(92 days)
The ARCHITECT CA 19-9xR assay is a chemiluminescent microparticle immunoassay (CMIA) for the quantitative determination of 1116-NS-19-9 reactive determinants in human serum or plasma on the ARCHITECT i System. The ARCHITECT CA 19-9xR assay is to be used as an aid in the management of pancreatic cancer patients in conjunction with other clinical methods.
Patients known to be genotypically negative for the Lewis blood group antigen will be unable to produce the CA 19-9 antigen even in the presence of malignant tissue. Phenotyping for the presence of the Lewis antigen may be insufficient to detect true Lewis antigen negative individuals. Even patients who are genotypically positive for the Lewis antigen may produce varying levels of CA 19-9 based on gene dosage effect.
The ARCHITECT CA 19-9xR assay is a two-step immunoassay for the quantitative determination of 1116-NS-19-9 reactive determinants in human serum or plasma using CMIA technology with flexible assay protocols, referred to as Chemiflex®.
In the first step, sample and 1116-NS-19-9 coated paramagnetic microparticles are combined. 1116-NS-19-9 reactive determinants present in the sample bind to the 1116-NS-19-9 coated microparticles. After washing, 1116-NS-19-9 acridinium-labeled conjugate is added to create a reaction mixture in the second step. Following another wash cycle, pretrigger and trigger solutions are added to the reaction mixture. The resulting chemiluminescent reaction is measured as relative light units (RLUs). A direct relationship exists between the amount of 1116-NS-19-9 reactive determinants in the sample and the RLUs detected by the ARCHITECT i System optics.
The ARCHITECT® CA 19-9™xR Assay is a chemiluminescent microparticle immunoassay (CMIA) for the quantitative determination of 1116-NS-19-9 reactive determinants in human serum or plasma. It is intended to be used as an aid in the management of pancreatic cancer patients in conjunction with other clinical methods.
1. Table of Acceptance Criteria and Reported Device Performance:
The document primarily focuses on demonstrating substantial equivalence to a predicate device and summarizing performance characteristics rather than explicit 'acceptance criteria' in the format of a clinical trial endpoint table. However, based on the reproducibility and comparison studies, we can infer performance targets and the achieved results.
| Study/Metric | Acceptance Criteria (Inferred) | Reported Device Performance |
|---|---|---|
| Reproducibility | Total precision %CV less than or equal to 10% | Total precision %CV was determined to be less than or equal to 10%. |
| Comparison (Correlation with Predicate) | High correlation coefficient (e.g., typically >0.90 for substantial equivalence in quantitative assays) and acceptable slope/intercept with the predicate device. | Correlation coefficient of 0.96 with the Fujirebio Diagnostics, Inc. CA 19-9 RIA. Slope of 1.2 (99% CI: 1.08, 1.37). Y-axis intercept of -5.1 U/mL (99% CI: -7.4, -3.4). |
| Reference Ranges (Apparently Healthy) | A high percentage of apparently healthy individuals should fall below a specified cutoff (e.g., 37 U/mL). | 94.4% of apparently healthy subjects (n=360) had values of 37 U/mL or less. |
| Association between Marker Change and Disease State Change (Concordance) | Demonstrated reasonable concordance rates (Total, Positive, Negative) between CA 19-9 value changes and disease state changes. | * Overall (Observation Pairs): Total Concordance (C) = 61.0%, Positive Concordance (C+) = 48.5%, Negative Concordance (C-) = 63.6% (based on 187 observation pairs). |
- Per-Patient Analysis: Total Concordance (C) = 68.92% (95% CI: 57.10% - 79.17%), Positive Concordance (C+) = 68.18% (95% CI: 45.13% - 86.14%), Negative Concordance (C-) = 69.23% (95% CI: 54.90% - 81.28%) (based on 74 patients). |
2. Sample Size Used for the Test Set and Data Provenance:
- Reproducibility: Not explicitly stated as a test set size, but samples were tested consisting of two panels of standards, one panel of serum with added determinants, and controls. Each sample was tested two separate times per day for 20 nonconsecutive days, using two lots of reagents, in replicates.
- Comparison Study: 259 serum specimens. Data provenance is not specified, but typically for clinical studies supporting US regulatory submissions, samples would be sourced from a diverse US population or explicitly identified as international if applicable. The document does not state whether it was retrospective or prospective.
- Reference Ranges (Apparently Healthy Population): 360 serum specimens from apparently healthy individuals. Data provenance not specified.
- Patient Groups (Disease Distribution): 978 individual serum samples from various non-malignant and malignant disease groups. Data provenance not specified.
