The Aptima Trichomonas vaginalis (TV) assay is an in vitro qualitative nucleic acid amplification test (NAAT) for the detection of ribosomal RNA (rRNA) from Trichomonas vaginalis to aid in the diagnosis of trichomoniasis using the Panther system.

The assay may be used to test the following specimens from symptomatic or asymptomatic individuals: clinician-collected endocervical swabs, clinician-collected and patient-collected vaginal swabs (in a clinical setting), female and male urine, and specimens collected in PreservCyt Solution.

The Aptima TV assay involves the technologies of target capture, transcription-mediated amplification (TMA), and hybridization protection assay (HPA). Specimens are collected and transferred into their respective specimen transport tubes. The transport solution in these tubes releases the rRNA target and protects it from degradation during storage. When the Aptima TV assay is performed in the laboratory, the target rRNA is isolated from the specimens using a specific capture oligomer and magnetic microparticles in a method called target capture. The capture oligomer contains a sequence complementary to a specific region of the target molecule as well as a string of deoxyadenosine residues. During the hybridization step, the sequence-specific region of the capture oligomer binds to a specific region of the target molecule. The capture oligomer:target complex is then captured out of solution by decreasing the temperature of the reaction to room temperature. This temperature reduction allows hybridization to occur between the deoxyadenosine region on the capture oligomer and the poly-deoxythymidine molecules that are covalently attached to the magnetic particles. The microparticles, including the captured target molecule bound to them, are pulled to the side of the reaction vessel using magnets and the supernatant is aspirated. The particles are washed to remove residual specimen matrix that may contain amplification inhibitors. After the target capture steps are completed, the specimens are ready for amplification.

The provided text describes the analytical and clinical studies performed for the Aptima Trichomonas vaginalis Assay. Here's a breakdown of the requested information:

1. Table of Acceptance Criteria and Reported Device Performance

The document does not explicitly state pre-defined acceptance criteria values for sensitivity, specificity, PPA, or NPA. Instead, it presents the achieved performance. However, implicit acceptance criteria for NAAT assays generally involve high sensitivity and specificity. The reproducibility study tables show numerical targets for agreement (e.g., >95% positivity for LoD, 90.7% to 100% agreement for reproducibility).

2. Sample Size Used for the Test Set and the Data Provenance

• Clinical Study 2 (Primary Test Set):
  • Total evaluable specimens: 5502 specimens from 3820 evaluable subjects.
  • Breakdown by specimen type:
    • 1785 patient-collected vaginal swab specimens
    • 1782 female urine specimens
    • 1935 male urine specimens
  • Data Provenance: Prospective, multicenter clinical study conducted at 11 geographically and ethnically diverse US clinical sites (obstetrics and gynecology, family planning, and STI clinics).
  • Retrospective/Prospective: Primarily prospective. Samples were collected from consenting subjects in a prospective study.

3. Number of Experts Used to Establish the Ground Truth for the Test Set and the Qualifications of Those Experts

The document does not mention the use of human experts (e.g., radiologists, pathologists) to establish the ground truth. For this in vitro diagnostic (IVD) device, the ground truth was established by molecular testing.

4. Adjudication Method for the Test Set

The ground truth for the clinical test set was established using a "composite comparator method" or "patient infected status (PIS)" / "composite comparator algorithm (CCA)" based on results from up to three cleared NAATs.

• Method:
  • Specimens were categorized as infected (PIS) or positive (CCA) if a positive result occurred in at least two of the comparator NAATs.
  • Specimens were categorized as not infected or negative if at least 2 of the comparators results were negative.
  • A third (tie-breaker) comparator was only required if the first 2 comparator results were discordant.
  • Specimens that could not be categorized due to missing results from comparator assays were excluded.

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 applicable and therefore not performed. This is an in vitro diagnostic (IVD) device for the detection of ribosomal RNA from Trichomonas vaginalis. It is a lab-based assay, not an imaging device that requires human readers to interpret results, or AI assistance for human reader improvement.

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

Yes, the performance reported (sensitivity, specificity, PPA, NPA) for the Aptima TV assay is its standalone performance. The assay itself is the "algorithm" in this context; it processes specimens and provides a qualitative result (positive/negative) without direct human interpretation of the assay's raw output for diagnosis. The study evaluates the device's ability to accurately detect the target in various specimens against the established ground truth.

7. The Type of Ground Truth Used

The ground truth used was established using a composite comparator method (sometimes referred to as a "gold standard" or "reference standard" in IVD studies), which relied on the results of multiple (up to three) cleared Nucleic Acid Amplification Tests (NAATs). This is a form of expert consensus among molecular assays.

8. The Sample Size for the Training Set

The document does not explicitly state the sample size for a separate "training set." For IVD assays, particularly those based on well-established molecular biology principles (like NAATs), development and optimization (analogous to "training") often use specific analytical panels or earlier development runs rather than a distinct, large "training set" of clinical samples as seen in machine learning/AI models. The studies described are primarily for validation.

9. How the Ground Truth for the Training Set Was Established

Since a distinct "training set" with established ground truth is not explicitly mentioned as a separate phase of the pivotal study, the establishment of ground truth for any developmental or optimization work would likely follow similar principles as the validation ground truth: using reference materials, spiked samples, or well-characterized clinical samples confirmed by established laboratory methods or multiple comparator assays. The document focuses on the validation and reproducibility of the assay.

Ask a specific question about this device

The Xpert® Xpress MVP test, performed on the GeneXpert® Xpress System, is an automated qualitative in vitro diagnostic test for the detection of DNA targets from anaerobic bacteria associated with bacterial vaginosis (BV), Candida species associated with vulvovaginal candidiasis, and Trichomonas vaginalis. The Xpert Xpress MVP test uses clinician-collected and self-collected vaginal swabs (collected in a clinical setting) from patients who are symptomatic for vaginitis/vaginosis. The Xpert Xpress MVP test utilizes real-time polymerase chain reaction (PCR) for the amplification of specific DNA targets and utilizes fluorogenic target-specific hybridization probes to detect and differentiate DNA from:

• Organisms associated with bacterial vaginosis (detected organisms not reported individually) .
  • Atopobium spp. (Atopobium vaginae, Atopobium novel species CCUG 55226) O
  • Bacterial Vaginosis-Associated Bacterium 2 (BVAB2) O
  • Megasphaera-1 O
• Candida spp. (C. albicans, C. tropicalis, C. parapsilosis, C. dubliniensis, species not differentiated) .
• Candida glabrata/Candida krusei (species not differentiated) ●
• . Trichomonas vaginalis

The Xpert Xpress MVP test is intended to aid in the diagnosis of vaginal infections in women with a clinical presentation consistent with bacterial vaginosis, vulvovaginal candidiasis, or trichomoniasis.

The Xpert® Xpress MVP test is an automated in vitro diagnostic test for qualitative detection of DNA targets from anaerobic bacteria associated with bacterial vaginosis (BV), Candida species associated with vulvovaginal candidiasis, and Trichomonas vaginalis, the agent of trichomoniasis. In the CLIA-waived environment, the Xpert Xpress MVP test is performed on the GeneXpert® Xpress System.

The latest Hub configuration of the GeneXpert Xpress System consists of a GeneXpert IV instrument that executes sample preparation, nucleic acid amplification and real-time fluorescent signal detection for the tests, and a GeneXpert Hub with preloaded GeneXpert Xpress software for running the tests and viewing the test results. The GeneXpert Hub accessory integrates the computer, touchscreen monitor and barcode scanner. Each of the GeneXpert modules in the GeneXpert IV instrument can perform independent sample preparation and testing.

The Xpert Xpress MVP test is a PCR-based Nucleic Acid Amplification Test. Each test requires the use of a single-use disposable GeneXpert cartridge that contains all necessary reagents for the detection of DNA from BV organisms, Candida species, and Trichomonas vaginalis. A Sample Processing Control (SPC) and a Probe Check Control (PCC) are also included in the cartridge serving as internal controls. The SPC is present to control for adequate sample processing, to monitor PCR conditions, the presence of potential inhibitor(s) and possible reagent degradation. The PCC verifies reagent rehydration. PCR tube filling, and confirms that all reaction components are present in the cartridge including monitoring for probe integrity and dye stability. Because the cartridges are self-contained, the risk of cross- contamination between samples is minimized.

The Xpert Xpress MVP test is designed for use with the following specimens collected from symptomatic individuals: self-collected vaginal swabs (collected in a clinical setting) and clinician-collected vaginal swabs. The ancillary specimen collection kit for use with the Xpert Xpress MVP test is the Xpert Swab Specimen Collection Kit. The swab and the transport reagent included in the Xpert Swab Specimen Collection Kit are designed to collect and preserve patient specimens to allow transport to the testing site prior to analysis with the Xpert Xpress MVP test.

Here's a summary of the acceptance criteria and study details for the Cepheid Xpert Xpress MVP device based on the provided text:

1. Table of Acceptance Criteria and Reported Device Performance:

The document doesn't explicitly state "acceptance criteria" in a numerical, threshold-based format. Instead, it presents the clinical performance results (PPA/NPA, Sensitivity/Specificity) for the Xpert Xpress MVP test. The implication is that these reported performance metrics meet internal or regulatory acceptance thresholds for substantial equivalence.

*Target includes C. albicans, C. tropicalis, C. parapsilosis, and C. dubliniensis

2. Sample Size for the Test Set and Data Provenance:

• Sample Size:
  • Clinical Study: 1,275 female patients (18 to ≥50 years of age, plus two patients 14-17 years old). A total of 2,544 vaginal swabs were tested (likely one clinician-collected and one self-collected per patient).
• Data Provenance: Retrospective and prospective. The clinical study was conducted at 9 geographically diverse sites in the U.S.

3. Number of Experts Used to Establish Ground Truth and Qualifications:

The document does not specify the number or qualifications of experts used to establish the ground truth for the clinical study. It refers to "reference/comparator methods" for ground truth.

