Search Results

Alinity m Resp-4-Plex is a multiplexed real-time in vitro reverse transcription polymerase chain reaction (RT-PCR) assay for use with the automated Alinity m System for the qualitative detection and differentiation of Severe Acute Respiratory Syndrome Coronavirus 2 (SARS-CoV-2), influenza A virus, influenza B virus and Respiratory Syncytial Virus (RSV) in nasopharyngeal swab specimens collected from patients with signs and symptoms of respiratory tract infection. Clinical signs and symptoms of respiratory tract infection due to SARS-CoV-2, influenza B, and RSV can be similar.

The Alinity m Resp-4-Plex assay is intended for use in the differential detection of SARS-CoV-2, influenza B and/or RSV RNA and aids in the diagnosis of COVID-19, influenza and/or RSV infections if used in conjunction with other clinical and epidemiological information, and laboratory findings. SARS-CoV-2, influenza B and RSV viral RNA are generally detectable in nasopharyngeal swab specimens during the acute phase of infection. This test is not intended to detect influenza C virus infections.

Positive results are indication of the identified virus, but do not rule out bacterial infection or co-infection with other pathogens not detected by the test. The agent(s) detected by the Alinity m Resp-4-Plex assay may not be the definite cause of disease.

Negative results do not preclude SARS-CoV-2, influenza B and/or RSV infections and should not be used as the sole basis for diagnosis, treatment or other patient management decisions.

The Alinity m Resp-4-Plex assay requires two separate assay-specific kits: Alinity m Resp-4-Plex AMP Kit and Alinity m Resp-4-Plex CTRL Kit. The assay utilizes real-time PCR to amplify and detect genomic RNA sequences of influenza A (flu A), influenza B (flu B), RSV, and/or SARS-CoV-2 from nasopharyngeal (NP) swab specimens. The assay targets 2 different genes within the SARS-CoV-2 genome. Fluorescently labeled probes allow for simultaneous detection and differentiation of amplified products of all 4 viruses and Internal Control (IC) in a single reaction vessel. All steps of the assay procedure are executed automatically by the Alinity m System, which is a continuous random-access analyzer. The system performs automated sample preparation using magnetic microparticle technology. The IC is introduced into each specimen at the beginning of sample preparation. Purified RNA is combined with activation and amplification/detection reagents and transferred to a reaction vessel for reverse transcription, PCR amplification, and real-time fluorescence detection. A positive and negative control are tested to ensure performance. Patient results are automatically reported. The assay also utilizes the Alinity m Resp-4-Plex Assay Application Specification File, Alinity m System and System Software, Alinity m Sample Prep Kit 2, Alinity m Tubes and Caps, and Alinity m System Solutions.

The provided text is a 510(k) Summary for the Abbott Molecular Inc. Alinity m Resp-4-Plex assay, a multiplexed real-time RT-PCR assay for the qualitative detection and differentiation of SARS-CoV-2, influenza A, influenza B, and RSV.

Here's an analysis of the acceptance criteria and the study that proves the device meets them, based on the provided document:

Acceptance Criteria and Reported Device Performance

The acceptance criteria for this device are demonstrated through various analytical and clinical studies, primarily focusing on analytical sensitivity (Limit of Detection), inclusivity, precision, reproducibility, analytical specificity (interfering substances and cross-reactants), competitive interference, carryover, and clinical performance (Positive Percent Agreement - PPA, and Negative Percent Agreement - NPA).

The document implicitly defines the acceptance criteria as the successful demonstration of performance metrics that are typically expected for such in vitro diagnostic devices, often compared to highly sensitive FDA-cleared or EUA assays. While specific numerical acceptance criteria (e.g., PPA > X%, NPA > Y%) are not explicitly stated as "acceptance criteria," they are demonstrated by the reported results. The conclusion statement (Section 5.0) explicitly states that the "analytical and clinical study results demonstrate that the Alinity m Resp-4-Plex assay... performs comparably to the predicate device... the results support a substantial equivalence decision." This implies that the demonstrated performance values met the FDA's criteria for substantial equivalence to the predicate device.

Here's a table summarizing the reported device performance, which implicitly met the acceptance criteria:

Table 1: Acceptance Criteria (as Demonstrated Performance) and Reported Device Performance

Study Details

Here's the breakdown of the study details as requested:

1. A table of acceptance criteria and the reported device performance: Already provided above.

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

   • Analytical Studies (Test Set):

