(406 days)
The EUROIMMUN Anti-SLA/LP ELISA (IgG) test kit is intended for the qualitative detection of IgG class autoantibodies against SLA/LP in human serum and plasma. It is used as an aid in the diagnosis of autoimmune hepatitis, type 1, in coniunction with other laboratory and clinical findings.
The EUROIMMUN Anti-SLA/LP ELISA (IgG) consists of a microwell ELISA plate coated with SLA/LP antigen, calibrator, positive and negative control, peroxidase-labelled anti-human IgG conjugate, sample buffer, wash buffer concentrate, TMB chromogen/substrate solution and stop solution.
Here's a detailed breakdown of the acceptance criteria and the study proving the device meets them, based on the provided document:
Acceptance Criteria and Device Performance Study for EUROIMMUN Anti-SLA/LP ELISA (IgG)
The EUROIMMUN Anti-SLA/LP ELISA (IgG) test kit is intended for the qualitative detection of IgG class autoantibodies against SLA/LP in human serum and plasma, used as an aid in the diagnosis of autoimmune hepatitis, type 1, in conjunction with other laboratory and clinical findings.
1. Table of Acceptance Criteria and Reported Device Performance
The document does not explicitly state pre-defined acceptance criteria with specific numerical thresholds for each performance characteristic (e.g., "Sensitivity must be >X%"). Instead, it presents the results of various studies as proof of the device's performance, implying that these results are considered acceptable for market clearance. The comparison to a predicate device and the clinical study results form the basis of showing substantial equivalence.
Based on the provided information, the following table summarizes the key performance metrics and the reported results:
| Acceptance Criterion (Implied) | Reported Device Performance |
|---|---|
| Precision/Reproducibility | |
| Intra-assay reproducibility (no positive sample found negative and vice versa). | Based on 20 determinations for 8 samples across different ratio ranges: |
- Samples with Mean Ratios 0.2 to 0.9 (Negative Expected): 0% positive, 100% negative.
- Samples with Mean Ratios 1.1 to 4.9 (Positive Expected): 100% positive, 0% negative.
Result: Sufficient, as no positive sample was found negative and vice versa. |
| Inter-assay reproducibility (no positive sample found negative and vice versa). | Based on 24 determinations over 6 different runs for 8 samples across different ratio ranges: - Samples with Mean Ratios 0.2 to 0.8 (Negative Expected): 0% positive, 100% negative.
- Samples with Mean Ratios 1.2 to 5.4 (Positive Expected): 100% positive, 0% negative.
Result: Sufficient, as no positive sample was found negative and vice versa. |
| Lot-to-Lot reproducibility (no positive sample found negative and vice versa). | Based on QC samples using different lots over the measurement range: - Positive Samples (Mean Ratios 5.6, 7.6, 9.7, 0.9, 3.1, 3.9): 100% positive, 0% negative.
- Negative Samples (Mean Ratios 0.1): 0% positive, 100% negative.
Result: Sufficient, as no positive sample was found negative and vice versa. |
| Analytical Specificity / Cross-reactivity | - Panel of 18 sera serologically positive for antibodies against LKM were tested.
Result: All 18 sera were negative in the Anti-SLA/LP ELISA (IgG), indicating no cross-reactivity. |
| Interference | - Concentrations of up to 1000 mg/dL for hemoglobin, 2000 mg/dL for triglyceride, and 40 mg/dL for bilirubin were tested.
Result: Individual recovery was within 90 - 119%. No significant interference observed. |
| Method Comparison with Predicate Device | |
| Negative Agreement with predicate device | Result: 96.9% (154/159), 95% C.I.: 92.8% - 99.0% |
| Positive Agreement with predicate device | Result: 100.0% (31/31), 95% C.I.: 88.8% - 100.0% |
| Overall Agreement with predicate device | Result: 97.4% (185/190), 95% C.I.: 94.0% - 99.1% |
| Matrix Comparison (Serum vs. Plasma) | The 95% C.I. of the slope contains 1.0 and the 95% C.I. of the intercept contains 0 for EDTA plasma, Heparin plasma, and Citrate plasma when compared to serum. - EDTA plasma: y = -0.02 + 1.00x (95% C.I. intercept: -0.14 to 0.18, 95% C.I. slope: 0.94 to 1.06)
- Heparin plasma: y = 0.02 + 1.00x (95% C.I. intercept: -0.06 to 0.21, 95% C.I. slope: 0.94 to 1.03)
- Citrate plasma: y = 0.04 + 0.99x (95% C.I. intercept: -0.18 to 0.16, 95% C.I. slope: 0.94 to 1.05)
Result: This condition indicates equivalence of concentration between serum and corresponding plasma matrices. Coefficients of determination were >0.99 and %BIAS from serum was 88-120%. |
| Clinical Sensitivity for AIH Type 1 | Result: 27.7% (95% C.I.: 17.3 - 40.2%) |
| Clinical Specificity | - Excluding AIH/PBC overlap samples: Result: 100.0% (95% C.I.: 99.2 - 100.0%) (435/435 negative out of 435 control samples).
- Including AIH/PBC overlap samples: Result: 99.3% (95% C.I.: 98.1 - 99.9%) (447/450 negative out of 450 control samples). |
| Expected values/Reference range for healthy blood donors | - 150 apparently healthy blood donors tested. - Cut-off ratio 1.0.
Result: All 150 blood donors found negative (0 positives, 150 negatives). Highest value was Ratio 0.8, Mean value 0.1. |
2. Sample Sizes Used for the Test Set and Data Provenance
- Precision/Reproducibility:
- Intra-assay: 20 determinations for each of 8 samples.
