(448 days)
The MLL (KMT2A) Breakapart FISH Probe Kit is a fluorescence in situ hybridization (FISH) Test used to detect rearrangement of the MLL (KMT2A) region on chromosome 11 at location 11g23.3 in fixed bone marrow specimens from patients with acute myeloid leukemia (AML) or myelodysplastic syndrome (MDS). The test is indicated for characterization of patient specimens consistent with World Health Organization (WHO) guidelines for Classification of Tumours of Haematopoietic and Lymphoid Tissues (Revised 4th Edition) and in conjunction with other clinicopathological criteria. The assay results are intended to be interpreted by a qualified pathologist or cytogeneticist. The test is not intended for use as a stand-alone diagnostic, disease screening, or as a companion diagnostic.
The P53 (TP53) Deletion FISH Probe Kit is a fluorescence in situ hybridization (FISH) Test used to detect deletion of the P53 (TP53) region on chromosome 7 at location 17p13 in fixed bone marrow specimens from patients with acute myeloid leukemia (AML) or myelodysplastic syndrome (MDS). The test is indicated for characterization of patient specimens consistent with World Health Organization (WHO) guidelines for Classification of Tumours of Haematopoietic and Lymphoid Tissues (Revised 4th Edition) and in conjunction with other clinicopathological criteria. The assay results are intended to be interpreted by a qualified pathologist or cytogeneticist. The test is not intended for use as a stand-alone diagnostic, disease screening, or as a companion diagnostic.
The Del(20q) Deletion FISH Probe Kit is a fluorescence in situ hybridization (FISH) Test used to detect deletion within the long arm of chromosome 20 at locations 20g12 and 20q13.1, in fixed bone marrow specimens from patients with myelodysplastic syndrome (MDS). The test is indicated for characterization of patient specimens consistent with World Health Organization (WHO) guidelines for Classification of Tumours of Haematopoietic and Lymphoid Tissues (Revised 4th Edition) and in conjunction with other clinicopathological criteria. The assay results are intended to be interpreted by a qualified pathologist or cytogeneticist. The test is not intended for use as a stand-alone diagnostic, disease screening, or as a companion diagnostic.
The CBFB (CBFB)/MYH11 Translocation, Dual Fusion FISH Probe Kit is a fluorescence in situ hybridization (FISH) Test used to detect rearrangement of the chromosome 16 causing the CBFB-MYH11 (CBFB-MYH11) fusion in fixed bone marrow specimens from patients with acute myeloid leukemia (AML). The test is indicated for characterization of patient specimens consistent with World Health Organization (WHO) guidelines for Classification of Tumours of Haematopoietic and Lymphoid Tissues (Revised 4th Edition) and in conjunction with other clinicopathological criteria. The assay results are intended to be interpreted by a qualified pathologist or cytogeneticist. The test is not intended for use as a stand-alone diagnostic, disease screening, or as a companion diagnostic.
The Del(5q) Deletion FISH Probe Kit is a fluorescence in situ hybridization (FISH) Test used to detect deletions within the long arm of chromosome 5 at location 5g31.2 in fixed bone marrow specimens from patients with acute myeloid leukemia (AML) or myelodysplastic syndrome (MDS). The test is indicated for characterization of patient specimens consistent with World Health Organization (WHO) guidelines for Classification of Tumours of Haematopoietic and Lymphoid Tissues (Revised 4th Edition) and in conjunction with other clinicopathological criteria. The assay results are intended to be interpreted by a qualified pathologist or cytogeneticist. The test is not intended for use as a stand-alone diagnostic, disease screening, or as a companion diagnostic.
The Del(7q) Deletion FISH Probe Kit is a fluorescence in situ hybridization (FISH) Test used to detect deletions within the long arm of chromosome 7 at locations 7g22 and 7q31.2 in fixed bone marrow specimens from patients with acute myeloid leukemia (AML) or myelodysplastic syndrome (MDS). The test is indicated for characterization of patient specimens consistent with World Health Organization (WHO) guidelines for Classification of Tumours of Haematopoietic and Lymphoid Tissues (Revised 4th Edition) and in conjunction with other clinicopathological criteria. The assay results are intended to be interpreted by a qualified pathologist or cytogeneticist. The test is not intended for use as a stand-alone diagnostic, disease screening, or as a companion diagnostic.
