(213 days)
The TriVerity test is an automated ve in vitro diagnostic test that measures the relative expression levels of host response genes in RNA isolated from whole blood collected in the PAXgene Blood RNA tube using reverse transcription loop-mediated isothermal amplification (RT-LAMP) on the Myrna instrument.
The TriVerity test is indicated for use in conjunction with clinical assessments and other laboratory findings as an aid to differentiate bacterial infections, viral infections, and non-infectious illness, as well as to determine the likelihood of 7day need for mechanical ventilation, vasopressors, and/or renal replacement with suspected acute infection or suspected sepsis presenting to the emergency department.
The test generates three scores that each fall within one of five discrete interpretation bands based on the increasing likelihood of
-
bacterial infection,
-
viral infection, and
-
severe illness, as defined by the need for mechanical ventilation, vasopressors, and/ or renal replacement therapy (RRT) within seven days.
The TriVerity test is an in vitro diagnostic test for simultaneous amplification and detection of 29 informative host response RNA transcripts and 3 housekeeping transcripts (listed in measurands) using total RNA extracted from human blood. The test has been designed, manufactured, and validated for use on the Myrna instrument. The TriVerity test is performed with a TriVerity cartridge, a single-use, disposable, multi-chambered fluidic cartridge that runs on the Myrna instrument. All processing steps are automated and occur within a TriVerity cartridge, including sample extraction/purification and qRT-LAMP for the detection and relative quantification of the 29 informative host response RNA transcripts and 3 housekeeping transcripts (listed in measurands). All cartridge steps in this process, following the addition of the sample, are fully automated and completely integrated. In approximately 30 minutes, the test generates three scores that each fall within one of five discrete interpretation bands based on the increasing likelihood of
-
- bacterial infection,
-
- viral infection, and
-
- severe illness, as defined by the need for mechanical ventilation, vasopressors, and/or renal replacement therapy (RRT) within seven days.
The specimen used for the TriVerity test is a sample of whole blood collected by venipuncture using the PAXgene blood collection tubes within the PAXgene Blood RNA System (K042613). The cartridge contains all the necessary reagents to perform RNA isolation and subsequent amplification from the sample.
The TriVerity test quantifies the amount of each transcript in the sample based on the detection of fluorescence by the Myrna instrument. The cartridge includes the reagents for reverse transcription and LAMP. All 32 transcripts (and two in cartridge controls) are amplified and quantified. These values are input to two fixed classifiers which output the three separate scores, each with five discrete interpretation bands. The bands reflect monotonically increasing likelihood of bacterial infection, viral infection, and severe illness, as defined by the need for mechanical ventilation, vasopressors, and/or renal replacement therapy (RRT) within seven days.
Here's a breakdown of the acceptance criteria and the study that proves the device meets them, based on the provided FDA 510(k) summary for the TriVerity device:
TriVerity Device Performance Study Summary
The TriVerity test is an automated in vitro diagnostic test that measures the relative expression levels of host response genes in RNA isolated from whole blood. It is indicated for use in conjunction with clinical assessments and other laboratory findings as an aid to differentiate bacterial infections, viral infections, and non-infectious illness, and to determine the likelihood of 7-day need for mechanical ventilation, vasopressors, and/or renal replacement therapy (RRT) in adult patients with suspected acute infection or suspected sepsis presenting to the emergency department. The device generates three scores (bacterial, viral, and severe illness) each falling within one of five discrete interpretation bands.
1. Table of Acceptance Criteria and Reported Device Performance
The document describes performance in terms of accuracy metrics (PPA, NPA, Likelihood Ratios, AUROC) and analytical performance (reproducibility, linearity, analytical sensitivity, detection limits, analytical specificity). The acceptance criteria for clinical performance are primarily articulated through performance targets for the "Very Low" and "Very High" interpretation bands, aiming to minimize false negatives and false positives, respectively.
Acceptance Criteria for Clinical Performance (Qualitative, as stated):
- "Very Low" Band PPA ≥95%: Minimize false negative patients.
- "Very High" Band NPA ≥95%: Minimize false positive patients.
