(147 days)
The QMS Plazomicin Immunoassay is intended for the quantitative determination of plazomicin in human K2-EDTA plasma on automated clinical chemistry analyzers. The assay results obtained should only be used as an aid in the management of patients with complicated urinary tract infection (cUTI) receiving plazomicin therapy.
The assay should only be used in conjunction with information available from clinical evaluations and other diagnostic procedures.
The OMS Plazomicin Immunoassay system is a homogeneous assay utilizing particle agglutination technology and it is based on the competitive binding principle. The assay consists of liquid ready-to-use reagents R1 (anti-plazomicin mouse monoclonal antibody) and R2 (plazomicin-coated microparticles).
Here's a breakdown of the acceptance criteria and the study proving the device's performance, based on the provided document:
1. Table of Acceptance Criteria and Reported Device Performance
The acceptance criteria for this device are not explicitly laid out as a numbered list with specific thresholds before the results are presented. Instead, the document describes the types of studies conducted to evaluate performance, and the results of those studies demonstrate that the device meets the "clinically appropriate" standards as determined by the FDA. The special controls outlined in Section P provide a retrospective view of what was deemed acceptable.
Therefore, the table below consolidates the implicit acceptance criteria based on the special controls and the corresponding reported performance.
| Performance Characteristic | Acceptance Criteria (Implied from Special Controls & Document Context) | Reported Device Performance |
|---|---|---|
| Precision | Clinically appropriate precision, including near medical decision points throughout the therapeutic range. Must include a minimum of three samples with different concentrations, including clinical specimens from patients taking plazomicin. | Internal Precision: QC materials, spiked plasma samples, and patient plasma pools showed %CVs generally ranging from 2% to 7% for total-run variability. |
| Multi-Laboratory Precision: Overall reproducibility %CVs ranged from 4% to 8% across three sites. (See tables in Section L.1.a) | ||
| Linearity/Assay Reportable Range | Evidence of linearity across the claimed measuring range. | Supported for 0.8 to 34.0 ug/mL. Regression analysis showed slopes close to 1 (0.98-1.04) and high correlation (r=1.00) for both spiked and patient pooled plasma. |
| Analytical Recovery | Acceptable recovery throughout the claimed measuring interval. | Recovery ranged between 97% and 104% throughout the claimed measuring interval (0.3 to 34.0 ug/mL). |
| Traceability | Traceable to reliable reference. | Traceable to plazomicin reference calibrators, gravimetrically prepared and value confirmed by LC-MS/MS. |
| Detection Limit (LoB, LoD, LoQ) | Clearly defined and clinically appropriate detection and quantitation limits. | LoB: (b)(4) ng/L (specific value redacted) |
| LoD: 0.4 ug/mL | ||
| LoQ: ≤ 0.8 ug/mL (defined as lowest concentration with interassay precision ≤ 20%CV and bias ≤ 15%) | ||
| Analytical Specificity (Endogenous Interference) | Free from clinically significant interference from endogenous substances. | All tested endogenous substances (Albumin, Bilirubin, Cholesterol, Creatinine, Gamma Globulin, HAMA, Hemoglobin, Rheumatoid Factor, Triglyceride, Uric Acid) resulted in ≤ 10% bias. |
| Analytical Specificity (Exogenous Interference) | Free from clinically significant interference from co-administered medications relevant to cUTI patients and structurally similar compounds. | None of the listed concomitant medications or structurally similar compounds, at clinically relevant concentrations, caused ≥ 10% bias in plazomicin measurement. |
| Method Comparison | Clinically appropriate accuracy against a comparator method. Data collected at a minimum of three laboratory sites. | Comparison with validated LC-MS/MS method for 134 K2-EDTA plasma samples from patients. |
| Deming: Slope = 1.007, Intercept = 0.72 | ||
| Passing-Bablok: Slope = 1.039, Intercept = 0.41 | ||
| Correlation (R): 0.983 | ||
| (Data across 3 test laboratories, as required). |
2. Sample Sizes Used for the Test Set and Data Provenance
-
Test Set (Analytical Studies):
- Precision (Internal):
- Quality Control (QC) and Spiked samples: n=80 each (2 replicates/run, 2 runs/day, 20 days).
- Patient Plasma Pools: n=20 each (2 replicates/run, 2 runs/day, 5 days).
- Precision (Multi-Laboratory):
- QC and Spiked samples: n=240 each (n=80 per site x 3 sites).
- Patient Pools: n=60 each (n=20 per site x 3 sites).
- Linearity: 9 levels of samples, tested in 5 replicates in a single run.
- Analytical Recovery: Each sample panel analyzed over 5 days with 4 replicates/day (total 20 measurements per sample).
- Detection Limit (LoB, LoD, LoQ): Specific sample numbers for LoB/LoD/LoQ are redacted (b)(4). LoB used (b)(4) K2-EDTA samples and LoD used (b)(4) blank plasma K2-EDTA samples.
- Method Comparison: 134 K2-EDTA plasma samples from patients taking plazomicin.
- Interference (Endogenous/Exogenous): Not explicitly stated, but implies a sufficient number of spiked samples to test the indicated concentrations.
- Precision (Internal):
-
Data Provenance:
- Country of Origin: Not explicitly stated, but the multi-laboratory precision study and method comparison study imply a multi-site approach, possibly within a single country or across different countries, but this is not detailed.
