Search Results

The IDS Cortisol assay is an in vitro diagnostic device intended for the quantitative determination of cortisol in human serum and plasma on the IDS system. Results are to be used in conjunction with other clinical and laboratory data to assist clinicians in the diagnosis and treatment of disorders of the adrenal gland.

Device Description

The IDS Cortisol assay consists of a reagent cartridge. The reagent cartridge contains multiple reagents:

MPE1:Magnetic particles coated rat anti-mouse monoclonal antibody in a phosphate buffer with Proclin as preservative.
CONJ: Cortisol coupled with an acridinium ester derivative in = phosphate buffer with Proclin as a preservative.
mAb: Mouse anti-cortisol monoclonal antibody in phosphate buffer with Proclin as a preservative .;
BUF: HEPES buffer containing Proclin as preservative .

AI/ML Overview

This document describes the analytical performance of the IDS Cortisol assay, an in vitro diagnostic device, and demonstrates its substantial equivalence to a predicate device (Roche Elecsys Cortisol II). The acceptance criteria and the study proving the device meets these criteria are detailed below.

1. Table of Acceptance Criteria and Reported Device Performance

The acceptance criteria for this in vitro diagnostic device are primarily based on demonstrating analytical performance that is comparable to, or meets specified standards relative to, established laboratory methods and a predicate device.

Performance Characteristic	Acceptance Criteria (from context/implied standard)	Reported Device Performance (IDS Cortisol)
Precision	Repeatability (Within-run): Lower CV%	Within-Run / Repeatability (n=80 per sample, 1 lot, 1 system): - 0.94 µg/dL: 7.8% CV - 1.84 µg/dL: 4.6% CV - 5.75 µg/dL: 2.4% CV - 13.06 µg/dL: 2.4% CV - 19.94 µg/dL: 1.8% CV - 44.63 µg/dL: 1.9% CV
	Intermediate Precision (Within-System/Total): Lower CV%	Within-System (n=80 per sample, 1 lot, 1 system): - 0.94 µg/dL: 16.2% CV - 1.84 µg/dL: 10.9% CV - 5.75 µg/dL: 5.2% CV - 13.06 µg/dL: 3.9% CV - 19.94 µg/dL: 5.1% CV - 44.63 µg/dL: 4.2% CV Total (Combind 3 lots, 3 systems, n=240 per sample): - 0.88 µg/dL: 15.3% CV - 1.78 µg/dL: 10.1% CV - 5.75 µg/dL: 4.5% CV - 13.09 µg/dL: 3.3% CV - 20.22 µg/dL: 4.8% CV - 44.48 µg/dL: 5.0% CV
Linearity/Reportable Range	Data should demonstrate linearity across the claimed measuring range.	Measuring Range: 0.59 to 45.00 µg/dL. Regression: Observed = 1.01 * Expected + 0.01 µg/dL; R²: 1.00 (Accepted based on R² close to 1 and slope ~1, intercept ~0)
Detection Limits (LoB, LoD, LoQ)	Specific quantifiable low limits.	LoB: 0.10 µg/dL LoD: 0.24 µg/dL LoQ: 0.59 µg/dL
Analytical Specificity (Interference)	Non-significant bias (<10%) for tested interfering substances at specified concentrations.	Interference (<10% bias): - Acetaminophen: 200 µg/mL - Bilirubin (Conj/Unconj): 40 mg/dL - Biotin: 6 µg/mL - Carbamazepine: 30 µg/mL - Haemoglobin: 500 mg/dL - HAMA: 1000 ng/mL - Ibuprofen: 500 µg/mL - Phenytoin: 50 µg/mL - RhF: 2000 IU/mL - Total protein: 12 g/dL - Triglycerides: 3000 mg/dL (All tested compounds found not to interfere significantly.)
Method Comparison (vs. Predicate Device)	Correlation coefficient (r) close to 1, slope close to 1, intercept close to 0, indicating strong agreement.	vs. Roche Elecsys Cortisol II (K152227): - N: 194 samples - Slope: 1.06 (95% CI: 1.04 to 1.07) - Intercept: -0.10 µg/dL (95% CI: -0.39 to 0.04) - Correlation Coefficient (r): 0.99 (Demonstrates strong agreement with predicate)
Matrix Comparison	Slopes close to 1 and intercepts close to 0, with high correlation coefficients, for various sample types vs. control.	vs. Red Top Serum: - SST (n=45): Slope 1.02, Int. -0.08, r 1.00 - K2 EDTA (n=45): Slope 1.03, Int. -0.11, r 1.00 - K3 EDTA (n=45): Slope 1.02, Int. -0.10, r 1.00 - Lithium Heparin (n=45): Slope 1.00, Int. 0.15, r 1.00 - Sodium Heparin (n=45): Slope 1.01, Int. 0.00, r 1.00 (All demonstrate strong agreement across sample types)
Stability	Demonstrates shelf-life stability.	Shelf life: 12 months (based on accelerated stability studies).

