(230 days)
Lumipulse G CA125II Immunoreaction Cartridges For in vitro diagnostic use. Lumipulse G CA125II is a Chemiluminescent Enzyme Immunoassay (CLEIA) for the quantitative determination of CA125 in human serum and plasma (sodium heparin, 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 ovarian cancer. Serial testing for patient CA125 assay values should be used in conjunction with other clinical methods used for monitoring ovarian cancer. Lumipulse G CA125II Calibrators Lumipulse G CA125II Calibrators are for use in the calibration of the LUMIPULSE G System for the quantitative measurement of CA125 in human serum or plasma (sodium heparin, or dipotassium EDTA). LUMIPULSE G1200 System LUMIPULSE G1200 is intended for in vitro diagnostics use, and is designed to perform automated chemiluminescence immunoassays of specimens using Lumipulse G reagents, conducting various processes such as dispensing, agitation and photometric measurement.
Lumipulse G CA125Il is an assay system, including a set of immunoassay reagents, for the quantitative measurement of CA125 in specimens based on CLEIA technology by a two-step sandwich immunoassay method on the LUMIPULSE G System. Lumipulse G CA125II Immunoreaction Cartridges consists of 3 x 14 tests. Each kit contains Antibody-Coated Particle Solution and Enzyme-Labeled Antibody Solution. Lumipulse G CA125II Calibrators Each calibrator kit contains one bottle each of Calibrators 1 and 2. The calibrator kit is packaged separately. LUMIPULSE G1200 System LUMIPULSE G1200 is intended for in vitro diagnostics use, and is designed to perform automated chemiluminescence immunoassays of specimens using Lumipulse G reagents, conducting various processes such as dispensing, agitation, and photometric measurement. The chemiluminescent enzvme immunoassay system is carried out using the ferrite particle coated with antigen or antibody and conjugate with alkaline phosphatase and the chemical luminescent substrate. The luminescence which is produced by the chemiluminescent enzyme immunoassay is measured by photometric detector.
The provided text describes the Lumipulse G CA125II Immunoreaction Cartridges, Lumipulse G CA125II Calibrators, and LUMIPULSE G1200 System, which are used for the quantitative determination of CA125 in human serum and plasma to aid in monitoring recurrence or progressive disease in patients with ovarian cancer.
Here's an analysis of the acceptance criteria and study data provided:
1. A table of acceptance criteria and the reported device performance
Unfortunately, the document does not explicitly state pre-defined "acceptance criteria" in a structured table. However, performance goals can be inferred from the study results and general clinical practice for such assays. The study's conclusion states that the assay "is substantially equivalent to the performance of the Siemens ADVIA Centaur CA 125II assay," implying that meeting the performance of the predicate device serves as an acceptance criterion.
Based on the provided performance characteristics, here's a reconstructed table with inferred performance goals and reported performance:
| Performance Characteristic | Inferred Performance Goal (Acceptance Criteria) | Reported Device Performance (Lumipulse G CA125II) |
|---|---|---|
| Precision (Within-laboratory Total %CV) | ≤ 10% (as stated in section 1.a) | ≤ 2.6% |
| Precision (Site-to-site Total %CV) | Not explicitly stated, but clinical acceptability implied by study. | ≤ 6.4% |
| Precision (Lot-to-lot Total %CV) | Not explicitly stated, but clinical acceptability implied by study. | ≤ 5.2% |
| Linearity/Reportable Range | Demonstrates linearity across the intended measurement range (2.5 to 1000 U/mL) | Demonstrated linearity from 2.5 to 1000 U/mL (R-squared: 0.9999 for Serum, 0.9996 for Plasma) |
| Recovery | 100 ± 15% (as stated in section on recovery) | 97% to 115% |
| High Dose Hook Effect | No high dose hook effect within the expected range of clinical samples. | No effect observed up to 200,000 U/mL. |
| Limit of Blank (LoB) | Not explicitly stated as acceptance criterion, but low value is desirable. | 0.1 U/mL |
| Limit of Detection (LoD) | ≤ 2.0 U/mL (as stated in detection limit section) | 0.5 U/mL |
| Limit of Quantitation (LoQ) | ≤ 2.0 U/mL (as stated in detection limit section) | 0.5 U/mL |
| Analytical Specificity (Interference) | Average interference ≤ 10% (as stated in section 1.e) | Average interference ≤ 10% for listed endogenous and therapeutic compounds |
| Method Comparison (Correlation with Predicate) | High correlation (r) and acceptable bias compared to predicate device. | For 102 samples: r=0.9745, Slope=1.13, Average Bias=25.6 U/mL |
| For 120 samples: r=0.9829, Slope=1.08, Average Bias=18.2 U/mL | ||
| Matrix Comparison (Slope vs. Control) | 95% CI for slope within 0.9 to 1.1, and correlation coefficients ≥ 0.9. | Slopes for each tube type had 95% CI entirely within 0.9 to 1.1; correlation coefficients ≥ 0.9. |
| Clinical Performance (Sensitivity in Monitoring) | Not explicitly stated a priori as acceptance criterion, but demonstrated effectiveness is key. | 67.3% (95% CI: 29.9-100.0) |
| Clinical Performance (Specificity in Monitoring) | Not explicitly stated a priori as acceptance criterion, but demonstrated effectiveness is key. | 76.0% (95% CI: 61.0-90.9) |
| Clinical Performance (Total Concordance in Monitoring) | Not explicitly stated a priori as acceptance criterion, but demonstrated effectiveness is key. | 74.4% (95% CI: 60.9-87.9) |
| Clinical Performance (AUC for progression diagnosis) | Statistically significantly better than no-association (0.5 AUC). | 0.728 (SE 0.047), highly statistically significant. |
2. Sample size used for the test set and the data provenance (e.g. country of origin of the data, retrospective or prospective)
- Precision/Reproducibility (Within-laboratory): 8 human serum-based samples (specimen pools) and 2 commercially available serum-based controls. Each sample tested in replicates of two at two separate times of the day for 20 days (n=80 for each sample).