4. Adjudication Method for the Test Set:

• For BV, Candida group, Candida glab-krus, and TV, the performance was determined relative to specific reference/comparator methods (see point 7).
• For discrepant results, "investigation of discrepant results was performed by testing specimens with another FDA-cleared NAAT." This indicates a form of discrepancy resolution rather than a multi-expert adjudication on all cases. The exact adjudication method (e.g., 2+1, 3+1) for discrepant cases is not detailed, but it involves re-testing with an FDA-cleared NAAT.

5. Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study:

No MRMC comparative effectiveness study was mentioned. The device is an in vitro diagnostic test, which typically does not involve human readers interpreting images or data to the same extent as AI-assisted diagnostic tools. Performance is typically compared against reference methods.

6. Standalone (Algorithm Only) Performance:

Yes, the entire clinical study and analytical studies described are standalone performance evaluations of the Xpert Xpress MVP device (an automated qualitative in vitro diagnostic test) without human-in-the-loop assistance in its diagnostic output. Its output is a qualitative detection result (Positive/Negative/Not Detected).

7. Type of Ground Truth Used:

• Bacterial Vaginosis (BV): An FDA-cleared nucleic acid amplification test (NAAT).
• Candida group (C. albicans, C. tropicalis, C. parapsilosis, C. dubliniensis): Yeast culture followed by mass spectrometry for species identification.
• Candida glabrata/Candida krusei: Yeast culture followed by mass spectrometry for species identification.
• Trichomonas vaginalis (TV): A patient infected status (PIS) algorithm that included results from an FDA-cleared NAAT and TV culture.
• Discrepant Results: Re-tested with another FDA-cleared NAAT.

8. Sample Size for the Training Set:

The document does not explicitly mention a "training set" in the context of machine learning model development. This device is a PCR-based NAAT, not an AI/ML-driven diagnostic. Therefore, the concept of a training set for an algorithm is not directly applicable in the same way as for an image-based AI device. Analytical studies (e.g., Limit of Detection, Analytical Reactivity, Analytical Specificity) and reproducibility studies served to characterize the device's performance chemically and biologically.

9. How the Ground Truth for the Training Set Was Established:

As noted above, the device is a PCR-based NAAT, not an AI/ML system, so a "training set" for an algorithm in the traditional sense is not discussed. The development and optimization of the assay's chemical and molecular components would have been guided by fundamental scientific principles and laboratory testing, rather than an algorithmic training process using labeled data. Benchmarking for analytical characteristics (like LoD) would involve preparing samples with known concentrations of organisms.

Ask a specific question about this device

The Aptima Combo 2® assay is a target amplification nucleic acid probe test that utilizes target capture for the in vitro qualitative detection and differentiation of ribosomal RNA (rRNA) from Chlamydia trachomatis (CT) and/or Neisseria gonorrhoeae (GC) to aid in the diagnosis of chlamydial and/or gonococcal disease using the Panther® system as specified.
On the Panther system, the assay may be used to test the following specimens from symptomatic and asymptomatic individuals: clinician-collected endocervical, PreservCyt® Solution liquid Pap specimens, vaginal, throat, rectal, and male urethral swab specimens; patient collected vaginal swab specimens , and female and male urine specimens.
¹Patient-collected vaginal swab specimens are an option for screening women when a pelvic exam is not otherwise indicated. The Aptima Multitest Swab Specimen Collection Kit has not been evaluated for home use.

The Aptima Combo 2® assay is a target amplification nucleic acid probe test that utilizes target capture for the in vitro qualitative detection and differentiation of ribosomal RNA (rRNA) from Chlamydia trachomatis (CT) and/or Neisseria gonorrhoeae (GC) to aid in the diagnosis of chlamydial and/or gonococcal urogenital disease using the Tigris® DTS® Automated Analyzer. The assay may be used to test the following specimens from symptomatic individuals: clinician-collected endocervical, vaginal and male urethral swab specimens; and female and male urine specimens. The assay may be used to test the following specimens from asymptomatic individuals: clinician-collected endocervical, vaginal and male urethral swab specimens; patient-collected vaginal swab specimens'; and female and male urine specimens. The assay is also intended for use with the testing of gynecological specimens, from both symptomatic and asymptomatic patients, collected in the PreservCyt® Solution.
1 Patient-collected vaginal swab specimens are an option for screening women when a pelvic exam is not otherwise indicated. The Aptima Multitest Swab Specimen Collection Kit is not for home use.

The Aptima Trichomonas vaginalis Assay is an in vitro qualitative nucleic acid amplification test (NAAT) for the detection of ribosomal RNA (rRNA) from Trichomonas vaginalis to aid in the diagnosis of trichomoniasis using the Panther System. The assay may be used to test the following specimens from symptomatic or asymptomatic women: clinician-collected endocervical swabs, clinician-collected vaginal swabs, and specimens collected in PreservCyt Solution.

The Aptima Trichomonas vaginalis Assay is an in vitro qualitative nucleic acid amplification test (NAAT) for the detection of ribosomal RNA (rRNA) from Trichomonas vaginalis to aid in the diagnosis of trichomoniasis using the Tigris® DTS® System. The assay may be used to test the following specimens from symptomatic or asymptomatic women: clinician-collected endocervical swabs, clinician-collected vaginal swabs, female urine specimens, and specimens collected in PreservCyt Solution.

The Aptima Combo 2 Assay (AC2) combines the technologies of target capture, TMA, and DKA. Specimens are collected and transferred into their respective specimen transport tubes. The transport solutions in these tubes release the rRNA targets and protect them from degradation during storage. When the Aptima Combo 2 Assay is performed in the laboratory, the target rRNA molecules are isolated from specimens by use of capture oligomers via target capture that utilizes magnetic microparticles. The capture oligomers contain sequences complementary to specific regions of the target molecules as well as a string of deoxyadenosine residues. A separate capture oligomer is used for each target. During the hybridization step, the sequence specific regions of the capture oligomers bind to specific regions of the target molecules. The capture oligomer:target complex is then captured out of solution by decreasing the temperature of the reaction to room temperature. This temperature reduction allows hybridization to occur between the deoxyadenosine region on the capture oligomer and the polydeoxythymidine molecules that are covalently attached to the magnetic particles. The microparticles, including the captured target molecules bound to them, are pulled to the side of the reaction vessel using magnets and the supernatant is aspirated. The particles are washed to remove residual specimen matrix that may contain amplification reaction inhibitors. After the target capture steps are completed, the specimens are ready for amplification.
Target amplification assays are based on the ability of complementary oligonucleotide primers to specifically anneal and allow enzymatic amplification of the target nucleic acid strands. The Aptima Combo 2 Assay replicates a specific region of the 23S rRNA from CT and a specific region of the 16S rRNA from GC via DNA intermediates. A unique set of primers is used for each target molecule. Detection of the rRNA amplification product sequences (amplicon) is achieved using nucleic acid hybridization. Single-stranded nucleic acid chemiluminescent probes, which are complementary to a region of each target amplicon, are labeled with different acridinium ester molecules. The updated version of the Aptima Combo 2 assay incorporates a second CT probe, complementary to a unique region of the existing CT amplicon. This tandem probe provides detection coverage for the variant strains of C. trachomatis that emerged in 2019. The labeled probes combine with amplicon to form stable hybrids. The Selection Reagent differentiates hybridized from unhybridized probe, eliminating the generation of signal from unhybridized probe. During the detection step, light emitted from the labeled hybrids is measured as photon signals in a luminometer and are reported as Relative Light Units (RLU). In DKA, differences in the kinetic profiles of the CT and GC labeled probes allow for the differentiation of signal; kinetic profiles are derived from measurements of photon output during the detection read time. The chemiluminescent detection for CT signal has very rapid kinetics and has the "flasher" kinetic type. The chemiluminescent detection for GC signal is relatively slower and has the "glower" kinetic type. Assay results are determined by a cut-off based on the total RLU and the kinetic curve type.

The Aptima Trichomonas vaginalis Assay (ATV) involves the technologies of target capture, transcription-mediated amplification (TMA), and hybridization protection assay (HPA). Specimens are collected and transferred into their respective specimen transport tubes. The transport solution in these tubes releases the rRNA target and protects it from degradation during storage. When the Aptima Trichomonas vaginalis Assay is performed in the laboratory, the target rRNA is isolated from the specimens by the use of a specific capture oligomer and magnetic microparticles in a method called target capture. The capture oligomer contains a sequence complementary to a specific region of the target molecule as well as a string of deoxyadenosine residues. During the hybridization step, the sequence-specific region of the capture oligomer binds to a specific region of the target molecule. The capture oligomer:target complex is then captured out of solution by decreasing the temperature of the reaction to room temperature. This temperature reduction allows hybridization to occur between the deoxyadenosine region on the capture oligomer and the poly-deoxythymidine molecules that are covalently attached to the magnetic particles. The microparticles, including the captured target molecule bound to them, are pulled to the side of the reaction vessel using magnets and the supernatant is aspirated. The particles are washed to remove residual specimen matrix that may contain amplification inhibitors. After the target capture steps are completed, the specimens are ready for amplification.
Target amplification assays are based on the ability of complementary oligonucleotide primers to specifically anneal and allow enzymatic amplification of the target nucleic acid strands. The Hologic TMA reaction amplifies a specific region of the small ribosomal subunit from T. vaginalis via DNA and RNA intermediates and generates RNA amplicon molecules. Detection of the rRNA amplification product sequences is achieved using nucleic acid hybridization (HPA). A single stranded chemiluminescent DNA probe, which is complementary to a region of the target amplicon, is labeled with an acridinium ester molecule. The labeled DNA probe combines with amplicon to form stable RNA:DNA hybrids. The Selection Reagent differentiates hybridized from unhybridized probe, eliminating the generation of signal from unhybridized probe. During the detection step, light emitted from the labeled RNA:DNA hybrids is measured as photon signals in a luminometer and are reported as Relative Light Units (RLU).

