(142 days)
The RLP-Cholesterol Assay is intended for use in the quantitative determination of cholesterol contained in remnant lipoproteins in human serum or plasma. The test results are used in combination with total serum triglyceride measurements to aids in the diagnosis of familial type III hyperlipoproteinemia in patients with serum total cholesterol concentration ≥ 200 mg/dL and triglyceride concentrations between 200 and 800 mg/dL.
The RLP-Cholesterol Assay uses two mouse monoclonal antibodies (Mab) to isolate remnant lipoproteins. The first one (JI-H) is raised against human apo B-100. This Mab recognizes an epitope near the apo B-51 region to remove lipoproteins containing apo B-100, such as LDL, Lp(a) and nascent VLDL. The second one (H-12) is raised against human apo A-I. This Mab removes lipoproteins containing apo A-I, such as HDL. The monoclonal antibodies are conjugated to sepharose-4B beads and to separate bound lipoproteins from the remnant lipoproteins that remain in the unbound fraction. Cholesterol in the unbound fraction (RLP-C) is then quantified by an enzymatic assay.
Here's an analysis of the acceptance criteria and study details for the RLP-Cholesterol Immunoseparation Assay, based on the provided 510(k) notification:
Acceptance Criteria and Reported Device Performance
The acceptance criteria for the RLP-Cholesterol Assay are primarily focused on its ability to accurately diagnose Type III hyperlipoproteinemia using the RLP-C to total TG ratio, comparing its performance against a laboratory reference method (ultracentrifugation) and a predicate device (REP Kit).
Table of Acceptance Criteria and Reported Device Performance:
| Acceptance Criteria Category | Specific Metric | Predicate/Reference Method Performance | Acceptance Criteria | Reported Device Performance (RLP-Cholesterol Assay) |
|---|---|---|---|---|
| Diagnosis of Type III Hyperlipoproteinemia (Clinical Study - Type IIb, III, and IV Study) | Overall proper diagnosis rate for Type III and non-Type III subjects (N=51) | Laboratory Reference Method (Ultracentrifugation): 90.2% | At least as good as the laboratory reference method, and similar to the predicate device. | 94.1% |
| Proper diagnosis of Type III subjects (N=20) using RLP-C/TG ratio ≥ 0.10 | Reference method (VLDL-C/TG > 0.30): 16/20 (80%) | |||
| B-VLDL presence: 20/20 (100%) | Not explicitly stated as a numerical threshold, but implied to be high and consistent with reference methods. | 19/20 (95.0%) | ||
| Proper diagnosis of non-Type III subjects (N=31) using RLP-C/TG ratio 0.90) | 0.96 | |||
| Overall Agreement (Clinical Utility Study) | Overall agreement between RLP-C to TG ratio and VLDL-C to TG ratio in hyperlipidemic subjects (N=140) | Not applicable | High agreement | 98.6% |
| Analytical Performance (Nonclinical Tests) | Detection Limit | Not specified for predicate | Sufficiently low for clinical utility. | 3.1 mg/dL |
| Linear Range | Not specified for predicate | Sufficiently wide for clinical utility. | 3.1 to 90 mg/dL | |
| Within-run imprecision (at ~4.82 mg/dL) | Comparable to other devices of the same type. | Comparable to other devices of the same type. | 5.0% | |
| Within-run imprecision (at ~24.87 mg/dL) | Comparable to other devices of the same type. | Comparable to other devices of the same type. | 3.4% | |
| Between-run imprecision (at ~4.82 mg/dL) | Comparable to other devices of the same type. | Comparable to other devices of the same type. | 10.9% | |
| Between-run imprecision (at ~24.87 mg/dL) | Comparable to other devices of the same type. | Comparable to other devices of the same type. | 5.7% | |
| Interference from common substances (Hemoglobin, Bilirubin, Ascorbic Acid, Glucose) | Not specified for predicate | No significant interference. | No interference observed at specified concentrations. | |
| Cut-off Value Establishment (Reference Range Study) | 95th percentile for RLP-C to total TG ratio | Not applicable | A clinically appropriate cut-off for distinguishing Type III hyperlipoproteinemia. | 0.08 (95th percentile from healthy volunteers), adjusted to > 0.10 for clinical cut-off. |
Study Details:
-
Sample sizes used for the test set and the data provenance:
- Reference Range Study:
- Test Set Size: 729 healthy volunteers.
- Data Provenance: Not explicitly stated, but implied to be from a general population. Prospective (implied, as they "enrolled" volunteers).
