(108 days)
The PARAGON CZE® 2000 Urine Protein Electrophoresis (UPE) Kit is intended for use with the Paragon CZE 2000 Capillary Electrophoresis System for the electrophoretic separation of proteins in human urine.
The PARAGON CZE® 2000 Urine Immunofixation by Subtraction (U-IFE/s) Electrophoresis Kit is intended for use with the Paragon CZE 2000 Capillary Electrophoresis System for the immunologic identification of monoclonal components in human urine.
The PARAGON CZE® 2000 UPE and U-IFE/s kits are designed for optimal performance on the PARAGON CZE® 2000. The UPE kits contain one Segment package containing 20 Segments, five Desalting Column packages with 4 columns per package, and four 500 mL Desalting Reagent bottles. The U-IFE/s kits contain two Segment packages containing 10 Segments, one Desalting Column package with 4 columns per package, and one 500 mL Desalting Reagent bottle.
This document describes the acceptance criteria and supporting studies for the Beckman Coulter PARAGON CZE® 2000 Urine Protein Electrophoresis (UPE) Kit and Urine Immunofixation By Subtraction (U-IFE/s) Kit.
Acceptance Criteria and Device Performance
The provided document focuses on demonstrating substantial equivalence to predicate devices rather than establishing specific acceptance criteria for a novel device. The performance data presented in Section 8.0 serves to show that the new devices perform comparably to their gel electrophoresis predicate counterparts.
Here's a table summarizing the performance data, which implicitly acts as evidence for meeting "acceptance criteria" of comparable performance to the predicate:
| Test | Acceptance Criteria (Implied by Comparison) | Reported Device Performance (PARAGON CZE 2000) |
|---|---|---|
| UPE Kit Method Comparison | High agreement with the PARAGON Gel Electrophoresis – SPE Kit | 95% Agreement, 4% Partial Agreement, 1% Disagreement (n=100) |
| U-IFE/s Kit Method Comparison | High agreement with the PARAGON Gel Electrophoresis – IFE Kit | 92.7% Full Agreement, 7.3% Disagreement (n=82) |
| UPE Imprecision (Albumin) | Low Coefficient of Variation (C.V.) for various imprecision types | Within-Run (System): 2.5% C.V. |
| Within-Run (Desalting): 3.0% C.V. | ||
| Total (Level 1): 2.5% C.V. | ||
| Total (Level 2): 2.2% C.V. | ||
| Total (Level 3): 9.2% C.V. | ||
| UPE Imprecision (BJP) | Low Coefficient of Variation (C.V.) for various imprecision types | Within-Run (System): 7.1% C.V. |
| Within-Run (Desalting): 5.3% C.V. | ||
| Total (Level 1): 5.9% C.V. | ||
| Total (Level 2): 12.7% C.V. | ||
| Total (Level 3): 2.0% C.V. | ||
| U-IFE/s Reproducibility | Visual agreement of electropherograms, no observable differences in morphology or subtraction, and Bence-Jones protein subtraction as expected, meeting an 80% specification for visual agreement. | Reproducibility meets the 80% specification for visual agreement for the replicate segments. No observable difference in morphology or subtraction was seen. BJP subtraction with kappa and lambda was as expected. |
Study Information
The document describes studies conducted to demonstrate substantial equivalence of the new PARAGON CZE® 2000 UPE and U-IFE/s Kits to their respective predicate gel electrophoresis kits.
2. Sample Sizes Used for the Test Set and Data Provenance
- UPE Kit Method Comparison:
- Test Set Sample Size: 100 samples (indicated by 'n=100').
- Data Provenance: Not explicitly stated, but clinical laboratory samples are generally assumed to be retrospective or a mix of prospective/retrospective from a specific clinical setting. Given the year 2005, it's highly likely to be retrospective samples from a clinical lab. No country of origin is specified.
- U-IFE/s Kit Method Comparison:
- Test Set Sample Size: 82 samples (indicated by 'n=82').
- Data Provenance: Not explicitly stated, but similar to the UPE kit, likely retrospective clinical samples. No country of origin is specified.
- UPE Imprecision Studies:
- System Reproducibility: 21 runs for each urine pool (Albumin and BJP).
- Desalting Reproducibility: 21 runs for each urine pool (Albumin and BJP).
- Total Imprecision (EP 10-A): 15 runs for each of the three urine levels (Albumin and BJP).
- Data Provenance: These are laboratory studies, likely conducted internally by the manufacturer. No country of origin is specified.
- U-IFE/s Reproducibility Studies:
- Test Set Sample Size: Not explicitly stated as a number, but the description mentions "the replicate segments tested."
- Data Provenance: These are laboratory studies, likely conducted internally by the manufacturer. No country of origin is specified.