- Monotonicity: Likelihood ratios should monotonically increase across interpretation bands.
- Non-overlapping CIs: 80% CIs for likelihood ratios in non-adjacent bands should not overlap.
| Metric (Outcome) | Interpretation Band | Expected Result (Acceptance Criteria per document) | Reported Device Performance (Forced Adjudication) | Reported Device Performance (Consensus Adjudication) |
|---|---|---|---|---|
| Bacterial Infection | ||||
| PPA | Very Low | ≥95% | 95.7% | 97.2% |
| NPA | Very High | ≥95% | 94.2% | 95.5% |
| LR (Bacterial) | All Bands | Monotonically Increasing; Non-overlapping CIs | 0.16 (VL) to 5.24 (VH); CIs non-overlapping | 0.08 (VL) to 8.04 (VH); CIs non-overlapping |
| AUROC | Overall | High | 0.76 | 0.83 |
| Viral Infection | ||||
| PPA | Very Low | ≥95% | 88.7% | 95.5% |
| NPA | Very High | ≥95% | 97.7% | 98.6% |
| LR (Viral) | All Bands | Monotonically Increasing; Non-overlapping CIs | 0.25 (VL) to 20.04 (VH); CIs non-overlapping | 0.09 (VL) to 40.93 (VH); CIs non-overlapping |
| AUROC | Overall | High | 0.83 | 0.91 |
| Severe Illness | (Prognostic Endpoint) | N/A | ||
| PPA | Very Low | ≥95% (Implicitly, to minimize false negatives) | 91.8% | N/A |
| NPA | Very High | ≥95% (Implicitly, to minimize false positives) | 98.7% | N/A |
| LR (Severity) | All Bands | Monotonically Increasing; Non-overlapping CIs | 0.22 (VL) to 11.33 (VH); CIs non-overlapping | N/A |
| AUROC | Overall | High | 0.78 | N/A |
Note: "Very Low" (VL), "Low" (L), "Moderate" (M), "High" (H), "Very High" (VH). PPA (Positive Percent Agreement), NPA (Negative Percent Agreement), LR (Likelihood Ratio), AUROC (Area Under the Receiver Operating Characteristic curve).
2. Sample Sizes and Data Provenance
- Test Set (Clinical Study - SEPSIS-SHIELD; NCT04094818):
- Total Consented: 1,441 participants.
- Evaluable for Diagnostic Endpoint: 1,222 participants (for bacterial and viral infection differentiation).
- Evaluable for Prognostic Endpoint: 1,120 participants (for severe illness prediction).
- Provenance: Prospective, multi-center clinical study conducted in twenty-one geographically diverse locations throughout the United States plus one site in Europe.
- Data Type: Clinical data and biological samples collected from participants with suspected acute infection or suspected sepsis presenting to the emergency department.
3. Number of Experts and Qualifications for Ground Truth
- The document mentions "consensus and forced clinical adjudication" for establishing ground truth for bacterial and viral infections, but it does not specify the number or qualifications of the experts involved in this adjudication process. This information is typically detailed in the study protocol or full clinical study report.
4. Adjudication Method
- Forced Adjudication: All patients were adjudicated as "Yes" and/or "Probable" vs. "No" and/or "Unlikely" for bacterial and viral infection.
- Consensus Adjudication: Only participants with a "certain (Yes or No)" infection status were included. This inherently excludes "Probable" and "Unlikely" cases, leading to a smaller, more definitive ground truth dataset. The document explicitly states that "higher accuracy results were observed when applying the consensus adjudication as the reference standard... due to the exclusion of participants with uncertain (Probable and Unlikely) bacterial infection status."
5. Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study
- No MRMC comparative effectiveness study was mentioned in this summary. The device is an in vitro diagnostic test that provides scores, not an AI-assisted diagnostic tool for human readers (e.g., radiology AI). The study focuses on the standalone performance of the device against clinical adjudication and compares its performance to traditional biomarkers (PCT, CRP, WBC, Lactate, qSOFA).
