. Instead, it presents performance characteristics and implies that the observed results met the requirements for substantial equivalence. The key performance indicators evaluated were precision, interference, linearity/recovery, stability, calibration duration, and method comparison.

For the purpose of this analysis, I will infer the acceptance criteria from common expectations for FDA-cleared diagnostic devices, particularly for qualitative and semi-quantitative assays, and present the reported device performance against these implicit criteria.

Table of Acceptance Criteria and Reported Device Performance

Acceptance Criteria CategorySpecific Criterion (Inferred)Reported Device Performance
PrecisionQualitative: Accurate classification (Negative/Positive) at and around the cutoff (10 ng/mL).Qualitative:
* Below Cutoff (-100% to -25%): 60/60 Negative (100% accuracy)
* At Cutoff (10 ng/mL): 29 Neg / 31 Pos (indicating expected variability at the cutoff)
* Above Cutoff (+25% to +100%): 60/60 Positive (100% accuracy)
Semi-Quantitative: Mean concentration values close to expected, with accurate classification at and around the cutoff.Semi-Quantitative:
* Below Cutoff (-100% to -25%): Mean concentrations very close to expected, 60/60 Negative.
* At Cutoff (10 ng/mL): Mean concentration 10.1 ng/mL, 30 Neg / 30 Pos.
* Above Cutoff (+25% to +100%): Mean concentrations very close to expected, 60/60 Positive.
Specificity & Cross-ReactivityNo significant interference or false positives/negatives from structurally similar compounds or other common substances.Data reported in K181135 (predicate device submission). Implies acceptable performance for the candidate device.
Interference (Other)No significant interference from structurally unrelated compounds, endogenous compounds, or exogenous compounds (e.g., orally used products).Orally Used Exogenous Compounds: No interference observed with any of the tested compounds (e.g., Teeth Whitener, Hydrogen Peroxide, Cigarette, Hard Candy, Chewing Gum, Cough Syrup) at tested concentrations, for both qualitative and semi-quantitative modes (data for Quantisal II device reported in K181135).
Other Interferents: Data for Structurally Unrelated, Endogenous, and pH interference reported in K181135. Implies acceptable performance.
Linearity/RecoveryRecovery within an acceptable range (e.g., 80-120%) across the linear range.Linear range confirmed to be 4-40 ng/mL. Recovery percentages ranged from 97.5% to 111.4%, indicating good recovery across the tested range. (Data for Quantisal II device reported in K181135).
PCP StabilityOral fluid samples containing PCP remain stable for a specified period under recommended storage conditions.Oral fluid samples containing PCP are stable for up to 12 months at 2°C - 8°C in Quantisal II Oral Fluid Collection Device. Data for 10-day storage at ambient temperature (8°C - 25°C) in Quantisal II were reported in K183048 and K200801.
Calibration DurationDevice maintains performance within acceptable limits for a specified duration between calibrations.The recommended frequency of calibration is 14 days, as test results met acceptance criteria at each time point in a study up to 14 days. Data for qualitative mode reported in K181135.
Method ComparisonHigh agreement rates (e.g., usually >95% or 98%) with a confirmed analytical method (LC-MS/MS) for both positive and negative samples.100% Agreement for both Qualitative and Semi-Quantitative modes across all observed PCP concentration ranges (Quantisal: 40/40 Positive, 40/40 Negative; Quantisal II A: 40/40 Positive, 40/40 Negative; Quantisal II B: 40/40 Positive, 40/40 Negative). This indicates excellent concordance with LC-MS/MS. This high agreement suggests the device effectively identifies PCP presence and absence relative to the gold standard.

Study Details

The provided document describes a series of laboratory performance studies to demonstrate the substantial equivalence of the SEFRIA PCP Oral Fluid Enzyme Immunoassay.