Acceptance Criteria and Device Performance for Aptima Combo 2 Assay and Aptima Trichomonas Vaginalis Assay (RMR Probe Reagent Update)

This document describes the acceptance criteria and the studies that demonstrate the device meets those criteria, specifically concerning the manufacturing change to the Ready-Made Reagents (RMR) Probe Reagent for the Aptima Combo 2 Assay (AC2) and Aptima Trichomonas Vaginalis Assay (ATV). The core of this submission is to prove that the removal of the lyophilization step for the RMR Probe Reagent does not negatively impact assay performance compared to the previously cleared predicate devices.

1. Table of Acceptance Criteria and Reported Device Performance

The acceptance criteria are implicitly defined by demonstrating comparability to the predicate devices. The key performance metrics evaluated are Intended Use (overall agreement with expected positivity) and Limit of Detection (LoD). For clinical performance, the acceptance criteria are 100% agreement between the predicate and the modified RMR assay. For LoD, the acceptance criterion is that the RMR assay's LoD is within ½ log of the predicate assay's LoD.

2. Sample Size for the Test Set and Data Provenance

Intended Use Study:

• Aptima Combo 2 Assay (Panther): Negative and positive panels (exact number of panels or individual samples not specified, but stated to include CT positive, FI-nvCT positive, and CT/GC dual positive panels).
• Aptima Combo 2 Assay (Tigris): Negative and positive panels (exact number of panels or individual samples not specified, but stated to include CT positive, FI-nvCT positive, and CT/GC dual positive panels).
• Aptima Trichomonas Vaginalis Assay (Panther): Negative, TV positive, and TV low positive panels (exact number of panels or individual samples not specified).
• Aptima Trichomonas Vaginalis Assay (Tigris): Negative, TV positive, and TV low positive panels (exact number of panels or individual samples not specified).
• Data Provenance: Not explicitly stated, but these appear to be contrived panels (controlled positive and negative samples) rather than directly clinical specimens for this specific study.

Limit of Detection (LoD) Study:

• Aptima Combo 2 Assay (Panther & Tigris): Stocks of CT and GC organisms, and FI-nvCT in vitro transcript. These were run in negative clinical liquid pap specimens collected in PreservCyt solution (ThinPrep).
• Aptima Trichomonas Vaginalis Assay (Panther & Tigris): Stocks of TV organisms in negative clinical liquid pap specimens collected in PreservCyt solution (ThinPrep).
• Data Provenance: The base matrix used for spiking was "negative clinical liquid pap specimens," suggesting these are retrospective clinical samples from an unspecified origin, used in a prospective manner for LoD determination.

Clinical Performance Study:

• Aptima Combo 2 Assay (Panther): 219 remnant clinical swab specimens.
• Aptima Combo 2 Assay (Tigris): 200 remnant clinical swab specimens.
• Data Provenance: "remnant clinical swab specimens". This indicates these are retrospective samples collected from prior clinical testing, with the country of origin not specified.

3. Number of Experts Used to Establish the Ground Truth for the Test Set and Their Qualifications

The ground truth for these studies is established by the performance of the predicate AC2 assay or ATV assay, which was previously cleared by the FDA. The document does not describe the involvement of additional human experts for the ground truth of these specific comparability studies, as the goal is to show the new RMR Probe Reagent performs equivalently to the already established predicate. The reliability of the predicate assays themselves would have been established through prior studies reviewed by the FDA, presumably involving expert consensus or validated methods.

4. Adjudication Method for the Test Set

Adjudication methods were not applicable in these studies. The assessment compares the performance of the modified AC2/ATV RMR assays directly against the predicate AC2/ATV assays. The predicate assay's result is used as the reference/ground truth for comparison. There is no mention of a separate expert adjudication process for discordant results between the predicate and the modified device in these comparability studies. In the LoD studies, the ground truth for spiked samples is defined by the known concentration of the spiked organisms/transcripts.

5. Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study

No MRMC comparative effectiveness study was done. This study concerns an in vitro diagnostic (IVD) device (nucleic acid amplification test) for direct detection of pathogens, not an AI-assisted diagnostic tool for human readers. Therefore, there is no human-in-the-loop performance to evaluate, and thus no effect size for human reader improvement with AI assistance.

6. Standalone (Algorithm Only Without Human-in-the-Loop Performance)

The studies presented are effectively standalone performance evaluations of the modified IVD assays. The assays are fully automated on the Panther and Tigris systems, and the results are interpreted based on predefined cut-offs (Relative Light Units and kinetic curve type). There is no human intervention in the result determination process once the assay is run. The comparison is between two versions of an automated assay.

7. Type of Ground Truth Used

• Intended Use Study: The "expected positivity results" indicate that calibrated positive and negative control panels (contrived ground truth) were used. These panels simulate clinical conditions but are created in a controlled laboratory setting.
• Limit of Detection (LoD) Study: The ground truth for LoD was spiked negative clinical specimens with known concentrations of target organisms/transcripts. This is a form of contrived ground truth based on quantitative standards.
• Clinical Performance Study: The ground truth for the clinical comparability study was the result of the predicate AC2 assay. This means the predicate assay's output was considered the reference standard, rather than an independent expert consensus or pathology review of the remnant specimens themselves for this specific comparability study. The performance of the predicate itself would have been validated against clinical outcomes or a gold standard during its initial clearance.

8. Sample Size for the Training Set

No training set is explicitly mentioned or relevant for this submission. This submission describes a manufacturing change to a reagent component of already cleared IVD assays. The assays are based on established molecular biology principles and do not involve machine learning algorithms that require a training set to "learn" patterns or make predictions. The "development activities" mentioned relate to design control and verification testing, not algorithmic training.

9. How the Ground Truth for the Training Set Was Established

As there is no training set for an algorithm, this question is not applicable.

Ask a specific question about this device

The BD CTGCTV2 assay incorporates automated DNA extraction and real-time polymerase chain reaction (PCR) for the direct, qualitative detection of DNA from:

• Chlamydia trachomatis (CT)
• . Neisseria gonorrhoeae (GC)
• . Trichomonas vaginalis (TV)

The assay may be used for detection of CT, GC and/or TV DNA in patient- or clinician-collected vaginal swab specimens (in a clinical setting), and male and female urine specimens. The assay may also be used for the detection of CT and GC DNA in endocervical swab and Liquid-Based Cytology (LBC) specimens in ThinPrep PreservCyt Solution using an aliquot that is removed prior to processing for the ThinPrep Pap test.

The assay is indicated for use with asymptomatic and symptomatic individuals to aid in the diagnosis of chlamydial urogenital disease, gonococcal urogenital disease and/or trichomoniasis.

The BD CTGCTV2 assay is available for use on the BD MAX System or the BD COR System.

As with the existing BD CTGCTV2 for BD MAX System, K182692, the BD COR PX/MX (BD COR) high throughput system conducts sample extraction steps to isolate and concentrate DNA which is then amplified to detect specific sequences for diagnostic purposes.

The BD COR System is designed to allow the user to place clinical specimens directly into designated transport racks to be loaded into the System. Once the specimens are loaded, the System will perform the necessary pre-analytical steps such as vortexing, aliquoting into a molecular tube with the correct diluent, sorting/grouping of the secondary samples for testing by assay, pre-warming and cooling of the sample (where required), and transport of the sample into a molecular analyzer, where extraction, amplification and detection will take place.

Additionally, the steps of ordering tests on the instrument for specific samples will be managed directly by the user interaction with the Laboratory Information System (LIS), which communicates with the instrument.

Once the clinical specimens are received in the laboratory and loaded into the transport racks, the user will not be required to directly handle the specimen again prior to result reporting and removal from the system.

Here's a breakdown of the acceptance criteria and study details for the BD CTGCTV2 assay, based on the provided document:

Acceptance Criteria and Device Performance

The core of this submission focuses on demonstrating the substantial equivalence of the BD CTGCTV2 assay when run on the BD COR System to its previously cleared performance on the BD MAX System. Therefore, the acceptance criteria are implicitly tied to demonstrating comparable analytical and clinical performance between the two platforms.

Key Performance Metrics (Implicit Acceptance Criteria) and Reported Device Performance:

Study Information: BD CTGCTV2 Assay on BD COR System

This submission pertains to the BD CTGCTV2 assay being used on the BD COR System. The key study is a clinical agreement study and analytical performance studies designed to demonstrate that the performance of the assay on the BD COR System is equivalent to its already cleared performance on the BD MAX System (K182692).

1. A table of acceptance criteria and the reported device performance:
* See table above.

2. Sample size used for the test set and the data provenance (e.g. country of origin of the data, retrospective or prospective):
* Analytical Performance (Precision & Reproducibility Test Set):
* Precision (within-laboratory): 72 replicates for each target (CT, GC, TV) at each of 4 concentration levels (MP, LP, HN, TN) for PreservCyt samples and 4 concentration levels for Urine samples. Total N is not explicitly stated as a single number but is derived. For example, for CT in PreservCyt, it's 72 * 4 = 288 data points.
* Reproducibility (multi-site): For each target and concentration level, 36 replicates per site across 3 sites (36 * 3 = 108 total replicates per target/level).
* Analytical Sensitivity Confirmation: 4 panel members (A, B, C, D) created with 1.5x LoD and 3x LoD for various target organisms and strains in pooled female urine, pooled vaginal swab, and pooled PreservCyt LBC matrix. The study compared performance between BD COR and BD MAX. The exact number of replicates for each panel member for this confirmation study is not explicitly stated as a total N, but implied to be sufficient for statistical comparison (mean Ct.Scores and 95% CIs are reported).
* Cross-Contamination: 540 positive samples and 540 negative samples for a total of 1080 samples.
* Clinical Agreement Study (Test Set):
* Sample Size: 433 independent panel members.
* Provenance: "Remnant urine specimens from the previous clinical trial for BD CTGCTV2 on BD MAX as well as urine specimens obtained from both internal and external collections were used for the comparison study." This suggests a retrospective collection of remnant urine specimens combined with potentially prospective collections from internal and external sources to create the clinical panels. The country of origin is not explicitly stated, but given the FDA submission, it can be inferred to be primarily US-based or at least compliant with US regulations.