- Type IIb, III and IV Study:
- Test Set Size: 62 subjects met inclusion criteria (24 Type III, 38 Type IIb or IV). For the diagnostic comparison, 51 subjects were analyzed (20 Type III, 31 non-Type III) who had total cholesterol > 200 mg/dL and TG between 200-800 mg/dL.
- Data Provenance: Conducted at two centers specializing in familial lipid disorders. Implied to be prospective, as subjects were "enrolled." Country of origin not specified, but the submission is from Japan with an official correspondent in the US, suggesting international or US-based studies.
- Clinical Utility Study:
- Test Set Size: 724 subjects in total. 140 of these were hyperlipidemic (used for agreement analysis).
- Data Provenance: Not explicitly stated but seemed to involve different subject groups (healthy, CAD, diabetes). Implied to be prospective.
- Reference Range Study:
-
Number of experts used to establish the ground truth for the test set and the qualifications of those experts:
- The document does not explicitly state the number of experts or their qualifications for establishing the ground truth.
- Ground truth for Type III hyperlipoproteinemia diagnosis was primarily established using:
- Laboratory Reference Method: Ultracentrifugation (specifically VLDL-C to total TG ratio > 0.30 and presence of ß-VLDL). This is a well-established laboratory method using instrumental analysis, not solely expert interpretation.
- Predicate Device: REP Kit (implies a comparison to its established diagnostic capability, which likely relied on visual patterns or quantitative results interpreted by laboratory personnel).
- The "two centers specializing in familial lipid disorders" in the Type IIb, III, and IV Study suggest that qualified clinicians and laboratory professionals were involved in patient selection and ground truth determination, but specific expert count or qualifications aren't provided.
-
Adjudication method for the test set:
- No specific adjudication method (e.g., 2+1, 3+1) is mentioned.
- The ground truth relied on established laboratory methodologies (ultracentrifugation, predicate device results for comparison). For the Type IIb, III, and IV Study, the types of hyperlipoproteinemia were determined by these methods.
-
If a multi reader multi case (MRMC) comparative effectiveness study was done, if so, what was the effect size of how much human readers improve with AI vs without AI assistance:
- No MRMC comparative effectiveness study was done.
- This device is an in-vitro diagnostic (IVD) assay that produces quantitative results, not an AI-powered image analysis or diagnostic assist tool for human readers. Therefore, the concept of "human readers improve with AI" does not apply here.
-
If a standalone (i.e. algorithm only without human-in-the-loop performance) was done:
- The RLP-Cholesterol Assay is inherently a "standalone" device in that it generates quantitative results (RLP-C values and RLP-C to total TG ratios) without human interpretation as part of its core function. Human clinicians then use these numerical results, in combination with other clinical findings, to make a diagnosis. The studies evaluate the analytical and diagnostic performance of the assay itself.
-
The type of ground truth used (expert consensus, pathology, outcomes data, etc.):
- Primary Ground Truth: Established laboratory reference method, specifically ultracentrifugation, to determine VLDL-C concentrations and the presence of ß-VLDL, which are the gold standard for diagnosing Type III hyperlipoproteinemia.
- Secondary Ground Truth/Comparator: Performance of a legally marketed predicate device (REP Kit).
-
The sample size for the training set:
- The document implies that there was no distinct "training set" in the context of machine learning. This is an immunoassay device, not an AI/ML algorithm.
- The "Reference Range Study" with 729 healthy volunteers helped establish a clinical cut-off value, which could be considered an initial calibration or parameter setting. The "Type IIb, III and IV Study" and "Clinical Utility Study" served as validation/test sets to demonstrate performance.
-
How the ground truth for the training set was established:
- As there was no explicit "training set" for an AI/ML algorithm, this question isn't directly applicable.
- For the "Reference Range Study," the 95th percentile of the RLP-C to total TG ratio from a large cohort of healthy volunteers was used to establish an initial cut-off. This is a statistical approach based on a healthy population, not a "ground truth" derived from expert labeling for individual cases. The diagnosis of hyperlipoproteinemia types in the Type IIb, III, and IV Study was based on the ultracentrifugation method.
§ 862.1175 Cholesterol (total) test system.
(a)
Identification. A cholesterol (total) test system is a device intended to measure cholesterol in plasma and serum. Cholesterol measurements are used in the diagnosis and treatment of disorders involving excess cholesterol in the blood and lipid and lipoprotein metabolism disorders.(b)
Classification. Class I (general controls). The device is exempt from the premarket notification procedures in subpart E of part 807 of this chapter subject to § 862.9.