Search Results

The MolecularMD MRDx BCR-ABL Test is an in vitro diagnostic test for the quantitative detection of BCR-ABL 1 transcripts (e13a2/b2a2 and/or e14a2/63a2) and the ABL1 endogenous control mRNA in peripheral blood specimens from patients previously diagnosed with t(9:22) positive chronic myeloid leukemia (CML). The ratio of BCR-ABL1 is calculated and reported on the WHO International Scale. The test utilizes quantitative, real-time reverse transcription polymerase chain reaction performed on the Applied Biosystems 7500 Fast Dx instrument.

The MolecularMD MRDx BCR-ABL Test is intended to measure BCR-ABL mRNA transcript levels in patients diagnosed with t(9;22) positive CML during monitoring of treatment with Tyrosine Kinase Inhibitors (TKIs).

The device is also intended to be used in the serial monitoring for BCR-ABL mRNA transcript levels as an aid in identifying CML patients in the chronic phase being treated with nilotinib who may be candidates for treatment discontinuation and for monitoring of treatment-free remission.

The test does not differentiate between e13a2 or e14a2 fusion transcripts and does not monitor other rare fusion franscripts resulting from t(9:22). The test is not intended for the diagnosis of CML.

The MolecularMD MRDx® BCR-ABL Test is a quantitative, real-time polymerase chain reaction test that provides quantitation of BCR-ABL1 (hereafter BCR-ABL) transcript e13a2/b2a2 or e14a2/b3a2 and ABL1 (hereafter ABL) transcript levels in RNA extracted from peripheral blood samples collected from CML patients. Peripheral blood is collected in either EDTA or PAXgene Blood RNA Tubes. Each collection tube type requires a specified RNA extraction method and RNA input amount. Total RNA is extracted from peripheral blood and serves as the template for RT-qPCR. The test is performed using a onestep RT-qPCR protocol wherein the reverse transcription and quantitative, real-time PCR reactions are performed in the same well. BCR-ABL and ABL amplicons are generated and detected in real-time using TaqMan® MGB probes. The MolecularMD MRDx BCR-ABL Test is performed on the Applied Biosystems 7500 Fast Dx Real-Time PCR Instrument. The instrument integrates a thermal cycler, a fluorometer and application specific software. The ABI 7500 Fast Dx instrument software v1.4.1 calculates the BCR-ABL and ABL copy numbers using a standard curve generated with calibrators on each individual plate. The data are exported as a CSV file for further analysis in the MRDx® BCR-ABL Test Software. The BCR-ABL/ABL ratio is calculated and converted to the International Scale by the MRDx BCR-ABL Test Software. The MRDx BCR-ABL Test Software is used to analyze all test results. This software, provided with the MRDx BCR-ABL Test, is used to calculate the BCR-ABL/ABL % IS and MR value for patient sample using the conversion factor for the MRDx BCR-ABL Test after validating each MRDx BCR-ABL Test result against the run acceptance criteria and sample acceptance criteria.

The MolecularMD MRDx BCR-ABL Test is an in vitro diagnostic test for the quantitative detection of BCR-ABL1 transcripts and ABL1 endogenous control mRNA in peripheral blood specimens from patients previously diagnosed with t(9;22) positive chronic myeloid leukemia (CML). It calculates the ratio of BCR-ABL1 to ABL1 and reports it on the WHO International Scale. The device is intended to measure BCR-ABL mRNA transcript levels in CML patients during monitoring of treatment with Tyrosine Kinase Inhibitors (TKIs). It is also intended for serial monitoring of BCR-ABL mRNA transcript levels to aid in identifying CML patients in the chronic phase being treated with nilotinib who may be candidates for treatment discontinuation and for monitoring of treatment-free remission.