3. Number of experts used to establish the ground truth for the test set and the qualifications of those experts (e.g. radiologist with 10 years of experience):
* For the Precision, Reproducibility, Analytical Sensitivity, and Cross-Contamination studies: The ground truth was established by the known concentrations or presence/absence of organisms in contrived samples or spiked matrices. These are analytical studies, not clinical studies requiring expert interpretation.
* For the Clinical Agreement Study: The BD MAX System results served as the reference/ground truth. The positive or negative status of a panel member was defined by "≥2 out of 3 evaluable results obtained on the BD MAX." This is a comparator method, not direct expert consensus on patient samples. Therefore, no human experts were used to establish the ground truth for the clinical agreement study beyond the definition of the BD MAX as the reference standard.

4. Adjudication method (e.g. 2+1, 3+1, none) for the test set:
* For the Clinical Agreement Study: The "ground truth" (reference comparator result from BD MAX) was established by "≥2 out of 3 evaluable results obtained on the BD MAX". This functions as a form of "consensus" or adjudication among multiple runs (3 aliquots) on the reference system.
* For the analytical studies (precision, reproducibility), ground truth was based on known concentrations, so no adjudication by a panel of experts was necessary.

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) assay that detects nucleic acids. It's an automated molecular system, not an AI-assisted diagnostic tool that human readers would interpret. Therefore, the concept of "human readers improve with AI vs without AI assistance" is not applicable here.

6. If a standalone (i.e. algorithm only without human-in-the-loop performance) was done:
* Yes, this is a standalone performance study in the context of an IVD assay. The BD CTGCTV2 assay on the BD COR System is an automated system for qualitative detection of DNA. The studies described (analytical and clinical agreement) evaluate the performance of this automated system directly, without human interpretation of its diagnostic output. The output itself (positive/negative) is the final result.

7. The type of ground truth used (expert consensus, pathology, outcomes data, etc):
* Analytical Studies (Precision, Reproducibility, Cross-Contamination, Analytical Sensitivity): The ground truth was based on known concentrations of purified organisms or specific strains spiked into negative matrices. This is an analytical ground truth.
* Clinical Agreement Study: The ground truth was the result from the legally marketed predicate device, the BD MAX System, determined by a "consensus" of ≥2 out of 3 evaluable results from the BD MAX. This is a (predicate) comparator ground truth.

8. The sample size for the training set:
* The document describes studies for validation and equivalency demonstration of the BD CTGCTV2 assay on the BD COR System compared to the BD MAX System. It does not mention "training sets" in the context of machine learning or AI models.
* For IVD assays, "training" typically refers to the initial development and optimization of the assay performed by the manufacturer. The data used for this developmental phase is not typically detailed in 510(k) summaries, which focus on formal validation studies.
* The assay itself incorporates "automated DNA extraction and real-time PCR," which are well-established molecular biology techniques, not typically "trained" in the AI sense.

9. How the ground truth for the training set was established:
* As noted above, the document does not describe a "training set" in the context of an AI/machine learning model. Therefore, this question is not applicable. The development of the PCR assay and its parameters would have involved extensive laboratory work by the manufacturer, but the "ground truth" for those developmental phases would be based on well-characterized materials and samples, similar to the analytical studies described for validation.

Ask a specific question about this device

The Xpert Xpress MVP test, performed on the GeneXpert Instrument Systems, is an automated qualitative in vitro diagnostic test for the detection of DNA targets from anaerobic bacteria associated with bacterial vaginosis (BV), Candida species associated with vulvovaginal candidiasis, and Trichomonas vaginalis. The Xpert Xpress MVP test uses clinician-collected and self-collected vaginal swabs (collected in a clinical setting) from patients who are symptomatic for vaginitis/vaginosis. The Xpert Xpress MVP test utilizes real-time polymerase chain reaction (PCR) for the amplification of specific DNA targets and utilizes fluorogenic target-specific hybridization probes to detect and differentiate DNA from:

• . Organisms associated with bacterial vaginosis (detected organisms not reported individually)
  • o Atopobium spp. (Atopobium vaginae, Atopobium novel species CCUG 55226)
  • Bacterial Vaginosis-Associated Bacterium 2 (BVAB2) o
  • Megasphaera-1 o
• Candida spp. (C. albicans, C. tropicalis, C. parapsilosis, C. dubliniensis, species not differentiated)
• Candida glabrata/Candida krusei (species not differentiated)
• . Trichomonas vaginalis

The Xpert Xpress MVP test is intended to aid in the diagnosis of vaginal infections in women with a clinical presentation consistent with bacterial vaginosis, vulvovaginal candidiasis, or trichomoniasis.

The Xpert® Xpress MVP test is an automated in vitro diagnostic test for qualitative detection of DNA targets from anaerobic bacteria associated with bacterial vaginosis, Candida species associated with vulvovaginal candidiasis, and Trichomonas vaginalis, the agent of trichomoniasis. The Xpert Xpress MVP test is performed on GeneXpert Instrument Systems.

The GeneXpert Instrument Systems automate and integrate sample preparation, nucleic acid extraction and amplification, and detection of the target sequences in simple or complex samples using real-time PCR assays. The systems consist of an instrument, computer, and preloaded software for running tests and viewing the results. The systems require the use of single-use disposable cartridges that hold the PCR reagents and host the PCR process. Because the cartridges are self-contained, cross-contamination between samples is minimized.

The Xpert Xpress MVP test includes reagents for the detection of DNA from BV organisms, Candida species, and Trichomonas vaginalis from vaginal swab samples. A Sample Processing Control (SPC) and a Probe Check Control (PCC) are also included in the cartridge utilized by the GeneXpert System instrument. The SPC is present to control for adequate processing of the sample and to monitor for the presence of potential inhibitor(s) in the PCR reaction. The SPC also ensures that the PCR reaction conditions (temperature and time) are appropriate for the amplification reaction and that the PCR reagents are functional. The PCC verifies reagent rehydration, PCR tube filling, and confirms that all reaction components are present in the cartridge including monitoring for probe integrity and dye stability.

The Xpert Xpress MVP test is designed for use with the following specimens collected from symptomatic individuals: self-collected vaginal swabs (collected in a clinical setting) and clinician-collected vaginal swabs. The swab transport reagent included in the Xpert Swab Specimen Collection Kit is designed to collect and preserve patient specimens to allow transport to the laboratory prior to analysis with the Xpert Xpress MVP test.

Here's an analysis of the acceptance criteria and study proving the device meets those criteria, based on the provided text:

1. Table of Acceptance Criteria and Reported Device Performance

The acceptance criteria are not explicitly stated as distinct numerical targets for sensitivity, specificity, and agreement rates in the provided document. However, the performance outcomes of the Xpert Xpress MVP test are presented and can be interpreted as the device meeting the performance standards considered acceptable for its intended use, especially given the FDA's 510(k) clearance based on "substantial equivalence." The document compares the device's performance to an FDA-cleared predicate device.

For the purpose of this table, "Acceptance Criteria" will be inferred from the reported performance, as it highlights what the device achieved and what was deemed sufficient for clearance.

2. Sample Size Used for the Test Set and Data Provenance

• Sample Size (Clinical Study/Test Set):
  • Patients: 1,478 female patients (14 to ≥ 50 years of age).
  • Vaginal Swabs Tested: 2,947 (one self-collected vaginal swab (SVS) and five clinician-collected vaginal swab (CVS) specimens per patient).
• Data Provenance:
  • Country of Origin of the Data: United States (multi-site clinical study with 12 sites from geographically diverse locations in the U.S.).
  • Retrospective or Prospective: Prospective observational, method comparison clinical study.

3. Number of Experts Used to Establish the Ground Truth for the Test Set and Their Qualifications

The document does not specify the number or qualifications of experts (e.g., radiologists, pathologists) used to establish the ground truth for the clinical test set. The ground truth for the clinical study was established by comparator methods (FDA-cleared NAATs, yeast culture followed by mass spectrometry, and a PIS algorithm). These methods are analytical laboratory tests, not dependent on expert visual review.

4. Adjudication Method for the Test Set

• The document states: "When applicable, investigation of discrepant results was performed by testing specimens with another FDA-cleared NAAT."
• This indicates a form of adjudication for discrepant results, where a third, independent, FDA-cleared NAAT was used to resolve disagreements between the Xpert Xpress MVP test and the initial comparator method. The specific rule (e.g., 2 out of 3 agreement) for this discrepancy resolution is not detailed, but the use of an independent NAAT as a tie-breaker or confirming tool is implied.

5. If a Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study Was Done

• No, an MRMC comparative effectiveness study was not done.
• This study evaluates an in vitro diagnostic (IVD) device that detects nucleic acid sequences from microorganisms using real-time PCR. It is not an imaging-based AI device that would typically involve human readers interpreting images. Therefore, the concept of human readers improving with AI vs. without AI assistance does not apply to this type of device.

6. If a Standalone (i.e., algorithm only without human-in-the-loop performance) Was Done

• Yes, the primary performance evaluation of the Xpert Xpress MVP test in the clinical study was standalone.
• The device is an automated, qualitative in vitro diagnostic test that performs sample preparation, nucleic acid extraction and amplification, and detection, and provides results "within 60 minutes." The clinical performance tables (Table 5-13) represent the direct output of the device compared to reference methods, without human interpretation of the device's signal directly impacting the final result reported by the device itself.
• Human involvement is in specimen collection, loading the cartridge, and reviewing the system's final reported result for the pathogen. The device's diagnostic output for a given sample is fully automated.

