THE MANAGEMENT OF PATIENTS WITH LEUKAEMIA: THE ROLE OF CYTOGENETICS IN THIS MOLECULAR ERA
2000; Wiley; Volume: 108; Issue: 1 Linguagem: Inglês
10.1046/j.1365-2141.2000.01801.x
ISSN1365-2141
Autores Tópico(s)Acute Lymphoblastic Leukemia research
ResumoIn this era of advancing molecular technologies, cytogenetic investigation remains an essential requirement in the management of patients with leukaemia. Metaphase chromosome analysis of bone marrow cells enables the entire genome to be screened for evidence of chromosomal changes, which provide the chromosomal landmarks for the genes involved in leukaemogenesis. The majority of newly diagnosed leukaemias show chromosomal abnormalities. Up to 80% of patients with acute lymphoblastic leukaemia (ALL) and 70% of patients with acute myeloid leukaemia (AML) have abnormal karyotypes ( Heim & Mitelman, 1995). A large number of these chromosomal changes are recurrent and have been associated with specific morphological types in both myeloid and lymphoid lineages. As a result, cytogenetics plays a major role in diagnosis and, notably, in the myeloproliferative and myelodysplastic disorders, the finding of an abnormal cytogenetic result may provide the definitive diagnosis. The greatest impact on patient management has been the demonstration that the cytogenetic result is an independent prognostic indicator. Certain karyotypes are associated with a good prognosis, whereas others indicate a poor outcome, leading to the administration of alternative therapies. The demand for chromosomal analysis of the majority of new cases of haematological malignancies from major referral centres and district hospitals reflects the value placed on cytogenetics at diagnosis. The National External Quality Assessment Scheme (NEQAS) in Clinical Cytogenetics reported that, in 1997/98, more than 17 500 samples were referred for analysis to 36 cytogenetics laboratories in the UK, of which ≈2000 were new acute leukaemias ( Waters & Rodgers, 1998). Evidence from NEQAS data over a number of years has demonstrated that participation in this scheme has made a major contribution to improvements in cytogenetic preparation quality and the standard of interpretation and reporting of results. Cytogenetic analysis is now a requirement for entry of patients to the current UK Medical Research Council treatment trials for childhood acute lymphoblastic leukaemia (ALL 97), adult ALL (UKALL XII), acute myeloid leukaemia in children and adults (AML 12) and patients > 55 years (AML 14). Collection and co-ordination of cytogenetic results from patients entered into these trials is now well-established in the UK through karyotype databases for ALL and AML. These are funded by the Leukaemia Research Fund and the Kay Kendall Leukaemia Fund and are directed by experienced cytogeneticists. Their operation has led to a closer interaction between UK cytogenetics laboratories and the clinical trial co-ordinators, resulting in an improved understanding of the significance of chromosomal changes in acute leukaemia. One advantage of cytogenetic analysis is that it is sensitive enough to detect abnormalities in small populations of cells. In fact, two metaphases with the same chromosomal gain or structural rearrangement, or three cells with the same chromosomal loss define the presence of an abnormal clone ( ISCN, 1995). The examination of individual cells enables secondary chromosomal changes and patterns of karyotypic evolution to be identified. Cytogenetic analysis is highly informative in the majority of cases, although interpretation of the findings may be difficult if the karyotype is complex or multiple clones are present. In spite of continual technical developments in methodology, some samples fail to produce a successful result because of a low mitotic index or poor chromosome morphology. The growth of techniques in molecular cytogenetics, or fluorescence in situ hybridization (FISH), has overcome some of these limitations. FISH has produced major improvements in the sensitivity, accuracy and reliability of cytogenetic analysis for both research and routine applications ( Kearney, 1999). The introduction of interphase-FISH has made one of the most important contributions, allowing large numbers of non-dividing cells to be screened. Analysis is no longer restricted to cytogenetic cell suspensions ( Bentz et al, 1993 ). The increasing availability of commercial probes means that FISH has become accessible to most routine cytogenetics laboratories. Molecular cytogenetics has become an integral part of the discipline of cytogenetics and is indispensable in accurate chromosomal analysis of leukaemias. The expansion of molecular technologies in the 1980s was seen as a serious potential rival to routine cytogenetic analysis, particularly when the specificity and sensitivity of these techniques was realized. On the contrary, they were