(186 days)
Lumipulse G HE4 Immunoreaction Cartridges:
For in vitro diagnostic use.
Lumipulse G HE4 is a Chemiluminescent Enzyme Immunoassay (CLEIA) for the quantitative determination of HE4 in human serum and plasma (lithium heparin or dipotassium EDTA) on the Lumipulse G System.
The assay is to be used as an aid in monitoring recurrence or progressive disease in patients with epithelial ovarian cancer. Serial testing for patient HE4 values should be used in conjunction with other clinical methods used for monitoring epithelial ovarian cancer.
Lumipulse G HE4 Calibrators:
Lumipulse G HE4 Calibrators are for use in the calibration of the Lumipulse G System for the quantitative measurement of HE4 in human serum and plasma (lithium heparin or dipotassium EDTA).
The Lumipulse G HE4 is an assay system, including a set of immunoassay reagents, for the quantitative measurement of HE4 in specimens based on CLEIA technology by a two-step sandwich immunoassay method on the LUMIPULSE G System.
The provided document describes the analytical and clinical performance of the Lumipulse G HE4 Immunoassay for monitoring recurrence or progressive disease in patients with epithelial ovarian cancer.
Here's an analysis of the acceptance criteria and the study that proves the device meets them:
1. Table of Acceptance Criteria and Reported Device Performance
The acceptance criteria for this device are primarily demonstrated through its analytical performance, comparison to a predicate device, and clinical concordance in monitoring disease progression. While explicit "acceptance criteria" are not listed in a separate section, they are implicitly defined by the satisfactory results presented in the analytical and clinical studies.
| Category | Acceptance Criteria (Implicit) | Reported Device Performance |
|---|---|---|
| Precision | Total (within-laboratory) CV ≤10% (as stated by the manufacturer for Lumipulse G HE4). Excellent inter-laboratory and lot-to-lot precision. | Within-laboratory: Total %CV ≤3.5% (observed). |
| Serum 1: 1.9%, Serum 2: 1.9%, Serum 4: 1.6%, Serum 5: 2.3%, Serum 6: 2.1%, Serum 7: 3.5%, Serum 8: 1.4%, Serum 9: 2.2%, Control 1: 2.4%, Control 2: 1.8%. |
External Laboratory (Inter-site) Precision: Total %CV ≤6.1% (observed).
Serum 1: 3.8%, Serum 2: 3.7%, Serum 4: 4.2%, Serum 5: 5.1%, Serum 6: 5.7%, Serum 7: 6.1%, Serum 8: 5.4%, Serum 9: 5.2%, Control 1: 5.8%, Control 2: 4.0%.
Lot-to-Lot Precision: Total %CV ≤3.2% (observed).
Serum 1: 2.2%, Serum 2: 2.3%, Serum 4: 2.3%, Serum 5: 3.2%, Serum 6: 2.2%, Serum 7: 2.7%, Serum 8: 2.2%, Serum 9: 2.5%, Control 1: 2.3%, Control 2: 1.6%. |
| Linearity/Range | Demonstrated linearity across the assay range (20.0 to 1500.0 pM). Recovery 100 ± 10%. | Linearity: Found in the range of 20.0 to 1500.0 pM for both serum and plasma. R-squared = 1.0000 for linear regression.
Recovery: Ranged from 91% to 107% (within 100 ± 10%). |
| Detection Limits | LoB, LoD, and LoQ ≤ 20.0 pM (as stated by the manufacturer). | LoB: 0.1 pM.
LoD: 3.5 pM.
LoQ: 3.5 pM (since percent total error estimate is ≤ 30%, LoQ equals LoD). |
| Analytical Specificity/Interference | Average interference ≤ 10% for potentially interfering compounds. | Demonstrated average interference ≤ 10% for all tested endogenous (Bilirubin, Triglycerides, Hemoglobin, Human Serum Albumin, Immunoglobulin G, Biotin, HAMA, RF) and therapeutic drugs (Bevacizumab, Carboplatin, Cisplatin, Clotrimazole, Cyclophosphamide, Dexamethasone, Docetaxel, Doxil®, Doxorubicin, Erlotinib, Etoposide, 5-Fluorouracil, Gemcitabine, Leucovorin, Magestrol Acetate, Melphalan, Methotrexate, Olaparib, Paclitaxel, Rituximab, Tamoxifen, Topotecan, Trastuzumab). No high dose hook effect observed up to 300,000 pM. |
| Method Comparison | Substantial equivalence to predicate device (Fujirebio Diagnostics, Inc. HE4 EIA) indicated by strong correlation and acceptable bias. | Correlation Coefficient (r): 0.9891 (143 samples, range 35.2-969.5 pM) and 0.9917 (168 samples, range 35.2-4602.0 pM).
