Search Results

The Wako AFP-L3% assay is intended as a risk assessment test for the development of hepatocellular carcinoma (HCC) in patients with chronic liver diseases (CLD). Elevated AFPL3% values (≥ 10%) have been shown to be associated with a seven-fold increase in the risk of developing HCC within the next 21 months. Patients with elevated serum AFPL3% should be more intensely evaluated for evidence of HCC according to the existing HCC practice guidelines in oncology.

The Wako AFP-L3% device consists of reagent 1 (LCA and anion 1-conjugated anti-AFP mouse monoclonal antibody), reagent 2 (horseradish peroxidase (POD)labeled anti-AFP mouse monoclonal antibody and anion 2 conjugated anti-AFP mouse monoclonal antibody, substrate 1 (4 acetamidophenol in 2-propanol) and substrate 2 (hydrogen peroxide) and a column. Reagent 1, reagent 2 and the column are ready-to-use. Elution buffers A to C, sample cups, inside and outside cuvettes are sold separately from kit.

The Wako AFP-L3 Calibrator set and Control set are sold separately. The calibrator set consisted of Calibrator 1 and 2. Calibrator 1 contains human AFP -L1 fraction and Calibrator 2 has human AFP-L3 fraction. The control set consisted of Control 1 and 2, each containing different concentrations of human AFP-L1 and L-3.

Here's an analysis of the provided 510(k) summary, extracting the requested information about acceptance criteria and the supporting study:

Acceptance Criteria and Device Performance for Wako LBA AFP-L3

1. Table of Acceptance Criteria and Reported Device Performance

The 510(k) summary provided does not explicitly define acceptance criteria in a structured table for each performance characteristic. However, it presents measured performance values and, for linearity, implicitly sets criteria within the reported ranges for slope and R². For clinical performance, the intended use statement sets a clear threshold for risk assessment.

Here's an inferred table based on the provided data:

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

The primary clinical study involved 494 evaluable patients from:

• 6 US clinical sites (Lahey (MA), MCV (VA), Miami (FL), Mt. Sinai (NY), UCSF (SF), UPenn (PA))
• 1 Canadian clinical site (Toronto (Canada))

The study was a double-blind, multi-site prospective study. Serum samples were collected and stored frozen, and AFP-L3% tests were performed retrospectively by Wako Chemicals USA, Inc.

For the relative risk calculation, 57 patients from Group 4 (No HCC) were excluded due to less than 21 months of follow-up, resulting in a dataset of 312 patients (39 HCC, 273 No HCC) being used for the primary relative risk calculation.

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

The summary does not explicitly state the number or specific qualifications of experts used to establish the ground truth for the clinical study. However, the diagnosis of HCC for all enrolled patients was made by "at least one or a combination of the following observations," indicating clinical judgment by physician investigators:

• HCC result on a liver biopsy
• Enlarging mass by imaging (ultrasound, CT, MRI) with elevated serum total AFP
• Enlarging mass by CT or MRI in the setting of cirrhosis
• Very high serum total AFP (>400-500 ng/mL) alone
• At least 3 serial blood draws showing rising serum AFP in the setting of a liver mass
• Mass on CT scan enhancing in arterial phase and hypoattenuating in venous phase.

For evaluable subjects without HCC at enrollment, they were categorized by "the physician investigators" based on biopsy, explanted liver histology, and imaging results. This implies that the ground truth was established by the clinical team at each site.

4. Adjudication Method for the Test Set

The adjudication method for the test set is not explicitly described as a formal numerical system (e.g., 2+1, 3+1). Instead, the diagnosis of HCC was based on a combination of clinical observations and confirmatory tests, as determined by the "physician investigators." This suggests a consensus-based or standard clinical practice approach at each site, rather than a centralized, predefined adjudication panel for the study.

5. Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study

No MRMC comparative effectiveness study was done. This device is a quantitative immunoassay for a biomarker (AFP-L3%) and is not an imaging device or AI-assisted diagnostic tool that would typically involve human readers. The study focuses on the standalone performance of the test in assessing HCC risk.

6. Standalone Performance (Algorithm Only Without Human-in-the-Loop)

Yes, a standalone performance study was done. The entire clinical study described is a standalone evaluation of the Wako AFP-L3% assay. The device measures AFP-L3% in serum, and its performance is evaluated based on its correlation with subsequent HCC development (risk assessment). Human intervention in the diagnostic process occurred after the test result was obtained, not as part of the test interpretation itself. The AFP-L3% values and the 10% cutoff are directly used to stratify patient risk.

7. Type of Ground Truth Used

The ground truth for the clinical study was primarily established through clinical diagnosis based on a combination of pathology (liver biopsy, explanted liver histology), imaging results (CT, MRI, ultrasound), and serial biomarker measurements (total AFP), as determined by physician investigators and standard oncology practice guidelines. This represents a robust clinical ground truth.

8. Sample Size for the Training Set

The 510(k) summary does not specify a separate training set for the clinical performance evaluation. The clinical study described appears to be the primary validation study (test set) for the device's intended use claim. For the analytical performance (precision, linearity, etc.), specific samples (control sera, spiked samples, diluted samples) were used, but these are not referred to as a "training set" in the context of machine learning model development. Given this is an immunoassay, the "training" aspect is more about assay optimization and standard curve generation during assay development than a machine learning training phase.

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

As no dedicated "training set" is identified in the context of a machine learning algorithm, this question is not directly applicable. For the analytical studies, the ground truth was inherently established by:

• Known concentrations in spiked samples.
• Reference materials (e.g., 1st International standard for AFP from WHO) for calibrator and control value assignments.
• Manufacturer-defined preparations for control sera (e.g., purified AFP-L1 and L-3 spiked into pooled serum).

Ask a specific question about this device