This guidance was written prior to the February 27, 1997 implementation of FDA’s Good Guidance Practices, GGP’s. It does not create or confer rights for or on any person and does not operate to bind FDA or the public. An alternative approach may be used if such approach satisfies the requirements of the applicable statute, regulations, or both. This guidance will be updated in the next revision to include the standard elements of GGP’s.
DRAFT DOCUMENT FOR MANUFACTURERS VERSION: Original DATE: 7/15/91 REVIEW CRITERIA FOR ASSESSMENT OF CYTOGENETIC ANALYSIS USING AUTOMATED AND SEMI-AUTOMATED CHROMOSOME ANALYZERS. This is a flexible draft document representing the current major concerns and suggestions regarding in vitro automated and semi-automated chromosome analyzers for cytogenetic analysis. It is based on 1) current basic science, 2) clinical experience, 3) current standard laboratory practice, and 4) previous submissions by manufacturers to the FDA. As advances are made in science and medicine, these review criteria will be re-evaluated and revised as necessary to accommodate new knowledge. PURPOSE: The purpose of this document is to provide guidance on information to present to the Food and Drug Administration (FDA) before a device for chromosome analysis may be cleared for marketing. This information enables FDA to make better informed decisions based on a uniform data base. DEFINITION: The generic type device is intended for use in the cytogenetics laboratory to aid laboratory personnel in performing certain procedures used in karyotyping human metaphase/prometaphase cells for in vitro cytogenetic analysis. PRODUCT CODE: LNJ - 88 REGULATION NUMBER: CFR ~ 864.5260 Automated cell-locating device. PANEL: PATHOLOGY CLASS: II REVIEW REQUIRED: 510(k) REGULATORY ISSUES: Manufacturers are encouraged to develop devices that will be compatible with guidelines of laboratory regulatory organizations. Most cytogenetics laboratories subscribe to certain proficiency testing and accreditation agencies or are state regulated. The College oœ American Pathologist (CAP) and the Council of Regional Genetics Networks (CORN) offer voluntary proficiency testing. In New York State, the Department of Health is responsible for laboratory testing and certification; the State of Oregon licenses cytogenetics laboratories. In New York State, cytogenetics laboratories that use image analysis computer systems must develop a system for record retention and retrieval that meets laboratory licensure quality control standards. I. Background A. Cytogenetics in Clinical Practice Cytogenetic analysis is an in vitro clinical laboratory procedure that evaluates the chromosomes of a cell. Clinical cytogenetics is the study of chromosomes and their correlation to the phenotype (observable clinical characteristics). Certain clinical characteristics occur consistently in association with a particular chromosome abnormality. This phenotype-karyotype correlation is useful to the clinician in making a clinical genetic diagnosis and prognosis. There are two basic classes of chromosome aberrations: numeric and structural. Common abnormalities of chromosome number include aneuploidy, mosaicism, and polyploidy. Structural abnormalities include duplications, deletions, inversions, shifts, fragile sites, etc. Chromosome aberrations are found in association with many anomalies of sexual development such as the Turner syndrome (XO) and the Kleinfelter syndrome (XXY); mental retardation (Trisomy 21 in the Down syndrome and fragile X in the Fragile X syndrome); complex malformation syndromes (Trisomy 13); spontaneous abortions; and malignant disorders (chromosome translocation between chromosome 9 and 22 in chronic myeloid leukemia). In addition to studies for detection of classic chromosome abnormalities, more recent biotechnologies add a new dimension to traditional cytogenetic analysis. Cytogenetic procedures may be used to study cell cycle associated phenomenon, gene amplification (homogeneous staining regions and double minutes), clastogen challenge, chromosome breakage syndromes, chromosome fragile sites, and polymorphisms to monitor organ/tissue transplantation. DNA probes that are site specific are used for identification of chromosome abnormalities by in situ hybridization using chromosome specific DNA probes and for purposes of gene mapping. Clinicians request cytogenetic analysis for individuals with clinical findings or a medical or family history indicative of a chromosome abnormality. Requests for cytogenetic studies have increased in the last decade due to the demand for prenatal chromosome testing and because of the widespread use of chromosome analysis in cancel diagnosis and monitoring. An accurate cytogenetics analysis is essential for the clinician to make an accurate clinical genetic