7. The Type of Ground Truth Used

The ground truth varied by the target organism:

• Bacterial Vaginosis (BV): An FDA-cleared nucleic acid amplification test (NAAT).
• Candida group and Candida glabrata/krusei: Yeast culture followed by mass spectrometry for species identification. For Candida glabrata/krusei, there was also a "contrived" study, meaning the ground truth was based on known concentrations of the organisms.
• Trichomonas vaginalis (TV): A Patient Infected Status (PIS) algorithm that included results from an FDA-cleared NAAT and TV culture.
• Discrepancy Resolution: For all targets, a second FDA-cleared NAAT was used for investigation of discrepant results, effectively serving as an adjudication method to establish the clinical ground truth for those specific samples.

8. The Sample Size for the Training Set

The document describes the clinical study as a "performance evaluation" and "method comparison clinical study" used to demonstrate substantial equivalence. It does not explicitly reference or describe a separate "training set" for the device's algorithm in the context of machine learning, because this is a molecular diagnostic test based on PCR, not an adaptable AI algorithm that is trained on data in the traditional sense. The development of such a device involves assay design and optimization rather than machine learning training sets.

9. How the Ground Truth for the Training Set Was Established

Since there is no explicit mention of a "training set" in the context of an AI/ML algorithm for this PCR-based diagnostic device, the concept of establishing ground truth for a training set as typically described for AI/ML devices doesn't apply. The development and validation of the device would have involved extensive laboratory (non-clinical) studies, including analytical sensitivity (LoD), analytical reactivity (inclusivity), analytical specificity (exclusivity), microbial interference, competitive interference, interfering substances, and carry-over contamination studies (as described in Section 5.4), where the "ground truth" for these studies would be precisely controlled laboratory-prepared samples with known concentrations of target organisms and potential interferents.

Ask a specific question about this device

The Visby Medical Sexual Health Click Test is a single-use (disposable), fully-integrated, automated Polymerase Chain Reaction (PCR) in vitro diagnostic test intended for use in point of-care or clinical laboratory settings for the rapid detection and differentiation of DNA from Chlamydia trachomatis, Neisseria gonorrhoeae, and Trichomonas vaginalis in seff-collected female vaginal swab specimens using the Visby Medical Sexual Health Vaginal Specimen Collection Kit in a health care setting. The test results are to aid in the diagnosis of symptomatic infections with Chlamydia trachomatis, Neisseria gonorrhoeae, and Trichomonas vaginalis.

The test system includes the Visby Medical Sexual Health Click device, the Visby Medical power supply, the Visby Medical Vaginal Collection kit, and fixed-volume transfer pipettes. The device processes a vaginal swab sample by automatically performing all steps required to complete lysis, polymerase chain reaction, and amplicon detection.

The patient uses the Visby Medical Vaginal Collection Kit to self-collect a vaginal specimen with the provided flocked swab, and then the patient elutes the specimen into the Visby Medical Collection Media. The test operator transfers the collection media containing the patient specimen into the sample port of the device using the provided fixed-volume pipette where it rehydrates a lyophilized internal process control. The sample enters a lysis module, where the DNA of the sample and the internal process control are extracted using a combination of chemical lysis and high temperature. The extracted DNA enters a mixing chamber where it rehydrated lyophilized PCR reagents, followed by thermocycling to amplify target DNA. If present, the amplified pathogen target (CT, NG, and/or TV) and internal process control hybridize to specific probes located on a flow channel. Detection of the target-specific PCR product is accomplished via an enzyme-linked colorimetric assay using streptavidin bound horseradish peroxidase (HRP) and a colorimetric substrate that forms a purple precipitate. Test results can be expected in approximately 30 minutes: a green check mark will appear, and a purple color will appear in the "Control" spot, indicating a valid test. A purple spot adjacent to "Chlamydia," "Gonorrhoeae," and/or "Trichomonas" signifies the presence of amplified CT, NG, and/or TV DNA in the sample. Tests with invalid results due to an error (indicated by a red X status light) or failure to develop a purple spot in the "Control" spot, are retested with a new Visby device.

The Visby Medical Sexual Health Click Test is an automated Polymerase Chain Reaction (PCR) in vitro diagnostic test for the rapid detection and differentiation of DNA from Chlamydia trachomatis (CT), Neisseria gonorrhoeae (NG), and Trichomonas vaginalis (TV) in self-collected female vaginal swab specimens.

Here's an analysis of its acceptance criteria and the study that proves its performance:

1. Table of Acceptance Criteria and Reported Device Performance

The acceptance criteria are implicitly defined by the clinical performance observed in the study, aiming for high sensitivity and specificity. The reported device performance is based on the achieved Positive Percent Agreement (PPA) and Negative Percent Agreement (NPA) or Sensitivity and Specificity against a Composite Comparator Result (CCR) or Patient Infected Status (PIS).

2. Sample Size Used for the Test Set and Data Provenance

• Test Set Sample Size:
  • CT: 1774 subjects
  • NG: 1786 subjects
  • TV: 1765 subjects
  • Initially, 1899 subjects were enrolled, 1881 were eligible, and 1789 were included in the data analysis.
• Data Provenance: The study was conducted at 14 clinical sites geographically distributed across the United States. The study subjects were prospectively enrolled females.

3. Number of Experts Used to Establish Ground Truth and Their Qualifications

• The document does not specify the number of experts explicitly used to establish the ground truth.
• The ground truth was established by a Composite Comparator Result (CCR) for CT and NG, and a Patient Infected Status (PIS) algorithm for TV, which were comprised of three FDA-cleared NAAT assays.
• The qualifications of the individuals interpreting the results of these FDA-cleared NAAT assays are not explicitly stated, but it can be inferred that they would be trained laboratory personnel given the nature of NAAT testing.

4. Adjudication Method for the Test Set

• Adjudication Method: For all three organisms (CT, NG, TV), the ground truth (CCR/PIS) was determined as follows:
  • A study participant was considered infected if a positive result was reported from two of the three FDA-cleared NAAT tests.
  • If the test results of the first two NAATs were discordant (one positive, one negative), the CCR/PIS was determined by the third NAAT. This is often referred to as a "2 out of 3" or "tie-breaker" adjudication method.

5. Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study

• No MRMC comparative effectiveness study was done in the context of human readers improving with or without AI assistance. This device is a fully-automated in vitro diagnostic test, and the clinical study evaluated the device's performance directly against a laboratory-based ground truth, not human reader performance.
• However, a reproducibility study was conducted with "untrained users" (non-laboratorians) in CLIA waived settings, to assess the device's performance when used by typical healthcare professionals in such environments. This indirectly assesses the device's usability and reliability in a real-world setting where "human readers" (the operators) process samples and interpret simple visual results (green check mark for valid, purple spots for positive). The study demonstrated robust reproducibility and that untrained users could perform and interpret results accurately, but it wasn't a comparative effectiveness study of human reader improvement.

6. Standalone Performance Study (Algorithm Only Without Human-in-the-Loop Performance)

• Yes, the clinical performance described in section H is a standalone performance study (device only without human-in-the-loop performance influencing the result generation). The device processes a vaginal swab sample fully-integrated and automatically, performing lysis, PCR, and amplicon detection, and then provides a visual result (a green check mark for valid and purple spots for positive). The study evaluated how this automated device's results compared to the established ground truth.
• While human operators handled the samples and initiated the test, the test interpretation is designed to be straightforward and automatic (visual detection of colored spots), making the clinical performance metrics largely reflective of the algorithm's standalone capability to detect the presence of pathogens. The reproducibility study explicitly confirmed that "untrained users can perform the test and interpret the results accurately."

7. Type of Ground Truth Used

• The ground truth used was a Composite Comparator Result (CCR) for CT and NG, and a Patient Infected Status (PIS) algorithm for TV.
• This ground truth was derived from the results of three FDA-cleared NAAT (Nucleic Acid Amplification Test) assays. This is a common and robust method for establishing ground truth in molecular diagnostics, as NAATs are highly sensitive and specific.

8. Sample Size for the Training Set

• The document focuses on the validation of the device through non-clinical and clinical studies. It does not specify the sample size for a training set.
• As an in vitro diagnostic (IVD) device, its development likely involves extensive internal optimization and testing (which could be considered analogous to "training" in the context of AI/ML, though this device is PCR-based with colorimetric detection, not explicitly a machine learning algorithm) using characterized samples and analytical performance studies (Limit of Detection, Inclusivity, Cross-Reactivity, etc.). These analytical studies serve as a rigorous "training" and development phase, but specific "training set sizes" as might be reported for a machine learning model are not applicable or provided here.

9. How the Ground Truth for the Training Set Was Established

• Since a specific "training set" with established ground truth as in AI/ML model development is not explicitly mentioned or applicable to this type of PCR-based diagnostic, this question cannot be directly answered from the provided text.
• However, the analytical studies (LoD, Inclusivity, Cross-Reactivity, Competitive Interference, Interfering Substances, Reproducibility) described in section G involve using well-characterized organisms (known strains/serovars with specified concentrations) spiked into negative clinical vaginal sample matrix. The "ground truth" for these analytical samples is the known presence/absence and concentration of the spiked organisms. This rigorous analytical characterization serves to ensure the robust performance of the assay.

Ask a specific question about this device

The Aptima Combo 2® Assay is a target amplification nucleic acid probe test that utilizes target capture for the in vitro qualitative detection and differentiation of ribosomal RNA (rRNA) from Chlamydia trachomatis (CT) and/or Neisseria gonorrhoeae (GC) to aid in the diagnosis of chlamydial and/or gonococcal disease using the Panther® System as specified. On the Panther System, the assay may be used to test the following specimens from symptomatic and asymptomatic individuals: clinician-collected endocervical, vaginal, throat, rectal, and male urethral swab specimens, clinician-collected gynecological specimens collected in the PreservCyt® Solution, patient-collected vaginal swab specimens,1 and female and male urine specimens.