Slope: 1.0349 (95% CI: 1.0074-1.0624) and 1.0289 (95% CI: 1.0045-1.0533).
Average Bias: 8.8 pM and 13.8 pM. |
| Clinical Concordance | Reasonable sensitivity and specificity for monitoring disease progression/recurrence in epithelial ovarian cancer patients alongside clinical methods. | Sensitivity: 49.2% (95% CI: 27.8-70.6%).
Specificity: 79.6% (95% CI: 65.9-93.2%).
Total Concordance: 73.9% (95% CI: 62.6-85.3%).
Positive Predictive Value (PPV): 35.3% (95% CI: 19.9-50.7%).
Negative Predictive Value (NPV): 87.4% (95% CI: 72.3-102.4%). |
2. Sample Sizes Used for the Test Set and Data Provenance
-
Test Set (Clinical Study):
- Sample Size: 72 patients with epithelial ovarian cancer, providing a total of 330 pairs of observations (serial samples). The average number of observations per patient was 5.6.
- Data Provenance: Not explicitly stated regarding country of origin or specific institutions. The nature of the study (monitoring recurrence/progression) implies a retrospective collection of serial samples from patients already diagnosed, or a prospective collection over time. Given the context of a 510(k) submission, it is common for such clinical performance studies to be retrospective using existing clinical samples.
-
Test Set (Analytical Studies - Precision, Linearity, Detection Limits, Interference):
-
Precision (within-lab): 8 human serum-based samples (specimen pools) and 2 commercially available serum-based controls. Each assayed in replicates of two, twice a day for 20 days (n=80 for each sample).
-
Precision (external lab): Human serum-based samples (specimen pools) and 2 commercially available controls. Assayed in replicates of two, twice a day for 10 days at 3 external sites (n=40 for each sample per site, total n=120 per sample across sites).
-
Precision (lot-to-lot): Human serum-based samples (specimen pools) and 2 commercially available controls. Assayed in replicates of two, twice a day for 10 days with 3 lots of cartridges and calibrators (n=40 for each sample per lot, total n=120 per sample across lots).
-
Linearity/Recovery: Human serum and dipotassium EDTA plasma samples (pools). The document doesn't specify the exact number of unique "serum/plasma samples" but rather "specimen pools" and subsequent dilutions/additions. Tables show 3 serum and 3 plasma samples assessed for recovery, each with 6 target HE4 levels.
-
Detection Limits (LoB, LoD, LoQ): Seven low-level specimens were tested over 3 days using three LUMIPULSE G Systems and three Lumipulse G HE4 lots, yielding 180 determinations for each panel.
-
Analytical Specificity (Interference): Human serum specimens supplemented with potentially interfering compounds. The exact number of unique "specimens" is not given, but a long list of compounds tested.
-
Method Comparison: 143 samples for the common measuring range, and 168 samples including those requiring dilution.
-
Matrix Comparison: Not specified, but involved comparing SST, K2EDTA, and Lithium Heparin tube types against Red top serum.
-
Data Provenance (Analytical): Not explicitly stated, but typically these are laboratory-generated studies using human donor samples, commercial controls, or spiked samples. The CLSI guidelines followed (EP5-A3, EP6-A, EP7-A2, EP9-A3, EP17-A2) are standard for in vitro diagnostic device validation, implying controlled laboratory conditions.