Here's an overview of the acceptance criteria and study data:

1. Table of Acceptance Criteria and Reported Device Performance

The document does not explicitly present a consolidated table of acceptance criteria for all performance characteristics. However, individual acceptance criteria are mentioned within the description of each study. Below is a summary of acceptance criteria and the device's reported performance, extracted from the document:

2. Sample sizes used for the test set and data provenance

• Traceability to WHO IS: Tested over 20 days for each of 3 MRDx BCR-ABL Test kit lots. A different WHO primary panel (four levels A-D) was reconstituted daily, with each level tested multiple times (implicitly, across the 20 days). MolecularMD WHO BCR-ABL Reference Panel Secondary Standards were also tested.
• Accuracy (Correlation to ddPCR):
  • Test Set Size: 217 samples total. 119 evaluable EDTA samples and 98 evaluable PAXgene samples.
  • Data Provenance: Samples were from CML patients. The document does not specify the country of origin but implies clinical samples. The study involved comparison to a reference method, suggesting an analytical, retrospective-like analysis of patient samples.
• Limit of Blank (LoB): 30 BCR-ABL negative samples, tested in duplicate using 3 assay kit lots, totaling 180 replicates.
• Limit of Detection (LoD): 6 CML patient specimens (3 e13a2, 3 e14a2) used to create serial dilutions. Sufficient RNA for each panel (5 levels) was made to allow for testing of 60 total replicates (20 replicates x 3 lots) per level. 1725 valid EDTA samples and 1595 valid PAXgene samples were tested.
• Analytical Specificity (Interfering Substances): CML samples at 2 levels (MR3.0 and MR4.5) extracted from both EDTA and PAXgene tubes, spiked with test substances. No specific number of replicates or distinct samples given for this part.
• Analytical Specificity (Primer Specificity): RNA from six CML patient samples (three e13a2, three e14a2) were used. Additionally, ABL2 IVT RNA and RNA from an e19a2 positive cell line were tested.
• Analytical Specificity (Specimen Cross Contamination):
  • PAXgene: 30 replicates.
  • EDTA: 40 replicates.
• Precision (Repeatability): Contrived patient samples (e13a2 and e14a2 at MR3.0, MR4.0, MR4.5). Each MR sample level for each transcript was tested in 17 replicates on individual MRDx BCR-ABL Test plates.
• Precision (Reproducibility): Two series of samples (e13a2 and e14a2 CML patient samples diluted into non-diseased blood) at 5 levels each.
  • EDTA: 1800 total tests (60 replicates x 3 sites x 2 transcripts x 5 panel members). N for individual levels ranged from 149-152.
  • PAXgene: 900 total tests (90 measurements per panel member x 2 transcripts x 5 panel members, implying 90 measurements per level). N for individual levels ranged from 77-92.
• Linear Range (Assay Range): Two dilution series (one e13a2, one e14a2) for each tube type. Each series had 10 BCR-ABL target levels, with 5 replicates at each level.
• Kit Real-Time and Freeze-Thaw Stability: 6 kits from each of 3 lots for each time point were tested.
• Reagent Preparation and Reaction Intermediate Stability (In Use): 2 kits were run for each condition (standard/extreme) on each of 2 days, with 2 sets of calibration curves, RNA controls, and samples (total n=8 per condition).
• Specimen Stability:
  • PAXgene: Samples at three levels (MR3.0, MR4.0, MR4.5) tested in quadruplicate at 0, 24, 48, 60 hours (for 2-8°C storage) and at 0, 14, 28, 30 days (for -30 to -15°C storage).
  • EDTA: 17 CML patient specimens, tested in quadruplicate at 0, 24, 48, 60 hours.
• Shipping Stability: Multiple types of containers and shipping conditions, varying load. Functional testing using samples across the assay range. Specific N not provided for samples.
• PAXgene-EDTA Tube Equivalency: 100 CML patient paired blood samples.

3. Number of experts used to establish the ground truth for the test set and the qualifications of those experts

Not applicable for analytical performance studies. The "ground truth" for these studies is established through defined reference methods (e.g., WHO International Genetic Reference Panel, ddPCR, precisely prepared dilutions/spikes, or established consensus sequences/protocols), not expert consensus in the human interpretation sense.

For the clinical studies (ENESTfreedom and ENESTop), the efficacy outcome (treatment-free remission, loss of MMR) is based on the quantitative results of the MRDx BCR-ABL Test itself per pre-defined thresholds, rather than expert interpretation of individual clinical cases as a ground truth.

