Cytogendies, in situ Hybridization and Molecular Approaches in the Diagnosis of Cancer*

Transcription

1 ANNALS OF CLINICAL AND LABORATORY SCIENCE, Vol. 28, No. 6 Copyright 1998, Institute for Clinical Science, Inc. Cytogendies, in situ Hybridization and Molecular Approaches in the Diagnosis of Cancer* ARMAND B. GLASSMAN, M.D. Olla Stribling Distinguished Chair fo r Cancer Research, U.T. M.D. Anderson Cancer Center, Houston, TX ABSTRACT The past 100 years represent almost the entire history of the recognition of the role of genetics in human cancer. The purpose of this work is to: 1) review that history; 2) explore the techniques that have brought cancer genetics to its present state of knowledge; and 3) to provide preliminary data on how cytogenetics, fluorescence in situ-hybridization (FISH) and molecular techniques contribute to the diagnosis of chronic myelogenous leukemia (CML). Conventional cytogenetics provided the first chromosomal marker for malignancy in This was to be known as the so-called Philadelphia chromosome. Additional chromosomal changes associated with various hematologic malignancies followed in the 1970s after chromosomal banding was perfected. FISH, polymerase chain reaction (PCR), reverse transcriptase PCR (RT- PCR) and other molecular techniques followed. Using these techniques, the diagnosis and prognosis of CML continue to evolve. The current study evaluates 21 patients by conventional cytogenetics and compares the percentage of t(9;22) metaphases with FISH information on the bone marrow and peripheral blood specimens of the same patients. The correlation of the techniques in this small study is 100 percent. Depending on the cutoff for abnormal FISH, 10 percent or less of the FISH studies on bone marrow or peripheral blood are false negatives. Conventional cytogenetics is used as the present gold standard. FISH and molecular techniques are complementary and will provide additional significant information in the future. Introduction Genetic concepts (eg, avoiding family intermarriages) date from ancient times. Mendel in the mid-1800s defined general genetic prin * Address Reprint Requests to: A. B. Glassman, M.D., Pathology/Laboratory M edicine, U.T. M.D. Anderson Cancer Center, 1515 Holcombe Blvd., Box 73, Houston, TX This work is supported in part by the Olla Stribling Fund. 324 ciples through plant experiments. Late in the 19th century a German pathologist suggested that cancer and morphologic abnormalities of nuclei of neoplastic cells were related.1 Boveri, in 1914, expanded thinking of the role of genetics in the development of cancer.2 D uring the 1920s and 1930s techniques for the study of chromosomes in plants and insects were evolving. Human chromosome number was reported from counting the chromosomes of spermatogonia in thin slice preparations of the testes of executed criminals by Painter in /98/ $01.75 Institute for Clinical Science, Inc.

2 CYTOGENETICS The preparations were suboptimal and the hum an chromosome num ber was proposed, with some hedging, to be 48. Not until refined methods for hum an chromosome preparation were available in the mid-1950s, did Tijo and Levan establish the human chromosome complement as 46.3 Levan, in 1956, described chromosomal abnorm alities of human tumors and proposed applications.4 N ow ell an d H u n g e rfo rd a c c u ra te ly described the first specific chromosomal abnormality associated with cancer in I960.5 This came to be known as the Philadelphia chromosome, a shortened chromosome 22. Identification of human chromosomes was still primitive and it was not until the 1970s that banding techniques would permit identification of the translocation of genetic materials from chromosome 22 to chromosome 9 (table I). Originally, the translocation of chromosomes 9 and 22 [(t(9;22)] was thought to occur only in CML. It is now recognized to occur in