(199 days)
The eSensor® Warfarin Sensitivity Test is an in vitro diagnostic for the detection and genotyping of the *2 and * 3 alleles of the cytochrome P450 (CYP450) 2C9 gene locus and the Vitamin K epoxide reductase C1 (VKORC1) gene promoter polymorphism (-1639G>A) from genomic DNA extracted from fresh whole blood samples preserved with EDTA, as an aid in the identification of patients at risk for increased warfarin sensitivity. The eSensor® Warfarin Sensitivity Test is for Rx only professional use within the confines of a licensed laboratory, as defined by the Clinical Laboratory Improvement Amendments (CLIA) of 1988.
The eSensor® XT-8 Instrument is an in vitro diagnostic device intended for genotyping multiple mutations or polymorphisms in an amplified DNA sample utilizing electrochemical detection technology.
The eSensor® XT-8 System is an in vitro diagnostic device for performing hybridization and genotyping of multiple mutations and/or polymorphisms in an amplified DNA sample. The XT-8 Instrument is configured with one to three processing towers which perform up to 8 simultaneous tests per tower. The XT-8 System uses a single-use, disposable test cartridge to perform hybridization and genotyping in approximately 30 minutes per sample. The cartridge contains an EEPROM chip which transmits the cartridge lot number, expiration date and protocol identity to the instrument.
The analysis process for each sample consists of three steps: 1) Genomic DNA isolated from whole blood obtained using EDTA as anti-coagulant is combined with PCR Mix and Taq polymerase enzyme and is subjected to amplification of target sequences by PCR using a thermal cycler. 2) Amplified DNA is treated with exonuclease enzyme to generate single-stranded target DNA. 3) Single-stranded, amplified target DNA is mixed with hybridization and genotyping reagents and transferred to an eSensor® Warfarin Sensitivity Test cartridge, and the cartridge is inserted in the eSensor® XT-8 Instrument. The instrument controls the circulation of the sample inside the cartridge containing to allow hybridization at a controlled temperature, and then detects and genotypes the sample by voltammetry.
Genotyping of the test panel polymorphisms is achieved by a sandwich assay principle: 1) Each pair of electrodes contains a different synthetic oligonucleotide capture probe which is complementary to one of the target DNA fragments. 2) The hybridization reagents contain pairs of ferrocene-labeled synthetic oligonucleotide signal probes; one member of each pair is complementary to the major allele sequence of the target polymorphism, while the second member of the pair is complementary to the minor allele sequence. Each member of the probe pair has a ferrocene label with a different oxidation potential for each allele. 3) Single-stranded, amplified target DNA hybridizes to its specific capture probe, and in turn hybridizes to the allele-specific, ferrocene-labeled signal probe. 4) Each electrode of the array is analyzed by voltammetry; the target polymorphism is determined by the location of the electrode containing the capture probe, and the genotype is identified by the ratio of signals from the allele-specific ferrocene labels. The array also includes positive and negative controls to confirm the hybridization reaction and detect non-specific signals.
Upon completion of the test, the EEPROM chip on the cartridge contains information that prevents its re-use with a new sample. The instrument analyzes the results and provides a report of the test results. The operator removes the used cartridge from the slot of the XT-8 Instrument, and that slot is ready to accept a new test.
Here's a breakdown of the acceptance criteria and the study details for the eSensor® Warfarin Sensitivity Test and XT-8 System, based on the provided 510(k) summary:
Acceptance Criteria and Device Performance
| Acceptance Criteria Category | Specific Metric (Implied) | Acceptance Criteria (Implied from 100% agreement) | Reported Device Performance | Comments |
|---|---|---|---|---|
| Reproducibility | Inter-laboratory, Inter-Operator, Lot-to-Lot, Day-to-Day, Run-to-Run | 100% agreement with DNA sequencing after additional runs for no-calls | 100% agreement (95% LCB > 94.2% for individual sites/alleles, > 98.5% overall) for final results | Initial no-calls (9 total) were attributed to cartridge manufacturing error (1) or operator error (8), all resolved with additional runs. |
| Genomic DNA Extraction Reproducibility | Across different extraction methods and sites | 100% agreement with DNA sequencing for first-pass results | 100% agreement (95% LCB > 86.7%) for first-pass results | No no-calls or incorrect calls during first-pass. |
| Method Comparison to Bi-directional DNA Sequencing | Agreement with DNA sequencing | 100% agreement with DNA sequencing for first-pass sample results | 100% agreement (95% LCB > 98.1% per sample, > 99.4% per SNP) for first-pass results | No no-calls or miscalls during first-pass. |
| Limit of Detection (LOD) | Lowest and highest detectable DNA concentration | 0.1 ng to 1000 ng of purified DNA per reaction | Successfully genotyped at 0.1 ng to 1000 ng per reaction | Recommended input range: 10 to 1000 ng. |
| Interfering Substances | No impact on test performance | Test performance not affected by specified substances | No impact from human serum albumin, bilirubin, human immunoglobulin G, triglycerides, hemoglobin, warfarin, heparin sodium, or elevated EDTA. | Qualitative assessment based on accurate results. |
| Interfering Mutations and Polymorphisms | Accurate results despite certain known polymorphisms | Accurate results for specified CYP450 2C9 polymorphisms | Accurate results for CYP450 2C9 (*4, *5, *6, *11, *14, *15, *16) | VKORC1 additional polymorphisms (other than -1639G>A) are not detected. |
Study Details
-
Sample sizes used for the test set and the data provenance:
-
Reproducibility Study Test Set:
- Samples: 5 genomic DNA samples covering all possible genotypes for the three alleles (CYP2C92, CYP2C93, VKORC1).