The AML1/ETO (RUNX1/RUNX1T1) Translocation, Dual Fusion FISH Probe Kit is a fluorescence in situ hybridization (FISH) Test used to detect rearrangement involving the AML1 (RUNX1) region on chromosome 21 at location 21g22.1 and the ETO (RUNXITI) region on chromosome 8 at location 8q21.3 in fixed bone marrow specimens from patients with acute myeloid leukemia (AML). The test is indicated for characterization of patient specimens consistent with World Health Organization (WHO) guidelines for Classification of Tumours of Haematopoietic and Lymphoid Tissues (Revised 4th Edition) and in conjunction with other clinicopathological criteria. The assay results are intended to be interpreted by a qualified pathologist or cytogeneticist. The test is not intended for use as a stand-alone diagnostic, disease screening, or as a companion diagnostic.
The EVI1 (MECOM) Breakapart FISH Probe Kit is a fluorescence in situ hybridization (FISH) Test used to detect rearrangement involving the EVII (MECOM) region on chromosome 3 at location 3g26.2, in fixed bone marrow specimens from patients with acute myeloid leukemia (AML) or myelodysplastic syndrome (MDS). The test is indicated for characterization of patient specimens consistent with World Health Organization (WHO) guidelines for Classification of Tumours of Haematopoietic and Lymphoid Tissues (Revised 4th Edition) and in conjunction with other clinicopathological criteria. The assay results are intended to be interpreted by a qualified pathologist or cytogeneticist. The test is not intended for use as a stand-alone diagnostic, disease screening, or as a companion diagnostic.
For prescription use. For in vitro diagnostic use
Each Cytocell FISH Probe Kit device consists of one vial containing specific probes as described below. The probes are provided premixed in hybridization solution (formamide: dextran sulfate; saline-sodium citrate (SSC)) and are ready to use. The Kit also includes one vial of 4,6-diamidino-2-phenylindole (DAPI) counterstain. The kits are available in a 10-test format.
The MLL (KMT2A) Breakapart FISH Probe Kit consists of an 87kb probe, labeled in Texas red, covering a region telemetric to the MLL (KMT2A) gene including the marker SHGC-111513 and a FITC green probe covering a 170kb region centromeric to the MLL (KMT2A) gene spanning the CD3G and UBE4A genes.
The P53 (TP53) Deletion FISH Probe Kit consists of a 161kb probe, labeled in Texas red, covering the whole P53 (TP53) gene, extending 74kb telomeric to the gene and covering a region centromeric to the gene, to just beyond the marker D17S655; and a probe, labelled in FITC green, covering the chromosome 17 centromere (D17Z1) region.
The Del(20q) Deletion FISH Probe Kit consists of a 331kb probe, labeled in Texas red, covering a region within the PTPRT gene and including the D20S108 marker; and two (141kb and 174kb) probes labeled in FITC green covering the MYBL2 gene and including the D20S150 marker.
The CBFB (CBFB) /MYH11 Translocation, Dual Fusion FISH Probe Kit consists of a 617kb probe, labeled in Texas red, covering a region, within 16q22 including the CBFB gene; and a 621kb probe, labeled in FITC green, covering a region within 16p13.1 including the MYH11 gene.
The Del(5q) Deletion FISH Probe Kit consists of a 186kb probe, labeled in Texas red, covering a region within 5q31.2, including the D5S500 marker; and a 376kb probe, labeled in FITC green, within 5p15.3, including the D5S630 marker.
The Del(7q) Deletion FISH Probe Kit consists of a 396kb probe, labeled in Texas red, covering a region within 7q22 including the telomeric end of the RELN gene and extending beyond the D7S658 marker; and a 203kb probe, labeled in FITC green, covering a region within 7q31.2 including the TES gene.