some childhood acute lymphocytic leukemias (ALL) and occasionally in patients with acute myelogenous leukemia (AML). Refinements in cytogenetic and related molecular techniques have since became invaluable adjuncts to the diagnosis of hematologic malignancies.6 Other chromosomal abnormalities of cancer are noted. Heim and Mitelman6,7 divide these genetic changes into primary aberrations, frequently found as the sole kaiyotypic abnormality of a specific cancer type and secondary, rarely found alone and developing in cells that already have a prim ary abnormality.8 The Philadelphia chromosome found with CML is a primary aberration. The development of additional chromosomal changes in CML (eg + 19, +8, +22, il7 ) represents secondary changes known as clonal evolution. Cytogenetic noise is a term used for extensive nonclonal abnormalities such as is seen in many solid tumors. O b je c t iv e s The purpose of this work is to: 1) review briefly the past 100 years of human cancers through the perspective of the contributions of genetics; 2) explore some of the history of the methods of genetics; and 3) provide data using current cytogenetic techniques that contribute to the diagnosis and follow-up of CML. M e t h o d s From the initial counting, classification by size, p (short) and q (long) arm relationship of chromosomes in the 1950s and 1960s, methods for further identifying specific chromosome by banding evolved in the 1970s.9,10 In situ hybridization was first described in A radionuclide labeled complementary DNA applied to metaphase chromosomes was detected by autoradiography. Present techniques use a fluorescent reporter molecule to perform fluorescence in situ hybridization (FISH). FISH uses either site specific or general sequence probes. D epending on the 1894 Von Hausemann 1914 Boveri 1926 Painter 1954 Tijo and Levan 1956 Levan 1960 Nowel and Hungerford 1970 s 1980's and 1990's 1990's and beyond TABLE I One Hundred Years: Genetics and Cancer Nuclear abnormalities of cancer Cancer developed from genetic abnormalities 48 chromosomes in humans 46 chromosomes in humans Chromosomal abnormalities in cancer Philadelphia chromosome in CML Chromosome banding Molecular techniques Gene therapy

3 3 2 6 GLASS MAN specificity of the probe, it may be used for the evaluation of metaphase, interphase or both.12 FISH methods and probes are available for a and 3 satellite regions, whole chromosome identification, telomeric and specific gene sites or translocations (eg, bcr/abl).13 Comparative genomic hybridization (CGH) is an in situ hybridization technique where normal and tumor DNA are fluorescently labeled. Then, using special digitalized imaging techniques, labeling of tumor to normal tissue is compared (table II). CGH has potential as a tool for determining chromosomal abnormalities and aiding in the diagnosis of cancer, but has technical and method limitations.14 Conventional karyotyping can detect chromosomal changes of fairly large magnitude, eg, 5 to 10 megabases (mb). FISH techniques using specific probes may provide information on genetic sites of 2 to 5 mb or less. Determining changes at the kilobase (kb) or smaller regions requires more sensitive analytical techniques such as the polymerase chain reaction (PCR). PCR and the related reverse-transcriptase PCR (RT-PCR) are able to provide an analytical sensitivity of chromosomal material from as little as one cell in a million. The PCR technique uses oligonucleotide primers specifically selected to bind to certain DNA regions that are being sought to amplify.15 Amplification of the specific fragment of DNA is accomplished through a heat-stable DNA polymerase. Repeated cycles result in detectable amounts of DNA being generated. Each cycle consists of three steps. They are: first, denaturation of the double-stranded DNA; second, annealing of the primers; and, third, primer