 |
|
|
 |
 |
|
Applications of Fluorescence in Situ Hybridization in Cytopathology
A Review Andre M. Oliveira, M.D., and Christopher A. French, M.D.
Fluorescence in situ hybridization (FISH) is a molecular cytogenetic technique that is a powerful and versatile research tool and an important adjunct to tumor diagnosis and prognosis. Akin to the recent role of immunohistochemistry, research and clinical applications of FISH in cytopathology have been growing enormously over the last few years. Furthermore, due to its feasibility for virtually all types of cytologic specimens, a new and exciting era in both investigative and clinical cytopathology is expected. (Acta Cytol 2005; 49: 587–594) Keywords: hybridization, fluorescence in situ; FISH; molecular genetics.
Mammalian cytogenetics, which came to fruition in the 1950s,1 initiated a new era in terms of how pathologists and scientists look at and define diseases. It became evident that specific chromosomal aberrations were characteristic of certain tumors and could be used as diagnostic or prognostic markers.2-5 Today, traditional cytogenetic analysis remains a powerful tool because it reveals major chromosome aberrations and is used as a starting point for several lines of investigation. Fluorescence in situ hybridization (FISH), whereby fluorescent DNA probes are used to label defined chromosomal regions, was first developed for the chromosomal localization of a certain type of DNA, called satellite DNA.6 This technique could be used not only on metaphases but also on interphase nuclei to identify chromosomal aberrations more precisely. Though initially a research tool, FISH soon became a targeted approach to help diagnose certain diseases7,8 and in some cases provide important prognostic information.9 Until recently, FISH had limited applicability to cytologic specimens because methodologies had not been developed for its use on various specimen types. A number of methods were published in the 1990s, however, that provide reliable approaches to the use of interphase FISH on essentially all cytologic specimens.8,10,11 This review outlines the various uses of DNA FISH that are relevant to the practice of cancer cytopathology.  FISH is expected to greatly enhance the value of cytology as a screening, diagnostic and research tool, integrating it further with the evolving field of molecular pathology. Basic Principles
FISH, like several other molecular techniques, is based on the ability of complementary single-stranded nuclei acid molecules (DNA or RNA) to hybridize to each other.12 The most important feature of FISH is its ability to localize specific DNA sequences in intact nuclei and chromosome spreads. Since its inception, in the early 1970s,1 and its first clinical applications, in the early 1990s,13 FISH has been used for a variety of investigative and clinical purposes, from gene mapping studies to the identification of chromosome abnormalities in genetic disorders and cancer.
The FISH procedure essentially requires 2 types of DNA, target and probe. Target DNA is a heterogeneous population of DNA molecules immobilized on a solid surface (e.g., glass slides) as either chromosome spreads (e.g., metaphase spreads) or interphase cells (e.g., tissue sections or cytologic preparations). Probe DNA, by contrast, is a population of labeled DNA molecules specific to certain chromosomes, chromosome regions or genes.14 Probe DNA is made visible under the microscope by a procedure called nucleic acid labeling.15,16 In direct labeling, modified nucleotides containing a covalently attached chemical group that emits light (or fluorescence) when excited by light of a specific wavelength, known as fluorochromes, are incorporated into the probe DNA by an enzymatic reaction. Commonly used fluorochromes include fluorescein (green signal), rhodamine (red signal) and Texas Red (red signal). In indirect labeling, nucleotides containing covalently attached molecules, known as reporter molecules, are incorporated into the probe DNA and detected by an immunologic reaction using fluorescent-labeled antibodies directed against the reporter molecule. Commonly used reporter molecules include biotin and digoxigenin.12
FISH can be performed on either fresh or archival tissues. Unstained cytologic preparations, including air-dried smears and cytospins,10,17 ThinPrep (Cytyc Corp., Boxborough, Massachusetts, U.S.A.) slides,8 alcohol-fixed smears and cytospins,10,18 and paraffin-embedded cell blocks11 are very amenable to FISH analysis. FISH on destained archival preparations can be successfully performed but with less consistent results.19,20
A standard FISH protocol can be divided into 5 basic steps: (1) DNA unmasking, (2) probe and target DNA denaturation, (3) probe-target DNA hybridization, (4) detection, and (5) image analysis on a fluorescence microscope (Figure 1). The DNA unmasking step varies according to preparation and cell type and can involve minimal unmasking (cytogenetic preparations) or extensive unmasking, including digestion with proteases and boiling in concentrated sodium citrate or EDTA buffer (e.g., formalin-fixed, paraffin-embedded tissue).11 The detection step varies according to whether the labeling procedure is direct or indirect.