diagnosis and prognosis for patient management, pregnancy planning and prenatal diagnosis. An incorrect diagnosis, both false positive and false negative, will have far reaching medical and legal implications. Therefore it is imperative that any device used for purposes of cytogenetic analysis is sensitive, specific, safe, and effective. B. Historical Background The correct chromosome number in humans, 46, was determined in 1956 by Tjio and Levan2. Prior to this date the normal human chromosome number was considered to be 48. With this new knowledge came the recognition that certain chromosome abnormalities were associated with specific congenital defects. Improved techniques for handling mitotic chromosomes awakened interest in human cytogenetics. In 1959 Lejeune, et al., described the first chromosome abnormality associated with a clinical syndrome, trisomy 21 in the Down syndrome3. With the advent of chromosome banding techniques in the 1970s, it was possible to identify with certainty all chromosome pairs and to characterize more accurately abnormalities of chromosome number and structure. For a given banding technique, each chromosome pair exhibits a unique pattern of differential staining along the length of the chromosome. The traditional method for analyzing chromosomes is labor intensive. By the early 1960s, development of instruments for 4 automated analysis of metaphase chromosomes was well under way. The primary objective was to develop automated systems which perform as well as a cytogeneticist using conventional techniques and are faster and more cost effective. Because of the rapid increase in the work load of the cytogenetics laboratory and the improved computer capabilities for image processing, these devices are being used with increased frequency to automate parts of the manual procedure.5,6 Various computer hardware and software features are available which are designed to assist in one or more steps of the process outlined in II.C. C. Basic Steps in Routine Cytogenetic Analysis7,8,9 1. Specimen Types and Cell Preparation. (Automated chromosome analyzers are not involved in this step.) a. Obtain a specimen with a large population of cycling (dividing) cells Specimens that contain rapidly dividing cells (e g , bone marrow, solid tumors and chorionic villus) may be harvested without culturing (direct method). Other types of issue (e g., peripheral blood lymphocytes, skin fibroblasts and amniotic fluid cells) must be cultured in a nutritive media and controlled environment before harvesting. b. Harvest the cells Arrest cell division at the metaphase stage of the cell cycle by adding colcemid or a similar agent that inhibits spindle fiber formation. Treat the cells with a hypotonic agent to swell the cells and to facilitate a better spread of the metaphase chromosomes when they are dropped onto a microscope slide or are grown on coverslips. Treat them with a fixative to kill the cells, clarify the chromosome morphology, and enhance the basophilic property of the chromosomes. c. Prepare the slides and stain the cells 2. Selection and Analysis of cell with Metaphase Chromosomes (Automated Chromosome Analyzers may modify or aid in one or more of these processes.) a. Select a predetermined number of metaphase spreads of suitable quality for study and count the chromosomes to determine the modal chromosome number. b. Select several representative spreads for detailed analysis. The number of cells selected varies depending on individual laboratory practices and the clinical indication for testing. c. Photograph the representative metaphase spreads and make an appropriate number of photographic prints. d. Cut individual chromosomes from the prints. Pair and arrange them in a standard format following the International System for Human Cytogenetic Nomenclature (ISCN), 1985 guidelines10. e. Prepare a final report using standard nomenclature that includes a summary and interpretation of the reservations and the number of cells counted and analyzed. Send the report to the referring clinician. Keep permanent records including a copy of the final report, the original metaphase images and karyotypes, and the microscope slides on file for a period of time determined by the individual institution or other regulatory agency. II. Device Description A. General Principles and Features Key issues in the review of these devices center on specific intended use statements and claims dependent on the type of device manufactured. The following features/capabilities are representative of devices currently marketed or being developed in the United States for clinical use. Describe fully these and any other features for which claims are made in the labeling section of the 51OK submission. Work Station Hardware/Software Electronic camera with automatic focusing Monitor (color, monochrome) Word Processor Special Decision Making Features Automatic karyotyping Metaphase finding Chromosome finding Satellite analysis in situ hybridization analysis Sister chromatin exchange (SCE) analysis Automatic cutting and/or separation of chromosomes Image manipulation Positioning Rotating Moving Eliminating artifact (dodging) Enlarging selected metaphases, chromosome pairs, or individual chromosomes Enhancing Contrasting Image capturing process by a TV camera Resolving gray scale (levels of grayness) Printer interface Microscope interface Capacity for networking Data storage mechanism and capacity Ability to recognize and analyze prophase chromosomes Staining methods the instrument can utilize Chromosome recognition capability Training feature B. Description of Specific Features: 1. Metaphase Finders/Scanners aid the cytotechnologist to locate rapidly suitable metaphases for analysis. These instruments automatically scan the microscope slide to locate likely metaphase spreads. The instrument may rank the metaphase cells according to quality and store their microscope slide coordinates in the computer's data base. Metaphase finding/scanning instruments are not always accurate. In some cases their use is limited to specific types of staining and they may not be appropriate for use with certain types of disorders. 2. Chromosome Counters determine the chromosome number by automatically counting the number of chromosomes in a given metaphase spread. 3. Photomicroscopy and Photographic Dark Room Processes eliminate the need for photomicroscopy, photographic dark room work and cutting and pasting chromosomes when performing karyotyping. The process uses digital image processing to digitize the metaphase images by dividing the picture into a grid of pixels. The resolution and detail is determined by the number of pixels in the image and the range in the level of contrast (grayness). The level of grayness may theoretically range from 0 to 256. Optical information about each pixel as well as its location may be processed and stored in the computer data base. The metaphase chromosomes are manipulated (cut) and arranged (pasted) in pairs on the karyotype card (projected onto the computer monitor). Some instruments are designed to do the "cutting and pasting" automatically (see section, B.4.b., below) - others require the operator to manipulate the images (see section B.4.a., below). 4. Interactive and Automatic Karyotyping a. Interactive systems have no decision making ability and depend on the operator to classify the chromosomes from the computer screen and arrange them on the computerized karyotype sheet. b. Automatic karyotyping systems exist with varying amounts of decision making ability. Chromosomes are classified on the basis of chromosome dimension (e.g., ratio of short arm to long arm) and banding pattern profile. 5. Enhancement, Alteration and Manipulative Features. Some instruments have features that enhance or contrast chromosome images in metaphases and/or individual chromosomes to improve the banding characteristics of the chromosome(s). Other features alter or allow the operator to alter chromosome morphology and/or other cellular characteristics. These features include: straightening, enlarging, trimming, "mirror-image", "enhancement" (selectively altering staining pattern within a metaphase spread or within a chromosome region), and dodging/lifting the cytoplasmic background. 6. Specialized Analysis (in development) A few instruments have the ability to perform analysis of specialized studies such as chromosome satellites, in situ hybridization and sister chromatid exchange (SCE). Some chromosome analyzers automatically count the number of satellites or SCEs per metaphase cell or the number of hybridized probe sites per interphase or metaphase cell. Although this feature is currently being used for research purposes, it has not been cleared by the FDA for clinical use. 7. Hard Copy Prints. Most systems are capable of producing near photographic quality, printer generated hard copy of the metaphase images and karyotypes. 8. Generating Reports. Many instruments have capabilities of generating (preparation & printing) a final summary for the referring clinician and of handling the billing process and other bookkeeping. 9. Computerized Patient Data Storage, Retrieval and Archival Systems. A data base may include metaphase images and karyotypes, patient identifying information, and final reports. 10. Training Feature (in development). This feature permits the operator to teach the system to recognize the chromosomes preparations particular to a given laboratory. It allows the operator to "train" the instrument to recognize different staining preparations. Although this feature is currently being used for research purposes, it has not been cleared by the FDA for clinical use. 11. Telecommunication Features allows for site-to-site image transmission. 