responsible for the progression of cytogenetics through developments in FISH. The introduction of Southern blotting allowed genetic rearrangements to be identified in malignant tissue. However, it was the development of the polymerase chain reaction (PCR) ( Saiki et al, 1988 ) and, subsequently, reverse transcriptase–PCR (RT–PCR) that transformed molecular technology. Novel fusion mRNAs transcribed by the fusion genes at the site of chromosomal rearrangements provide tumour-specific markers suitable for PCR amplification. PCR is 400–4000 times more sensitive than Southern analysis, with the ability to detect one leukaemic cell in 105–106 normal bone marrow cells. It is now possible to detect breakpoints scattered over a large genomic region using long-range PCR ( Willis et al, 1997 ). Quantitative PCR assays, such as quantitative competitive PCR (QC-PCR) ( Cross et al, 1993 ) and, more recently, real time quantitative PCR (RQ-PCR) ( Heid et al, 1996 ) have been introduced for patient monitoring. This review will emphasize the complementary role of cytogenetics, FISH and molecular techniques in the management of patients with leukaemia. The expertise lies in the selection of the most appropriate procedure to be used, according to the individual disease type and chromosomal abnormality to be investigated. This will be illustrated with specific examples highlighting the advantages and disadvantages of the individual approaches. The most famous example of an acquired chromosomal change related to a haematological malignancy is the association of the Philadelphia chromosome (Ph) with chronic myeloid leukaemia (CML). It was the first specific chromosomal abnormality to be described in leukaemia ( Nowell & Hungerford, 1960), and it is now a diagnostic indicator, found in 95% of CML cases. In ALL, the Ph occurs in 2–3% of childhood cases ( GFCH, 1993) but, in adults, it is the most common cytogenetic change, the incidence of which increases with age ( Secker-Walker et al, 1997 ). It is associated with a poor outcome, with an event-free survival (EFS) in children of 15% at 5 years ( GFCH, 1993). The Ph provides an elegant example of how cytogenetic findings provided the starting point for understanding the genetic mechanisms involved in leukaemogenesis. The Ph normally arises from a reciprocal translocation that joins 3′ sequences of the tyrosine kinase ABL proto-oncogene on chromosome 9 to the 5′ sequences of the BCR gene on chromosome 22. In a small number of cases, variant or complex translocations occur, in which other chromosomes, in addition to 9 and 22, may also be involved in the Ph rearrangement. Breakpoints within BCR occur in a 5.8-kb region in CML, which has been termed the major breakpoint cluster region (M-BCR) between either exons 13 and 14 (b2a2) or exons 14 and 15 (b3a2). The formation of the Ph from t(9;22)(q34;q11) or variant translocations thus results in a BCR/ABL hybrid gene. This transcribes an aberrant 8.5-kb mRNA, encoding a chimaeric p210 protein with tyrosine kinase activity ( Groffen & Heisterkamp, 1987). In the majority of Ph-positive ALL cases, the breakpoint occurs in the first intron of the BCR gene, the minor cluster region (m-BCR), and between exons 1 and 2 and intron 2 of the ABL gene (e1a2). This results in the generation of a 7-kb mRNA and the expression of a p190 protein ( Clark et al, 1988 ). Both BCR/ABL fusion proteins (p190 and p210) possess enhanced tyrosine kinase activity ( Lugo et al, 1990 ) and provide examples of activation of an oncogene by the creation of a novel fusion product, leading to the generation of leukaemia. The Ph can usually be identified in CML and ALL by conventional cytogenetic analysis. However, 5% of CML and a small number of ALL cases are Ph negative, in which no Ph is visible, but they are positive for the BCR/ABL fusion ( Van Rhee et al, 1995 ). In these cases, the fusion occurs either as a submicroscopic rearrangement or is masked within a complex karyotype. An additional problem in ALL is that sometimes the cytogenetics fails, or the quality of the preparations is poor, which precludes accurate detection of the Ph. Although this applies to only a minority of cases, alternative molecular methods were developed to ensure that Ph-negative BCR/ABL-positive cases were not overlooked. These include the detection of rearrangements within M-BCR by Southern analysis, the BCR/ABL fusion transcript by RT–PCR ( Hooberman & Westbrook, 1990) or FISH ( Tkachuk et al, 1990 ). The demonstration of the BCR/ABL mRNA by RT–PCR is a particularly sensitive assay for the detection of the leukaemic clone in Ph-positive ALL ( Radich et al, 1994 ). Reliable FISH probes are now available commercially for detection of BCR/ABL (Appligene-Oncor and Vysis). Two locus-specific probes are used, one for BCR and one for ABL, labelled with two different coloured fluorochromes, which enable the