(88 days)
The CAPI 3 IMMUNOTYPING kit is designed for the qualitative detection and the characterization of monoclonal proteins (immunotyping) in human urine and serum with the CAPILLARYS 3 TERA instrument, SEBIA, for capillary electrophoresis. It is used in conjunction with the CAPI 3 PROTEIN(E) 6 kit, SEBIA, designed for proteins separation into 6 major fractions in alkaline buffer (pH 9.9).
The CAPILLARYS 3 TERA instrument performs all procedural sequences automatically to obtain a profile for qualitative analysis. Each urine or serum sample is mixed with individual antisera that are specific against gamma (Ig G). alpha (Ig A) and mu (Ig M) heavy chains, and kappa (free and bound) light chains and lambda (free and bound) light chains, respectively.
The proteins, separated in silica capillaries, are directly detected by their absorbance at 200 nm.
The electrophoregrams are evaluated visually to detect the presence of specific reactions with the suspect monoclonal proteins.
For In Vitro Diagnostic Use.
The capillary electrophoresis provides complete automation with fast separation and good resolution. This electrokinetic separation technique is carried out in a silica glass tube (i.e., capillary) with internal diameter lower than 100 um filled with a buffer composed of electrolytes.
The CAPILLARYS 3 TERA instrument uses the principle of capillary electrophoresis in free solution. With this technique, charged molecules are separated by their electrophoretic mobility in an alkaline buffer with a specific pH. Separation occurs according to the electrolyte pH and electroosmotic flow. The CAPILLARYS 3 TERA instrument has silica capillaries functioning in parallel allowing 12 simultaneous analyses.
In capillary electrophoresis, abnormal fractions detected in serum or urine protein electrophoregrams, primarily those in the beta globulin and gamma globulin zones, are always suspect of being monoclonal proteins (M-proteins, paraproteins, monoclonal immunoglobulins). With the CAPI 3 IMMUNOTYPING procedure, the immunotyping procedure uses specific antibodies to identify these abnormal fractions.
In immunotyping a sample dilution is prepared and injected at the anodic end of six capillaries. The reference pattern (ELP pattern), which is a complete electrophoretic pattern of the sample's proteins, is obtained by mixing the sample with the ELP solution and injection into the 1st capillary. The antisera patterns are obtained by sample aspiration into the 5 subsequent capillaries. Previously diluted samples are mixed with specific antisera against gamma (Ig G), alpha (lg A), mu (lg M) heavy chains, and free and bound Kappa and Lambda light chains. Protein separation is performed in a high voltage electrical field. The separated proteins are detected using absorbance at 200 nm at the cathodic end of the capillary. After the analysis, the capillaries are immediately washed with a wash solution and filled with buffer which prepares the capillaries for the next analysis.
The immunotyping is performed in four automated steps:
- Dilution of serum or urine samples with a specific diluent in the pre-dilution well of the reagent cup. This dilution is made according to the sample's immunoglobulin concentration.
- Mixing diluted serum sample with specific antisera. The antigen - antibody complex is formed rapidly in liquid medium without the need for extra incubation step or removal of the immune complexes.
- Injection of the prepared samples with simultaneous aspiration into 6 capillaries at the anodic end. Protein separation occurs when a high voltage field is applied to the alkaline buffer. The separated proteins are detected using absorbance at 200 nm at the cathodic end of the capillary.
- Overlay of the ELP pattern on the antisera patterns (Ig G, Ig A, Ig M, Kappa and Lambda) allows characterization of suspected monoclonal component.
Here's an analysis of the acceptance criteria and study data for the CAPI 3 IMMUNOTYPING device, based on the provided FDA 510(k) summary:
1. Table of Acceptance Criteria and Reported Device Performance
The FDA 510(k) summary does not explicitly list predefined quantitative acceptance criteria with specific numerical thresholds for all performance metrics (e.g., repeatability, reproducibility, accuracy). Instead, for qualitative assessments, the acceptance criterion appears to be "concordant results" or "100% agreement." For sensitivity, the acceptance criterion is implicitly shown by the reported detection limits.