12. Networking features provide networking between workstations and local areas. III. Specific Performance Characteristics Support specific parameters of importance to the operation of the instrument by data generated with the device. Demonstrate that the device is substantially equivalent to a legally marketed predicate device. Conduct performance studies to demonstrate that the device is safe and effective by comparing the device's performance to the manual reference method of chromosome analysis. See section II.B., "Device Description" for details of performance characteristics required for specific feature. Address all aspects of performance characteristics as stated in section III.C. in the Performance Characteristics section of the Labeling. Provide the following specific information on reproducibility/precision and accuracy for instruments with features that warrant such studies. Include a detailed study protocol, generated data and statistical analysis of the data in any submission to the include a summary of the performance data in the Performance Characteristics section of the Labeling A. Analytical/Laboratory/in vitro Studies 1. Reproducibility Studies Study a sufficient number of control specimens and test specimens with the types of chromosome abnormalities or characteristics for which claims are made (e.g., normal, aneuploidies, structural rearrangements, fragile sites, sister chromatin exchange, etc.) to demonstrate: a. Within Sample Reproducibility Does the device give the same results on repeated trials (analysis) of a given procedure? e.g., does it locate the same cells on a given slide, rank cells in the same manner, give the same chromosome count on a given cell, generate the same karyotype for a given cell, etc. b. Between Instrument Reproducibility Do different devices give the same results for a given procedure? e.g. do they locate the same cells on a given slide, rank the cells in the same manner, give the same chromosome count on a given cell, generate the same karyotype for a given cell, etc. 2. Comparison Studies Comparison studies provide data on the ability of the instrument to determine accurately specific results as compared to the manual method for chromosome analysis. (If a method other than the manual method is used justify the choice of the method and include pertinent references.) Perform the test on a sufficient number of specimens with and without chromosome aberrations and calculate the following parameters: a. Relative Diagnostic Sensitivity: the probability that the instrument will correctly identify an abnormality determined to be abnormal by the reference method. b. Relative Diagnostic Specificity: the probability that the instrument will correctly identify as normal a specimen determined to be normal by the reference method. 3. Specifications Describe the relative quality: Does the instrument achieve the same (or better band resolution compared to the reference method according to the ISCM (1985) guidelines10? B. Software Documentation All computer software should comply with the FDA's Policy for Regulation of Computer Products. For general information contact the FDA Division of Small Manufacturers (phone, 800-638-2041). For specific information contact the Division of Product Surveillance (phone, 301-427-8156). C. Special Considerations for Specific Features Described in II.B. All devices that have any decision making features should also have a feature which allows the operator to interact, edit and override the work generated by the device. Several features of imaging analysis computer systems need special consideration which are addressed below. For each of the following features, provide reproducibility and comparison studies, specifications, and software documentation unless otherwise specified. Incorporate the following considerations as appropriate in the Labeling (e.g., Intended Use, Methods, Limitations, etc.). 1. Metaphase Finders a. State which staining methods may be used with the device. State what types of preparations may be analyzed (e.g., air dried, grown on coverslips, primary colonies, etc.) b. Provide data to demonstrate that the device does not introduce bias in selection of metaphases with respect to chromosome number (e.g., aneuploidy, polyploidy), endoreduplication, poor chromosome morphology as in some malignant cells, structural chromosome abnormality (e.g., translocations, dicentrics, fragments, etc.), or chromosome staining factors. c. Provide data to demonstrate the performance of the metaphase ranking feature. State in the Labeling whether metaphase finders will or will not detect certain cellular abnormalities such as micronuclei, nucleolar organizing region (NOR) alterations and other abnormalities of the cytoplasm or nucleus which would be noted by an astute observer. d. Since these instruments are not always accurate for finding and ranking metaphases, place the following or (similar) statement in the Limitations Section of the Labeling. "The cytogenetic technologist/ cytogeneticist should always review slides independently of the metaphase scanner/finders." e. State whether or not it is appropriate to use the metaphase finder for studies such as drug sensitivity, tumor hard tissue, etc. State any limitations in the Limitations section of the Labeling. 