- Total Tests: 200 tests for each allele (CYP2C92, CYP2C93, VKORC1), for a grand total of 600 allele tests in the final analysis (5 samples * 4 operators/sites * 5 days * 2 runs per day = 200 tests per allele for an operator who performed 2 runs a day).
- Data Provenance: Three sites were used: one internal (likely Osmetech Molecular Diagnostics) and two external. The country of origin is not specified, but given the submission is to the FDA, it is likely US-based or recognized for regulatory purposes. The study appears prospective, as it involves controlled testing of specific samples under varied conditions.
-
Genomic DNA Extraction Reproducibility Test Set:
- Samples: 7 whole blood samples of different genotypes.
- Total Tests: 21 tests for each allele per site (7 samples * 3 replicates). With 3 sites, this totals 63 tests per allele, and 189 allele tests overall.
- Data Provenance: Three different sites, using different commercially available extraction methods. Similar to the above, country of origin is not specified but likely US-based, and the study is prospective.
-
Method Comparison Test Set:
- Samples: 157 samples.
- Total Tests: 157 samples tested on the eSensor device; 157 samples tested by DNA sequencing. On a per-SNP basis, this represents 471 data points (157 samples * 3 SNPs).
- Data Provenance: Not explicitly stated, but implies collected samples for method comparison. The nature (retrospective/prospective) isn't directly stated, but typically, method comparison studies utilize a representative set of existing or collected samples in a controlled manner, making them essentially prospective for the purpose of the comparison.
-
-
Number of experts used to establish the ground truth for the test set and the qualifications of those experts:
- The ground truth for all studies (reproducibility and method comparison) was established by DNA sequencing. The document does not specify the number of experts or their qualifications for interpreting the DNA sequencing results. DNA sequencing is generally considered a highly accurate gold standard for genotyping, and its interpretation often involves trained molecular biologists or geneticists, but no specific details are provided here.
-
Adjudication method for the test set:
- No formal adjudication method (like 2+1, 3+1 consensus) is explicitly mentioned for the test set.
- For the reproducibility study, "An additional run using the same kit lot and sample as for the first-pass test were performed for test that gave a no-call result." This implies a re-testing/re-run strategy for initial failures rather than human expert adjudication of output discrepancies. All no-calls were resolved to 100% agreement after additional runs.
- For the method comparison and extraction reproducibility, there were no initial no-calls or incorrect calls, so no adjudication or re-testing was necessary beyond the initial DNA sequencing ground truth.
-
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 performed or described. This device is an in vitro diagnostic for genotyping, meaning it produces a direct genetic result, and there is no "human reader" analogue in the typical sense of interpreting imaging or complex clinical data where AI assistance would be measured. The output is a genotype, which is then used by medical professionals.
-
If a standalone (i.e., algorithm only without human-in-the-loop performance) was done:
- Yes, the performance studies (reproducibility, extraction reproducibility, and method comparison) represent the standalone performance of the eSensor® Warfarin Sensitivity Test and XT-8 System. The device analyzes DNA samples and outputs a genotype. While human operators perform PCR and load samples, the critical genotyping step and result interpretation are performed by the instrument's software ("Assay signal results are interpreted by a software program and are assigned a genotype that is presented to the end-user in a report format"). The 100% agreement with DNA sequencing demonstrates this algorithmic performance.
-
The type of ground truth used (expert consensus, pathology, outcomes data, etc.):
- The ground truth used for all performance studies was bi-directional DNA sequencing. This is considered a highly reliable molecular diagnostic method.
-
The sample size for the training set:
- The document does not specify a separate "training set" or its size. As an in vitro diagnostic device, particularly for genetic testing, the development process generally involves analytical validation (like the studies described) rather than a machine learning training phase analogous to image analysis AI. The device's underlying "algorithm" is based on biochemical reactions and electrochemical detection, not a learned model from a dataset.
-
How the ground truth for the training set was established:
- Since no training set is explicitly mentioned in the context of machine learning, there's no ground truth establishment for such a set described. The "knowledge" or parameters for the device's operation would have been developed through biochemical and engineering principles, with validation done against known standards (DNA sequencing) as detailed in the performance studies.
§ 862.3360 Drug metabolizing enzyme genotyping system.
(a)
Identification. A drug metabolizing enzyme genotyping system is a device intended for use in testing deoxyribonucleic acid (DNA) extracted from clinical samples to identify the presence or absence of human genotypic markers encoding a drug metabolizing enzyme. This device is used as an aid in determining treatment choice and individualizing treatment dose for therapeutics that are metabolized primarily by the specific enzyme about which the system provides genotypic information.(b)
Classification. Class II (special controls). The special control is FDA's guidance document entitled “Class II Special Controls Guidance Document: Drug Metabolizing Enzyme Genotyping Test System.” See § 862.1(d) for the availability of this guidance document.