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

Precision Testing:
- Per sample/concentration level (for one representative lot on one system): N=80 replicates.
- Combined (3 lots on 3 systems): N=240 replicates per sample/concentration level.
- Provenance: Not explicitly stated, but clinical laboratory studies typically use samples from diverse patient populations to cover the measurement range. The study mentions "human serum samples" and "low level samples."
Linearity Testing:
- A high human serum sample and a low human serum sample were used, along with 12 evenly spaced dilutions.
- Provenance: "high patient sample with a low patient sample."
Detection Limit Testing (LoB, LoD, LoQ):
- 60 blank replicates and 13 low-level samples.
- Provenance: Not explicitly stated, but based on typical laboratory practices.
Analytical Specificity (Interference):
- Human serum samples with low and high cortisol concentrations.
- For Rheumatoid Factor, a serum sample with low Cortisol and high Rheumatoid Factor was diluted 1:2 and 1:4. Each was assayed in duplicate.
- For cross-reactivity, serum samples spiked with cross-reactants and unspiked controls were assayed in 26 replicates each.
Method Comparison (vs. Predicate Device):
- N: 194 samples.
- Provenance: Samples were "selected to represent a wide range of Cortisol concentrations" (0.64 to 44.66 ug/dL), implying patient samples. Country of origin not specified, but usually derived from laboratory settings.
Matrix Comparison:
- N: 45 samples (36 native, 9 spiked or diluted) per matrix type.
- Provenance: "human samples" to cover the range of 0.89 to 42.38 ug/dL.
Expected Values/Reference Range:
- N: 307 apparently healthy donors.
- Provenance: "serum samples collected from 307 apparently healthy donors" from 21 to 65 years of age. Details imply a prospective collection for this specific study, but geographical location not explicitly stated beyond "local population" for general reference ranges.

All studies mentioned state they followed CLSI (Clinical and Laboratory Standards Institute) guidelines, which implies these were prospective analytical performance studies conducted under controlled laboratory conditions.

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

For an in vitro diagnostic (IVD) device like the IDS Cortisol assay, "ground truth" for the test set is established through reference methods or highly accurate comparative assays, not typically by human expert consensus or adjudicated readings. The nature of this device is a quantitative immunoassay meant to measure a biomarker.

Ground Truth Establishment for Analytical Performance:
- Traceability: The IDS Cortisol assay is traceable to the LC-MS/MS Candidate Reference Measurement Procedure (cRMP) Total Serum Cortisol. This is a highly accurate chemical analytical method, considered a "higher-order" reference method. The ground truth in this context is the quantitative value determined by this cRMP, often involving specialized laboratories and expert analytical chemists.
- Method Comparison: The test device's performance is compared against the Roche Elecsys Cortisol II (K152227), a previously cleared and marketed predicate device. The values obtained from the predicate device serve as the comparative "truth" for demonstrating substantial equivalence.
- Precision, Linearity, Detection Limits, Analytical Specificity, Matrix Comparison: These performance characteristics are evaluated against pre-defined statistical criteria (e.g., CV thresholds, R² values, bias limits) using controlled samples, calibrators, and reference materials. The "truth" is the known concentration or behavior of these materials, verified through standard laboratory practices and reference methods.
Number of Experts & Qualifications: Not applicable in the context of human expert adjudication for image interpretation. The "experts" would be the skilled laboratory personnel and analytical chemists performing the LC-MS/MS and predicate device measurements, as well as those defining the CLSI protocols for analytical testing.

4. Adjudication Method for the Test Set

Not applicable. This is an in vitro quantitative diagnostic device, not an imaging device requiring human-in-the-loop reading and adjudication.

5. If a Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study Was Done

Not applicable. This is an in vitro quantitative diagnostic device, not an imaging device requiring human readers or MRMC studies.

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

Yes, the entire analytical performance evaluation (precision, linearity, detection limits, interference, method comparison, and matrix comparison) represents the standalone performance of the IDS Cortisol assay as an algorithm (chemical reaction + instrument measurement and calculation) without human-in-the-loop intervention for result generation. The human role is in operating the instrument and interpreting the results, but the measurement itself is automated.

7. The Type of Ground Truth Used

The primary ground truth used for this quantitative in vitro diagnostic device is:

Reference Measurement Procedure: The IDS Cortisol assay is traceable to the LC-MS/MS Candidate Reference Measurement Procedure (cRMP) Total Serum Cortisol, which represents a highly accurate and standardized method for determining cortisol concentrations. This is a form of outcomes data in the sense of a definitive analytical outcome.
Predicate Device Comparison: The values generated by the Roche Elecsys Cortisol II assay serve as a comparative ground truth for demonstrating substantial equivalence.
Known Concentrations of Reference Materials/Spiked Samples: For analytical performance studies like precision, linearity, and detection limit, the ground truth is often established by using precisely prepared calibrators or samples with known, verified concentrations of the analyte.

8. The Sample Size for the Training Set

This document describes the analytical validation of a commercial diagnostic kit, not a machine learning algorithm that typically undergoes distinct "training" datasets. Therefore, there isn't a "training set" in the sense of data used to iteratively optimize an AI model.

The "development" or "design" of the assay (reagents, methodology) would have involved extensive R&D and internal testing, which could be considered analogous to a training phase in a broader sense, but no specific sample size for such a phase is provided in this regulatory submission. The data presented are for the validation of the final device.

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

As noted in point 8, there isn't a distinct "training set" in the context of an AI/ML algorithm as described in this document. The "ground truth" for the development of the assay itself would have been established through a combination of:

Biochemical principles: Understanding the cortisol molecule and its interactions with antibodies.
Standard laboratory methods: Using established analytical techniques (like spectrophotometry, chromatography, or existing cortisol assays) to characterize reagents and ensure proper reaction kinetics.
Reference materials: Utilizing internationally recognized reference materials and calibrators for initial assay calibration and optimization.
Iterative development and testing: Repeated internal testing with various concentrations of cortisol in different matrices to optimize assay components (antibodies, conjugates, buffers) and instrument parameters to achieve desired performance characteristics (sensitivity, specificity, dynamic range). This iterative process involves comparing results against known concentrations or a "gold standard" method (like LC-MS/MS) during development.

Ask a Question

Ask a specific question about this device

Page 1 of 1