FISH probes can be of different types. The three most common types are alpha-satellite, telomeric and locus-specific probes.21 Alpha-satellite probes hybridize with DNA sequences present in the pericentromeric region of each chromosome.22 These sequences are comprised of 0.1–5-megabase-long tandem repeats of 171 base pair, A-T rich units, known as alphoid monomers; these vary from chromosome to chromosome. Alpha-satellite probes are used mainly for aneusomy and sex-determination studies. Telomeric probes are specific for TTAGGG(n) repeats of up to 10 kilobases, present at the terminus of all chromosomes.23 These probes have been used for the identification of chromosome terminal deletions. Locus-specific probes are specific to certain chromosome regions, groups of contiguous genes or even single genes; they are commonly used to identify gene amplifications, gene deletions or chromosome rearrangements. Currently, the most commonly used locus-specific probes are derived from yeast artificial chromosomes (300–1,500 kb)24 and bacterial artificial chromosomes (100–300 kb).25
Clinical Applications There exists a large variety of clinical applications for FISH. The following discussion addresses mainly its application for cancer diagnosis and prognosis and its relevance to cytology.
Gains and Losses FISH is commonly performed for the identification of abnormal chromosome numbers in cancer.26 Common examples include the identification of trisomy 8 in many hematologic tumors and sarcomas and trisomy 12 in chronic lymphocytic leukemia (CLL).27-29 These abnormalities are detected by using centromere-specific probes and may provide important prognostic information.30,31 From the perspective of diagnosis, as an example, substantial evidence suggests that detection of numeric chromosomal abnormalities; gain of chromosomes 3, 7 and 17; and loss of 9p21 help increase the sensitivity of urine cytology for the presence of transitional cell carcinoma.32-34
Chromosome Rearrangements Identification of chromosomal translocations by FISH has been used frequently for tumor diagnosis. The best known example is the detection of the t(9;22)(q34;22q11) in chronic myelogenous leukemia (CML). This cytogenetic abnormality occurs in >90–95% of CML and in a small subset of acute lymphoblastic leukemias and results in the fusion of the BCR to the ABL gene.35,36 A typical FISH strategy to detect this translocation involves the use of 2 differently labeled probes (dual-color FISH) for both BCR and ABL loci in a single experiment.37,38 The appearance of a fusion signal strongly suggests the presence of BCR-ABL and is diagnostic of CML (Figure 2).
From the cytopathologist’s perspective, frequent differential diagnoses can be resolved by FISH. In small round cell tumors, t(11;22)(q24;q12) identifies Ewing’s/primitive neuroectodermal tumor.13,39 In spindle cell tumors, t(X;18)(p11;q11) is characteristic of synovial sarcoma.40 In low to intermediate grade lymphoproliferative disorders, t(11;14)(q13;q32) is diagnostic of mantle cell lymphoma. t(14;18)(q32;q21) identifies follicular center cell lymphoma.41 In high grade lymphomas, rearrangements of MYC on chromosome 8 with either light or heavy chain immunoglobulin genes is diagnostic of Burkitt’s lymphoma.13,42 Nevertheless, these studies are not routinely performed, largely because of historical technical limitations. Now that FISH is becoming a more robust technique for most cytologic preparations, these assays are expected to be used more frequently, especially for difficult cases.
Gene Amplification Gene amplification is an oncogenic mechanism observed in most types of cancer.43-45 FISH has been used to detect gene amplification because the results may possess prognostic and therapeutic implications. FISH for autosomal gene loci on normal cells produces only 2 signals, which correspond to the 2 normal alleles. In cases of gene amplification, FISH produces multiple, often clustered signals (Figure 3). Some examples include the detection of multiple gene copies of HER2 in breast adenocarcinoma, C-MYC in carcinomas and leukemias (Figure 3), and N-MYC in neuroblastomas. The adopted strategy to identify gene amplification includes the use of probes specific to the gene locus or chromosome segment that is commonly amplified (amplicon). Recently, a good correlation of HER2 amplification in breast cytology and paired resection sections has been demonstrated,46,47 illustrating the potential utility of fine needle aspiration for prognosis and therapy in primary or metastatic breast cancer. Similarly, N-MYC and chromosome 17q gains not only provide important prognostic information for neuroblastoma but are an important means to identify rare cells disseminated to bone marrow not detectable by conventional cytomorphology.48,49
Gene Deletions
Inactivation of tumor suppressor genes, such as p53 and RB, is another rather commonly observed genetic event in cancer.50-53 Several mechanisms of tumor suppressor gene inactivation have been reported, including loss of a chromosome segment or locus containing the normal allele of a tumor suppressor gene that has already been inactivated in the homologous chromosome, usually by a point mutation.54 This is referred to as loss of heterozygosity and can be detected with FISH probes specific to the commonly deleted regions (Figure 4). FISH deletional analysis is particularly useful for cytologic specimens in which cytomorphologic overlap between reactive and neoplastic cells makes the diagnosis challenging. For example, 9p21 deletion can help identify transitional cell carcinoma in urine cytology,33 and deletion of 1p, 3p or 22q is potentially useful for the detection of malignant mesothelioma.55