fusion gene, BCR/ABL, to be visualized accurately in both metaphase and interphase cells, as illustrated by Kearney (1999). Cytogenetics plays a vital role in the early detection of transformation to acute phase in CML. In ≈ 80% of CML cases, transformation is accompanied by karyotypic evolution, in which a series of secondary chromosomal abnormalities arises ( Parreira et al, 1986 ). These are usually detectable before any haematological changes. Although most cases exhibit a defined pattern of abnormalities, it is the karyotypic evolution that is the important predictor of transformation, rather than the specificity of the changes themselves. There are no comparable molecular markers of karyotypic evolution. As variant and complex translocations and, sometimes, additional chromosomal changes are present in chronic phase, it is important to know the precise cytogenetic status of the patient at presentation before transformation can be accurately diagnosed. For these reasons, cytogenetic monitoring remains the method of choice for the detection of the Ph in chronic phase CML, with FISH and molecular techniques providing a back-up in those cases in which the cytogenetics fails or is inconclusive. There are now many specific chromosomal rearrangements in leukaemia known to give rise to fusion genes. The most well-known examples are shown in Table I ( Harrison & Secker-Walker, 1999). These include the t(8;21)(q22;q22), associated with acute myeloblastic leukaemia, AML M2 ( Rowley, 1973). This translocation results in the formation of the hybrid gene AML1/ETO by the fusion of the first five exons of the AML1 gene, localized to the chromosome band 21q22, with the entire coding sequence of ETO, located to 8q22 ( Miyoshi et al, 1993 ). The translocation, t(15;17) (q22;q11~21), is diagnostic for acute promyelocytic leukaemia, AML M3 ( Rowley et al, 1977 ). This translocation results in the fusion of the retinoic acid receptor alpha gene (RARA), located at 17q21, to PML on chromosome 15, producing the fusion gene PML/RARA on the derived chromosome 15 ( Grignani et al, 1994 ). Inversion of chromosome 16, inv(16)(p13q22), is associated with acute myelomonocytic leukaemia, AML M4, with eosinophilia ( Le Beau et al, 1983 ). This rearrangement fuses the CBFβ, a core binding factor gene, with MYH11, a smooth muscle myosin heavy-chain gene, giving rise to the hybrid gene CBFβ/MYH11 ( Liu et al, 1993a ). Results from the previous UK MRC treatment trial in AML, AML 10, associated these three rearrangements with a good prognosis. In the recent trial, AML 12, patients with these abnormalities are assigned to a good risk group ( Grimwade et al, 1998a ). Therefore, their accurate identification is critical. Cytogenetic analysis correctly identifies these rearrangements, their fusion transcripts can be detected reliably by RT–PCR ( Poirel et al, 1995 ; Andrieu et al, 1996 ; Grimwade et al, 1998b ) and probes have been developed to identify their fusion genes by FISH ( Liu et al, 1993b ; Mancini et al, 1995 ; Sacchi et al, 1995 ). FISH probes are now available commercially for PML/RARA and inv(16) (Appligene-Oncor and Vysis). AML1/ETO was detected at a higher incidence than t(8;21) in a series of AML M2 patients ( Andrieu et al, 1996 ) and MYH11/CBFB than inv(16) in a series of AML M4 with eosinophilia ( Tobal et al, 1995 ) in parallel cytogenetic and molecular studies. In the series of cases of AML M2 with no apparent t(8;21)(q22;q22) ( Andrieu et al, 1996 ), FISH confirmed the presence of the AML1/ETO hybrid gene masked as part of a complex karyotypic picture ( Harrison et al, 1999 ). Cryptic t(15;17) rearrangements have been disclosed similarly ( Grimwade et al, 1997 ). These positive FISH results confirm the accuracy of molecular techniques in the detection of these rearrangements. It has now been demonstrated conclusively that secondary karyotypic changes found in association with these good-risk abnormalities have no detrimental effect on outcome ( Grimwade et al, 1998a ). It may therefore be more appropriate to screen all cases of AML initially by PCR for the good-risk indicators. Cytogenetics can be applied as a follow-up in cases with no evidence of these markers or a discordant clinical picture, thus avoiding unnecessary duplication of results. The chromosome band 11q23 is the site of the MLL gene ( Tkachuk et al, 1992 ). Chromosomal rearrangements involving MLL are specifically associated with leukaemias of lymphoid, myeloid and bi-phenotypic origin in infants < 1 year old, accounting for 85% of these cases ( Secker-Walker, 1998). These infants have an extremely poor prognosis ( Heerema et al, 1994 ). Another group of infants, usually Down's syndrome, has transient abnormal myelopoiesis (TAM) with a good outcome and usually remit spontaneously ( Faed et al, 1990 ). Older children with MLL rearrangements also have a poor prognosis