4. Adjudication method for the test set

Not applicable. For the analytical studies, the ground truth is quantitative and objectively determined (e.g., standard concentrations, reference assay results, sequencing). For the clinical studies, outcomes are determined by the device's quantitative measurements against pre-defined clinical thresholds, not by human adjudication of conflicting interpretations.

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

Not applicable. This device is an in vitro diagnostic test (RT-qPCR assay) and not an AI-assisted diagnostic imaging or interpretation tool that involves human "readers." Therefore, no MRMC study, or assessment of human reader improvement with AI, was performed or is relevant to this device.

6. If a standalone (i.e. algorithm only without human-in-the-loop performance) was done

Yes, the analytical performance studies explicitly demonstrate the standalone performance of the MolecularMD MRDx BCR-ABL Test. This device is an automated quantitative assay (RT-qPCR) followed by software analysis. The outputs are quantitative values (BCR-ABL/ABL % IS and MR values), and the performance characteristics (accuracy, precision, LoD, LoQ, linearity, etc.) are reported for the assay system itself, without human-in-the-loop performance influencing the primary measurements or calculations. While interpretation of results for clinical decision-making is done by clinicians, the device's measurement function is standalone.

7. The type of ground truth used (expert consensus, pathology, outcomes data, etc.)

• Traceability to WHO IS: The ground truth was the 1st WHO International Genetic Reference Panel for quantitation of BCR-ABL translocation by RQ-PCR, which represents a harmonized international standard.
• Accuracy (Correlation to ddPCR): The ground truth was derived from a validated BCR-ABL reverse transcription droplet digital PCR (ddPCR) assay, which served as the reference method for comparison.
• Limit of Blank (LoB): "BCR-ABL negative samples" served as the negative ground truth.
• Limit of Detection (LoD), Linear Range: Ground truth was established using serial dilutions of CML patient RNA and non-diseased subject blood RNA, allowing for precise control of target concentrations.
• Analytical Specificity: Ground truth included precisely spiked interfering substances (endogenous and exogenous), known ABL2 IVT RNA, and e19a2 positive cell line RNA, and alternation of known high-positive and negative samples for cross-contamination.
• Precision (Repeatability and Reproducibility): Ground truth was created using contrived patient samples (CML patient RNA and non-diseased subject RNA) with targeted MR levels.
• PAXgene-EDTA Tube Equivalency: Ground truth was established by comparing results from paired CML patient blood samples processed via two different collection and extraction methods.
• Clinical Studies (ENESTfreedom and ENESTop): The clinical validity was demonstrated against clinical outcomes data from two prospective clinical trials (ENESTfreedom and ENESTop). The ground truth for identifying candidates for treatment discontinuation and monitoring remission relied on sustained molecular response (MR4.5/MR4.0) as defined by the measured BCR-ABL levels.

8. The sample size for the training set

The document does not explicitly specify a "training set" for the development of the device's algorithm in the way that would typically be described for a machine learning model. This is an RT-qPCR assay, not a machine learning-based AI device.

However, if "training set" is broadly interpreted as data used during the development and optimization phases:

• The quantitative methodology is based on established RT-qPCR principles and standard curve generation using calibrators.
• The "conversion factor" to the International Scale is derived from harmonization studies with the WHO International Genetic Reference Panel.
• The "MRDx BCR-ABL Test Software" uses established equations and validated run/sample acceptance criteria, which would have been developed and refined using internal data.
• The specific number of samples or data points used for these internal development and optimization processes is not provided.

9. How the ground truth for the training set was established

As noted above, there isn't a "training set" in the machine-learning sense for this device. The ground truth for analytical performance studies, which serve to validate the device's performance, is established through:

• International Reference Standards: The 1st WHO International Genetic Reference Panel for BCR-ABL quantitation.
• Reference Methods: A validated droplet digital PCR (ddPCR) assay.
• Controlled Experiments: Creation of precisely defined mock samples or panels with known concentrations or characteristics (e.g., serial dilutions of CML patient RNA, non-diseased RNA, spiked samples, targeted MR levels in contrived samples).
• Clinical Trial Data: Long-term clinical follow-up in the ENESTfreedom and ENESTop studies, where the MRDx BCR-ABL Test itself determined the molecular response status that defined the clinical ground truth for treatment discontinuation and remission monitoring.

Ask a specific question about this device