Disease Monitoring
The identification of residual malignant clones after chemotherapy for cancer can be extremely difficult by cytomorphology alone. Many of the strategies described above can be used to detect residual malignant clones with either numeric or structural chromosome aberrations. Identification of residual disease has been performed largely for hematologic malignancies and some small round cell tumors, including neuroblastoma.56,57 For example, abnormal and neoplastic cells of myelodysplastic syndrome and CML, especially after treatment, can be indistinguishable from normal bone marrow elements. In this circumstance, FISH can identify lesional cells that are otherwise morphologically normal. This is relevant to management of CML because reduced numbers of Philadelphia
chromosome–positive cells correlates with longer survival and better response to therapy.58 FISH, which is less sensitive than reverse transcriptase polymerase chain reaction for the detection of rare neoplastic cells, is generally not used to detect minimal residual disease when monitoring response to imatinib.59 Future Directions Chromogenic in Situ Hybridization (CISH)
CISH is a newer technique that is based on the same principles as FISH is except that the final image analysis can be performed with a regular light microscope.60-63 CISH differs from FISH in the detection step. Instead of using fluorochromes, CISH detection is carried out using similar systems that are commonly used in regular immunohistochemical reactions; DNA probes labeled with either biotin or digoxigenin (same probes used for FISH) are detected using streptavidin–horseradish peroxidase and anti–digoxigenin– alkaline phosphatase, respectively. Because the image analysis can be done with a common light microscope (bright field), correlative analysis between architectural and cytologic features and molecular alterations, which is rather challenging with FISH, can be readily performed by a cytopathologist at his/her own microscope. CISH can also be combined with immunohistochemistry, thereby allowing simultaneous analysis of molecular alterations and resultant protein expression.62 A remarkable demonstration of this is the concurrent presence of amplified Her2/neu gene copies detected with chromogen-labeled locus-specific probes and overexpression of Her2/neu protein detected by immunohistochemistry (Figure 5). Another important advantage of CISH is that slides can be archived for very long periods because probe signals are not subject to fluorescent fading. The main disadvantages of CISH as compared with FISH include its extended detection protocol and, due to limited available chromogens, poor color differentiation when used for dual-color analysis.
Multicolor FISH
Multicolor FISH, whereby multiple centromeric and/or locus-specific probes of different colors are used on a single slide (with multiple colors produced by several combinations of a limited number of fluorochromes, such as red, green, aqua and gold) to assess a panel of chromosome aberrations, is a potentially powerful technique.63-68 Unfortunately, the complexity of signals requires computer interpretation. Nevertheless, multicolor FISH has been demonstrated to improve sensitivity of urine screening for bladder cancer, cohybridizing with a combination of probes to chromosomes 3, 7 and 17 and 9p21.34
Several newly emerging techniques based on the same principles as FISH may eventually be applied to cytologic specimens, but at the present time they remain basic research tools.
Conclusion Currently, most cytologic specimens are readily amenable to FISH analysis, and the same FISH techniques that are applied to surgical pathology specimens are equally applicable to cytologic specimens. FISH can also add substantial sensitivity and specificity to the standard cytology approach for the identification of neoplastic cells; this is especially important when dealing with limited cellular material. Thus, FISH is expected to greatly enhance the value of cytology as a screening, diagnostic and research tool, integrating it further with the evolving field of molecular pathology.
Acknowledgments We thank Jonathan A. Fletcher, M.D., for his helpful suggestions.
References
From the Department of Pathology, Brigham and Women’s Hospital, Harvard Medical School, Boston, Massachusetts, and theDepartment of Laboratory Medicine and Pathology, Mayo Clinic and Mayo Foundation, Rochester, Minnesota, U.S.A.
Dr. Oliveira is Assistant Professor, Department of Laboratory Medicine and Pathology, Mayo Clinic and Mayo Foundation. Dr. French is Assistant Professor, Department of Pathology, Brigham and Women’s Hospital. Address correspondence to: Christopher A. French, M.D., Department of Pathology, Brigham and Women’s Hospital, 75 Francis Street, Boston, Massachusetts 02115 (cfrench@partners.org). Financial Disclosure: The authors have no connection to any companies or products mentioned in this article. Received for publication July 20, 2005. Accepted for publication July 27, 2005.
 0001-5547/05/4906-0587/$19.00/0 © The International Academy of Cytology
Acta Cytologica Send website comments to Webmaster@acta-cytol.com |
||||||||||||||||||||||||||
100). Courtesy of A. Roy, Department of Pathology, Brigham and Women’s Hospital.