The AML1/ETO (RUNX1/RUNX1T1) Translocation, Dual Fusion FISH Probe Kit consists of a 156kb probe labeled in Texas red, centromeric to the AMLI (RUNXI) gene, including the CLIC6 gene; a 169kb probe labelled in Texas red, telomeric to AMLI (RUNXI) gene, extending beyond the marker D21S1921; and two (151kb and 194kb) probes, labeled inn FITC green, on either side of the ETO (RUNXIT) gene.
The EVI1 (MECOM) Breakapart FISH Probe Kit consists of a 158kb probe, labeled in Texas red, telomeric to the D3S4415 marker and including the LRRC34 gene, a FITC green probe covering a 181kb region. including the entire EVI1 (MECOM) gene and flanking regions and a PF-415 blue probe, which covers a 563kb region centromeric to the EVI I (MECOM) gene, including the D3S1614 marker.
Acceptance Criteria and Study for Cytocell FISH Probe Kits
1. Table of Acceptance Criteria and Reported Device Performance
The acceptance criteria for these devices are defined by adherence to various analytical performance metrics and comparison to established literature prevalence ranges. The performance is reported across multiple tables within the document. Below is a summary, with specific numerical criteria inferred from the successful outcome of the studies.
| Category | Acceptance Criteria (Inferred from successful study outcome) | Reported Device Performance |
|---|---|---|
| Analytical Performance | ||
| Precision/Reproducibility (Intra-day, Inter-day, Between-site SD) | Acceptable precision in terms of standard deviation (SD) for negative, near cut-off/low positive, and high positive specimens. Agreement of replicates should be high, particularly for negative and high positive samples. | Tables 2-10 show standard deviations across intra-day, inter-day, and between-site measurements. For example, for MLL (KMT2A) Breakapart: Negative 1 (Total SD: 0.00-0.09), Near cut-off 1 (Total SD: 1.98), Positive 1 (Total SD: 6.60). Agreement replicates for negative and positive samples are generally 100%, with near cut-off samples showing lower but acceptable agreement (e.g., 63-90% for MLL, 27-43% for CBFβ/MYH11). Additional studies for near cut-off samples of CBFβ/MYH11 and EVI1 (MECOM) Breakapart Inversion also demonstrated acceptable precision. Each probe kit demonstrated acceptable precision. |
| Between-Lot Reproducibility | Acceptable standard deviation (SD) for intra-lot, inter-lot, and total SD using one normal, one low positive, and one high positive specimens across 3 different reagent lots. High agreement of replicates. | Tables 11-19 show acceptable total SDs for negative, near cut-off, and positive specimens. Agreement of replicates for negative and positive samples is generally 100%, with near cut-off samples showing lower but acceptable agreement (e.g., 67-100%). Each probe kit demonstrated acceptable lot-to-lot reproducibility. |
| Real-time Kit Stability | Probes should maintain performance (accuracy) for the claimed shelf-life (e.g., 24 months at -20°C). | Supported a claimed shelf life of 24 months when stored at -20°C. |
| Transport Stability | No significant change in device performance under defined stress conditions (e.g., 2 weeks at +40°C, then -20°C). | No change in device performance under the stress conditions. |
| Freeze/Thaw Stability | Probes should maintain performance after a specified number of freeze/thaw cycles (e.g., 10 cycles). | Ten freeze-thaw cycles were found to be acceptable. |
| Post Hybridization Stability | Hybridized slides should remain stable for a specified duration (e.g., 1 month). | Post-hybridization signal was determined to be stable to one month. |
| Photostability | Probes should maintain performance after a specified exposure to light (e.g., 24-36 hours). | Photo-stability of slides was supported to 24-36 hours. Prolonged exposure of slides to light should be avoided. |
| Analytical Sensitivity | Analytical sensitivity of greater than 95% (proportion of interphase nuclei showing the expected normal signal pattern from karyotypically normal samples). | All nine probes evaluated had an analytical sensitivity of greater than 95%, meeting the required acceptance criteria (Table 21). Ranges from 97.98% to 99.4% with 95% Confidence Intervals. |
| Analytical Specificity | 100% specificity (percentage of signals hybridizing to the correct locus and no other location) from normal male peripheral blood samples. | All probes met the required acceptance criteria for analytical specificity with all components having a specificity of 100% (Table 22). |