- Retrospective or Prospective: Most of the analytical performance studies (precision, linearity, recovery, detection limits, interference) appear to be prospective laboratory studies using prepared or collected samples. The patient samples used for precision and method comparison are collected from patients taking plazomicin, indicating a real-world context for some of the samples. The clinical trial data for plazomicin efficacy/nephrotoxicity (mentioned under "Clinical Studies") was retrospective regarding the device's evaluation, as TDM was not used to adjust dosing during that trial, but the data was later correlated.
3. Number of Experts Used to Establish the Ground Truth for the Test Set and Qualifications of Those Experts
- Number of Experts: This is a quantitative immunoassay, not an imaging device requiring expert reader interpretation. Therefore, the concept of "experts establishing ground truth" in the typical sense (e.g., radiologists reviewing images) does not directly apply.
- Qualifications of Experts: The ground truth for this device is established through:
- Reference Methods: The primary ground truth for the device's accuracy and traceability is a validated LC-MS/MS method. Personnel operating and validating LC-MS/MS are highly trained laboratory professionals with expertise in mass spectrometry and analytical chemistry, though their specific qualifications (e.g., years of experience, certifications) are not detailed in this document.
- Gravimetric Preparation: Plazomicin reference calibrators are gravimetrically prepared, meaning their concentration is determined by precise weighing, which implies a fundamental chemical and metrological expertise.
- CLSI Guidelines: The studies follow CLSI (Clinical and Laboratory Standards Institute) guidelines, which are established by consensus among experts in laboratory medicine.
4. Adjudication Method for the Test Set
- Since the ground truth is established by reference methods (LC-MS/MS, gravimetric preparation) and not by subjective expert interpretation, there is no "adjudication method" in the sense of resolving discrepancies between multiple human readers. The analytical methods themselves have defined reconciliation processes, but it's not a multi-reader adjudication.
5. If a Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study was done
- No, an MRMC study was NOT done. This device is a quantitative immunoassay, not an AI-powered diagnostic tool that assists human readers in interpreting complex data like medical images. Therefore, the concept of human readers improving with AI assistance is not applicable here.
6. If a Standalone (i.e. algorithm only without human-in-the loop performance) was done
- Yes, in essence, standalone performance was evaluated. The QMS Plazomicin Immunoassay is an automated laboratory test. Its performance characteristics (precision, linearity, accuracy against LC-MS/MS, detection limits, interference) are all measures of its intrinsic, standalone capability to quantify plazomicin in a sample. There isn't a "human-in-the-loop" component to its analytical function; rather, humans use the results obtained from the automated device.
7. The Type of Ground Truth Used
The ground truth used for evaluating the QMS Plazomicin Immunoassay's performance includes:
- Reference Measurement Procedure: A validated LC-MS/MS method was used as the comparator for method comparison studies, serving as the "gold standard" for quantifying plazomicin.
- Gravimetric Preparation: For calibrators and spiked samples, the known concentration derived from gravimetric preparation (weighing out precise amounts of plazomicin) served as the ground truth.
- Clinical Outcomes Data (Indirectly for clinical utility, not analytical performance): The clinical trial data (retrospectively analyzed post-drug approval) provided evidence for the correlation between plazomicin trough levels and nephrotoxicity, which established the clinical need and medical decision points for the assay, but not the direct analytical accuracy of the device itself.
8. The Sample Size for the Training Set
- Not Applicable in the traditional machine learning sense. This document describes a traditional in-vitro diagnostic (IVD) device (a chemical immunoassay), not an AI/Machine Learning algorithm that requires a "training set" to learn from data. The device's operational parameters (e.g., reagant concentrations, instrument settings) are determined through development and optimization processes, not by training on a large dataset in the way an AI model would be.
9. How the Ground Truth for the Training Set Was Established
- Not Applicable. As this is not an AI/ML device, there is no "training set" or corresponding ground truth establishment process in that context. The "ground truth" for the device's development primarily relies on established principles of analytical chemistry, reagent formulation, and calibration against reference methods like LC-MS/MS and gravimetric standards.
§ 862.3460 Plazomicin test system.
(a)
Identification. A plazomicin test system is a device intended to measure plazomicin in human specimens. Measurements obtained by this device are used in monitoring levels of plazomicin to ensure appropriate therapy in patients with complicated urinary tract infection.(b)
Classification. Class II (special controls). The special controls for this device are:(1) Design verification and validation must include the following:
(i) Precision study data that demonstrates clinically appropriate precision of the plazomicin test system. Precision studies must include a minimum of three samples containing different concentrations of plazomicin, including near medical decision points throughout the expected therapeutic range of plazomicin. Samples near the medical decision points must be clinical specimens collected from patients taking plazomicin.
(ii) Method comparison data that demonstrates clinically appropriate accuracy of the plazomicin test system, as determined by FDA. Method comparison data must be collected at a minimum of three laboratory sites.
(iii) Data from studies appropriate to demonstrate that the device is free from clinically significant interference from co-administered medications that are used in patients with complicated urinary tract infection, as determined by FDA.
(2) The device's labeling required under § 809.10 of this chapter must include a warning statement that explains: “This assay should only be used in conjunction with information available from clinical evaluations and other diagnostic procedures.”