2. Chromosome Counters a. State which staining methods may be used with the device. b. State limitations imposed by how well the chromosomes are spread (e.g., overlapping chromosomes, too much spread, broken cells, too many metaphases in one location, etc.). c. Provide data to demonstrate that bias is not introduced into chromosome counts by aneuploidy, polyploidy, endoreduplication, radial formations, chromosome pulverization, poor chromosome morphology as in some malignant cells, and structural chromosome abnormality (dicentrics, fragments, etc.). If any incorrect chromosome counts result, this should be declared in the limitations section of the package insert. d. Include the following or similar statement in the Limitations Section of the Labeling: "In general there will be one or more errors in determining the correct chromosome number for the population of cells studied. It is the responsibility of the operator to determine the correct modal chromosome number." 3. Interactive Karyotyping Systems (no decision making ability) Reproducibility and comparison studies are not required since this feature has no decision making ability. 4. Automated Karyotyping Reproducibility and comparison studies are required only for the decision making features of the device. Include the following or similar statement in the Limitations section of the Labeling: "In general, there will be one or more errors in the computer-generated karyotype Therefore, it must always be examined and edited as a final interactive manual operation by a qualified cytogeneticist or cytogenetic technologist." 5. Enhancement, Alteration and Manipulation Features Altering chromosome morphology in any way (aside from improving culturing and staining techniques) is not an accepted standard of professional practice in cytogenetic analysis. a. Comparison data are not required for enhancement, alteration or manipulation features. b. Any device with features that in any way alter chromosome morphology should also have a built-in feature that automatically and permanently marks/designates these alterations in the karyotype. c. Digitized straightening of chromosomes will often artificially induce extra bands and thus make it impossible to determine, unequivocally, whether the straightened chromosome is, indeed, normal. Designate any straightened chromosomes as stat. d. Assure that the device does not have the capacity to induce artifacts. e. Features that "lift" cellular background should not be automatic and should have a built-in feature that requires operator activation. Such a feature should not be used unless the operator has already examined the cell to assure that the material to be lifted is true artifact and not chromatin material such as double minutes. f. Include a statement about the potential for misuse of each feature under the Limitations section of the Labeling and elsewhere in the operators manual whenever use of these features are described. Use the following or a similar statement about features listed in this section: "It is the responsibility of the cytogenetics technician and/or the cytogeneticist to utilize all features in compliance with standard laboratory practice and regulatory guidelines." 6. Photomicroscopy and Dark Room Process a. Reproducibility and comparison data are not required. b. Compare the resolution of the digitized images to standard microscopic resolution in terms of the resolution standards (400, 500 and 850 bands) of the ISCN, 1985 guidelines10. c. For devices that have no decision making ability, include the following or similar statement in the Intended Use section of the Labeling: "The device does not locate metaphase spreads; it does not rank the given cells according to quality; it does not automatically classify chromosomes; it does require and relies completely on the operator to manipulate the digitized microscope images." 