| Assay Cut-off (Upper Reference Limit) | Normal cut-offs confirmed by analytical sensitivity data; an analytical result above the cut-off is positive, below is negative. | Clinical cut-offs were either generated by a central clinical laboratory using a large patient dataset or derived from analytical sensitivity data of 25 karyotypically normal bone marrow samples. The normal cut-offs were confirmed as none of the 25 normal samples showed an abnormal signal pattern at or above the established cut-offs (Table 23 shows the cut-offs for each probe). |
| Clinical Validation | ||
| Incidence Rate Comparison | Device-measured incidence rates (and 95% CI) from clinical specimens should fall within the expected prevalence range from peer-reviewed literature for the target population (AML/MDS patients). | For each probe, incidence rates from two data sources (GOP and YAL) using the Cytocell FISH probe kits were compared to expected ranges from 3-6 literature sources. In most cases, the observed incidence rates, including their 95% CIs, fell within or aligned closely with the expected literature ranges (Tables 25-32). For example, for MLL (KMT2A) Breakapart, GOP (2%) and YAL (1.45%) were within the literature range of 0.2%-4.5%. |
| Concordance with Karyotyping (for specific probes) | High percentage agreement between the FISH probe kit results and G-band karyotyping, especially for normal and confirmed abnormal specimens. Any discordance should be justifiable (e.g., near cut-off). | For AML1/ETO (RUNX1/RUNX1T1) and CBFβ (CBFB)/MYH11 probes, 100% agreement was observed for both data sets with G-band karyotyping, except for one specimen that fell into the re-test zone or was near the cut-off (Tables 33-34). This demonstrates acceptable clinical concordance. |
2. Sample Size Used for the Test Set and Data Provenance
- Analytical Performance Test Set:
- Precision/Reproducibility: For each probe kit, 6 specimens were used (2 normal, 2 near cut-off/low positive, 2 high positive). These were tested over 5 non-consecutive days in duplicates at 3 sites (n=30 per specimen). In some cases, additional specimens were re-assessed for "near cut-off" performance.
- Between-Lot Reproducibility: 3 specimens (1 normal, 1 low positive, 1 high positive) were tested with 3 replicates per lot across 3 reagent lots.
- Analytical Sensitivity: 25 karyotypically normal bone marrow samples.
- Analytical Specificity: 5 normal male peripheral blood samples.
- Clinical Performance Test Set:
- Incidence Rate Comparison:
- GOP data set: 100 known or suspected AML and MDS specimens in total.
- YAL data set: Re-analyzed specimen data ranging from 266 to 746 AML or MDS specimens depending on the probe.
- G-band Concordance: Additional clinical data sets (sample size not explicitly stated for each, but tables 33 and 34 imply sufficient numbers to show concordance for AML1/ETO and CBFβ/MYH11).
- Incidence Rate Comparison:
- Data Provenance: The document does not explicitly state the country of origin for the clinical data (GOP and YAL data sets). The analytical validation studies appear to have been conducted by the applicant, Cytocell, Ltd. (a UK-based company). The literature references used for prevalence rates are international. The clinical data sets (GOP and YAL) are implied to be from clinical laboratories. The studies mentioned (clinical and analytical) were prospective studies for the current submission.
3. Number of Experts Used to Establish the Ground Truth for the Test Set and Qualifications of Those Experts
- Analytical Studies (Analytical Sensitivity & Specificity): 2 independent analysts analyzed the interphase nuclei for analytical sensitivity. The qualifications of these analysts are not explicitly stated, but it is implied they are trained cytogeneticists or laboratory personnel performing FISH analysis.
- Clinical Performance Studies (Incidence Rate & Concordance): The ground truth for the clinical specimens used in the incidence rate studies (GOP and YAL data sets) and the G-band concordance studies was established by established alternative cytogenetic methods (competitor probe or G-banding) prior to distribution for analytical studies, or by standard clinical diagnostic procedures. Final assay results are intended to be interpreted by a qualified pathologist or cytogeneticist. The number of experts establishing the initial "known" status of the clinical samples is not specified, but the references to "established alternative cytogenetic method (competitor probe or G-banding)" imply experienced cytogenetic laboratories. For the concordance studies, the G-band result serves as the established ground truth.