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

    • Precision (Quantitative Test Set): For each concentration level (0 ng/mL, 2.5 ng/mL, 5 ng/mL, 7.5 ng/mL, 10 ng/mL, 12.5 ng/mL, 15 ng/mL, 17.5 ng/mL, 20 ng/mL), there were 60 determinations.
      • Data Provenance: Drug-free negative oral fluid was "spiked" to target concentrations. This is a controlled laboratory study, not directly from human patients. The "Data for the candidate device used with Quantisal II device were reported in K181135" indicates that some results, specifically for Quantisal II, were referenced from a previous submission, suggesting a mix of new and previously generated data on the device platform.
    • Interference - Orally Used Endogenous Compounds: For each compound tested (e.g., Teeth Whitener, Cigarette), samples were prepared by volunteers using the substance, then spiked with PCP at ±25% of the cutoff. The exact number of samples for each compound is not specified, but the conclusion states "No interference was observed with any of the compounds," implying sufficient testing.
      • Data Provenance: Oral fluid collected from "volunteers" after use of substances. This implies prospective collection in a controlled setting.
    • Linearity/Recovery: Multiple pools were created by serial dilution. Each pool was tested in triplicate. (Number of distinct samples not explicitly stated, but 13 concentration levels were tested in triplicate for 39 total runs of the assay).
      • Data Provenance: Drug-free oral fluid pools spiked with PCP. Controlled laboratory study.
    • PCP Stability in Oral Fluid: Samples spiked with PCP were stored and tested periodically. The exact number of samples per time point is not specified but referenced baseline concentration results.
      • Data Provenance: Spiked drug-free negative oral fluid. Controlled laboratory study.
    • Calibration Duration: Samples spiked with PCP at ±25% of the cutoff were tested at time points up to 14 days. Exact number of samples per time point not specified.
      • Data Provenance: Spiked drug-free negative oral fluid. Controlled laboratory study.
    • Method Comparison (Clinical Test Set): 80 deidentified, unaltered clinical oral fluid samples.
      • Data Provenance: Obtained from drug treatment facilities. This indicates retrospective collection from human subjects in a real-world clinical setting.

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

    • Precision, Interference, Linearity/Recovery, Stability, Calibration Duration: For these analytical studies, the ground truth was established by carefully controlled spiking concentrations of PCP, confirmed by mass spectrometry (LC-MS/MS) before collection (for precision) or by reference methods. This doesn't involve "experts" in the clinical sense, but rather relies on the accuracy of the laboratory's preparation and analytical techniques.
    • Method Comparison: For the 80 clinical samples, the ground truth was established by Liquid Chromatography-Tandem Mass Spectrometry (LC-MS/MS). This is a highly sensitive and specific confirmatory analytical method and acts as the gold standard, so no human experts are used for this type of ground truth.

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

    • Adjudication methods (like 2+1 or 3+1 consensus by experts) are typically used in studies involving subjective interpretation, such as imaging or pathology, where human readers create the initial "truth."
    • In this context, where the ground truth is established by objective analytical methods (LC-MS/MS, or precisely prepared spiked samples), there is no human adjudication process described or needed. The analytical results are the ground truth.

  4. If a multi-reader, multi-case (MRMC) comparative effectiveness study was done:

    • No, an MRMC comparative effectiveness study was not done. This type of study (comparing human readers with and without AI assistance) is not applicable to an in vitro diagnostic device like an enzyme immunoassay, which is a standalone automated analytical test. The "readers" here are the instruments and their output, not human interpreters.

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

    • Yes, the performance studies described are inherently "standalone." The SEFRIA PCP Oral Fluid Enzyme Immunoassay is an automated system (used with clinical analyzers like the Beckman Coulter AU480). All performance characteristics (precision, linearity, method comparison, etc.) evaluate the device's analytical output without human intervention in the interpretation of the primary result. Human intervention comes after the preliminary positive result, where confirmation by GC-MS or LC-MS/MS is required.

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

    • Primary Ground Truth: For the method comparison and for confirming spiked concentrations, the analytical gold standard was Liquid Chromatography-Tandem Mass Spectrometry (LC-MS/MS).
    • Other Ground Truths: For precision, linearity/recovery, stability, and calibration duration, the ground truth was established by precisely prepared spiked samples with known concentrations, often verified by LC-MS/MS.

  7. The sample size for the training set:

    • The document describes performance validation studies for the device, not the development or training of a machine learning model. Therefore, a "training set" in the context of AI/ML is not explicitly mentioned or relevant here. The device is an enzyme immunoassay, which operates on chemical reactions and optical detection, not a machine learning algorithm that requires a training set.

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

    • As there's no mention of a machine learning "training set," this question is not applicable. The assay's "learning" or optimization would have occurred during its internal development and formulation, not through an enumerated "training set" in the context of an FDA submission for an in vitro diagnostic immunoassay.

N/A