- Pancreatic Cancer Serial Specimens (Concordance Study): 74 patients, yielding 261 evaluable observations (average 3.5 observations per patient) for the observation-pair analysis, and 74 for the per-patient analysis. Data provenance not specified. The average age was 61.8 years (range 17-85 years), 55% were men, 45% women. Staging was available for 70 of the 74 patients.
3. Number of Experts Used to Establish Ground Truth for the Test Set and Their Qualifications:
The document describes an in vitro diagnostic (IVD) assay measuring a tumor marker, CA 19-9. The "ground truth" for such assays typically refers to the confirmed clinical diagnosis of pancreatic cancer and the progression/regression of the disease.
- The document does not explicitly state the number or qualifications of experts used to establish the clinical ground truth (e.g., pancreatic cancer diagnosis, disease progression/no progression) for the patient samples used in the reference ranges, patient group distribution, or concordance studies.
- However, for the "Association between Change in Marker Value and Change in Disease State" study, the ground truth for "Change in Disease State (W)" (Progression vs. No Progression) was established. This likely involved clinical assessment by medical professionals (e.g., oncologists, radiologists) based on various clinical methods like imaging, biopsy, and other clinical observations. The document itself is a 510(k) summary, which often omits granular details about clinical expert involvement if not directly related to the device's technical performance.
4. Adjudication Method for the Test Set:
- The document does not describe an adjudication method for establishing the "Change in Disease State (W)" in the concordance study. Clinical diagnoses and disease state changes are typically determined by treating clinicians based on standard clinical practice, which may inherently involve a form of consensus or judgment, but a specific, formalized adjudication process involving multiple independent reviewers is not mentioned.
5. If a Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study Was Done:
- No, an MRMC comparative effectiveness study was not done. This device is an in vitro diagnostic (IVD) assay that provides a quantitative measurement of a biomarker. MRMC studies are typically performed for imaging devices or algorithms where human readers interpret images or data, and the AI's impact on their performance is being evaluated. This document focuses on the analytical and clinical performance of the assay itself.
6. If a Standalone (i.e., algorithm only without human-in-the-loop performance) Was Done:
- Yes, this entire submission effectively describes a standalone performance evaluation of the ARCHITECT CA 19-9xR assay. As an IVD test, it generates results independently. The "performance" described (reproducibility, correlation, reference ranges, association with disease state) is the performance of the assay system itself, not in conjunction with human interpretation of its results in a structured reader study. Its intended use is "as an aid in the management of pancreatic cancer patients in conjunction with other clinical methods," implying its results are interpreted by clinicians.
7. The Type of Ground Truth Used:
- For the technical performance (reproducibility, comparison): The ground truth is the true concentration of CA 19-9 reactive determinants or reference method values.
- For the clinical performance (reference ranges, patient groups, concordance study): The ground truth for disease classifications and changes in disease state would be clinical diagnosis and disease progression/regression determined by treating physicians based on a combination of clinical methods (e.g., pathology reports, imaging studies, clinical assessment, treatment response, outcomes data). The document refers to "Change in Disease State (W)," which is derived from these clinical assessments.
8. The Sample Size for the Training Set:
- This submission describes a mature IVD product (ARCHITECT CA 19-9xR Assay) that is compared to an existing predicate device (Fujirebio Diagnostics, Inc. CA 19-9 RIA). IVD assays, especially for biomarkers, are typically developed using rigorous analytical validation rather than machine learning "training sets" in the conventional sense.
- Therefore, the concept of a "training set" as understood in AI/ML contexts does not directly apply to this type of device and study. The assay itself is a chemical measurement system, not a learning algorithm that requires a separate training phase. The studies described are for validation and performance assessment.
9. How the Ground Truth for the Training Set Was Established:
- As explained in point 8, there isn't a "training set" in the AI/ML sense for this device. The ground truth for developing and optimizing such an assay would involve extensive analytical validation against reference materials, known concentrations, and clinical samples with well-characterized disease states. This process is part of product development and analytical verification, not typically described as establishing ground truth for a "training set" in a 510(k) summary.
§ 866.6010 Tumor-associated antigen immunological test system.
(a)
Identification. A tumor-associated antigen immunological test system is a device that consists of reagents used to qualitatively or quantitatively measure, by immunochemical techniques, tumor-associated antigens in serum, plasma, urine, or other body fluids. This device is intended as an aid in monitoring patients for disease progress or response to therapy or for the detection of recurrent or residual disease.(b)
Classification. Class II (special controls). Tumor markers must comply with the following special controls: (1) A guidance document entitled “Guidance Document for the Submission of Tumor Associated Antigen Premarket Notifications (510(k)s) to FDA,” and (2) voluntary assay performance standards issued by the National Committee on Clinical Laboratory Standards.