The Aptima Combo 2® Assay is a target amplification nucleic acid probe test that utilizes target capture for the in vitro qualitative detection and differentiation of ribosomal RNA (rRNA) from Chlamydia trachomatis (CT) and/or Neisseria gonorrhoeae (GC) to aid in the diagnosis of chlamydial and/or gonococcal urogenital disease using the Tigris® DTS® Automated Analyzer or semi-automated instrumentation as specified. The assay may be used to test the following specimens from symptomatic individuals: clinician-collected endocervical, vaginal and male urethral swab specimens; and female and male urine specimens. The assay may be used to test the following specimens from asymptomatic individuals: clinician-collected endocervical, vaginal and male urethral swab specimens; patient-collected vaginal swab specimens1; and female and male urine specimens. The assay is also intended for use with the testing of gynecological specimens, from both symptomatic and asymptomatic patients, collected in the PreservCyt® Solution.

The Aptima Trichomonas vaginalis Assay is an in vitro qualitative nucleic acid amplification test (NAAT) for the detection of ribosomal RNA (rRNA) from Trichomonas vaginalis to aid in the diagnosis of trichomoniasis using the Panther System. The assay may be used to test the following specimens from symptomatic or asymptomatic women: clinician-collected endocervical swabs. clinician-collected vaginal swabs, and specimens collected in PreservCyt Solution.

The Aptima Trichomonas vaginalis Assay is an in vitro qualitative nucleic acid amplification test (NAAT) for the detection of ribosomal RNA (rRNA) from Trichomonas vaginalis to aid in the diagnosis of trichomoniasis using the Tigris® DTS® System. The assay may be used to test the following specimens from symptomatic or asymptomatic women: clinician-collected endocervical swabs, clinician-collected vaginal swabs, female urine specimens, and specimens collected in PreservCyt Solution.

The clearance of this Special 510(k) application will allow the use of a Ready-Made Reagent format for the Aptima Combo 2 assay (AC2) and the Aptima Trichomonas Vaginalis assay (ATV) on the Tigris and Panther systems. The use of Ready-Made Reagent assays does not change the principles of procedure, intended use, or primary technological characteristics.

Currently, each of the AC2 and ATV Amplification, Enzyme, and Probe reagents are provided in two parts: a lyophilized reagent (cake form) and a reconstitution solution (liquid form). Per the instructions provided in the respective assay's package inserts, the customers are instructed to prepare the reagents by reconstituting each reagent by combining the bottles of lyophilized reagent with the reconstitution solution and mixing reagents manually prior to placing on the Panther or Tigris system. Hologic developed "Ready Made Reagents" (RMRs), which are liquid format or pre-reconstituted Amplification, Enzyme, and Probe reagents available for customers to procure in the 250-Test Kit size available for use on both the Tigris and Panther systems.

Changes to the user interface are minimal as the RMRs are identical to the lyophilized reagents once they have been reconstituted at the laboratory. In order to prepare the current format reagents (lyophilized format), the laboratory personnel pairs each reconstitution solution (Amplification, Enzyme, and Probe) with its respective lyophilized reagent. Using the RMR format, the customer eliminates the reconstitution step and is only required to bring the three reagents to room temperature following the same process currently done for the previously reconstituted reagents. All subsequent steps by the operator are unchanged. All assay principles and processing steps on the Panther or Tigris systems remain unchanged. There are no changes to the instrument hardware or software based on this change.

The provided text describes a 510(k) summary for new "Ready-Made Reagents" (RMRs) for the Aptima Combo 2 Assay and Aptima Trichomonas Vaginalis Assay, designed to simplify reagent preparation for laboratory personnel. The submission aims to demonstrate substantial equivalence to previously cleared predicate devices.

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

1. Table of Acceptance Criteria and Reported Device Performance

The document doesn't explicitly state "acceptance criteria" in a numerical table format with pre-defined thresholds. However, the implicit acceptance criteria for demonstrating substantial equivalence are based on comparability to the predicate devices, particularly in the analytical performance studies (Limit of Detection and Intended Use) and clinical performance studies. The goal is to show that the RMR format performs equivalently to the existing lyophilized format.

Implicit Acceptance Criteria (based on study design and conclusions):

2. Sample Size Used for the Test Set and Data Provenance

• Intended Use Study:
  • The document mentions "negative and positive panels" and "CT positive, GC positive, and CT/GC dual positive panels" but does not specify the exact number of samples/panels used for each test.
  • Data provenance is implicitly laboratory-generated (panels with known analytes), not clinical patient samples for this specific study section.
• Limit of Detection Study:
  • The document mentions using "stocks of CT and GC organisms in negative clinical liquid pap specimens" and "stocks of TV organisms in negative clinical liquid pap specimens".
  • Does not specify a sample size (number of specimens/replicates) explicitly for the LoD determination.
  • Data provenance appears to be laboratory-prepared samples using clinical specimen matrices.
• Clinical Performance Study:
  • Sample Size: Three hundred (300) remnant clinical swab specimens.
  • Data Provenance: Retrospective, clinical samples ("remnant clinical swab specimens"). The country of origin is not specified but implicitly within the US as this is an FDA submission.

3. Number of Experts Used to Establish the Ground Truth for the Test Set and Qualifications of Those Experts

• Not applicable in the traditional sense of image-based AI studies using human expert consensus.
• For these in vitro diagnostic (IVD) assays, the "ground truth" for the analytical and clinical studies is established through:
  • Known concentrations in prepared panels (Intended Use, LoD).
  • Reference assay results (the "current AC2 assay" or "current ATV assay" is used as the baseline/reference result in the clinical comparability study). This assumes the predicate device's performance is the established truth for comparison.

4. Adjudication Method for the Test Set

• Not applicable in the context of human expert adjudication for a test set.
• The "adjudication" is essentially the comparison of the RMR assay results against the predicate assay results or against known panel concentrations. Discrepancies would be investigated, but there's no mention of a multi-reader/adjudicator process.

5. 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 is more common for imaging AI devices where human readers interpret images with and without AI assistance.
• This submission is for an in vitro diagnostic (IVD) assay where the device output is typically a qualitative (positive/negative) or quantitative result, not an interpretation by a human reader that is then augmented by AI. The comparison is directly between the new reagent format and the existing reagent format.

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

• Yes, in spirit, the studies are analogous to standalone performance. The performance evaluation of the RMR assays (the "device") is measured directly against established analytical and clinical benchmarks (predicate assay performance, known concentrations). There isn't a human-in-the-loop component for the performance evaluation of the assay itself, beyond the manual steps involved in sample preparation or loading described in the "Differences" section. The assay's output (presence/absence of target RNA) is the direct result.

7. The Type of Ground Truth Used

• Analytical Truth: Known concentrations of target organisms in prepared panels (for Intended Use and Limit of Detection studies).
• Comparative Truth: The previously cleared predicate devices (Aptima Combo 2 Assay and Aptima Trichomonas Vaginalis Assay using the lyophilized reagent format) served as the "reference result" or "baseline" for comparison in both analytical and clinical studies. This is a common approach in 510(k) submissions for modifications to existing devices.

8. The Sample Size for the Training Set

• Not applicable. This submission is for a modification to an existing IVD assay (a change in reagent format), not for a machine learning or AI algorithm that requires a "training set." The assays are nucleic acid amplification tests (NAATs) based on established biochemical principles, not on learned patterns from a dataset.

9. How the Ground Truth for the Training Set was Established

• Not applicable as there is no training set for this type of device.

Ask a specific question about this device

cobas TV/MG on the cobas 6800/8800 Systems is an automated, qualitative in vitro nucleic acid diagnostic test that utilizes real-time polymerase chain reaction (PCR), for the direct detection of Trichomonas vaginalis (TV) and Mycoplasma genitalium (MG) DNA in male or female urine, self-collected vaginal swab specimens (collected in a clinical setting), clinician-collected vaginal swab specimens, and endocervical specimens, all collected in cobas PCR Media (Roche Molecular Systems, Inc.). cobas TV/MG also detects TV DNA in cervical specimens collected in PreservCyt solution and MG DNA in self-collected meatal swab specimens (collected in a clinical setting) and clinician-collected meatal swab specimens. This test is intended as an aid in the diagnosis of TV and MG infections in individuals suspected to have TV or MG infection.

A vaginal swab (self-collected or clinician-collected) is the preferred specimen type for MG testing in females due to higher sensitivity compared to endocervical swabs and urine. For males, urine is the preferred specimen type due to higher sensitivity compared to meatal swabs. If vaginal swab or male urine is not used and MG testing, is negative, further testing with the preferred specimen type may be indicated if M. genitalium infection is strongly suspected.

cobas® TV/MG on the cobas® 6800/8800 Systems is an automated, qualitative in vitro nucleic acid diagnostic test for the direct detection of Trichomonas vaginalis (TV) and Mycoplasma genitalium (MG) DNA in male or female urine, self-collected vaginal swab specimens (collected in a clinical setting), clinician-collected vaginal swab specimens, and endocervical specimens collected in cobas® PCR Media (Roche Molecular Systems, Inc.). cobas® TV/MG also detects TV DNA in cervical specimens collected in PreservCyt® Solution and MG DNA in self-collected meatal swab specimens (collected in a clinical setting) and clinician-collected meatal swab specimens. The DNA Internal Control, used to monitor the entire sample preparation and PCR amplification process, is introduced into each specimen during sample processing. In addition, the test utilizes external controls (low titer positive control and a negative control).

cobas® TV/MG is based on fully automated sample preparation (nucleic acid extraction and purification) followed by PCR amplification and detection. The cobas® 6800/8800 Systems consist of the sample supply module, the transfer module, the processing module, and the analytic module. Automated data management is performed by the cobas® 6800/8800 software which assigns test results for all tests as positive, negative or invalid. Results can be reviewed directly on the system screen, exported, or printed as a report.

