(66 days)
The ARIES® C. difficile Assay is a real-time polymerase chain reaction (PCR) based qualitative in vitro diagnostic test for the direct detection of toxigenic Clostridium difficile (C. difficile) nucleic acid in unpreserved, unformed (liquid or soft) stool specimens obtained from patients suspected of having Clostridium difficile infection (CDI). The test targets the C. difficile toxin A gene (tcdA) and toxin B gene (tcdB) and is indicated for use as an aid in the diagnosis of C. difficile infection (CDI). The ARIES® C. difficile Assay is indicated for use with ARIES® Systems.
The ARIES® C. difficile Assay is a polymerase chain reaction (PCR)-based qualitative in vitro diagnostic test system which consists of the ARIES® System or the ARIES® M1 System with their included ARIES® Software, a stool resuspension kit, an assay-specific cassette, and an assay-specific protocol file. The ARIES® C. difficile Assay cassette is a disposable, single-use cassette containing nucleic acid purification reagents, internal sample process control (SPC), and an assay-specific master mix capable of performing the designated assay on one sample. The ARIES® C. difficile Assay cassette directly detects toxigenic Clostridium difficile (C. difficile) from unformed stool specimens obtained from patients suspected of having Clostridium difficile infection. Specifically, the ARIES® C. difficile Assay cassette detects the C. difficile toxin A gene (tcdA) and toxin B gene (tcdB) and a DNA Sample Processing Control. Unpreserved raw stool is processed using the ARIES® Stool Resuspension Kit. The ARIES® Stool Resuspension Kit includes a flocked swab, a tube containing preprocessing beads, and Stool Resuspension Buffer. The preprocessing method involves transfer of a swab of stool specimen into a tube containing preprocessing beads and Stool Resuspension Buffer. The swab is mixed in the tube by vortexing and then centrifuged. Finally, the preprocessed sample is added to the ARIES® C. difficile Assay cassette. The specimen is lysed and nucleic acid is extracted using an ARIES® instrument. An extractable sample processing control (SPC) target is also present in the ARIES® C. difficile Assay cassette and is processed with the specimen. The SPC controls for recovery of extracted nucleic acid, for inhibitory substances and for PCR reagent and instrument integrity. The Ct value of the SPC is designed to verify nucleic acid extraction, to identify PCR inhibition, if any, and verify proper function of the extraction system and real-time instrument. The Tm value of the SPC is used as a reference for determining the Tm of the tcdA and tcdB targets (if present). The extracted nucleic acid and SPC are transferred via magnetic beads through the cassette to the ARIES® C. difficile Assay lyophilized PCR reagents in the PCR tube that contain primer pairs specific to tcdA, tcdB, and the SPC sequence. Each of the primer pairs is labeled with a distinct fluorophore and detected in distinct channels of an ARIES® System. PCR amplification is performed and assay fluorescence is monitored. Incorporation of a quencher-labeled nucleotide results in a decrease in fluorescence for the associated primer pair. Following amplification, the reaction is slowly heated to separate the fluorescent-labeled strand from the quencher-labeled strand, a process that results in an increase in the fluorescence signal. The reaction fluorescence is measured during this process and the temperature at which the change in fluorescence is the maximum Tm of the amplicon. The strands of the amplicons will separate at a specific melting temperature (Tm) and an increase in fluorescence is observed. The instrument fluorescence output is analyzed and test results are determined using the ARIES® System software and the ARIES® C. difficile Assay protocol and run files. ARIES® C. difficile Assay results may be reported from the ARIES Software or from the optional SYNCT® Software.
Here's an analysis of the acceptance criteria and the study that proves the device meets them, based on the provided FDA 510(k) summary for the ARIES® C. difficile Assay:
1. Table of Acceptance Criteria and Reported Device Performance
The FDA 510(k) summary does not explicitly state pre-defined acceptance criteria values for clinical performance as a single table. Instead, it presents the assay's performance metrics against a reference method and implicitly indicates that these results were considered "acceptable" for substantial equivalence. Based on the clinical performance section, we can infer the key performance metrics reported.
| Metric | Acceptance Criteria (Inferred from reported performance deemed acceptable) | Reported Device Performance (ARIES® C. difficile Assay) |
|---|---|---|
| Positive Percent Agreement (PPA) vs. Direct Toxigenic Culture | High (e.g., typically >90%) | 98.1% (Lower Bound 95% CI: 93.3%) |
| Negative Percent Agreement (NPA) vs. Direct Toxigenic Culture | High (e.g., typically >90%) | 92.6% (Lower Bound 95% CI: 90.6%) |
| Sensitivity vs. Direct & Enriched Toxigenic Culture | High (e.g., typically >90%) | 90.5% (Lower Bound 95% CI: 84.6%) |
| Specificity vs. Direct & Enriched Toxigenic Culture | High (e.g., typically >90%) | 95.8% (Lower Bound 95% CI: 94.2%) |
| Overall Invalid Rate | Low | 0.5% (after re-test) |
2. Sample Size Used for the Test Set and Data Provenance
-
Sample Size:
- Clinical Study: 984 unique specimens were included in the data analysis. After exclusions and re-tests, 979 eligible specimens were used for performance calculations.
- Analytical Precision/Reproducibility:
- Within-Laboratory: 252 samples (36 for each of 7 panel members).
- Reproducibility (3 sites x 5 days x 6 panel members x 3 replicates): 270 samples (90 for each of 7 panel members, assuming negative also had 90 replicates).