| Performance Metric | Acceptance Criteria (Implicit from Document) | Reported Device Performance |
|---|---|---|
| Repeatability (Within-run and Between-capillaries) | All samples should give concordant results within run and between capillaries. | For each tested reagent (ELP, Anti-IgG, Anti-IgA, Anti-IgM, Anti-Kappa, Anti-Lambda), all 4 urine samples (including Bence Jones proteins) gave concordant results within run and between capillaries. All dilution programs were tested. |
| Reproducibility (Between lots and instruments) | All samples should give concordant results across different runs, instruments, and reagent lots. | All 3 urine samples (including Bence Jones proteins) gave concordant results for all 18 runs (over 5 working days, 2 times/day) on the 3 CAPILLARYS 3 TERA instruments and with the 3 lots of CAPI 3 IMMUNOTYPING kit. All dilution programs were tested. |
| Sensitivity (Detection Limit) | Monoclonal proteins in urine should be detectable at clinically relevant low concentrations. (No explicit numerical acceptance criterion is stated, but the ability to detect at specified g/L or mg/dL is the measure). | Detection Limits Reported: |
- Lambda free: 0.010 g/L (1.0 mg/dL)
- Kappa free: 0.030 g/L (3.0 mg/dL)
- IgG Lambda: 0.004 g/L (0.4 mg/dL)
- Lambda (part of IgG Lambda): 0.004 g/L (0.4 mg/dL) |
| Sample Stability (2-8 °C) | Results after 1 week storage at 2-8 °C should comply with "acceptance criteria defined by SEBIA." (Specific criteria not detailed in document). | The results obtained complied with the acceptance criteria defined by SEBIA. Conclusion: Urine samples can be stored for 1 week between 2 and 8 °C. |
| Sample Stability (-70/-80 °C) | Results after 1 month storage at -70/-80 °C should comply with "acceptance criteria defined by SEBIA." (Specific criteria not detailed in document). | The results obtained complied with the acceptance criteria defined by SEBIA. Conclusion: Urine samples can be stored for 1 month between -70 and -80 ℃. |
| Kit Stability (CAPI 3 IMMUNOTYPING sample diluent) | Stable for 2 years at 2 - 30 °C. (The acceptance criterion is the claimed stability period and conditions). | 2 years at 2 - 30 °C (Reported as claimed stability). |
| Kit Stability (CAPI 3 IMMUNOTYPING ELP solution) | Stable for 2 years at 2 - 8 °C. (The acceptance criterion is the claimed stability period and conditions). | 2 years at 2 - 8 °C (Reported as claimed stability). |
| Kit Stability (CAPI 3 IMMUNOTYPING antisera) | Stable for 2 years at 2 - 8 °C. (The acceptance criterion is the claimed stability period and conditions). | 2 years at 2 - 8 °C (Reported as claimed stability). |
| On-Board Stability (Sample Diluent, ELP, Antisera) | Stable for 2 months on-board the instrument. (The acceptance criterion is the claimed stability period). | 2 months (Reported as claimed stability). |
| Accuracy/Concordance | 100% agreement between the test technique (CAPI 3 IMMUNOTYPING on CAPILLARYS 3 TERA) and the reference technique (CAPILLARYS IMMUNOTYPING on CAPILLARYS 2) for qualitative results, including presence/absence and type of monoclonal component. | This study demonstrated 100% agreement between the two techniques for all 52 urine samples (8 without, 44 with monoclonal components of various types). This 100% agreement was confirmed for specific monoclonal protein types (IgG Lambda, IgG Kappa, IgG Lambda with Lambda free, IgG Kappa with Kappa free, Lambda free, Kappa free, IgM Kappa with Kappa free, IgA Lambda, IgA Lambda with Lambda free, and without Monoclonal). |
2. Sample Size Used for the Test Set and Data Provenance
- Repeatability: 4 different urine samples with monoclonal proteins (including Bence Jones proteins).
- Reproducibility: 3 different urine samples with monoclonal proteins (including Bence Jones proteins).
- Sensitivity: 3 pathological urine samples, serially diluted in normal urine.
- Sample Stability (2-8 °C): 10 urine samples (normal and pathological).
- Sample Stability (-70/-80 °C): 10 urine samples (normal and pathological).
- Accuracy/Concordance: 52 urine samples (8 without monoclonal component, 44 with monoclonal component including Bence Jones proteins).