- Precision (Site-to-site): Human serum-based samples (specimen pools) and 2 commercially available serum-based controls. Each sample tested in replicates of two at two separate times of the day at each of 3 external laboratory sites for 10 days (n=40 for each sample). The document does not specify the country of origin or whether the data was retrospective or prospective.
- Precision (Lot-to-lot): Human serum-based samples (specimen pools) and 2 commercially available serum-based controls. Each sample tested in replicates of two at two separate times of the day for each of 3 lots for 10 days (n=40 for each sample).
- Linearity/Assay Reportable Range: One human serum specimen pool and one K2 EDTA plasma specimen pool with high CA125 levels were diluted with low CA125 levels. Specific N for individual samples is not given, but the study method is described.
- Recovery: Not specified, but involved human serum and K2 EDTA plasma samples.
- Limit of Detection (LoD): Seven low-level specimens were tested over 3 days using two LUMIPULSE G1200 Systems and two Lumipulse G CA125II lots, resulting in 120 determinations for each panel.
- Analytical Specificity (Interference): Human serum and K2 EDTA plasma specimen pools supplemented with potentially interfering compounds.
- Method Comparison:
- 102 samples (range 15.5-822.2 U/mL) for direct comparison.
- 120 samples (range 15.5-1128.8 U/mL) including those requiring dilution.
The document does not specify the country of origin or whether the data was retrospective or prospective for these samples.
- Matrix Comparison: Not specified, but involved different tube types (SST, K2EDTA, Lithium Heparin, and Sodium Heparin) versus red top serum samples.
- Clinical Studies (Monitoring of Disease Status): 59 patients, with a total of 289 pairs of observations (average of 5.9 observations per patient). The data provenance (country, retrospective/prospective) is not specified.
- Expected values/Reference range:
- Healthy premenopausal women: N=120
- Healthy postmenopausal women: N=120
- Benign Gynecological Disease: N=260
- Other Benign Disease: N=40
- Congestive Heart Failure: N=40
- Hypertension: N=40
- Pregnant: N=40
- Ovarian Cancer: N=105
- Bladder Cancer: N=40
- Breast Cancer: N=40
- Endometrial Cancer: N=40
- GI Cancer: N=40
- Lung Cancer: N=40
The data provenance (country, retrospective/prospective) is not specified.
3. Number of experts used to establish the ground truth for the test set and the qualifications of those experts (e.g. radiologist with 10 years of experience)
For the "Monitoring of Disease Status in Patients Diagnosed with Ovarian Cancer" study, the ground truth ("disease progression") was determined by comparing "changes in CA125 levels in serial serum samples from 59 patients compared to changes in disease status." The document states that "Serial testing for patient CA125 assay values should be used in conjunction with other clinical methods used for monitoring ovarian cancer." However, it does not specify the number of experts or their qualifications who established the "disease status" ground truth (i.e., whether the disease had progressed or not, independently of the CA125 reading).
4. Adjudication method (e.g. 2+1, 3+1, none) for the test set
The document does not describe an adjudication method for establishing the ground truth for the clinical study. It implies that "disease status" was determined by "other clinical methods," but how conflicts or uncertainties in those methods were resolved is not detailed.
5. 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, a multi-reader multi-case (MRMC) comparative effectiveness study was not conducted. This device is an automated in vitro diagnostic assay, not an AI-assisted diagnostic tool that would involve human readers interpreting results. Its purpose is to quantify CA125 levels, and the clinical study evaluates its performance in monitoring disease status based on changes in these quantitative values, not reader performance.
6. If a standalone (i.e. algorithm only without human-in-the-loop performance) was done
Yes, the studies presented are effectively standalone performance evaluations of the Lumipulse G CA125II assay and the LUMIPULSE G1200 System. The precision, linearity, recovery, detection limits, analytical specificity, and method comparison studies evaluate the algorithm's (the assay's) performance in quantifying CA125 levels. The clinical monitoring study also evaluates the assay's ability to track disease status based on its quantitative output, independent of human interpretive reading of images or complex data (beyond simple interpretation of CA125 value changes).
7. The type of ground truth used (expert consensus, pathology, outcomes data, etc)
For the clinical monitoring study, the ground truth for "disease progression" was based on "changes in disease status," which is referred to as being determined by "other clinical methods used for monitoring ovarian cancer." This suggests a composite clinical ground truth, likely involving a combination of imaging, clinical assessment, and potentially other biomarkers or pathology, but the specific details are not provided. It is not explicitly stated as pathology or expert consensus, but rather a broader "clinical methods."
8. The sample size for the training set
The document does not describe a "training set" in the context of machine learning. The studies described are performance validation studies for an in vitro diagnostic device, which typically involves analytical validation and clinical validation using a distinct set of samples. The assay itself is a Chemiluminescent Enzyme Immunoassay (CLEIA), a biochemical measurement system, not a machine learning algorithm that requires a separate training set.
9. How the ground truth for the training set was established
As there is no "training set" for a machine learning algorithm, this question is not applicable to the provided document. The ground truth for the reference calibrators for the assay itself is established via traceability to "Fujirebio Diagnostics maintained reference preparation" and correlation to "Fujirebio Diagnostics' CA125II RIA."
§ 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.