     • LoD: For each virus (Flu A, Flu B, RSV, SARS-CoV-2), preliminary LoD involved testing a minimum of 3 levels, each in a minimum of 3 replicates. Final LoD confirmation involved testing a minimum of 3 panel members with target concentrations bracketing the preliminary LoD, each panel member in a minimum of 20 replicates. (Total specific sample numbers not provided per virus, but this describes the method and minimums). Specimens were pooled negative NP clinical specimens.
     • Inclusivity: Each individual virus isolate or strain was tested in replicates of 5. (Total specific sample numbers not provided per virus, but 16 Flu A strains, 9 Flu B, 6 RSV, 9 SARS-CoV-2). Specimens were pooled negative clinical NP swab matrix.
     • Precision: 5 panel members (1 negative, 4 positive) tested with 4 replicates twice each day for 5 days, on 3 Alinity m Systems operated by 3 operators using 3 reagent lots. This leads to:
       • Flu A: 120 positive replicates for each level, 360 negative replicates.
       • Flu B: 120 positive replicates for each level, 357 negative replicates.
       • RSV: 120 positive replicates for moderate, 117 for low, 360 negative.
       • SARS-CoV-2: 120 positive replicates for each level, 357 negative.
     • Reproducibility: 5 panel members tested at 3 external clinical testing sites. Each site tested 2 Alinity m Resp-4-Plex AMP Kit lots, on 5 non-consecutive days for each lot. Four replicates of each panel member were tested on each of 5 days. This leads to:
       • Flu A: 120 positive replicates for each level, 360 negative replicates.
       • Flu B: 120 positive replicates for each level, 359 negative replicates.
       • RSV: 120 positive replicates for moderate, 119 for low, 360 negative.
       • SARS-CoV-2: 120 positive replicates for moderate, 117 for low, 359 negative.
     • Analytical Specificity (Interfering Substances): 34 substances evaluated in 2 different positive panel members (PM1 & PM2), each containing multiple analytes at 3xLoD. (Replicate number not specified).
     • Analytical Specificity (Cross-Reactants): 74 microorganisms added to pooled negative clinical NP swab matrix (replicate number not specified) and also to positive samples (replicate number not specified).
     • Competitive Interference: 4 panel members, each containing 3 viruses at low concentrations and one at high concentration. (Replicate number not specified).
     • Carryover: Negative and high positive samples tested in alternating positions, across 3 Alinity m Systems. 360 negative samples total.

   • Clinical Performance (Test Set):

     • Prospective Clinical Study:
       • Flu A/B, RSV: 2,753 valid results initially, 2,504 (Flu A), 2,710 (Flu B), 2,700 (RSV) used in analysis.
         • Data Provenance: Multicenter study using prospectively collected nasopharyngeal swab specimens. 4 US clinical sites for testing. Specimens collected during 2021-2022 flu season at 7 geographically distributed locations in the US and during the 2020 flu season at 1 location in the Southern Hemisphere.
       • SARS-CoV-2: 826 valid results initially, 698 used in analysis.
         • Data Provenance: Specimens collected at 10 geographically distributed locations in the US over 2 time periods (Dec 2020 - Feb 2021 and May 2023).
     • Retrospective Clinical Study:
       • Flu A/B, RSV: 515 valid results initially, 506 (Flu A), 504 (Flu B), 505 (RSV) used in analysis.
         • Data Provenance: Preselected archived flu B positive NP swab specimens in UVT or UTM collected during the 2017-2018 and 2019-2020 flu seasons. Randomly mixed with known negative specimens.

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

   • The ground truth for the clinical test sets (both prospective and retrospective) was established using a Composite Comparator (CC). This CC was based on results from "2 to 3 FDA cleared assays for flu A, flu B, and RSV" and "2 to 3 highly sensitive EUA SARS-CoV-2 molecular assays."
   • The document does not specify the number or qualifications of human experts (e.g., radiologists, pathologists) involved in establishing this ground truth. The ground truth method described is entirely based on laboratory comparator assays, not expert human interpretation of results.

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

   • The ground truth was established by a Composite Comparator (CC) method:
     • A specimen was categorized as CC positive if a minimum of 2 comparator positive results were reported.
     • A specimen was categorized as CC negative if a minimum of 2 comparator negative results were reported.
     • A specimen was categorized CC indeterminate if a CC could not be determined due to missing results from the comparator assays.
   • This functions as a type of "majority rule" adjudication or consensus, but strictly between other molecular assays, not human experts.

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, an MRMC comparative effectiveness study was not done.
   • This device is an in vitro diagnostic (RT-PCR assay) that provides a qualitative (positive/negative) result directly. It does not involve human "readers" interpreting images or other complex data that would typically benefit from AI assistance or an MRMC study design. Therefore, there's no data on human reader improvement with or without AI assistance.

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

   • Yes, the primary performance evaluation is a standalone algorithm-only performance.
   • The Alinity m Resp-4-Plex assay is an automated RT-PCR system. Its performance (PPA, NPA) in both analytical and clinical studies is the performance of the "algorithm only" in generating positive/negative results from the sample. Human intervention is limited to sample collection and system operation, not interpretation of the primary diagnostic output.

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

   • The ground truth for the clinical studies was established using a Composite Comparator (CC) method based on the results of other FDA-cleared molecular diagnostic assays (or EUA high-sensitivity molecular assays for SARS-CoV-2).
   • It was not based on expert consensus, pathology, or outcomes data directly.

8. The sample size for the training set:

   • The document describes a de novo device, or a device for which substantial equivalence is being sought, not an AI/ML device that requires a distinct "training set" and "test set" in the context of model development.
   • The studies presented are primarily verification and validation studies to demonstrate the device's analytical and clinical performance after its development.
   • Therefore, the concept of a separate "training set" as understood in machine learning (where data is used to train a model) is not applicable to this RT-PCR assay. The assay's "knowledge" is embedded in its reagents, primers, probes, and system parameters, which were likely optimized during development using various analytical samples, but these are not typically referred to as a "training set" in the context of a 510(k) for an RT-PCR assay.

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

   • As explained in point 8, there isn't a "training set" in the AI/ML sense for this RT-PCR assay. The ground truth for any samples used during the development or optimization phases would similarly be established using well-characterized samples (e.g., cultured viruses, positive clinical samples confirmed by reference methods, synthetic nucleic acids, negative clinical samples).

Ask a specific question about this device