- Limit of Detection (LoD): Not specified precisely, but implied to be multiple replicates at various dilutions.
- Analytical Reactivity: 15 distinct strains of toxigenic C. difficile (in addition to those in LoD).
- Analytical Specificity (Cross-Reactivity/Interference): 61 microorganisms/viruses + human DNA, each tested with three replicates of two C. difficile strains (BAA-1870 and BAA-1871) and three negative replicates.
- Carry-Over/Cross-Contamination: 30 high positive and 30 negative samples.
-
Data Provenance:
- Clinical Study: Prospective. The specimens were collected from 31-October-2016 to 21-February-2017 at four geographically distinct clinical sites within the United States. The specimens consisted of excess leftover de-identified, unpreserved, unformed stool specimens from patients suspected of having Clostridium difficile infection (CDI).
3. Number of Experts Used to Establish Ground Truth for the Test Set and Qualifications of Experts
The document does not explicitly state the "number of experts" or their specific "qualifications" for establishing the ground truth using the reference method.
- Ground Truth Method: The reference method for the clinical study was "direct and enriched toxigenic culture." This is a laboratory-based method.
- Location of Ground Truth Establishment: "Reference method testing was performed at a centralized testing facility." It is implied that qualified laboratory personnel perform these culture tests, but specific number or qualifications are not provided.
4. Adjudication Method for the Test Set
The document does not explicitly describe an adjudication method (like 2+1, 3+1, none) for discordant results between the ARIES® C. difficile Assay and the reference method.
However, for discordant cases, further characterization was performed:
- For specimens negative by direct toxigenic culture but positive by ARIES® C. difficile Assay (False Positives based on direct culture), enriched toxigenic culture was used, and for some, bi-directional sequencing with analytically validated primers was performed.
- For specimens positive by direct toxigenic culture but negative by ARIES® C. difficile Assay (False Negatives based on direct culture), bi-directional sequencing with analytically validated primers was used.
- Similar follow-up testing (bi-directional sequencing) was done for discords when comparing to "Direct and Enriched Toxigenic Culture."
This suggests a form of further characterization for discordant results rather than an adjudication panel of experts re-reading primary data.
5. If a Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study was done
No, an MRMC comparative effectiveness study involving human readers with and without AI assistance was not done. This device is an in vitro diagnostic (IVD) assay, not an AI imaging or diagnostic algorithm designed to assist human readers. Its performance is evaluated stand-alone against a reference laboratory method.
6. If a Standalone (i.e. algorithm only without human-in-the-loop performance) was done
Yes, the clinical performance study evaluated the ARIES® C. difficile Assay as a standalone algorithm/device. The assay generates a qualitative result (POSITIVE, NEGATIVE, or INVALID) based on its internal processing and software analysis, without human interpretation of raw data for diagnosis. The ARIES® System software automatically determines and reports results.
7. The Type of Ground Truth Used
The primary ground truth used for the clinical performance study was direct and enriched toxigenic culture.
- Direct Toxigenic Culture: This involves culturing the C. difficile from the stool sample directly to identify toxigenic strains.
- Enriched Toxigenic Culture: This involves an enrichment step before culturing to increase the yield of C. difficile, potentially improving sensitivity.
For discordant results, further analysis using bi-directional sequencing with analytically validated primers was employed.
8. The Sample Size for the Training Set
The document does not specify a distinct training set sample size. For IVD submissions like this one, it's common that assay parameters (like Ct cut-offs and Tm windows) are established during "internal optimization studies" and "internal verification studies." The document explicitly states:
- "Initial Assay Protocol File parameters were set during internal optimization studies."
- "The final Assay Protocol File parameters were then established during internal verification studies."
The sizes of these internal optimization and verification studies are not provided but would have involved various analytical samples (LoD, inclusivity, exclusivity, etc.) rather than a dedicated "training set" of clinical samples in the same way an AI model might be trained. The clinical trial data (979 samples) served as a test set for validation, not for training the assay's parameters.
9. How the Ground Truth for the Training Set Was Established
Since no distinct "training set" of clinical samples is explicitly described for algorithm development in the AI sense, the ground truth establishment for the internal optimization and verification studies likely involved:
- Carefully characterized reference strains of C. difficile (toxigenic and non-toxigenic).
- Spiked stool matrixes with known concentrations of C. difficile to determine detection limits and reactivity.
- Potentially, a smaller set of pre-characterized clinical samples to refine initial assay parameters.
Again, these studies would rely on established microbiological and molecular biology techniques (culture, PCR, sequencing) to define the presence/absence and characteristics of C. difficile in the samples used for internal development.
§ 866.3130 Clostridium difficile toxin gene amplification assay.
(a)
Identification. AClostridium difficile toxin gene amplification assay is a device that consists of reagents for the amplification and detection of target sequences inClostridium difficile toxin genes in fecal specimens from patients suspected of havingClostridium difficile infection (CDI). The detection of clostridial toxin genes, in conjunction with other laboratory tests, aids in the clinical laboratory diagnosis of CDI caused byClostridium difficile. (b)
Classification. Class II (special controls). The special controls are set forth in FDA's guideline document entitled: “Class II Special Controls Guideline: Toxin Gene Amplification Assays for the Detection ofClostridium difficile; Guideline for Industry and Food and Drug Administration Staff.” See § 866.1(e) for information on obtaining this document.