extension using heat-stable DNA polymerase. H eat denaturation is accomplished at C, followed by a lower temperature (50-70 C) to permit annealing to allow the oligonucleotide primers (usually 15 to 25 nucleotides long) to anneal to the single stranded DNA. Primer extension results in the synthesis of the complementary DNA by 5' to 3' extension of the DNA: primer hybrids through the action of heat-stable DNA polymerase from the organism Thermus aquaticus (Taq). Typically this procedure is repeated for 20 to 40 cycles, resulting in approximately 1 -million-fold amplification. The DNA fragment size amplified is usually 60 to 4000 bp. While PCR starts with DNA isolated from cells, RT-PCR starts with purified and isolated RNA. Reverse transcriptase and appropriate primers are used to generate complementary DNA (cdna). Then the cdna and primers are used in a PCR procedure. Proper negative and positive controls should be run with the PCR and RT-PCR methods to ensure the validity of the results as misleading results can occur w ith th ese com plex p ro ced u res. Recently additional molecular diagnostic tests for determ ining unknown mutations have evolved from these basic methods.16 The methods used to detect genetic mutations at the molecular level utilize chemical identificaiton of nucleic acid duplex mismatches, electrophoretic separation of singleor double-stranded DNA and/or the sequencing of DNA. Specific techniques include the older Southern (for DNA) and Northern blots. TABLE II Methods in the Study of Chromosomes and Genes 1950 s and 1960 s Chromosome growth and counting 1970 s Banding and modern conventional cytogenetics 1970 s and 1980 s In situ hybridization Gene sequencing and function Polymerase chain reactions and related methods 1990 s Comparative genomic hybridization Human Genome Project

4 CYTOGENETICS Newer approaches include allele-specific oligonucleotide hybridization, allele-specific amplification, ligase amplification, prim er extension and DNA sequencing, microsatellite analysis and positional cloning.17 As genes are cloned and sequenced, it is becoming evident that many (tens to hundreds depending upon the gene) of mutations of the same gene may be involved in disease causation. Often a candidate site for gene cloning is identified through conventional karyotype methods. M utation information has begun to yield important genotype-phenotype correlations, eg, specific BRCA1 mutations associated with a higher incidence of ovarian carcinoma. Advances in molecular genetic information continue to occur. The Human Genome Project (HGP), which had its official beginning in 1990, dates back to That year (1984) discussion in Alta, UT, centered around DNA analysis as a means of detecting mutations in atomic bomb survivors. HGP is scheduled to complete its work by the year The project has begun to produce genetic and physical maps of each human chromosome. A major benefit has been the development of new and improved technologies and methods for DNA sequencing. As a byproduct, certain model organisms (eg, E. coli and S. cerevisiae) will be sequenced as a prelude to human genomic sequencing. Associated concerns include ethical, legal and social issues, education and training of the public and individuals interested in genetic research and technology transfer to related sciences. It is estimated that the HGP is on schedule for completion early in the next century. Will the HGP provide all the answers for genetic related malignancies? Additional inform ation will probably be needed. The HGP will provide wonderful avenues for further understanding, basic investigation and clinical relevance (table II). Applications and Results in CML Applications of cytogenetic techniques in the study of the CML provide the following data. Conventional cytogenetics techniques are credited with the first chromosomal abnormality