(96 days)
The QUANTA Lite™ h-tTG Screen is an enzyme-linked immunosorbent assay (ELISA) for the semi-quantitative detection of IgA and IgG antibodies to human tissue transglutaminase (htTG) in human serum. The presence of these antibodies can be used in conjunction with clinical findings and other laboratory tests to aid in the diagnosis of both IgA sufficient and IgA deficient celiac disease as well as dermatitis herpetiformis.
Each device contains the following: polystyrene microplate strips with breakaway (12 (1x8) microwells coated with human tissue transglutaminase antigen with holder; high positive, low positive and negative controls (human serum); HRP wash concentrate; HRP sample diluent; HRP IgG and IgA (goat) anti-human conjugate; TMB chromogen; and HRP stop solution (0.344M sulfuric acid).
Below is the information regarding the acceptance criteria and the study proving the device meets those criteria, extracted from the provided text.
QUANTA Lite™ h-tTG Screen
1. Table of Acceptance Criteria and Reported Device Performance
The document does not explicitly state pre-defined acceptance criteria with pass/fail thresholds in a table format. However, it reports performance characteristics that serve as the basis for a substantial equivalence decision. These are presented below, along with the device's reported performance. The implied acceptance is that these values demonstrate performance comparable to the predicate devices and are clinically acceptable for the intended use.
| Performance Metric | Reported Device Performance (QUANTA Lite™ h-tTG Screen) |
|---|---|
| Analytical Performance | |
| Intra-Assay Precision (high conc.) | %CV: 1.2-7.7% (32.6-46.8 units) |
| Intra-Assay Precision (near cut-off) | %CV: 2.1-5.8% (18.3-22.3 units) |
| Intra-Assay Precision (negative) | %CV: 5.6-7.2% (7.7-13.5 units) |
| Inter-Assay Precision (high conc.) | %CV: 2.2-9.3% (31.3-49.3 units) |
| Inter-Assay Precision (near cut-off) | %CV: 4.8-11.3% (16.0-23.0 units) |
| Inter-Assay Precision (negative) | %CV: 6.1-18.4% (5.2-14.5 units) |
| Assay Specificity (healthy individuals) | 97.9% (373/381) |
| Cut-off value | 20.0 units |
| Cross-reactivity (other autoantibodies) | All 44 samples negative (mean value 4.6 U/mL) |
| Method Comparison with Predicate | |
| Positive Percent Agreement (PPA) | 100.0% (52/52) |
| Negative Percent Agreement (NPA) | 97.1% (451/454) |
| Overall Agreement | 97.4% (493/506) |
| Clinical Performance | |
| Clinical Sensitivity | 87.8% (36/41) |
| Clinical Specificity | 97.1% (462/476) |
2. Sample size used for the test set and the data provenance
- Analytical Performance (Assay Cut-off): 381 random asymptomatic healthy individuals residing in the United States.
- Method Comparison with Predicate Device: 506 samples in total. This included 125 samples from four celiac disease reference labs and 81 normal samples. The document doesn't explicitly state the country of origin for all samples beyond "United States" for the assay cut-off derivation. The study is retrospective, utilizing existing samples.
- Clinical Sensitivity and Specificity: 517 clinically defined samples. This included 23 Celiacs untreated, 5 Celiac IgA Deficient, 18 Celiac 1st degree relatives, 13 Dermatitis Herpetiformis, 44 Disease Controls, and 414 Healthy individuals. The provenance is implied to be from a clinical setting, likely retrospective as samples are "clinically defined."
3. Number of experts used to establish the ground truth for the test set and the qualifications of those experts
The document does not specify the number of experts used or their qualifications for establishing the ground truth for the test set.
4. Adjudication method for the test set
The document does not describe an adjudication method (e.g., 2+1, 3+1) for the test set in the context of expert review. Clinical diagnoses and reference lab results likely served as the ground truth.
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
This is an ELISA-based diagnostic assay for antibodies, not an imaging or interpretive device that would typically involve human "readers" or an "AI assistant" in the sense of an MRMC study. Therefore, no MRMC comparative effectiveness study was done, and terms like "human readers improve with AI" are not applicable here. The device itself is an automated test.
6. If a standalone (i.e. algorithm only without human-in-the-loop performance) was done
Yes, the performance data presented (precision, assay specificity, method comparison, clinical sensitivity/specificity) reflects the standalone performance of the QUANTA Lite™ h-tTG Screen ELISA assay. It's a laboratory test, not an AI algorithm, and its output (semi-quantitative detection of antibodies) is the direct result of the assay, to be interpreted by a healthcare professional.
7. The type of ground truth used (expert consensus, pathology, outcomes data, etc)
- Assay Cut-off: Derived from 381 asymptomatic healthy individuals, implying a ground truth of "absence of disease" or "normal" based on clinical status. One "strong positive" sample showing a value of 54.9 units was "believed to be from a true celiac patient based on a positive IgA anti-h-tTG result of 72.4 units," indicating a combination of existing diagnostic results and clinical judgment.
- Method Comparison: Ground truth was established by comparing against predicate devices (QUANTA Lite™ h-tTG IgA and QUANTA Lite™ h-tTG IgG) and samples from "four celiac disease reference labs." This indicates ground truth based on established laboratory methods and clinical diagnoses from reference centers.
- Clinical Sensitivity and Specificity: Ground truth was based on "clinically defined samples" with diagnoses such as "Celiacs untreated," "Celiac IgA Deficient," "Dermatitis Herpetiformis," "Disease Controls," and "Healthy individuals." This suggests a reliance on clinicians' diagnoses, potentially supported by other tests, pathology (e.g., biopsy for celiac disease), or expert consensus.
8. The sample size for the training set
The document does not explicitly mention a "training set" in the context of machine learning. For an ELISA assay, the equivalent of a training set would be samples used during the development and optimization of the assay, including optimization of reagents, reaction conditions, and establishment of controls and calibrators. The document mentions that "positive and negative controls are prepared in-house and arbitrary units are assigned during the development process," which would involve a set of samples, but a specific "sample size for the training set" is not provided.
9. How the ground truth for the training set was established
As noted above, a formal "training set" with ground truth in the machine learning sense is not applicable. For the development process (equivalent to "training"), ground truth for controls was established by preparing them "in-house." For assay development and optimization, various samples (e.g., known positive, known negative) would have been used. The "assay cut-off" was established from 381 healthy individuals, providing a benchmark for defining "negative." Clinical samples with established diagnoses would have been used to guide the development of an assay that accurately reflects disease status. However, the specific details of ground truth establishment for all developmental samples are not provided.
§ 866.5660 Multiple autoantibodies immunological test system.
(a)
Identification. A multiple autoantibodies immunological test system is a device that consists of the reagents used to measure by immunochemical techniques the autoantibodies (antibodies produced against the body's own tissues) in serum and other body fluids. Measurement of multiple autoantibodies aids in the diagnosis of autoimmune disorders (disease produced when the body's own tissues are injured by autoantibodies).(b)
Classification. Class II (performance standards).