Nucleic acid from patient samples and added internal control DNA (DNA-IC) molecules are simultaneously extracted. In summary, nucleic acid is released by addition of proteinase and lysis reagent to the sample. The released nucleic acid binds to the silica surface of the added magnetic glass particles. Unbound substances and impurities, such as denatured protein, cellular debris and potential PCR inhibitors are removed with subsequent wash steps and purified nucleic acid is eluted from the magnetic glass particles with elution buffer at elevated temperature. External controls (positive and negative) are processed in the same way with each cobas® TV/MG run.

Selective amplification of target nucleic acid from the sample is achieved by the use of targetspecific forward and reverse primers for TV and MG which are selected from highly-conserved regions within the respective target organism. TV is detected by one selective set of primers and a probe, while MG is detected by using two sets targeting separate regions (dual-target). Selective amplification of DNA IC is achieved by the use of sequence-specific forward and reverse primers which are selected to have no homology with either the TV or MG target regions. A thermostable DNA polymerase enzyme is used for PCR amplification. The target and DNA-IC sequences are amplified simultaneously utilizing a universal PCR amplification profile with predefined temperature steps and number of cycles. The master mix includes deoxyuridine triphosphate (dUTP), instead of deoxythymidine triphosphate (dTTP), which is incorporated into the newly synthesized DNA (amplicon). Any contaminating amplicon from previous PCR runs is eliminated by the AmpErase enzyme, which is included in the PCR master mix, during the first thermal cycling step. However, newly formed amplicons are not eliminated since the AmpErase enzyme is inactivated once exposed to temperatures above 55°C.

The cobas® TV/MG master mix contains one detection probe specific for the TV target sequence, two detection probes specific for the MG target sequences and one for the DNA-IC. The probes are labeled with target specific fluorescent reporter dyes allowing simultaneous detection of TV target, MG targets and DNA-IC in three different target channels. When not bound to the target sequence, the fluorescent signal of the intact probes is suppressed by a quencher dye. During the PCR amplification step, hybridization of the probes to the specific single-stranded DNA template results in cleavage of the 5' to 3' exonuclease activity of the DNA polymerase resulting in separation of the reporter and quencher dyes and the generation of a fluorescent signal. With each PCR cycle, increasing amounts of cleaved probes are generated and the cumulative signal of the reporter dye increases concomitantly. Real-time detection and discrimination of PCR products is accomplished by measuring the fluorescence of the released reporter dyes for the TV and MG targets and DNA-IC, respectively.

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

Device: cobas TV/MG for use on cobas 6800/8800 systems

Acceptance Criteria and Reported Device Performance (Summary derived from Clinical Performance Evaluation):

The acceptance criteria are implicitly defined by the reported sensitivity and specificity values necessary for the device to be considered "substantially equivalent" to predicate devices and acceptable for aiding diagnosis. The tables below show the detailed performance for various specimen types and patient populations.

1. Table of Acceptance Criteria and Reported Device Performance:

Note: The document implies these performance characteristics are the acceptance criteria by evaluating the device against them for substantial equivalence.

2. Sample Size and Data Provenance:

• Test Set (Clinical Study):
  • Total subjects enrolled: 2,194 (1,108 females and 1,046 males evaluable for TV and/or MG analyses).
  • Total samples tested: 6,807 (with 5,285 TV results and 5,382 MG results across all specimen types from evaluable subjects).
  • Data Provenance: Multi-site, prospective study from 10 geographically diverse sites in the US.

3. Number of Experts and Qualifications for Ground Truth:

• Number of Experts: Not explicitly stated as human experts in the context of adjudication for the clinical study. The ground truth (Patient Infected Status - PIS) was established using a combination of FDA-cleared and laboratory-developed tests, not through expert consensus on images or clinical assessments by a panel of medical professionals in an adjudication process.
• Qualifications of Experts: Not applicable in the traditional sense of human readers adjudicating data. The "experts" are the established reference methods (FDA-cleared TV NAATs, TV culture, and 3 laboratory developed MG NAATs).

4. Adjudication Method for the Test Set:

• Trichomonas vaginalis (TV) PIS:
  • Combination of FDA-cleared TV NAATs and TV culture.
  • If Culture (+) or FDA-Cleared NAAT (+), then PIS = Infected.
  • If both Culture (-) and FDA-Cleared NAAT (-), then PIS = Non-Infected.
  • If either is invalid and the other is positive, PIS = Infected. If the valid test is negative with an invalid, PIS = Indeterminate.
• Mycoplasma genitalium (MG) PIS:
  • A combination of 3 laboratory developed MG NAATs.
  • "Any 2 positive tests out of the 3 MG reference NAATs used" determined as composite reference standard for positivity. (This implicitly means 2/3 or 3/3 positive for infected, and likely 0/3 or 1/3 positive for non-infected, though not explicitly detailed for all negative scenarios).
  • For the specimen-specific agreement section (Table 41), the ground truth for MG was defined as an "anatomic site-specific composite reference standard" with "any 2 positive tests out of the 3 MG reference NAATs used."

5. Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study:

• No MRMC comparative effectiveness study was done. This study is for an in vitro diagnostic (IVD) device, not an AI imaging or clinical decision support system that would typically involve human readers. The performance evaluation is against a defined gold standard (PIS), not a comparison of human reader performance with and without AI assistance.

6. Standalone (Algorithm Only) Performance:

• Yes, a standalone performance was done. The entire study is a standalone (device only) performance study, as the cobas TV/MG system is "an automated, qualitative in vitro nucleic acid diagnostic test." Its performance is evaluated directly against the Patient Infected Status (PIS), without human interpretation in the loop.

7. Type of Ground Truth Used:

• Composite Reference Standard (CRS) / Patient Infected Status (PIS).
  • For TV: Combination of FDA-cleared TV NAATs and TV culture.
  • For MG: Combination of 3 laboratory-developed MG NAATs (specifically, any 2 of 3 positive).

8. Sample Size for the Training Set:

• Not Applicable / Not Provided. This document describes a PMA (premarket approval) or 510(k) submission for a diagnostic test. For such devices, while methods are developed and validated, a distinct "training set" for an AI algorithm is not typically detailed in these regulatory summaries in the same way it would be for an AI/ML-based device. The analytical studies (LoD, inclusivity, precision, specificity, interference) serve as a comprehensive "training" and "validation" of the assay's chemical and molecular components. The "clinical study" acts as the independent test set for regulatory submission.

9. How the Ground Truth for the Training Set was Established:

• Not Applicable / Not Provided. As noted above, typical diagnostic assays do not have a "training set" in the common AI/ML sense with explicitly defined ground truth established for it. The development and optimization of the assay's chemistry and PCR targets would involve internal studies and potentially reference materials to achieve performance targets, but these are part of the overall assay design process rather than a distinct "training set with ground truth."

Ask a specific question about this device

The BD MAX™ CTGCTV2 assay, performed on the BD MAX™ System, incorporates automated DNA extraction and real-time polymerase chain reaction (PCR) for the direct, qualitative detection of DNA from:

• Chlamydia trachomatis (CT) ●
• Neisseria gonorrhoeae (GC) ●
• Trichomonas vaginalis (TV) ●

The assay may be used for detection of CT, GC and/or TV DNA in patient- or clinician-collected vaginal swab specimens (in a clinical setting), and male and female urine specimens. The assay may also be used for the detection of CT and GC DNA in endocervical swab and Liquid-Based Cytology (LBC) specimens in PreservCyt® Solution using an aliquot that is removed prior to processing for the ThinPrep™ Pap test.

The assay is indicated for use with asymptomatic and symptomatic individuals to aid in the diagnosis of chlamydial urogenital disease, gonococcal urogenital disease and/or trichomoniasis.

The BD MAX System and the BD MAX CTGCTV2 are comprised of an instrument with associated hardware and accessories, disposable microfluidic cartridges, master mixes, unitized reagent strips, and extraction reagents. The instrument automates sample preparation including target lysis, DNA extraction and concentration, reagent rehydration, and target nucleic acid amplification and detection using real-time PCR. The assay includes a Sample Processing Control (SPC) that is present in the Extraction Tube. The SPC monitors DNA extraction steps, thermal cycling steps, reagent integrity and the presence of inhibitory substances. The BD MAX System software automatically interprets test results. A test result may be called as POS, NEG, UNR, IND or INC for each of the assay's targets, based on the amplification status of the target and of the Sample Processing Control. IND (Indeterminate) or INC (Incomplete) results are due to BD MAX System level failure.

The provided information describes the BD MAX CTGCTV2 assay, a diagnostic device for detecting Chlamydia trachomatis (CT), Neisseria gonorrhoeae (GC), and Trichomonas vaginalis (TV).

Here's an analysis of the acceptance criteria and study details:

1. Table of Acceptance Criteria and Reported Device Performance

The acceptance criteria are generally established for diagnostic assays in terms of sensitivity (Positive Percent Agreement - PPA) and specificity (Negative Percent Agreement - NPA). The presented clinical performance results in Table 21 for CT and GC, and Table 22 for TV, serve as the reported device performance against these criteria.

2. Sample Size Used for the Test Set and Data Provenance

The clinical performance study included 2,547 female subjects and 1,159 male subjects from North America.
The study was prospective, as indicated by the enrollment of subjects and collection of specimens for the trial.
The data provenance is North America, specifically from "Twelve geographically diverse clinical sites."

3. Number of Experts Used to Establish the Ground Truth for the Test Set and Their Qualifications

The document does not explicitly state the "number of experts" used to establish ground truth or their individual qualifications (e.g., radiologist with 10 years of experience). Instead, the ground truth was established using an algorithmic approach with multiple FDA-cleared NAATs (Nucleic Acid Amplification Tests). This is a common and accepted method for determining the true infection status in diagnostic studies for infectious diseases, rather than relying on individual expert interpretation.

