JBAIDS TULAREMIA DETECTION KIT

• K Number: K131936
• Device Name: JBAIDS TULAREMIA DETECTION KIT
• Manufacturer: BIOFIRE DIAGNOSTICS, INC.
• Date Cleared: 2013-07-31 ( 34 days)
• Product Code: OEH
• Regulation Number: 866.3280
• Type: Traditional
• Panel: MI
• Reference & Predicate Devices: K072547

Intended Use

The Joint Biological Agent Identification and Diagnostic System (JBAIDS) Tularemia Detection Kit is a real-time polymerase chain reaction (PCR) test kit intended for the qualitative in vitro diagnostic (IVD) detection of target DNA sequences of Francisella tularensis. The system can be used to test human whole blood collected in sodium citrate or sputum collected aseptically from individuals greater than 18 years of age suspected of having tularemia. In addition, positive blood cultures and colonies may be tested. This assay is intended to aid in the diagnosis of individual presenting with signs and symptoms of pneumonic or typhoidal tularemia. It is not intended to aid in the diagnosis of glandular, ulceroglandular, oculoglandular, or oropharyngeal tularemia.

The JBAIDS Tularemia Detection Kit is run on the JBAIDS instrument using the Diagnostic Wizard. Results are for the presumptive identification of F. tularensis in conjunction with culture and other laboratory tests. The definitive identification of F. tularensis from colony growth. liguid blood culture growth, blood specimens, or sputum specimens requires additional testing and confirmation procedures in consultation with public health or other authorities for whom reports are required.

The diagnosis of tularemia must be made based on history, signs, symptoms, exposure likelihood, and other laboratory evidence in addition to the identification of the target either from colonies, blood culture whole blood or sputum specimens.

The JBAIDS Tularemia Detection Kit is intended for use by trained clinical laboratory personnel who have received specific training on the use of the JBAIDS Tularemia Detection Kit. The level of F. tularensis that would be present in blood or sputum from individuals with early systemic or pneumonic infection is unknown. Due to the difficulty in obtaining clinical specimens, this assay was not evaluated with blood or sputum from individuals presenting with signs and symptoms of tularemia who have subsequently developed pneumonic or typhoidal tularemia pneumonic or typhoidal tularemia.

Device Description

The Joint Biological Agent Identification and Diagnostic System (JBAIDS) Tularemia Detection System is a fully integrated IVD system composed of the portable JBAIDS instrument, laptop computer and software and the JBAIDS Tularemia Detection Kit with one freeze-dried PCR assay for detection of Francisella tularensis DNA. The system has been validated using four different sample preparation kits for isolating DNA from whole blood (IT /-2-3TM Platinum Path and QFLOWana Sample Purification Kits), sputum (IT I-2-3TM Platinum Path and IT 1-2-3TM VIBE Sample Purification Kits), positive blood cultures (IT 1-2-3 TM SWIPE Sample Purification Kit), and plate cultures (IT /-2-3TM Platinum Path and IT 1-2-3TM SWIPE Sample Purification Kits). Use of the JBAIDS DNA Extraction Control Kit is also recommended.

Prior to testing, specimens are processed using BioFire Diagnostic's IT I-2-3 Sample Purification Kits. The resulting purified sample is added to Unknown and Inhibition Control vials, along with reconstitution buffer. Positive Control and Negative Control vials are prepared using reconstitution buffer and water. When F. tularensis DNA is present, a fragment of F. tularensis DNA is amplified. The amplicon is detected by fluorescence using a specific hydrolysis probe. Each probe is labeled on one end with a fluorescent reporter moiety (6-carboxyfluorescein (6-FAM)) and elsewhere with a quencher moiety (carboxy tetramethyIrhodamine (TAMRA)). When the probe is intact, the quencher absorbs the light emitted by the reporter moiety. During PCR, the probe hybridizes to the target sequence before the exonuclease activity of Taq polymerase hydrolyzes the probe, separating the fluorophore from the quencher and permitting detection of the fluorescent signal generated by the reporter. The fluorescent signal increases as additional templates are amplified and more probes are hydrolyzed.