related to a specific malignancy. This was the association of the Philadelphia chromosome with CML in In 1973, after chromosomal banding had been introduced, the exchange of genetic material between chromosomes 9 and 22 was recorded by Rowley.19 This translocation of chromosomes 9 and 22 [t(9;22)] is noted in approximately 93 percent of the CMLs seen at UTMDACC. Patients with ALL, about 10 percent, and occasional patients with AML also manifest the t(9;22). Molecular biology studies have demonstrated that this translocation yields a fusion oncogene, the result of Abelson (abl) gene activity on chromosome 9 being altered by the presence of the breakpoint cluster region gene (bcr) on 22. Additional work has revealed that the CML break of bcr is in the region of the 12 to 16 exons (Mbcr), while that associated with ALL is in the exon 1 to 2 region (mbcr). The longer bcr break results in an oncoprotein of 210 kd which can be identified by western blotting. A kd protein is the result of the shorter break. The cytogenetics laboratory at UTMDACC has been carrying out a study to examine the utility of FISH in the diagnosis and follow-up of CML. A commercially available probe and method are used. Using this probe resulted in a level of bcr/abl positive cells of 6 ± 1.5 (1 SD) determined by FISH for ambulatory volunteers with normal conventional karyotypes and no clinical evidence of CML. Applying this technique to 21 patients with known CML and conventional karyotypes displaying the t(9;22), the following results were obtained. Conventional karyotypes showed a range of 5 to 100 percent t(9;22) positivity of 21 to 30 metaphase counts for the 20 CML patients. These karyotypes were obtained from bone marrow specimens. FISH performed on the same bone marrow specimens ranged from 7.5 to 95 percent on 200 interphase cells. The 7.5 percent interphase count occurred in the patient with only one of 20 metaphases (5 percent demonstrating t(9;22). The patient with 100 percent metaphases containing t(9;22) had

5 Patients TABLE III CML Comparison of Karyotypes and Fluorescence Insitu Hybridization Karyotypes BM (200 cells) Interphase FISH P>B (200 cells) Interphase FISH 1 46,XY,t(9;22([3] 16.0% Ph+ 14.0% Ph+ 46,XY,[17] 84.0% Ph- 86.0% Ph- 2 46,XX,t(9;22)[17] 76.5% Ph+ 49.0% Ph+ 46,XX[3] 23.5% Ph- 51.0% Ph- 3 46,XY,t(9;22)[2] 26.5% Ph+ 14.5% Ph+ 46,XYI18] 73.5% Ph- 85.5% Ph- 4 46,XX,t(9;22)[6] 13.5% Ph+ 17.5% Ph+ 46,XX,[14] 86.5% Ph- 82.5% Ph- 5 46,XY,t(9;22)[1] 9.50% Ph+ 8.50% Ph+ 46,XY[19] 90.5% Ph- 91.5% Ph- 6 46,XX,del(5q),t(9;22)[7] 55.0% Ph+ 38.0% Ph+ 46,XX[3] 45.0% Ph- 62.0% Ph- 7 46,XY,t(9;22)[19] 85.5% Ph+ 75.0% Ph+ 46,XY,[1] 14.5% Ph- 25.0% Ph- 8 46,XY,t(9;22)[7] 16.5% Ph+ 14.5% Ph+ 46,XY[13] 83.5% Ph- 85.5% Ph- 9 46,XX,t(9;22)[1] 21.0%Ph+ 19.5% Ph+ 46,XX[15] 79.5% Ph- 80.5% Ph ,XX,t(9;22)[14] 46.5% Ph+ 44.0% Ph+ 46,XX,[6] 53.5% Ph- 56.0% Ph ,XY,t(9;22)[20] 47.0% Ph+ 23.0% Ph+ 53.0% Ph- 56.0% Ph ,XX,t(9;22)[1] 21.0%Ph+ 16.0% Ph+ 46,XX[3] 79.0% Ph- 84.0% Ph ,XX,t(9;22)[14] 35.5% Ph 16.0% Ph+ 46,XX,[6] 65.5% Ph- 84.0% Ph ,XY,t(9;22;17),+22q-{17] 95.0% Ph+ 85.1% Ph+ 46,XY[1] 5.0% Ph- 14.9% Ph ,XX,t(9;22)[2] 13.5% Ph+ 13.0% Ph+ 46,XX[14] 86.5% Ph- 87.0% Ph ,XX,t(9;22)[10] 42.0% Ph+ 27.5% Ph+ 46,XX,[10] 58.0% Ph- 72.5% Ph ,XX,t(9;22)[19] 80.5% Ph+ 47.5% Ph+ 46,XX[1] 19.5% Ph- 52.5% Ph ,XY,t(9;22)[15] 67.0% Ph+ 66.0% Ph+ 46,XY[1] 3.5% Ph++ 3.0% Ph++ 47,XY,t(9;22),+22q[1 ] 29.5% Ph- 31.0% Ph- 48,XY+8,t(9;22),+22q-[3] 19 46,XY,t(9;22)[17] 40.0% Ph+ 21.0% Ph+ 46,XY[3] 60.6% Ph- 79.0% Ph ,XY,t(9;22)[2] 7.5% Ph+ 4.5% Ph+ 46,XY[18] 92.5% Ph- 95.5% Ph ,XY,t(9;22)[9] 20.0% Ph+ 22.5% Ph+ 46,XY[11] 80.0% Ph- 77.5% Ph- BM = Bone marrow. PB = Peripheral blood. FISH = Fluorescence in situ hybridization. [ ] = Number of metaphases counted with cytogenetic finding. Ph+ = t(9;22) Philadelphia chromosome positive. Ph- = Philadelphia chromosome negative.