and, in the current childhood ALL treatment trial, ALL 97, are treated on a high-risk protocol. Therefore, firstly, to differentiate TAM patients from infant leukaemias and, secondly, to ensure that children are assigned to the correct treatment protocol, accurate identification of MLL rearrangements is essential. There are a number of well-defined chromosomal abnormalities involving 11q23 with known fusion partners, including, t(4;11)(q21;q23), t(6;11)(q27;q23), t(9;11) (p21;q23), t(10;11)(p12;q23) and t(11;19)(q23;p13) ( Secker-Walker, 1998), which can be reliably identified by cytogenetics and PCR. In order to ensure that the fusion transcripts for all these MLL rearrangements are detected, multiple PCR applications, in the form of multiplex-PCR, are required. Although accurate, this procedure is highly labour intensive ( Pallisgaard et al, 1998 ). Some problems arise in the accurate characterization of all MLL rearrangements, as MLL is a 'promiscuous' gene, having rearrangements with a large number of chromosomal partners, some of which have been reported in small groups of patients ( Bernard & Berger, 1995; Harrison et al, 1998 ). Molecular analysis is limited in those groups in which the fusion partners have not been identified. An additional limitation is that cryptic MLL rearrangements exist. Therefore, no visible evidence of a cytogenetic rearrangement of 11q23 does not rule out the involvement of MLL. Conversely, the presence of a visible chromosomal abnormality of 11q23 does not necessarily conclude that MLL is involved. The majority of breaks in MLL occur within an 8.3-kb breakpoint cluster region (bcr) between exons 5 and 11; these rearrangements of MLL can be identified most accurately by Southern blotting ( Thirman et al, 1993 ). As an alternative detection method, a FISH probe for MLL is commercially available (Appligene-Oncor), which spans the bcr covering most of the gene. One limitation of this MLL probe is that it does not detect rearrangements occurring 3′ of MLL, which account for ≈ 20% of cases. However, when a break occurs within the bcr as a result of a translocation, the probe hybridizes to the two parts of MLL and is seen as a 'split' signal. Three MLL signals are seen in abnormal metaphase and interphase cells (Fig 1). However, caution is required in the interpretation of positive results, as three signals also indicates three intact copies of MLL as a result of a trisomy or an unbalanced rearrangement giving rise to duplication of the 11q23 chromosomal region. Unbalanced translocations resulting from complex karyotypes involving MLL also exist, in which one of the reciprocal products is lost. In such cases, one of the 'split' signals is missing, and the rearrangement remains undetected by FISH. It has been reported that cases with the unbalanced forms of such translocations have an improved prognosis over the balanced forms ( Raimondi et al, 1995 ). Therefore, detection methods are required that include unbalanced translocations. An additional consideration is that trisomy 11 is invariably associated with 'self-fusion' of MLL ( Schichman et al, 1994 ). FISH cannot detect this type of partial duplication reliably. 9 (red). (4) Interphase cells from the bone marrow of a patient after bone marrow transplantation with a sex-mismatched donor. Probes for the centromeric regions of X and Y chromosomes are labelled in different colours; X is red and Y is green. This sample shows the presence of both male and female cells. (5) A colour-banded karyotype from a normal male. Each chromosome pair shows an individual colour banding pattern. A new probe kit using dual-colour FISH with two probes, 5′ and 3′ of MLL bcr, labelled in different colours (Vysis), is now available. As a result of a rearrangement involving MLL, the two probes become separated and are seen as two independent signals in metaphase and interphase cells (Fig 3). A pilot study has demonstrated that, by using this probe, rearrangements are distinguished from duplications (Fig 2) of the intact MLL gene, and unbalanced forms can be identified. Thus, this dual-colour approach may help to overcome some of the problems associated with single-colour FISH (C. J. Harrison, unpublished observations). The prognostic significance attached to MLL rearrangements demands accurate identification. All positive FISH results should be confirmed by examination of signals in metaphase cells wherever possible, complemented by further FISH tests, at least to exclude trisomy 11. It is advisable to apply cytogenetic analysis, FISH and Southern blotting in a complementary fashion to overcome the limitations of the individual techniques and to have confidence in a positive result. The translocation, t(1;19)(q23;p13) is strongly associated with a pre-B immunophenotype. Balanced, t(1;19)(q23;p13), and unbalanced, der(19)t(1;19)(q23;p13), forms of the translocation