JBAIDS Software analyzes the fluorescence amplification curves and reports results as positive, negative, uncertain or inhibited. A failure of the Positive or Negative Control will result in the entire run being called invalid. Retesting is required to resolve uncertain, invalid or inhibited results.

AI/ML Overview

Here's a breakdown of the acceptance criteria and study information for the JBAIDS Tularemia Detection Kit, based on the provided text:

Acceptance Criteria and Device Performance

The study aimed to demonstrate that the modified JBAIDS Tularemia Detection Kit, when used with the IT 1-2-3™ Platinum Path Sample Purification Kit, performs equivalently to the predicate device (JBAIDS Tularemia Detection Kit) using its established sample purification methods. The acceptance criteria are implicitly defined by achieving high agreement (PPA and NPA) with the results obtained from the predicate methods for surrogate clinical specimens and confirming Limit of Detection (LoD).

Performance Metric	Acceptance Criteria (Implicit)	Reported Device Performance (Platinum Path)
Whole Blood: Positive Percent Agreement (PPA) vs. QFLOWdna processed samples	High PPA (e.g., in the range of 95-100%)	100% (50/50), 95% CI = 92.9-100%
Whole Blood: Negative Percent Agreement (NPA) vs. QFLOWdna processed samples	High NPA (e.g., in the range of 95-100%)	100% (50/50), 95% CI = 92.9-100%
Sputum: Positive Percent Agreement (PPA) vs. VIBE processed samples	High PPA (e.g., in the range of 95-100%)	100% (49/49), 95% CI = 92.8-100%
Sputum: Negative Percent Agreement (NPA) vs. VIBE processed samples	High NPA (e.g., in the range of 95-100%)	96.1% (49/51), 95% CI = 86.5-99.5%
LoD for Whole Blood	100% detection at the established LoD	1500 CFU/mL: 20/20 (100.0%) detected
LoD for Sputum	≥95% detection at the established LoD	2000 CFU/mL: 19/20 (95.0%) detected
Reproducibility (Whole Blood - Medium Positive)	100% agreement with expected results in a multicenter study	100% (96/96) across all sites
Reproducibility (Whole Blood - Low Positive)	Expected strong agreement, acknowledges potential for variability near LoD (observed 86.5% overall due to likely under-spiking to 0.5x LoD)	86.5% (83/96) overall across all sites
Reproducibility (Whole Blood - Negative)	100% agreement with expected results in a multicenter study	100% (96/96) across all sites
Reproducibility (Sputum - Medium Positive)	100% agreement with expected results in a multicenter study	100% (90/90) across all sites
Reproducibility (Sputum - Low Positive)	≥95% agreement with expected results in a multicenter study	97.8% (88/90) overall across all sites
Reproducibility (Sputum - Negative)	100% agreement with expected results in a multicenter study	100% (90/90) across all sites
Direct Culture Detection (F. tularensis)	100% detection	10/10 detected
Direct Culture Detection (Non-F. tularensis)	0% detection	0/10 detected

Study Details

1. Sample Size and Data Provenance for Test Set:

   • Whole Blood Surrogate Specimens (Clinical Performance): 100 specimens (50 spiked with inactivated F. tularensis, 50 unspiked).
     • Data Provenance: Prospectively collected from febrile volunteers from November 2012 to April 2013. The country of origin is not specified, but the submission is to the U.S. FDA.
   • Sputum Surrogate Specimens (Clinical Performance): 100 specimens (50 spiked with inactivated F. tularensis, 50 unspiked).
     • Data Provenance: Frozen residual sputum specimens. The country of origin is not specified.
   • Limit of Detection (LoD) - Whole Blood: 20 independent whole blood specimens.
   • Limit of Detection (LoD) - Sputum: 20 independent sputum specimens.
   • Reproducibility (Whole Blood): A panel of 12 distinct samples, tested twice a day for 4 days at 3 sites (total 96 observations per category for positive/negative conclusions). Effectively, 4 medium positive, 4 low positive, and 4 negative samples were used.
   • Reproducibility (Sputum): A panel of 9 distinct samples, tested twice a day for 5 days at 3 sites (total 90 observations per category for positive/negative conclusions). Effectively, 3 medium positive, 3 low positive, and 3 negative samples were used.
   • Direct Culture Detection: 10 F. tularensis colonies and 10 non-F. tularensis colonies.