4. Adjudication Method (e.g., 2+1, 3+1, none) for the Test Set
- For the analytical studies (e.g., Analytical Sensitivity), the document states: "Each analyst scores independently 100 nuclei for each sample. In some cases, depending on the number of abnormal nuclei each analyst has seen, a third reader may be required." This suggests an adjudication method of 2+1 (two initial independent reads, with a third if there is disagreement or uncertainty regarding the number of abnormal nuclei for a final count near a threshold).
5. If a Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study Was Done, and Effect Size
- No, a traditional MRMC comparative effectiveness study comparing AI-assisted vs. unassisted human readers was not performed or described in this document.
- This document describes the validation of FISH probe kits, which are reagents for a laboratory test, not an AI-powered diagnostic device. The "device" in question refers to these biological reagents for in-situ hybridization.
- Therefore, the concept of "how much human readers improve with AI vs. without AI assistance" is not applicable to this submission.
6. If a Standalone (i.e., algorithm only without human-in-the-loop performance) Was Done
- No, this is not applicable. The Cytocell FISH Probe Kits are biological reagents, and the process of FISH analysis inherently requires human interpretation (qualified pathologist or cytogeneticist) of the hybridized signals under a microscope. This is not an algorithmic or AI-based device intended for standalone performance.
7. The Type of Ground Truth Used
- Analytical Studies:
- Analytical Sensitivity: Karyotypically normal bone marrow samples were used. The ground truth was "normal" based on karyotyping.
- Analytical Specificity: Normal male peripheral blood samples were used. The ground truth was "normal" for the target chromosome locations.
- Clinical Studies:
- Incidence Rate Comparison: "Known or suspected AML or MDS specimens" were used. The ground truth for identifying the presence/absence of specific chromosomal rearrangements was established by "established alternative cytogenetic method (competitor probe or G-banding)" or by "clinical thresholds defined in the literature." The literature itself acts as a form of "expert consensus" on prevalence.
- Concordance with Karyotyping: G-band karyotyping served as the ground truth for confirming the presence or absence of specific rearrangements (e.g., t(8;21) or inv(16)/t(16;16)).
8. The Sample Size for the Training Set
- The document describes the analytical and clinical validation of the FISH probe kits. It does not describe an AI/algorithm training set as this is not an AI-based device.
- However, for the establishment of cut-off values, the following "datasets" were used, which could be considered analogous to "training data" for establishing thresholds:
- For seven of the nine probes, a "large patient dataset exceeding the ACMG minimum sample size guidelines" was used by a central clinical laboratory to generate the clinical cut-off. The exact number is not explicitly stated.
- For the other two probes (AML1 and EVI1), "≥ 25 karyotypically normal bone marrow samples" were enumerated for establishing normal cut-offs.
9. How the Ground Truth for the Training Set Was Established
- As per point 8, this is not an AI device with a traditional training set.
- For the establishment of cut-off values:
- The "large patient dataset" used by the central clinical laboratory likely had ground truth established through comprehensive clinical and cytogenetic diagnoses (including techniques like G-banding and potentially other FISH probes) based on WHO guidelines.
- For the "≥ 25 karyotypically normal bone marrow samples," the ground truth was "karyotypically normal" established through standard karyotyping procedures (e.g., G-banding).
- These established cut-off values represent a consensus or standard derived from expert interpretation of a larger dataset over time.
§ 864.1880 Fluorescence in situ hybridization (FISH)-based detection of chromosomal abnormalities from patients with hematologic malignancies.