4. Adjudication Method for the Test Set

The adjudication method for establishing the Patient Infected Status (PIS) was as follows:

• Female PIS for C. trachomatis and N. gonorrhoeae: Established by testing urine and cervical specimens with two different FDA cleared NAATs. Females were designated as infected if at least two different reference NAATs were positive for urine and cervical specimens. For swab specimens, if only urine was positive with two comparator NAATs (and swabs were negative with both), they were considered non-infected.
• Male PIS for C. trachomatis and N. gonorrhoeae: Established by testing urine specimens using up to three different FDA cleared NAATs. Males were designated as infected if 2 out of 3 reference NAAT results were positive. They were categorized as non-infected if 2 out of 3 reference NAAT results were negative.
• Female PIS for T. vaginalis: Established by testing vaginal specimens with two different FDA cleared molecular tests, across three different instrument platforms. Females were designated as infected if 2 out of 3 reference test results were positive. They were categorized as non-infected if 2 out of 3 reference test results were negative.
• Male PIS for T. vaginalis: Established by testing urine specimens using up to three different FDA cleared NAATs. Males were designated as infected if 2 out of 3 reference test results were positive. They were categorized as non-infected if 2 out of 3 reference test results were negative.
• Female Urine CCA for C. trachomatis, N. gonorrhoeae, and T. vaginalis: Urine samples were used from up to three reference NAATs. Designated as positive if 2 out of 3 reference NAAT results were positive, and negative if 2 out of 3 reference NAAT results were negative.

5. Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study

No MRMC comparative effectiveness study was done. This type of study is more common for image-based diagnostic aids where human interpretation is a primary component, and the AI assists human readers. For an automated molecular diagnostic assay like the BD MAX CTGCTV2, the device's performance is standalone.

6. Standalone (Algorithm Only) Performance

Yes, the study primarily presents standalone performance of the BD MAX CTGCTV2 assay. The assay is an "automated DNA extraction and real-time polymerase chain reaction (PCR)" system, and its interpretation is done "automatically" by the BD MAX System software. The clinical performance tables (Table 21, 22, 23) reflect the algorithm-only performance against the established PIS.

7. Type of Ground Truth Used

The ground truth used was expert consensus derived through an algorithm based on multiple FDA-cleared Nucleic Acid Amplification Tests (NAATs). This is a common and robust method for establishing ground truth in molecular diagnostics, as these reference NAATs are highly sensitive and specific.

8. Sample Size for the Training Set

The document does not explicitly mention a separate "training set" sample size for the development of the BD MAX CTGCTV2 assay. This is typical for a 510(k) submission for an in vitro diagnostic (IVD) device, where the focus is on the validation of the finalized product's performance rather than the detailed development (training) process. The analytical studies (LoD, inclusivity, specificity) demonstrate the characteristics of the assay, which would have been refined during development.

9. How the Ground Truth for the Training Set Was Established

As no specific training set and its ground truth establishment are detailed, this information is not available in the provided text. The analytical studies like Limits of Detection (LoD), inclusivity, and analytical specificity use well-characterized microbial suspensions and established methods for their testing. These analytical studies are part of the broader validation, implying that the assay's design was based on such principles during its development.

Ask a specific question about this device

The BD MAX CT/GC/TV assay, as performed using the BD MAX System incorporates automated DNA extraction and real-time polymerase chain reaction (PCR) for the direct, qualitative detection of DNA from Chlamydia (CT), Neisseria gonorrhoeae (GC) and/or Trichomonas vaginalis (TV). The assay may be used for detection of CT and/or GC DNA in male urine specimens, and the detection of CT, GC and/or TV DNA in female urine specimens, cliniciancollected female endocervical swab speciment-collected vaginal swab specimens (in a clinical setting). The assay is indicated for use to aid in the diagnosis of chlamydial urogenital disease, gonococcal urogenital disease and/or trichomoniasis in asymptomatic and symptomatic individuals.

The BD MAX System and the BD MAX CT/GC/TV are comprised of an instrument with associated hardware and accessories, disposable microfluidic cartridges, master mixes, unitized reagent strips, and extraction reagents. The instrument automates sample preparation including target lysis, DNA extraction and concentration, reagent rehydration, and target nucleic acid amplification and detection using real-time PCR. The assay includes a Sample Processing Control (SPC) that is present in the Extraction Tube. The SPC monitors DNA extraction steps, thermal cycling steps, reagent integrity and the presence of inhibitory substances. The BD MAX System software automatically interprets test results. A test result may be called as POS, NEG or UNR for each of the assay's targets, based on the amplification status of the target and of the Sample Processing Control. IND (Indeterminate) or INC (Incomplete) results are due to BD MAX System failure.

The BD MAX CT/GC/TV assay, performed using the BD MAX System, is an in vitro diagnostic device for the direct, qualitative detection of DNA from Chlamydia trachomatis (CT), Neisseria gonorrhoeae (GC), and Trichomonas vaginalis (TV).

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

1. Table of Acceptance Criteria and Reported Device Performance

The document does not explicitly present a table of acceptance criteria with corresponding performance metrics. However, the "Clinical Performance Studies" section implicitly establishes the performance expectations through its reported sensitivity and specificity values. The acceptance criteria can be inferred as the achieved sensitivity and specificity values for each pathogen across different specimen types and patient populations (asymptomatic and symptomatic).

For the purpose of this response, I will present the key performance metrics from the "Clinical Performance Studies" (Table 18) as the reported device performance, which are implicitly the performance targets the device met. The clinical performance is compared against a "Patient Infected Status (PIS)," which is a composite reference method.

Table: Acceptance Criteria (Inferred from Achieved Performance) and Reported Device Performance

Note: The "Acceptance Criteria" values are inferred as the lower bounds of the reported 95% Confidence Intervals for Sensitivity and Specificity, representing the minimum performance demonstrated to be acceptable.

2. Sample Size Used for the Test Set and Data Provenance

• Test Set Sample Size:

  • Female Subjects: 2,114 evaluable subjects for Chlamydia trachomatis (CT) and Neisseria gonorrhoeae (GC), and 1,291 of these for Trichomonas vaginalis (TV).
  • Male Subjects: 892 evaluable subjects for CT and GC.
  • Specimens for CT: 1,836 patient-collected vaginal swabs, 1,831 endocervical swabs, 1,849 female urine, and 830 male urine specimens.
  • Specimens for GC: 1,836 patient-collected vaginal swabs, 1,824 endocervical swabs, 1,849 female urine, and 840 male urine specimens.
  • Specimens for TV: 1,048 patient-collected vaginal swabs, 1,039 endocervical swabs, and 1,047 female urine specimens.
  • Total specimens initially evaluated: 6,573 specimens.

• Data Provenance: Retrospective and Prospective. The study mentions that it was a "multicenter study where clinical sites enrolled subjects and also performed testing." This suggests prospective collection of real-world samples for the clinical performance evaluation. The provenance is from "nine (9) geographically diverse clinical sites in North America."

3. Number of Experts Used to Establish the Ground Truth for the Test Set and Their Qualifications

The document refers to a "Patient Infected Status (PIS)" as the reference method for establishing ground truth. The PIS is described as a "composite reference method algorithm." This suggests that the ground truth was not established by a panel of individual experts directly reviewing each case. Instead, it was based on results from multiple reference methods integrated by an algorithm. The specific number of experts or their qualifications involved in developing or validating this algorithm (if any) is not specified in this document.

4. Adjudication Method for the Test Set

The ground truth was established by a "composite reference method algorithm" (PIS). This implies an algorithmic adjudication rather than human expert adjudication. The study states, "This multicenter study evaluated results obtained with the BD MAX CT/GC/TV compared to reference methods defining the Patient Infected Status (PIS)." No specific human adjudication method (e.g., 2+1, 3+1) is mentioned, as the PIS serves as the "truth."

5. If a Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study Was Done

No, an MRMC comparative effectiveness study was not explicitly mentioned or presented in the document. The study focuses on the standalone performance of the BD MAX CT/GC/TV assay against a defined Patient Infected Status (PIS). There is no comparison of human readers with versus without AI assistance.

6. If a Standalone (i.e., algorithm only without human-in-the-loop performance) Was Done

Yes, the study primarily presents standalone performance of the BD MAX CT/GC/TV assay. The device description states, "The BD MAX System software automatically interprets test results. A test result may be called as POS, NEG or UNR for each of the assay's targets..." and "The BD MAX System performs results interpretation automatically." The clinical performance tables (Table 18) present the sensitivity and specificity of the device's automated results against the PIS, which represents a standalone evaluation of the algorithm's performance.

7. The Type of Ground Truth Used

The type of ground truth used is a "Patient Infected Status (PIS)" which is described as a "composite reference method algorithm." This indicates that the truth was derived from the results of multiple established laboratory reference methods, combined using a predefined algorithm, rather than directly from pathology, individual expert consensus, or outcomes data solely.

8. The Sample Size for the Training Set

The document does not explicitly state the sample size used for the training set for the BD MAX CT/GC/TV assay. This type of information (training set details) is often omitted in premarket notification summaries which focus on clinical validation data for regulatory approval. The "Analytical Performance" section (Precision, Reproducibility, Analytical Sensitivity, Analytical Specificity, Interfering Substances, Carryover/Cross-Contamination, Mixed Infection/Competitive Interference) describes laboratory-based studies used to characterize the assay's analytical capabilities, which might be considered part of the development and optimization process, but not a distinct "training set" in the context of machine learning model development. For in vitro diagnostic assays like this, the "training" aspect is more about optimizing reaction parameters and thresholds rather than training a machine learning model on patient data.

9. How the Ground Truth for the Training Set Was Established

Since a distinct "training set" for a machine learning model is not explicitly mentioned, and the device is an IVD assay based on PCR and automated interpretation, the concept of "ground truth for the training set" as it applies to AI/ML devices is not directly applicable here. For the analytical studies, the "ground truth" (e.g., in LoD or analytical specificity) would have been established by precisely creating samples with known concentrations of organisms or known non-target organisms, based on established laboratory methods and controls.

Ask a specific question about this device