7. Hard Copy Prints a. Reproducibility and comparison data are not required. b. Describe the quality and resolution of the hard copy print in the Principle of the Procedure section of the Labeling. State whether the quality/resolution of the computer generated print is equivalent to (as good as) standard photomicroscopy (levels of gray) and whether the device meets the minimum resolution standard set by the ISCN, 1985 guidelines. c. Describe the composition, quality and durability of the photo image paper and whether contrast deteriorates with age. i. A precaution/warning statement in the Precautions section of the Labeling will be required for paper containing mercury or other toxic substances. ii. State how long the prints will remain of archival quality. The durability of these hard copy prints is important since some proficiency/licensing agencies require long term storage of metaphase images (up to 25 years for New York State). 8. Computerized Patient Data Storage, Retrieval and Archival System. a. Reproducibility and comparison data are not required. b. This feature should comply with section III.B. of this document. The following issues are of special concern and should be addressed in the Principles of Procedure section of the labeling: i. adequate security control which may require several levels of "password" security to assure protection and confidentially of patient information; ii. adequate safe guards to protect against accidental or virus generated deletion of the data iii. an explanation of where and how the data are stored; and iv. a recommendation of multiple identifying codes (e.g., for patient identifying data, metaphases, karyotypes, final report, etc.) to assure correct and usable storage and retrieval of information from the data base and archival system. IV. Labeling Considerations The Labeling (Operator's Manual or Package Insert) should include all information listed in the in vitro diagnostic Labeling regulations 21 CFR ~ 809.10(b)(6) plus additional pertinent headings found under 21 CFR ~ 809.10(b): A. Intended Use Statement Describe concisely the functions/features of the device. State clearly that a qualified cytogenetic technologist and/or cytogeneticist must edit and/or confirm all computer-generated data/results and make the final judgment/decision. For devices that have no decision making ability, include the following or similar statement in this section: "The device does not locate metaphase spreads; it does not rank the given cells according to quality; it does not automatically classify chromosomes; it does require and relies completely on the operator to manipulate the digitized microscope images." B. Limitations of the Device Include the following: 1. a statement that decision making capabilities of the instrument do not relieve the cytogeneticist/cytogenetics technologist of the responsibility to review and edit all work generated by the device and that the final decision must be made by a qualified cytogeneticist; 2. appropriate statements of precaution outlined for each feature in the Performance Characteristics section (III-C). In some cases, limitation statements for more than one features may be combined; and 3. these general statements for all devices that have features with decision making ability: "In general, there will be one or more errors in the computer-generated data/karyotypes. Therefore, it must always be examined and edited as a final interactive manual operation by qualified cytogenetic personnel." "The final clinical diagnosis must be made by qualified medical personnel." C. Performance Characteristics Provide a summary of all reproducibility and comparison studies (sensitivity and specificity) when performed as requested by Section III. Also address other performance characteristics that relate to specific features (as described in section III-C). V. Bibliography 1. Letter. September 1989 Cytogenetic Proficiency Test, General Comments. New York State Department of Health. 2. Tjio, JH and Levan, A: The chromosome number in man. Hereditas, 42:1, 1956. 3. Lejeune, J, Gautier, M, and Turpin, MR: Etude dais chromosomes somatiques de neut enfants mongoliens. C.R. Acad. Sci. (Paris) 248, 1721-1722, 1959. 4. Lubs, HA and Ledley, RS: Automated analysis of differentially stained human chromosomes. Nobel. 23:61-76, 1973.] 5. Lifshitz, MS and DeCresce, RP: Genetiscan Digital Karyotype System. Lab Med, 18(6):402-403, 1987. 6. Bender MK: Karyotyping System with Special Reference to the KARYOTEC 100. Presentation: 34th Annual Scientific Meeting of The Royal College of Pathologists of Australia. Hong Kong, 1989. 7. Rooney DE and Czepulkowski BH: Human Cytogenetics: A Practical Approach. Oxford, England: ILR Press; 1986. 8. Verma RS and Buba A. Chromosomes: A manual of Basic Techniques. New York, NY: Pergamon; 1989. 9. Priest JH: Medical Cytogenetics and Cell Culture. 2nd ed. Philadelphia, PA: Lea and Fabiger; 1977. 10. Harden DG and Klinger HP: An International System for Human Cytogenetic Nomenclature (1985). Basel, Switzerland: S Karger; 1985.
This document was still considered current as of July 1997.
It will be reviewed again in July 1998.