Search Results

The ADVIA Centaur® Erythropoietin (EPO) assay is for in the quantitative measurement of exythropoitin in pediatric and adult human serum or plasma (K2-EDTA, lithium heparin) using the ADVIA Centaur XP system. Measurement of erythropoietin is used as an aid in the diagnosis of anemias and polycythemias.

The ADVIA Centaur EPO assay is a fully automated, one-step sandwich immunoassay using direct chemiluminescent technology. The assay utilizes an acridinium-ester-labeled monoclonal mouse anti-EPO antibody in the Lite Reagent. The Solid Phase consists of mouse anti-EPO monoclonal antibody-coated paramagnetic microparticles.

Here's an analysis of the provided text to extract the acceptance criteria and study details for the ADVIA Centaur Erythropoietin (EPO) assay:

1. Table of Acceptance Criteria and Reported Device Performance

This table compiles information primarily from the "Summary of Non-Clinical Performance" and "Summary of Clinical Performance" sections.

2. Sample Size Used for the Test Set and Data Provenance

• Linearity: Not specified, but involved three high EPO samples mixed with low EPO human serum.
• Dilution Recovery: 10 samples (containing high EPO levels: 618.63-986.07 mIU/mL).
• Detection Capability (LoD): 323 determinations using 10 low-level samples.
• Cross-reactivity: Not explicitly stated as a "sample size," but involved numerous cross-reactants (e.g., various plasma proteins, epoetin alfa, darbepoetin alfa).
• Interference: Not explicitly stated as a specific "sample size" for each interferent, but involved various substances tested at different concentrations.
• Precision: 7 pooled serum samples. For each sample, there were 80 observations (replicates of 2, in 2 runs/day, over 20 days).
• Specimen Collection Comparison: 65 samples (serum EPO values ranging from 4.39 - 707.81 mIU/mL).
• Method Comparison:
  • Study 1: 216 human serum samples (range: 3.29 – 691.60 mIU/mL).
  • Study 2: 100 human serum samples from US population (range: 4.45 - 407.74 mIU/mL).
  • Study 3 (Multi-site): 327 human serum samples (range: 3.55 - 596.81 mIU/mL), with ≥ 100 samples per site.
• Expected Values (Adult): 251 apparently healthy subjects (128 males, 123 females), older than 21 years of age.
• Expected Values (Pediatric): 266 apparently healthy children (2 to <13 years) and adolescents (13 to <22 years).

Data Provenance:
The document does not explicitly state the country of origin for all data sets, but mentions:

• Method Comparison Study 2: "100 human serum samples from US population".
• Method Comparison Study 3: "3 sites, 2 within Europe and 1 within the US".
• Most studies imply retrospective (samples collected before the study) or prospective (samples collected during the study for specific evaluation), but this is not consistently specified for each study. For establishing expected values, samples were collected prospectively following specific criteria (e.g., healthy subjects, specific age ranges, exclusion criteria).

3. Number of Experts Used to Establish the Ground Truth for the Test Set and Their Qualifications

• This device is an in vitro diagnostic assay that measures erythropoietin levels. The "ground truth" for such assays typically comes from the actual concentration of the analyte in the biological sample, often confirmed by a reference method or standardized preparations.
• The document describes traceability to World Health Organization (WHO) International Reference Preparations and Standards (NIBSC codes 67/343 and 11/170). These international standards are established through collaborative studies involving multiple expert laboratories and scientists, but the immediate "ground truth" for the test samples in this submission would be their assigned values based on these standards or the results from the predicate device.
• No "experts" in the sense of radiologists or other clinicians interpreting images or assessments are directly involved in establishing the ground truth for individual test samples for this type of device. The ground truth is biochemical measurement.

4. Adjudication Method for the Test Set

• Not applicable. This is a quantitative diagnostic assay. "Adjudication" typically refers to resolving discrepancies between multiple human readers in diagnostic imaging or clinical assessment studies. For an immunoassay, the "ground truth" is a measured concentration.

5. If a Multi Reader Multi Case (MRMC) Comparative Effectiveness Study Was Done, and the Effect Size of How Much Human Readers Improve with AI vs Without AI Assistance

• Not applicable. This is an immunoassay, not an AI-assisted diagnostic imaging or clinical decision support device involving human readers.

6. If a Standalone (i.e. algorithm only without human-in-the-loop performance) Was Done

• Yes, the performance studies described (Linearity, Dilution Recovery, Measuring Interval, Detection Capability, High Dose Hook, Cross-reactivity, Interference, Precision, Specimen Collection Comparison, Method Comparison, Expected Values) represent the standalone performance of the ADVIA Centaur Erythropoietin (EPO) assay itself, meaning the algorithm/instrument's measurement capabilities without direct human interpretation of the final measurement result for diagnostic purposes. The device directly produces a quantitative EPO value.

7. The Type of Ground Truth Used

• Reference materials/standards: The assay is traceable to WHO International Reference Preparations/Standards for Erythropoietin.
• Predicate device comparison: For method comparison studies, the results from the legally marketed predicate device (Beckman Coulter Access EPO Assay) serve as a comparative "ground truth" to demonstrate substantial equivalence.
• Expert consensus (indirectly): The establishment of reference intervals (expected values) involved defining "healthy" populations based on standard medical criteria, which would implicitly rely on shared medical understanding and the consensus of medical professionals. The CLSI protocols referenced (e.g., EP28-A3) also represent a form of expert consensus on how to establish such values.
• Biochemical measurement: The fundamental ground truth is the actual concentration of EPO in the samples, determined through rigorous biochemical methods and validated against international standards.

8. The Sample Size for the Training Set

• For an immunoassay like this, there isn't a "training set" in the machine learning sense. The device is not learning from data in the same way an AI algorithm would. Instead, the assay's reagents, calibration curves, and analytical procedures are developed and optimized through extensive R&D and internal validation studies. The studies described in the 510(k) are primarily for verification and validation (V&V) of the final device's performance characteristics. Therefore, a specific "training set sample size" as might be provided for an AI/ML device is not applicable or provided.

9. How the Ground Truth for the Training Set Was Established

• As above, the concept of a "training set" and its "ground truth" in the context of machine learning is not directly applicable to an immunoassay. The development of the assay involves extensive characterization of reagents, optimization of reaction conditions, and establishment of calibration curves using purified EPO standards of known concentrations, which are themselves traceable to international standards. The "ground truth" during this development phase would be the known concentrations of these standards.

Ask a specific question about this device