(a)
Identification. A fluorescence in situ hybridization (FISH)-based detection of chromosomal abnormalities from patients with hematologic malignancies is used to detect chromosomal abnormalities in human specimens from patients with hematologic malignancies. The test is indicated for the clinical management of patients consistent with internationally accepted guidelines (e.g., World Health Organization guidelines for Classification of Tumours of Haematopoietic and Lymphoid Tissues) and in conjunction with other clinical and clinicopathological criteria. The results are to be interpreted by a pathologist or equivalent professional.(b)
Classification. Class II (special controls). The special controls for this device are:(1) Design verification and validation must include:
(i) A detailed description of all probes included in the kit;
(ii) Purpose of each probe;
(iii) Probe molecular specificity;
(iv) Probe specificity;
(v) Probe limits;
(vi) Probe sensitivity;
(vii) Specification of required ancillary reagents, instrumentation, and equipment;
(viii) Specification of the specimen collection, processing, storage and slide preparation methods;
(ix) Specification of the assay procedure;
(x) Specification of control elements that are incorporated into the recommended testing procedures;
(xi) Specification of the criteria for test result interpretation and reporting;
(xii) Documentation demonstrating analytical validation that includes:
(A) Device analytical sensitivity data with a minimum of 25 specimens from karyotypically normal males.
(B) Device analytical specificity data with a minimum of five specimens from karyotypically normal males.
(C) Description of how the clinical threshold was assigned and verification of the assigned clinical threshold.
(D) Device precision/reproducibility data with a minimum of six clinical specimens including two negative specimens, two positive specimens near the clinical decision threshold (cut-off) and two positive specimens. The data must include results obtained from three sites (as applicable), with two operators at each site, with the assay run for a minimum of 3-5 non-consecutive days and each specimen run in duplicate for a minimum of 30 replicates.
(E) Between-reagent lot reproducibility using three reagent lots and three clinical specimens representing negative, near cut-off/low positive, and positive.
(F) Device stability data to include:
(
1 ) Real-time stability,(
2 ) Freeze-thaw stability,(
3 ) Transport and temperature stability, as applicable,(
4 ) Post-hybridization signal stability, and(
5 ) Photostability of probe.(xiii) Documentation demonstrating the clinical validity of the device that includes:
(A) A summary of the prevalence and clinical thresholds reported in three peer-reviewed published literature references for the intended use population of the device and device performance data demonstrating conformance with the published prevalence as reported in peer-reviewed published literature references based on testing clinical specimens, selected without bias (
e.g., consecutively selected) from the intended use population using the specific device seeking marketing clearance. A minimum number of clinical specimens must be tested to ensure sufficient positives are evaluated by the device, or alternatively, in the absence of a sufficient number of positives, an additional comparison of results obtained with the device to clinical truth (e.g., confirmed clinical diagnosis and/or G-banded karyotyping) with an independent specimen set must be conducted.(B) Documentation for peer-reviewed published literature references must include the following elements:
(
1 ) Whether the specific device was used in the literature reference;(
2 ) Number and type of specimens;(
3 ) Target population studied;(
4 ) Upper reference limit; and(
5 ) Prevalence range estimated based on the number of positive probe results.(C) In the absence of clinical data obtained from paragraphs (b)(1)(xiii)(A) and (B) of this section, clinical data obtained from a method comparison to the predicate with positives and negative clinical specimens.
(2) The intended use required on the label under § 809.10(a)(4) of this chapter and on the labeling required under § 809.10(b)(5)(ii) of this chapter must include a statement that “The test is not intended for use as a stand-alone diagnostic, disease screening, or as a companion diagnostic.”
(3) The labeling required under § 809.10(b) of this chapter must include information that demonstrates the performance characteristics of the test, including a detailed summary of the performance studies conducted and their results, as described in paragraphs (b)(1)(iv) through (xiii) of this section. The labeling required under § 809.10(b) of this chapter must include the pre-specified acceptance criteria for these performance studies, justification for the pre-specified acceptance criteria, and whether the pre-specified acceptance criteria were met.
(4) The labeling required under § 809.10(b) of this chapter must include the following limiting statements:
(i) “Reporting and interpretation of FISH results should be consistent with professional standards of practice and should take into consideration other clinical and diagnostic information. This kit is intended as an adjunct to other diagnostic laboratory tests and therapeutic action should not be initiated on the basis of the FISH result alone. Failure to adhere to the protocol may affect the performance and lead to false results.”
(ii) “Each lab is responsible for establishing their own cut-off values. Each laboratory should test sufficiently large number of samples to establish normal population distribution of the signal levels and to assign a cut-off value. The product is for professional use only and is intended to be interpreted by a qualified pathologist or cytogeneticist.”