- Inter-assay: 24 determinations for each of 8 samples over 6 different runs.
- Lot-to-Lot: 6 determinations (3 lots x 2 runs) for 4 samples and 16 to 38 determinations (n lots x 1 run) for 5 samples.
- Provenance: Not specified, but likely internal lab samples/controls as part of device development and validation.
- Analytical Specificity (Cross-reactivity):
- 18 sera positive for antibodies against LKM.
- Provenance: Not specified.
- Interference:
- 5 different specimens at different anti-SLA/LP concentrations (ratio 0.4 - 10.0), spiked with interfering substances.
- Provenance: Not specified.
- Method Comparison with Predicate Device:
- Total samples: 190 samples (167 initial, plus 23 additional samples created by mixing positive samples to cover the lower range).
- 58 from AIH patients
- 66 from PBC patients
- 15 from patients with AIH/PBC overlap syndrome
- 28 from patients with viral hepatitis
- Demographics: 30 men and 137 women, age 1 to 87 years (average 50 years).
- Provenance: "A study was performed in cooperation with a university clinical hospital," indicating a retrospective cohort from a clinical setting. Country of origin not specified, but likely within the geographic region of the university clinical hospital.
- Total samples: 190 samples (167 initial, plus 23 additional samples created by mixing positive samples to cover the lower range).
- Matrix Comparison:
- 16 sample pairs (serum and corresponding plasma).
- Provenance: Not specified.
- Clinical Study:
- Total samples: 515 clinically characterized samples.
- 65 from AIH-1 patients
- 68 from AIH-2 patients
- 15 from patients with AIH/PBC overlap syndrome
- 367 from other control groups (including viral hepatitis, toxic liver damages, steatohepatitis, PBC, PSC, and other autoimmune diseases like MCTD, celiac disease, Diabetes Type I, Hashimoto, Grave's disease, ulcerative colitis).
- Provenance: "Performed in cooperation with several university, hospital and private laboratories," indicating a retrospective cohort drawn from multiple clinical sites. Country of origin not specified.
- Total samples: 515 clinically characterized samples.
- Expected Values/Reference Range (Healthy Donors):
- 150 apparently healthy blood donors (mixed age and sex).
- Provenance: Not specified, but likely a healthy donor pool (e.g., blood bank).
3. Number of Experts Used to Establish the Ground Truth for the Test Set and Qualifications
- The document does not explicitly state the number of experts or their specific qualifications for establishing the ground truth for any of the test sets (e.g., for the clinical study patients).
- For the "Method Comparison with Predicate Device" and "Clinical Study," samples were "clinically characterized" or from specific patient groups (e.g., "AlH patients," "AlH-1 patients"). This implies that the diagnosis (ground truth) was established through standard clinical practice by treating physicians and potentially confirmed by specialists, but the specific number and qualifications of these "experts" in the context of the study are not provided.
4. Adjudication Method for the Test Set
The document does not specify an adjudication method (e.g., 2+1, 3+1) for establishing the ground truth diagnoses or for interpreting the results during the studies. The clinical characterization of samples likely represents the consensus diagnosis from the participating clinical sites.
5. Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study
No, a multi-reader multi-case (MRMC) comparative effectiveness study comparing human readers with and without AI assistance was not done. This device is an in vitro diagnostic (IVD) ELISA kit for qualitative detection of autoantibodies, not an imaging or decision support AI device that would typically involve human-in-the-loop performance studies or require effect size reporting for human reader improvement.
6. Standalone (Algorithm Only Without Human-in-the-loop Performance)
Yes, the performance data presented (precision, analytical specificity, clinical sensitivity, specificity, method comparison) reflects the standalone (algorithm-only) performance of the ELISA kit. As an IVD kit, its output is a quantitative ratio which is then interpreted qualitatively (positive/negative) based on a predefined cut-off. There is no human-in-the-loop directly affecting the assay's output. Clinicians would then interpret these results in conjunction with other clinical findings.
7. The Type of Ground Truth Used
- For Analytical Studies (Precision, Specificity, Interference): The "ground truth" was based on predefined characteristics of the samples (e.g., known positive/negative samples, samples spiked with interferents, samples with known LKM antibodies).
- For Method Comparison with Predicate Device: The ground truth was the result obtained from the FDA-cleared Inova Quanta Lite SLA ELISA (the predicate device), alongside the clinical diagnoses of the patients from whom the samples were drawn.
- For Clinical Studies: The ground truth was based on clinical characterization and diagnosis of the patients (e.g., "Autoimmune hepatitis type 1 (AIH-1) patients", "AIH/PBC overlap syndrome", "Viral hepatitis"). This type of ground truth represents outcomes data and expert clinical judgment based on recognized diagnostic criteria for these conditions, rather than a single gold-standard pathological finding for every case.
8. The Sample Size for the Training Set
The document does not explicitly mention a "training set" in the context of machine learning or AI development. This device is an ELISA kit, which is a biochemical assay, not an AI/ML algorithm that undergoes a distinct training phase in the same manner. The development and optimization of the assay (e.g., antigen selection, reagent concentrations, incubation times) would be analogous to a "training" process, but this is done through biochemical and analytical validation, not with a distinct data-driven training set in the AI sense.
9. How the Ground Truth for the Training Set Was Established
As noted above, a "training set" in the AI/ML sense is not applicable here. The "ground truth" for the development of the assay would come from internal studies using known positive and negative samples, characterized by established clinical diagnoses or reference methods, to optimize the assay's performance characteristics (e.g., cut-off determination, reagent concentrations). The clinical study and predicate device comparison data then serve as the independent validation of the optimized assay.
§ 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).