2. Number of Experts and Qualifications for Ground Truth:
   The document does not mention the use of human experts to establish ground truth for the test sets in the typical sense of clinical expert review. For the "clinical performance" studies, the ground truth was established by:

   • Spiking: Preparing specimens by spiking with inactivated F. tularensis at known concentrations (for positive samples) or leaving them unspiked (for negative samples).
   • Comparison to Predicate: The results from the existing, cleared sample purification methods (QFLOWdna for whole blood, VIBE for sputum) were considered the "correct result" a comparative effectiveness study, implying these methods served as the de facto reference standard for comparison with the new Platinum Path method.

3. Adjudication Method for Test Set:
   No formal adjudication method (e.g., 2+1, 3+1) is described for resolving discrepancies, as the "ground truth" was largely established by the known spiked status of the samples or by the result from the predicate device's processing method. The JBAIDS system itself reports "uncertain" results, which require retesting by the trained laboratory personnel.

4. Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study:
   No MRMC comparative effectiveness study was done to assess how human readers improve with AI vs. without AI assistance. This device is an in vitro diagnostic (IVD) PCR test kit, not an AI-assisted diagnostic imaging or interpretative technology for human readers. However, a multicenter reproducibility study was performed with JBAIDS operators, effectively involving multiple readers/sites, but not in the context of comparing human performance with and without AI.

5. Standalone Performance (Algorithm Only):
   Yes, this study primarily assesses the standalone performance of the JBAIDS Tularemia Detection Kit (the algorithm/assay) when used with the new Platinum Path sample purification kit. The "JBAIDS operators were blinded to the analyte content of the samples," indicating a focus on the assay's performance independent of operator knowledge of the expected outcome. The device's software provides the final interpretation (positive, negative, uncertain, or inhibited).

6. Type of Ground Truth Used:

   • Surrogate Clinical Specimens: Ground truth was established by spiking known concentrations of inactivated F. tularensis into prospectively collected or residual human specimens. For comparative purposes, the results from the predicate device's purification method were also used as a reference for "correctness."
   • Limit of Detection (LoD): Ground truth was based on the known, spiked concentrations of F. tularensis.
   • Reproducibility: Ground truth was based on the known, spiked concentrations (medium positive, low positive) or unspiked status (negative).
   • Direct Culture Samples: Ground truth was based on the known identity of the colonies (F. tularensis or non-F. tularensis).

7. Sample Size for Training Set:
   The document does not specify a separate "training set" sample size. As this is a modification to an existing, cleared diagnostic kit (K072547), the current study focuses on validation of the modification (addition of a new sample purification kit). The development and initial validation of the JBAIDS Tularemia Detection Kit itself would have involved training data, but that information is not part of this 510(k) summary for the modification.

8. How Ground Truth for Training Set Was Established:
   Since no training set details are provided in this document, the method for establishing ground truth for any potential training set for the original device is not described here. Typically, for IVD assays, training data would involve well-characterized samples (e.g., clinical samples confirmed by culture or other gold-standard methods, or contrived samples with known analyte concentrations).

Regulation Number and Section

§ 866.3280

Francisella tularensis serological reagents.(a)
Identification. Francisella tularensis serological reagents are devices that consist of antigens and antisera used in serological tests to identify antibodies toFrancisella tularensis in serum or to identifyFrancisella tularensis in cultured isolates derived from clinical specimens. Additionally, some of these reagents consist of antisera conjugated with a fluorescent dye (immunofluorescent reagents) used to identifyFrancisella tularensis directly from clinical specimens. The identification aids in the diagnosis of tularemia caused byFrancisella tularensis and provides epidemiological information on this disease. Tularemia is a desease principally of rodents, but may be transmitted to humans through handling of infected animals, animal products, or by the bites of fleas and ticks. The disease takes on several forms depending upon the site of infection, such as skin lesions, lymph node enlargements, or pulmonary infection.(b)
Classification. Class II (special controls). The device is exempt from the premarket notification procedures in subpart E of part 807 of this chapter subject to § 866.9.