(105 days)
The Access SHBG assay is a paramagnetic particle, chemiluminescent immunoassay for the quantitative determination of Sex Hormone Binding Globulin levels in human serum and plasma using the Access Immunoassay Systems. The Access SHBG assay is indicated for use in the assessment of androgen disorders.
The Access SHBG assay is a sequential two-site immunoenzymatic ("sandwich") assay. The Access SHBG assay consists of the reagent pack, calibrators and QCs. Other items needed to run the assay include substrate and wash buffer. The Access SHBG assay reagent pack, Access SHBG assay calibrators, Access SHBG QCs, along with the UniCel Dxl Wash Buffer II are designed for use with the Dxl 9000 Access Immunoassay Analyzer in a clinical laboratory setting.
The FDA 510(k) clearance letter and accompanying summary describe the Acceptance Criteria and the study proving the Access SHBG assay meets these criteria.
Acceptance Criteria and Device Performance
The provided document focuses on the analytical performance of the Access SHBG assay on the Dxl 9000 Access Immunoassay Analyzer, comparing it to a previously cleared predicate device (Access SHBG on Access 2 instrument). The acceptance criteria are largely derived from CLSI (Clinical and Laboratory Standards Institute) guidelines for various analytical performance characteristics.
Here is a table summarizing the acceptance criteria and the reported device performance for the analytical studies:
| Performance Characteristic | Acceptance Criteria (from CLSI guidelines or internal design) | Reported Device Performance (Access SHBG on Dxl 9000) |
|---|---|---|
| Method Comparison (compared to predicate) | R$^2$ $\ge$ 0.95 with a slope equal to 1.00 ± 0.09 | R = 1.00, R$^2$ = 0.99, Slope = 1.01 (95% CI: 1.00-1.03), Intercept = -0.019 (95% CI: -0.46 - 0.29) |
| Precision (Within-Laboratory Imprecision) | $\le$ 0.14 nmol/L SD at concentrations $\le$ 2 nmol/L | |
| $\le$ 7.0% CV at concentrations > 2 nmol/L | Sample 1 (0.82 nmol/L): SD = 0.04, CV = 4.6% | |
| Sample 2 (18 nmol/L): SD = 0.5, CV = 2.7% | ||
| Sample 3 (47 nmol/L): SD = 1.3, CV = 2.7% | ||
| Sample 4 (90 nmol/L): SD = 2.6, CV = 2.9% | ||
| Sample 5 (198 nmol/L): SD = 5.1, CV = 2.6% | ||
| Linearity | The assay must demonstrate linearity throughout its analytical measuring interval. | Linear throughout the analytical measuring interval of 0.33 nmol/L - 200 nmol/L. |
| Limit of Blank (LoB) | Not explicitly stated as a number, but derived from the study. | 0.005 nmol/L (maximum observed) |
| Limit of Detection (LoD) | Not explicitly stated as a number, but derived from the study. | 0.01 nmol/L |
| Limit of Quantitation (LoQ) | Not explicitly stated as a number, but derived from the study. | 0.06 nmol/L |
Study Details
The provided document describes analytical verification studies for the Access SHBG assay, not a clinical study involving human readers or AI. Therefore, some of the requested information (e.g., MRMC studies, human reader improvement with AI, ground truth for training AI models, number of experts for AI ground truth) is not applicable or cannot be extracted from this document as it pertains to an immunoassay device, not an AI/ML-based diagnostic system.
Here's the relevant information that can be extracted:
2. Sample Size and Data Provenance:
- Method Comparison: A total of 151 samples were evaluated. The data provenance is implied to be clinical samples (patient samples are mentioned in the CLSI guideline for method comparison), though specific country of origin or whether they were retrospective/prospective is not stated. Given the context of a 510(k) submission for an in-vitro diagnostic, it is highly likely these were de-identified retrospective clinical samples.
- Precision: Five serum samples with varying SHBG concentrations were used. Each sample was assayed in duplicate with two runs per day, over 20 days, on three Dxl 9000 Access Immunoassay Analyzer systems, three reagent lots, and three calibrator lots. This totals 80 measurements per sample (5 samples * 2 duplicates * 2 runs/day * 20 days / 1 instrument * 1 reagent lot * 1 calibrator lot used for reported results).
- Linearity: A native low sample and a spiked high sample were used, along with seven mixtures in between.
- LoB, LoD, and LoQ: Four distinct blank samples were used for LoB. Six to seven samples were used for LoD. 12-13 serum samples were used for LoQ.
3. Number of Experts and Qualifications for Test Set Ground Truth:
- This is an immunoassay device, not an image-based AI system. The ground truth for analytical studies like linearity, precision, and limits of detection is established by the assay itself (measurements of known concentrations, serial dilutions, etc.) and validated against established analytical method guidelines (CLSI). No human experts are involved in establishing ground truth for these analytical performance characteristics in the way they would be for image interpretation tasks.
4. Adjudication Method for Test Set:
- Not applicable. Adjudication methods (e.g., 2+1, 3+1) are typically used in clinical studies where human readers provide interpretations (e.g., radiology reads) that need to be reconciled to form a ground truth. For analytical performance studies of an immunoassay, the "truth" is based on the chemical and instrument measurements and statistical analysis against predefined acceptance criteria.
5. Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study:
- No, an MRMC comparative effectiveness study was not done. This type of study is relevant for AI systems assisting human readers in diagnostic tasks, such as radiology image interpretation. This document describes the analytical performance of an immunoassay for quantitative determination of SHBG levels, which does not involve human readers interpreting AI output.
6. Standalone Performance:
- Yes, the entire document focuses on the "standalone" analytical performance of the Access SHBG assay on the Dxl 9000 Access Immunoassay Analyzer. The performance described (method comparison, precision, linearity, limits) is the direct output of the instrument and reagents, without human interpretation "in the loop" beyond standard laboratory procedures for operating the instrument.
7. Type of Ground Truth Used:
- For analytical performance studies, the "ground truth" is established through:
- Reference measurements/Predicate device: For method comparison, the results from the previously cleared predicate device (Access SHBG on Access 2 instrument) served as the comparative "truth."
- Known concentrations/mixtures: For linearity, samples with known or precisely diluted concentrations were used.
- Replicate measurements: For precision, repeated measurements establish the variability.
- Blank and low-level samples: For LoB, LoD, and LoQ, these are the "truth" against which the assay's detection capabilities are measured.
- CLSI Guidelines: The studies were designed and evaluated according to CLSI guidelines, which represent a form of accepted scientific and statistical "truth" for validating laboratory assays.
8. Sample Size for Training Set:
- This document describes the validation of an immunoassay device, not an AI/ML model. Therefore, there isn't a "training set" in the context of machine learning. The device's underlying chemistry and physics are "trained" during its development and manufacturing process, but not through a data-driven training set in the way an AI algorithm is.
9. How Ground Truth for Training Set was Established:
- Not applicable, as there is no AI/ML training set in the context of this immunoassay device.
§ 862.1680 Testosterone test system.
(a)
Identification. A testosterone test system is a device intended to measure testosterone (a male sex hormone) in serum, plasma, and urine. Measurement of testosterone are used in the diagnosis and treatment of disorders involving the male sex hormones (androgens), including primary and secondary hypogonadism, delayed or precocious puberty, impotence in males and, in females hirsutism (excessive hair) and virilization (masculinization) due to tumors, polycystic ovaries, and adrenogenital syndromes.(b)
Classification. Class I.