6 CYTOGENETICS percent of the interphases positive for the fusion signals. FISH evaluation of 200 cells from the peripheral blood taken at the same time as the bone marrow specimens had a range of 4.5 to 85.1 percent fusion signal positives. Using the karyotype as the gold standard and the upper limit of acceptable values for the laboratory, there were two possible false negatives by FISH for two patients, each of whom had only one (5 percent positive t(9;22) metaphases by conventional karyotyping. By convention, the same chromosomal abnormality in two of 20 metaphases is required to determine a positive cytogenetic finding (table III). For both of these patients, one karyotype of 20 was considered abnormal because of prior positive t(9;22) studies performed earlier. Correlation of 10 patients with CML by karyotyping, FISH and molecular studies (looking for bcr rearrangem ent and RT- PCR of the fusion protein) revealed consistency throughout. Discussion How these preliminary data will be used to diagnose, follow therapy, look for minimal residual disease or recurrence remains for future determination. The HGP should result in advances in technology and scientific application of genetic information. Genetic information is currently being used for the classification of hematologic and lymphoproliferative malignancies. Future progress will occur through integration of information of genotype-phenotype relationships, understanding of molecular interactions, genetic effects on biochemistry, physiology, immunology and combinations of these factors. The science of genetics will continue to progress. Diagnostic capabilities follow. Therapeutic approaches using gene therapy are evolving. Hematologic and lymphoproliferative disease treatm ent has benefited from these endeavors. Genetic diagnosis of solid tumors such as cancer of the breast, prostate, lung, kidney and gastrointestinal tract is at a very early stage. Screening for genetic abnormalities known to be related to malignancies remains controversial. The past 100 years have taken us on a journey from linking morphologic nuclear changes to multiple aberrations of a single gene resulting in cancer. A new millenium will see expansion of the burgeoning field of genetic engineering and therapy. C ytogenetics, in situ hybridization and molecular genetic techniques have and will continue to contribute to the diagnosis and therapy of malignancies. Acknowledgm ent My thanks and recognition of professional competence are extended to Mrs. Kimberly Hayes, Ms. V. Hopwood and the cytogenetics technologists of the Cytogenetics Laboratory o f the U.T. M.D. Anderson Cancer Center. The secretarial assistance of Ms. Lynn Benson is gratefully acknowledged. References 1. Von Hansenmann D. U eber asymmetrische Zelltheilung in Epithelkrebsen und deren biologischen Bedeutung. Virchow Arch Pathol Anat 1890; 119: Boveri T. Zur Frage der E ntstehung M aligner Tumoren. Jena:Gustaav Fischer, Tijo JH, Levan A. The chromosome number of man. Hereditas 1956;42: Levan A. Chromosomal studies on som e human tumors and tissues o f normal origin, grown in vivo and in vitro at the Sloan-Kettering Institute. Cancer 1956;9: Nowell S, Hungerford DA. A minute chromosome in human chronic granulocytic leukemia. Science 1960; 132: Heim S, Mitelman F. Primary chromosome abnormalities in human neoplasia. Adv Cancer Res 1989; 52:1^4. 7. Heim S, Mitelman F. Chromosome patterns in human cancer and leukem ia. E ncycl Hum Biol 1991; 2: Henri S. New frontiers in cancer causation: exploring new frontiers. In: Inverson, OH, editor. Tumor progression-karyotypic keys to multistage pathogenesis. Washington: Hemisphere Publishing Corp, 1993: Casperson T, Zech L, Johansson C. Differential binding of alkylating fluorochromes in human chromosomes. Exp Cell Res 1970;60: Seabright M. A rapid banding technique for human chromosomes. Lancet 1971;2: Zhao L, Hayes KJ, Glassman AB. A single efficient method of sequential G-banding and fluorescence in situ hybridization. Cancer Genet Cytogenet 1998;101: Zhao L, Khan Z, Hayes KJ, Glassman AB, Interphase fluorescence in situ hybridization analysis: a study

7 3 3 0 GLASSMAN using centromeric probes 7, 8, and 12. Ann Clin & Lab Sei 1998;28: Glassman AB. Cytogenetics, gene fusions and cancer. Ann Clin & Lab Sei 1995;25: Chang SS, Mark HF. Emerging molecular cytogenetic technologies. Cytobios 1997;90(360) Komminoth P, Long AA. In situ polymerase chain reaction. An overview o f methods, applications and limitations of a new molecular technique. Virchows Archiv B Cell Path 1993;64: Chang X-Y, Glassman AB, Bueso-Ramos C. Applicability of direct in situ reverse transcriptase PCR on bone marrow smears. Ann Clin & Lab Sci 1998;28: Grompe M. The rapid detection of unknown mutations in nucleic acids. Nat Genet 1993;5: Human Genome Project Information:wysiwyg://26/ h ttp ://w w w.o r n d.g o v /T e c h R e so u r c e s/h u m a n _ Genome/home.html. 19. Rowley JD. A new consistent chromosomal abnormality in chronic myelogenous leukemia identified by quinacrine and Giemsa staining. Nature 1973;243: