Chromosomal studies of malignancies pose a particular technical challenge. As the results are so unpredictable, there is no a single technique that can be guaranteed to work consistently and reliably. Therefore, every laboratory should adopt a slight variation of the basic protocol. In addition to setting up a bone marrow suspension cell culture of hematologic neoplasms, peripheral blood can also be investigated, if it contains more than 10% circulating blasts or immature cells. In addition, a lymph node biopsy is required, as lymph nodes are the preferred tissue for studies of most lymphomas [8]. The duration of the cell cycle in malignant cells varies greatly among patients; a range of 16 hr to 292 hr was obtained in a series of 37 patients with AML [9]. Therefore, one of the most significant factors in obtaining a successful result is setting up multiple cultures to maximize the chances of obtaining optimal malignant cell divisions: 1) direct harvest of bone marrow cells, 2) overnight culture, and 3) overnight culture with synchronization (by blocking at S-phase of the cell cycle) [10].

High-resolution banding of long chromosomes with good morphology can be achieved by applying synchronization techniques [10]. It enables the identification of subtle structural chromosome aberrations that are commonly found in malignant cells. However, it has been reported that fluorodeoxyuridine synchronization cultures are inferior to short-term cultures for chromosome analysis in ALL [11]. ALL is a frustrating disease for most cytogeneticists, as it has several technical challenges, including frequent poor chromosome morphology, low mitotic index, and samples that have a marked tendency to clot during harvest.

In cases where the cytogenetic analysis reveals only abnormal metaphases, especially in those with a balanced translocation, it may be necessary to rule out a constitutional chromosomal anomaly. In addition, the constitutive heterochromatic regions just below the centromeres of chromosomes 1, 9, and 16 and the telomeric end of the long arm of the Y-chromosome can vary widely in size among individuals and can sometimes appear abnormal (Fig. 2A). Constitutional chromosomal anomalies are best differentiated from differentiate acquired abnormal clones through cytogenetic examination of phytohemagglutinin (PHA)-stimulated lymphocytes from peripheral blood in disease remission. The Cytogenetics Resource Committee (CyRC) of the College of American Pathologists (CAP)/American College of Medical Genetics & Genomics (ACMG) has asked participants not to include heteromorphisms when responding to a challenge of the cytogenetics survey. However, the guidelines for reporting heteromorphisms have not been standardized. Of 223 cytogenetic laboratories, 136 (61%) stated that they would include selected heteromorphism information identified by routine G-banding in a clinical cytogenetic report in a CyRC survey on cytogenetic heteromorphisms [12]. The majority of the clinical cytogenetics surveyed would not include the more common chromosomal variants (such as prominent short arms, large or double satellites, and increased stalk length or double stalks on acrocentric chromosomes, and long arm heterochromatin variations of chromosomes 1, 9, 16, and Y), but would report most pericentric inversions [such as inv(9)(p11q13); Fig. 2B] and other rare heteromorphisms [12].

Constitutional pericentric inversion of chromosome 9 [inv(9)(p11q13)] occurs in 0.8% to 2% of the normal population and has long been considered a normal variant. Whether constitutional inv(9) is a predisposing factor for cancer remains controversial. We were the first group to document a case of an acquired pericentric inv(9) in essential thrombocythemia [13]. Since then, several other hematological malignancies with acquired inv(9) have been reported. However, inv(9) is not over-represented in patients with hematological malignancy; therefore, there is no evidence to suggest that the presence of constitutional inv(9) increases the risk of hematological malignancy [13, 14].

Furthermore, some abnormalities are both acquired and inherited. For example, trisomy 21 is common both as an inherited abnormality in Down syndrome and as an acquired abnormality in AML, MDS, and childhood ALL [15, 16, 17]. Trisomy 21 is the second most common trisomy in AML and MDS after trisomy 8 [18]. Therefore, if trisomy 21 is found in a karyotype, it is necessary to consider both inherited and acquired abnormalities. However, most individuals with Down syndrome have characteristic physical features.

The standard FISH protocol is illustrated in Fig. 3. Briefly, it includes five steps: 1) sample pretreatment, 2) denaturation of probe and sample, 3) hybridization of probe to target cells or metaphase spreads (annealing), 4) post-hybridization washing, and 5) detection using a simple epifluorescence microscope with appropriate filter sets. When a FISH test is initially implemented, the assay performance characteristics assessed should include sensitivity, accuracy, precision, and specificity [43]. The upper cutoff for normal results in a FISH assay can be determined by calculating the 95% confidence interval for probe signal patterns detected in normal control samples that are representative of the sample type to be analyzed.

ACMG sets internationally accepted standards for FISH analysis to ensure that FISH results are clear and interpretable [44]. Furthermore, ongoing monitoring of interobserver reproducibility, accomplished in part by having two laboratory personnel read every case, can help detect changes in assay performance or loss of consistency in applying scoring criteria. The standard FISH nomenclature has been simplified and expanded in the latest edition of the International System for Human Cytogenetic Nomenclature, ISCN 2013 [45]. However, since the use of the full FISH ISCN is likely to make it more difficult for physicians to understand, it is not recommended by the European Myeloma Network [46].

There are three broad types of FISH probes used in clinical genetics laboratories, each with a different application: 1) whole-chromosome painting (WCP) probes for deciphering cytogenetic aberrations [8, 34, 47, 48]; 2) repetitive sequence probes for chromosome enumeration [21, 25, 26, 49, 50, 51]; and 3) locus-specific identifier (LSI) probes for gene fusions [17, 33, 52], gene deletions [35, 53, 54] or duplications [55]. These probes can also be used in various combinations when investigating complex chromosome abnormalities [56, 57].

WCP probes are designed to mark the entire chromosome of interest and are useful for deciphering cytogenetic aberrations that are difficult to resolve on morphological grounds, such as marker chromosomes of uncertain nature or complex changes. However, the use of WCP probes in interphase cells is very limited, as the chromosomes are dispersed in the cell nucleus during interphase and they do not form discrete units.

Centromeric enumeration probes (CEPs) hybridize to alpha (or beta) satellite repeat sequences within the specific centromeric regions of each chromosome and are used for chromosomal enumeration. Interphase FISH analysis using dual color CEPs for chromosomes X and Y is an efficient method to predict relapse or failure of engraftment in patients after sex-mismatched bone marrow transplantation (Fig. 4A) [58].

There are two main LSI FISH probe systems for the detection of gene rearrangements in oncology, dual-color translocation probes and dual-color break-apart probe. The dual-color translocation probes are designed to detect chromosomal translocations involving know partner genes, such as BCR-ABL1 that result from t(9;22)(q34;q11.2) in CML (Fig. 4B). Furthermore, dual-color break-apart probes are useful for detecting chromosomal translocations that involve genes with unknown or multiple translocation partners, such as MLL, ALK, and RARα (Fig. 4C). Break-apart translocation FISH probes conveniently provide important information on the presence of gene rearrangements, although they are unable to identify the specific partner gene. Recently, with the use of long distance inverse-polymerase chain reaction (LDI)-PCR, it has become possible to identify unknown translocation partners and to map breakpoints at the base-pair level. LDI-PCR requires only approximate sequence information for one partner, rendering it ideal for use in combination with FISH to extend and refine cytogenetic breakpoint data [59]. LSI FISH probes have also been used to identify the origin of the amplified DNA that constitutes homogeneously staining regions (HSRs) and double minutes (DMs), which occur in a variety of tumor cells (Fig. 4D) [60].

Numerous methodological advances in molecular cytogenetic technology were initiated in the early 1990s, including comparative genomic hybridization (CGH) [54], spectral karyotyping (SKY) [62], multicolor FISH (MFISH) [63], multicolor banding (mBAND) [64], and array CGH (aCGH) [65]. All of these cytogenetic techniques add colors to the monotonous world of conventional banding. Two multicolor fluorescence technologies have been introduced, MFISH [63] and SKY [62]. These technologies are based on simultaneous hybridization of 24 chromosome-specific composite probes. This technique is suitable for the identification of subtle chromosomal aberrations that include an unidentified chromosome (marker chromosome) and an unbalanced chromosomal translocation. Regarding probe design, these chromosome-painting probes are generated from flow-sorted human chromosomes. Unique chromosome-specific colors are produced by labeling each chromosome library with either a single fluorochrome or specific combinations of multiple fluorochromes.

mBAND has been developed to facilitate the identification of intrachromosomal rearrangements and to map the exact breakpoint by using human overlapping microdissection libraries that are differentially labeled [64]. The color bands have great value for delineating intrachromosomal exchanges, such as inversions, deletions, duplications, and insertions [6].

CGH is based on quantitative dual-color FISH along each chromosome [61]. CGH can be used to detect genetic imbalances in test genomes, and to determine the chromosomal map positions of gains and losses of entire chromosomes or chromosomal subregions present in normal reference metaphase preparations. A distinct advantage of CGH is that tumor DNA is the only requirement for this analysis. Thus, archived, formalin-fixed and paraffin-embedded tissues can be used as well. CGH is useful for cancer research, especially for determining the low mitotic index of malignant cells with poor chromosome morphology and resolution [66, 67, 68].

Tremendous technical advances in cytogenetics have changed the approaches used for clinical diagnostics and research, in particular, the development of array CGH (aCGH) technology for "molecular karyotyping" with a resolution of 100 kb to 1 Mb [65]. aCGH greatly improves the resolution of this technique by substituting the hybridization targets, the metaphase chromosome spread, with genomic segments spotted in an array format. The complexity of the genomic aberrations in most human tumors hampers delineation of the genes that drive the tumorigenic process. Interestingly, cognate mouse models have been shown to recapitulate these genetic alterations with unexpected fidelity [69]. These results indicate that cross-species aCGH analysis is a powerful strategy to identify the responsible genes and assess their oncogenic capacity in the appropriate genetic context [69]. Recently developed genomic microarray methodologies, including aCGH and single-nucleotide polymorphism (SNP)-based arrays, are innovative methods that provide genomic data for multiple neoplastic disorders [70, 71, 72]. In addition, these methods can reveal additional important information about the genetics of specific disorders, such as leukemia with normal cytogenetic and FISH analyses [73, 74]. In 2013, ACMG developed professional standards and guidelines to assist clinical laboratories in the validation, consistent use, and the interpretation and reporting of results from these microarray methodologies [75].

The Mitelman Database of Chromosome Aberrations and Gene Fusions in Cancer (http://cgap.nci.nih.gov/Chromosomes/Mitelman) includes a comprehensive database of all published neoplasia-associated karyotypes and their corresponding gene fusions. The available information on chromosome abnormalities in human neoplasias has steadily increased over the past three decades. The data in this database have been manually culled from the literature by Felix Mitelman, Bertil Johansson, and Fredrik Mertens. The total number of tumor cases, in which clonal cytogenetic aberrations have been reported has reached 64,319, and the database was updated with the addition of 2,072 chimeric fusion genes in May 2014 [76].

The Atlas of Genetics and Cytogenetics in Oncology and Haematology (http://atlasgeneticsoncology.org/) [77], which was established in 1997, is a peer-reviewed, open access, online journal, encyclopedia, and database that is devoted to genes, cytogenetics, and clinical entities in cancer and cancer-prone diseases. Approximately 2,200 authors have contributed to the Atlas so far, making 2,167 review articles available. The Atlas contains peer-reviewed articles on 1,135 genes, 503 leukemia entities, 177 solid tumors, and 104 cancer-prone inherited diseases. It also contains "automated cards" on 8,190 other genes that are potentially implicated in cancer [78].

In the medical genetics community, interpretation and scientific communication is often facilitated by universally accepted nomenclature with precisely defined terms and syntax conventions that minimize complexity and add precision to the process. Cytogenetic nomenclature is based on the reports of an international committee that was established in 1960, known as the International System for Human Cytogenetic Nomenclature (ISCN) [79]. The nomenclature is updated periodically, most recently in 2013 with expanded guidelines for cancer cytogenetics, FISH, and microarray [45]. In cancer cytogenetic, an abnormal cytogenetic clone is defined as a population of cells with the same chromosome complement that is derived from a single progenitor. A clone must have at least two cells with the same aberration if the aberration is a chromosome gain or a structural rearrangement. If the abnormality is loss of a chromosome, the same loss must be present in at least three cells to be accepted as clonal. However, in the current version of the ISCN (2013), two cells with identical losses of one or more chromosomes and the same structural aberration(s) may be considered clonal and newly included [45]. However, whether two cells with the same loss of a single chromosome or one cell with a gain of a single chromosome in a composite karyotype should be counted and included in the size of the clone has not been mentioned in the current nomenclature system.

Recently, we conducted a pilot survey of the reporting practices for the random loss or gain of a single chromosome in clinical cytogenetics reports. The questionnaire, which consisted of two questions, was sent to 19 cancer cytogenetic laboratories in 13 counties and cities, and we invited them to report to us using their usual reporting system (Table 1). In question 1, participants were presented with a bone marrow sample from an AML patient with non-clonal (random) loss of chromosome(s), and asked whether these metaphases should be considered as normal and included in the clone size of the normal karyotype. In question 2, participants were presented with a marrow blood sample from a CML patient with 19 metaphases of t(9;22)(q34; q11.2) and one metaphase with only -5. We asked whether the normal metaphase with a loss of a chromosome 5 should be reported in final karyotype in addition to the 19 metaphases of t(9;22)(q34;q11.2). To our surprise, the responses were not unanimous. For question 1, 11 of the 19 laboratories (57.9%) reported 46,XX[20], and the remaining 8 (42.1%) reported 46, XX[9] (Table 1). Therefore, 57.9% of the cytogenetic laboratories assumed that metaphases with non-clonal loss of chromosomes were normal, and they included all these metaphases in the clone size of normal clone. Most laboratory accreditation bodies require the "total number of metaphases analyzed" in the final report. Thus, for the 42.1% of cytogenetic laboratories that reported 46,XX[9], there is a general understanding by the clinicians that the patient has 11 metaphases with non-clonal loss or gain of chromosomes among the 20 metaphases analyzed. For question 2, 8 of the 19 laboratories (42.1%) reported 46, XX,t(9;22)(q34;q11.2) [19] and omitted the only metaphase with -5; 7 laboratories (36.8%) reported 46,XX,t(9;22)(q34;q11.2)[19]/46,XX[1] and counted the only metaphase with -5 as a normal clone; and 4 laboratories (21.1%) reported 46,XX,t(9;22)(q34;q11.2)[19]/45,XX,-5 [1] and simply wrote down the only metaphase with -5 in the final karyotype (Table 1). Taken together, the ISCN standing committee should continue to discuss such discrepancies and make efforts to align the reporting system used by cytogenetic laboratories [80].

Interestingly, a review report of the CAP FISH proficiency surveys between 1997 and 2001 showed that syntax errors were far more common than diagnostic errors [81]. In addition, the syntax errors that are common in the proficiency surveys are also common in FISH surveys. Recently, since the full FISH ISCN is likely more difficult for physicians to understand, the use of FISH ISCN is not recommended by the European Myeloma Network [46]. Therefore, continuous improvement of the nomenclature guidelines to address some of these issues and identify the cause of these variations is essential [78, 82].