-
3. Number of Experts Used to Establish the Ground Truth for the Test Set and Their Qualifications
- For the clinical study (monitoring disease status): The ground truth was defined by "changes in disease status," which is a clinical outcome. The document states: "A positive change in HE4 was defined as an increase in the value that was at least 18% greater than the previous value of the test. This level of change takes into account the variability of the assay. Forty-nine percent (49%) or 30/61 of the patient samples with a positive change correlated with the disease progression while eighty percent (80%) or 214/269 of the patient serial samples with no significant change in HE4 value correlated with no progression."
- This implies that the "disease progression" ground truth was established by independent clinical assessments, likely by treating physicians or oncology teams using standard clinical methods (imaging, physician assessment, other biomarkers, symptoms).
- The document does not specify the number of experts or their specific qualifications (e.g., "Radiologist with 10 years of experience"). This is typical for IVD submissions where the ground truth is derived from standard clinical practice rather than a specific expert consensus panel for the study.
4. Adjudication Method for the Test Set
- For the clinical study: Not explicitly stated as a formal adjudication process involving multiple independent reviewers in the sense of medical imaging reader studies. The ground truth for "disease progression" appears to be derived from the patient's overall clinical status as assessed by their primary care or oncology team. The "correlation" mentions comparing HE4 changes with observed disease progression/no progression. There's no indication of multiple independent adjudicators for the disease status itself within the context of this study.
5. Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study
- No, an MRMC comparative effectiveness study was not done. This is an in vitro diagnostic (IVD) device, specifically an immunoassay for measuring a biomarker (HE4). MRMC studies are typically performed for imaging devices (e.g., AI for radiology) where human readers (e.g., radiologists) interpret images with and without AI assistance to measure improvement in diagnostic performance. This submission focuses on the analytical and clinical performance of a quantitative laboratory test.
6. Standalone Performance
- Yes, standalone performance was done. The entire analytical performance section (Precision, Linearity, Detection Limits, Analytical Specificity) describes the performance of the Lumipulse G HE4 assay as a standalone system (Lumipulse G System + HE4 kit).
- The clinical study also reports the performance characteristics (sensitivity, specificity, concordance) of the HE4 assay itself in correlating with disease progression, essentially its standalone diagnostic aid performance. The indication for use clearly states it is an "aid in monitoring recurrence or progressive disease," implying its utility as a standalone test in this context, albeit "in conjunction with other clinical methods."
7. Type of Ground Truth Used
- Analytical Studies:
- Known concentrations: For linearity, recovery, and detection limits, the ground truth is established by preparing samples with known concentrations of HE4.
- Reference materials/specimen pools: For precision and interference, ground truth relies on the characteristics of the prepared specimen pools or commercial controls.
- Clinical Study:
- Clinical Outcomes/Disease Status: The ground truth for the clinical study was the "change in disease status" (progression or no progression). This is derived from an unstated combination of "other clinical methods used for monitoring epithelial ovarian cancer" as applied by the treating clinicians. This likely includes imaging, physical exams, symptoms, and potentially other biomarker trends, but the document does not specify.
8. Sample Size for the Training Set
- The document describes a 510(k) submission for an in vitro diagnostic assay, not an AI/machine learning algorithm that requires a "training set" in the computational sense.
- For an immunoassay, the equivalent of "training" would be the initial development and optimization phase, which uses various characterized samples (e.g., patient samples, spiked solutions, controls) to establish the assay's operational parameters, calibrate it, and define its intended performance characteristics before formal validation studies. The exact sample sizes for this development phase are typically not detailed in a 510(k) summary. The studies presented are validation studies (test sets) for regulatory approval.
9. How the Ground Truth for the Training Set Was Established
- As above, the concept of a computational "training set" ground truth doesn't directly apply here. For the development and initial calibration of the immunoassay:
- Ground truth for calibrators is established via a hierarchical process traceable to an in-house reference preparation, which correlates to the predicate device (HE4 EIA). This is a metrological traceability chain, not an expert panel.
- Ground truth for developing the assay's measurement range, precision, and specificity would involve using characterized human samples (e.g., samples from healthy individuals, individuals with known HE4 levels, and individuals with confounding conditions) and spiked samples (where known amounts of HE4 are added to a matrix). The "ground truth" (i.e., the target value) for such samples is determined by precise laboratory methods, often against a reference standard or by the known amount of analyte added.
§ 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.