6. Standalone Performance
- Yes, standalone performance was done. The entire clinical study (SEPSIS-SHIELD) evaluates the diagnostic and prognostic performance of the TriVerity test on its own, providing scores that are interpreted by clinicians as an aid for diagnosis and prognosis. The device outputs scores, and its accuracy is measured against clinical ground truth.
7. Type of Ground Truth Used
- Expert Consensus/Clinical Adjudication: For bacterial and viral infection differentiation. This involves clinical assessment and other laboratory findings.
- Outcomes Data: For severe illness prediction, defined as the "need for mechanical ventilation, vasopressors, and/or renal replacement therapy (RRT) within seven days."
8. Sample Size for the Training Set
- The document states that "Illness Severity Score thresholds were set (tuned) using predicted probabilities and ground truths for 399 patient samples collected from Emergency Department settings."
- "Bacterial and Viral Score thresholds were tuned using predicted probabilities and ground truths for 472 patient samples collected from Emergency Department and ICU studies."
- These tuning studies were described as "independent from the final clinical study (INF-04)".
9. How Ground Truth for the Training Set was Established
- The ground truth for the training (tuning) sets was established using "predicted probabilities and ground truths" from patient samples collected from Emergency Department and ICU studies. While the document doesn't explicitly detail the method of ground truth establishment for these specific tuning sets (e.g., expert adjudication, culture results), it implies a similar clinical assessment as used for the test set, given they are designated as "ground truths." It also notes these samples were collected in PAXgene blood tubes and processed on the Myrna instrument using GMP-grade TriVerity cartridges, indicating they were real patient samples with associated clinical outcomes or diagnoses.
§ 866.3215 Device to detect and measure non-microbial analyte(s) in human clinical specimens to aid in assessment of patients with suspected sepsis.
(a)
Identification. A device to detect and measure non-microbial analyte(s) in human clinical specimens to aid in assessment of patients with suspected sepsis is identified as an in vitro device intended for the detection and qualitative and/or quantitative measurement of one or more non-microbial analytes in human clinical specimens to aid in the assessment of patients with suspected sepsis when used in conjunction with clinical signs and symptoms and other clinical and laboratory findings.(b)
Classification. Class II (special controls). The special controls for this device are:(1) Premarket notification submissions must include the device's detailed Indications for Use statement describing what the device detects and measures, the results provided to the user, whether the measure is qualitative and/or quantitative, the clinical indications for which the test is to be used, and the specific population(s) for which the device use is intended.
(2) Premarket notification submissions must include detailed documentation of the device description, including (as applicable), all device components, software, ancillary reagents required but not provided, explanation of the device principle and methodology, and for molecular devices include detailed documentation of the primer/probe sequence, design, and rationale for sequence selection.
(3) Premarket notification submissions must include detailed documentation of applicable analytical studies, such as, analytical sensitivity (Limit of Detection, Limit of Blank, and Limit of Quantitation), precision, reproducibility, analytical measuring range, interference, cross-reactivity, and specimen stability.
(4) Premarket notification submissions must include detailed documentation of a prospective clinical study or, if appropriate, results from an equivalent sample set. This detailed documentation must include the following information:
(i) Results must demonstrate adequate device performance relative to a well-accepted comparator.
(ii) Clinical sample results must demonstrate consistency of device output throughout the device measuring range likely to be encountered in the Intended Use population.
(iii) Clinical study documentation must include the original study protocol (including predefined statistical analysis plan), study report documenting support for the Indications for Use(s), and results of all statistical analyses.
(5) Premarket notification submissions must include evaluation of the level of the non-microbial analyte in asymptomatic patients with demographic characteristics (
e.g., age, racial, ethnic, and gender distribution) similar to the Intended Use population.(6) As part of the risk management activities performed under 21 CFR 820.30 design controls, you must document an appropriate end user device training program that will be offered as part of your efforts to mitigate the risk of failure to correctly operate the instrument.
(7) A detailed explanation of the interpretation of results and acceptance criteria must be included in the device's 21 CFR 809.10(b)(9) compliant labeling, and a detailed explanation of the interpretation of the limitations of the samples (
e.g., collected on day of diagnosis) must be included in the device's 21 CFR 809.10(b)(10) compliant labeling.