exist. Studies in which the two forms were combined associated this group with a poor prognosis. When separated, the unbalanced form in children was shown to have a significantly better outcome than the balanced form ( Secker-Walker et al, 1992 ). The translocation results in the formation of the E2A/PBX1 fusion transcript ( Hunger et al, 1991 ). Molecular analysis has been of value in showing the presence of the rearrangement in samples with a failed karyotype ( Devaraj et al, 1995 ). A small number of cases, in which t(1;19) was demonstrated by cytogenetics, failed to show the fusion gene by PCR, suggesting that an alternative breakpoint outside E2A may exist ( Privitera et al, 1992 ). The association with a poor prognosis, described previously, appears to be mitigated by aggressive chemotherapy ( Privitera et al, 1996 ). This highlights the urgency for cytogenetic analysis at presentation to identify these cases in sufficient time to administer the appropriate intensive therapy. As no FISH probes are in routine use for the detection of this translocation, in cases with a failed or normal cytogenetic result, molecular analysis is vital. The development of FISH probes directed to E2A and PBX1 would be valuable to ensure detection of cases with variant molecular breakpoints. Similarly, the translocation, t(8;14)(q24;q32), and variant forms, t(2;8)(p13;q24) and t(8;22)(q24;q11), in mature B-cell ALL (ALL L3) ( Berger & Bernheim, 1982) were associated with an extremely poor prognosis. They were also found to respond well to intensive chemotherapeutic regimens, showing a significant increase in overall survival ( Hoelzer et al, 1996 ). Thus, urgent cytogenetic confirmation of these abnormalities is required. The development of FISH technologies has revealed the existence of cryptic chromosomal rearrangements, which are not visible by conventional cytogenetic analysis. The most famous of these is the translocation, t(12;21)(p13q22), which was uncovered by chromosome painting ( Romana et al, 1994 ). It was later demonstrated to involve the formation of a hybrid gene between the ETV6 gene at 12p13 and the AML1 gene at 21q22 ( Romana et al, 1995a ), and to occur in up to 25% of childhood B-lineage ALL ( Romana et al, 1995b ). A commercial probe is now available for detection of the ETV6/AML1 fusion gene (Vysis). As it is invisible by cytogenetics, RT–PCR and/or FISH remain the only reliable methods for the detection of this abnormality. The important secondary event that seems to be required to trigger the process of leukaemogenesis is the deletion of the normal ETV6 gene from the chromosome 12, which is not involved in the translocation ( Greaves, 1999). FISH and loss of heterozygosity (LOH) studies have shown that the extent of this deletion can be highly variable ( Raynaud et al, 1996 ). Although the whole of ETV6 is deleted in the majority of cases, in a small number of cases, the deletion is intragenic and only detectable with probes directed to individual exons. It is important to be aware that the commercial probe, which spans exons 1–4 of ETV6, would be unable to detect these small intragenic deletions. The size of the population carrying the deletion is also variable, and small populations are not discernible by LOH. The translocation t(12;21) has been associated with an overall good prognosis. This relationship is under review, as an increasing number of patients have been shown to relapse ( Harbott et al, 1997 ). Additional chromosomal changes accompanying t(12;21) ( Martineau et al, 1998 ), including the extent of the deletion of the second ETV6 gene and the size of this population, may have a role to play in determining the likelihood or timing of relapse. Although PCR will identify the ETV6/AML1 transcript accurately, FISH, with parallel cytogenetic analysis, is the most appropriate method to demonstrate the presence of the fusion gene and, at the same time, investigate the role of associated genetic changes. Another example of a cryptic abnormality is a dicentric translocation between chromosomes 9 and 20, dic(9;20) (p11~13;q11), found in a small group of ALL patients ( Rieder et al, 1995 ; Slater et al, 1995 ; Heerema et al, 1996 ). It involves an overall loss of chromosomal material from the short arm of chromosome 9 and the long arm of chromosome 20, manifesting as monosomy 20 in G-banded karyotypes. The precise breakpoints involved have not been identified. It therefore remains uncertain whether or not it gives rise to a fusion gene. It may be that the formation of the dicentric chromosome and the interaction between the two centromeres is the important event in the process of leukaemogenesis ( Berger & Busson-Le Coniat, 1999). FISH analysis, with whole-chromosome paints and centromeric probes for chromosomes 9 and 20, remains the only reliable method for accurate identification of this cryptic rearrangement. The recent discovery of a cryptic translocation, t(5;11) (q35;p15.5), in three of four cases of childhood AML with apparent deletions of 5q ( Jaju et al, 1999 ) in addition to the two examples above raises the question as to how many other cryptic rearrangements remain undiscovered. In addition to the balanced structural rearrangements observed in haematological malignancies, numerical abnormalities, duplications and deletions, which give rise to the gain or loss of chromosomal material, have also been found to be associated with specific disease types. For example, trisomy 8 is a common finding in all types of myeloid disorders. Loss of chromosomes 5 and 7, or deletions of their long arms, are also frequent findings in AML, particularly in elderly patients and those with secondary leukaemia ( Le Beau et al, 1986 ), and are usually associated with a poor outcome. These poor cytogenetic indicators, in addition to chromosomal rearrangements involving 3q, have been assigned to a poor risk group in AML 12 ( Grimwade et al, 1998b ). Owing to the variety and complexity of these high-risk abnormalities in AML, cytogenetics remains the only reliable detection method. One numerical abnormality of particular significance in childhood ALL is a high hyperdiploid karyotype, with 50–65 chromosomes. It is found in ≈ 30% of childhood ALL cases and is associated with a good prognosis, having an event-free survival of > 80% at 5 years in children ( Secker-Walker et al, 1978 ). Conversely, near haploidy (≈ 23 chromosomes) in ALL is associated with a poor prognosis ( Gibbons et al, 1991 ). Children with this numerical change in ALL 97 are treated on a high-risk protocol. In general, molecular methods cannot be used reliably to detect numerical chromosomal changes. As a result, cytogenetic analysis plays a critical role in the direction of suitable therapy in these patients. FISH, using centromeric probes, can be used to detect changes in chromosome number in interphase cells or poorly defined metaphases ( Anastasi et al, 1990 ), thus providing a complementary approach in cases with a normal or failed cytogenetic result. Specific centromeric probes can be applied individually in the investigation of trisomy or monosomy. For example, in the search for trisomy 12 in chronic lymphocytic leukaemia, interphase-FISH using a probe specific for the centromere of chromosome 12 is more appropriate than metaphase analysis, as the abnormality occurs in the non-dividing mature B-cell population, which requires to be stimulated into cell division ( Anastasi et al, 1992 ). In the cytogenetic study of myeloid disorders, the finding of a single cell with trisomy 8 or monosomy 7 is insufficient to distinguish between a clonal abnormality at a low level or a random change. Interphase-FISH, with appropriate centromeric probes, can confirm or refute the clonal nature of such abnormalities. A device has been developed, known as Chromoprobe Multiprobe (Cytocell, UK), which enables simultaneous hybridization of probes for all 24 centromeres to be carried out on a single slide. This device is highly effective in revealing hyperdiploid clones in cases of ALL with failed or normal karyotypes. In a recent study, eight out of 48 childhood ALL cases with a failed or normal karyotype were shown to have a high hyperdiploid clone present in the non-dividing population (A. Kasprzyk, personal communication). The same device has been used successfully to screen for the poor-risk, near-haploid clones (R. Clark, personal communication). Therefore, all children entered into ALL 97 with a failed or normal karyotype are now being routinely screened for evidence of hyperdiploidy or near haploidy using the Chromoprobe Multiprobe device. Chromosomal deletions, for example deletion of the long arm of chromosome 5, del(5q), del(7q) and del(11q) in AML and del(6q) in ALL, are detectable by cytogenetic analysis, unless the deleted region is very small. However, cytogenetic analysis is often not sensitive enough to define the deletion breakpoints precisely. Chromosome painting has demonstrated that deletions of chromosomes 5 and 7 may arise as the result of cryptic unbalanced translocations ( Tosi et al, 1996 ). FISH has also been effective in the accurate recharacterization of deletion breakpoints at the same time as identifying a commonly deleted region on 5q ( Jaju et al, 1998 ) and 6q ( Sherratt et al, 1997 ). It is postulated that these regions may be the sites of tumour-suppressor genes (TSGs). FISH, with appropriate probes, is being used in the detection of deletions as a research tool but, until a TSG has been identified, routine molecular screening is not applicable because of the enormous variability of the breakpoints involved. Increasingly, patients with acute leukaemia are being monitored during the course of their management for evidence of minimal residual disease
Referência(s)