Data Provenance: The document does not explicitly state the country of origin of the data or whether it was retrospective or prospective. Given that it's a 510(k) submission from Sebia (manufacturing in France, submitter in USA), and refers to "urine samples," these are likely clinical samples, but details on their origin are not provided. The phrase "analyzed at the beginning of the study (reference) and after X storage (test)" suggests a prospective element for the stability studies on stored samples.
3. Number of Experts Used to Establish Ground Truth for the Test Set and Qualifications
The document states: "All electrophoregrams were evaluated visually for the qualitative results." However, it does not specify the number of experts used for ground truth establishment or their qualifications (e.g., "radiologist with 10 years of experience"). This is a significant omission from the perspective of external validation of the ground truth. It is implied that visual evaluation by trained personnel (likely laboratory professionals or experts in electrophoresis interpretation) was the method, but no further details are provided.
4. Adjudication Method for the Test Set
The document does not detail an adjudication method (such as 2+1 or 3+1 consensus) for the visual evaluation of electrophoregrams. It simply states they were "evaluated visually." This implies a single evaluator, or an internal process where disagreement (if any) was resolved, but the process is not described.
5. Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study
No multi-reader multi-case (MRMC) comparative effectiveness study was described. The accuracy/concordance study compares the new device's visual interpretation results to an existing, cleared device's visual interpretation results, which is a comparison of two methods, not an assessment of human readers with and without AI assistance. The device itself is a qualitative detection and characterization system, and the evaluation stated is "electrophoregrams were evaluated visually," indicating human interpretation is still involved in the final result.
6. Standalone Performance Study
The studies described (repeatability, reproducibility, sensitivity, stability, accuracy/concordance) evaluate the performance of the integrated system (CAPI 3 IMMUNOTYPING kit on CAPILLARYS 3 TERA instrument). The "electrophoregrams are evaluated visually" suggests human-in-the-loop performance, rather than a purely standalone algorithm. The device produces a "protein profile for qualitative analysis," which is then visually interpreted. Therefore, a standalone (algorithm only) performance study, in the sense of an AI algorithm making a definitive diagnosis without human oversight, does not appear to have been performed or is not directly applicable to the described use case which involves visual evaluation.
7. Type of Ground Truth Used
The ground truth for the test sets (e.g., for accuracy, repeatability, reproducibility studies) appears to be established by the visual evaluation of electrophoregrams performed by an expert or experts using a "reference technique" or the device itself.
- For the accuracy/concordance study, the "reference technique" (CAPILLARYS IMMUNOTYPING URINE procedure performed with the CAPILLARYS 2 instrument) served as the comparator for establishing ground truth, assuming the predicate device's results are considered the established truth.
- For other studies like repeatability and reproducibility, the "ground truth" seems to implicitly be the correct identification of presence/absence and type of monoclonal protein based on the expected outcome for the known pathological samples, assessed by visual evaluation.
- No mention of pathology, outcomes data, or independent gold standard beyond another electrophoretic method is provided.
8. Sample Size for the Training Set
The document does not mention a "training set" or explicitly describe the development of an algorithm that learns from data. This device is an immunodiagnostic test system based on capillary electrophoresis and specific antisera, producing profiles that are then visually interpreted. It is not presented as an AI/ML-driven device requiring a training set in the conventional sense. The "training" for such a system would typically involve its chemical and mechanical design, calibration, and validation against a standard.
9. How the Ground Truth for the Training Set Was Established
As no training set is described for an AI/ML algorithm, this question is not applicable. The assay's "truth" is inherent in its biochemical principle (antigen-antibody reaction, electrophoretic separation) and the expertise of interpreting the resulting electrophoregrams.
§ 866.5510 Immunoglobulins A, G, M, D, and E immunological test system.
(a)
Identification. An immunoglobulins A, G, M, D, and E immunological test system is a device that consists of the reagents used to measure by immunochemical techniques the immunoglobulins A, G, M, D, an E (serum antibodies) in serum. Measurement of these immunoglobulins aids in the diagnosis of abnormal protein metabolism and the body's lack of ability to resist infectious agents.(b)
Classification. Class II (performance standards).