3. Number of Experts and Qualifications for Establishing Ground Truth for the Test Set
The studies outlined (method comparison and imprecision) do not involve human expert consensus for "ground truth" in the way an imaging AI study would. Instead, the "ground truth" for the method comparison studies is established by the results from the predicate devices: the PARAGON Gel Electrophoresis – SPE Kit (for UPE) and the PARAGON Gel Electrophoresis – IFE Kit (for U-IFE/s).
- Number of Experts: Not applicable in the context of this type of IVD device study. The predicate device's results are considered the reference.
- Qualifications of Experts: Not applicable. The "ground truth" is derived from a previously cleared diagnostic method.
4. Adjudication Method for the Test Set
- Method Comparison Studies (UPE and U-IFE/s): The comparison method involves results from the new device being assessed against the predicate device. The results are categorized as "Agreement," "Partial Agreement," or "Disagreement" for UPE, and "Full Agreement" or "Disagreement" for U-IFE/s. It is implied that laboratory personnel or study investigators evaluated these agreements based on established protocols for comparing quantitative or qualitative results from two different methods. The specific adjudication method (e.g., 2+1, 3+1) is not specified as it's not a common paradigm for these types of IVD performance studies where instrument results are directly compared.
- Imprecision and Reproducibility Studies: These are quantitative measurements. Adjudication is statistical (e.g., calculation of Mean, SD, %C.V.). For U-IFE/s visual reproducibility, it states "electro-pherograms were visually inspected," but it doesn't specify if multiple readers were involved or an adjudication process for discrepancies.
5. If a Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study was Done
No, an MRMC comparative effectiveness study was not done. This type of study (comparing human readers with and without AI assistance) is typically performed for AI/CADe (Computer-Aided Detection/Diagnosis) devices, especially in medical imaging. The devices in this submission (UPE and U-IFE/s Kits) are in vitro diagnostic kits that automate protein analysis, not an AI/CADe system for human interpretation enhancement. The comparison is between a new automated method (capillary electrophoresis) and an older automated/semi-automated method (gel electrophoresis).
6. If a Standalone (i.e. algorithm only without human-in-the loop performance) was Done
Yes, the studies presented are essentially standalone performance evaluations of the devices. The "device performance" refers to the automated analysis performed by the PARAGON CZE® 2000 system using these kits. While a human operator initiates the test and interprets the final electropherograms (especially for U-IFE/s with visual inspection), the studies quantify the device's ability to perform the separation and detection reliably and accurately relative to a predicate, independent of variable human interpretation during the core analytical process.
7. The Type of Ground Truth Used
- For Method Comparison studies: The "ground truth" is implicitly the analytical result obtained from the predicate devices (PARAGON Gel Electrophoresis – SPE Kit for UPE, and PARAGON Gel Electrophoresis – IFE Kit for U-IFE/s).
- For Imprecision and Reproducibility studies: The ground truth is the expected performance within laboratory quality control parameters, with various urine pools (with known or characterized levels of Albumin and BJP) serving as reference samples.
8. The Sample Size for the Training Set
The document is a 510(k) submission for in vitro diagnostic kits, not an AI/machine learning device. Therefore, the concept of a "training set" for an algorithm, as understood in AI development, does not apply here. The kits are reagents and consumables for an existing analytical instrument (PARAGON CZE® 2000). The development of the methodology and reagents would have involved internal validation and optimization, but there isn't a "training set" in the sense of data used to train a predictive model.
9. How the Ground Truth for the Training Set Was Established
As stated above, the concept of a "training set" for an algorithm is not applicable to the development and validation of these IVD kits. The "ground truth" in the development of such kits would involve:
- Analytical Chemistry Principles: The fundamental principles of capillary electrophoresis and immunofixation.
- Reference Materials: Using certified reference materials or well-characterized patient samples with known protein compositions/concentrations.
- Comparison to Established Methods: Extensive internal testing and optimization against existing, validated methods (like the predicate gel electrophoresis systems) to fine-tune reaction conditions, reagent formulations, and instrument parameters to achieve desired performance characteristics.
This information is derived from the provided 510(k) summary document.
§ 866.5150 Bence-Jones proteins immunological test system.
(a)
Identification. A Bence-Jones proteins immunological test system is a device that consists of the reagents used to measure by immunochemical techniques the Bence-Jones proteins in urine and plasma. Immunoglobulin molecules normally consist of pairs of polypeptide chains (subunits) of unequal size (light chains and heavy chains) bound together by several disulfide bridges. In some cancerous conditions, there is a proliferation of one plasma cell (antibody-producing cell) with excess production of light chains of one specific kind (monoclonal light chains). These free homogeneous light chains not associated with an immunoglobulin molecule can be found in urine and plasma, and have been called Bence-Jones proteins. Measurement of Bence-Jones proteins and determination that they are monoclonal aid in the diagnosis of multiple myeloma (malignant proliferation of plasma cells), Waldenstrom's macroglobulinemia (increased production of large immunoglobulins by spleen and bone marrow cells), leukemia (cancer of the blood-forming organs), and lymphoma (cancer of the lymphoid tissue).(b)
Classification. Class II (performance standards).