The Leukaemia Research Fund/United Kingdom Cancer Cytogenetics Group Karyotype Database in acute lymphoblastic leukaemia: a valuable resource for patient management

Chromosomal abnormalities that occur in the malignant bone marrow cells of patients with acute lymphoblastic leukaemia (ALL) are intimately associated with the biology of the disease and indicate the location of genes involved in leukaemogenesis. They define patient subgroups and have important prognostic implications with a direct influence on the choice of therapy, albeit in a small number of patients. Therefore, accurate identification of the cytogenetic and genetic changes in the malignant cells plays a vital role in diagnosis and patient management. The Leukaemia Research Fund (LRF)/United Kingdom Cancer Cytogenetics Group (UKCCG) Karyotype Database (Database) is a computerized registry of the chromosomal findings in patients with ALL who have been entered in the Medical Research Council (MRC) treatment trials over the past 10 years. The original remit of the Database was to provide, in close association with the local cytogenetics centres, the most accurate cytogenetic result possible. Karyotype review by re-examination of G-banded slides produced a significant improvement. During the first 5 years of the Database, standard chromosomal banding techniques were used. However, by the time of the 1997 childhood ALL trial, ALL97, remarkable new molecular cytogenetics techniques had been sufficiently developed to be used in analysis. These techniques of fluorescence in situ hybridization (FISH) using chromosome- and gene-specific probes had a dramatic effect on the accuracy and sensitivity of cytogenetics. These procedures have been integrated into the Database protocol and are being applied to an increasing number of cases. As a result, known cytogenetic rearrangements have been confirmed and unexpected cryptic rearrangements have been revealed that were not apparent using ‘conventional’ cytogenetic analysis. The Database has now expanded its brief to provide a diagnostic interphase-FISH service for detecting chromosomal abnormalities of prognostic significance in childhood ALL patients. The Database contains a comprehensive profile of the cytogenetic records of more than 4000 patients diagnosed with ALL in the UK. For the previous childhood ALL trial, UKALLXI, the Database records are complete, with successful analyses of 1658 of the 2089 children. By the end of September 2000, records of 91 infants, a further 1307 children and 741 adults had been entered. An exceptionally large number of successfully analysed cases with a uniquely high incidence of abnormal clone detection makes this collection of karyotypes superior to those in any comparable published study in the world. The success of the Database has improved liaisons between cytogeneticists, molecular biologists and haematologists in the UK. Studies arizing from the Database have revealed rare ALL-associated chromosomal abnormalities (Martineau et al, 1996; Clark et al, 2000), provided a strategy for the detection of hidden hyperdiploid clones (Moorman et al, 1996) and allowed highly complex karyotypes to be completely characterized. The Database contributes to a large number of research projects of which cytogenetics is an integral part. Published studies include the role of particular genes in leukaemogenesis (Greaves, 1999; Wiemels & Greaves, 1999), the detection of minimal residual disease (Kasprzyk & Secker-Walker, 1997; Foroni et al, 1999), the identification of tumour suppressor genes (Jackson et al, 2000) and epidemiological studies (UK Childhood Cancer Study Investigators, 2000). Looking ahead, the cumulative effect of additional cytogenetic information will increasingly add to the validity of scientific research and the targeting of therapy to specific abnormalities. During the 1980s, there was a growing awareness of the importance of cytogenetics in the management of patients with haematological disorders. In ALL, it was realized that the cytogenetic results were of paramount significance for prognosis (Secker-Walker et al, 1978; Third International Workshop on Chromosomes in Leukaemia (IWCL3), 1981; Chessells et al, 1997). These observations led to the decision of the trial coordinators to require cytogenetic analysis of all patients with acute leukaemia entered into MRC trials. This resulted in an excessive demand for cytogenetic studies in both National Health Service (NHS) and research-funded cytogenetics laboratories. The analysis of bone marrow samples from patients with leukaemia is highly labour-intensive, requiring great attention to detail and the precise analysis of metaphases. Significant proportions of leukaemia samples produce metaphases of poor quality, increasing the difficulty in analysis and the need for careful review. Thus, escalating workloads in inadequately staffed laboratories led to disastrous consequences for the proper analysis of haematological samples. The number of cells analysed had to be restricted, the detection of abnormal clones was lower than expected, many specimens failed and, in some laboratories, cytogenetic reports were delayed for many months. Three initiatives helped to improve the quality of leukaemia cytogenetics. Firstly, the Association of Clinical Cytogeneticists (ACC) formed a Leukaemia Working Party and extended the UK National External Quality Assessment Scheme (NEQAS) in Clinical Cytogenetics to include leukaemia samples. Secondly, the coordinators of the adult trial, UKALLXA, designated the cytogenetics laboratory at the Royal Free Hospital School of Medicine as a coordinating and referral centre. Thirdly, coordinators of the much larger childhood ALL trial, UKALLX, met with cytogeneticists to devise a strategy to improve cytogenetic results. A panel of experienced haematology cytogeneticists was appointed to validate the analysis of each case and to code the chromosomal findings for computerized entry to the database of the Clinical Trials Service Unit (CTSU). It was then that the idea of a karyotype database was first proposed. In October 1988, the group called the UKCCG was formed. It was appreciated at that time that little of the data on cancer cytogenetics generated in NHS laboratories was ever published and that much valuable information, of interest to research, was thus lost to the world. The aim of the UKCCG was to provide a forum for collaborative research studies between all interested cytogenetics laboratories in the UK and Southern Ireland. This greatly improved the interaction between laboratories and provided a welcome focus for cytogeneticists to discuss problems as well as topics of mutual interest. The UKCCG considered the problems of cytogenetics in relation to the ALL trials. When comparing leukaemia cytogenetics in the UK with that elsewhere in the world, two major differences stood out. Firstly, the UK was the only country in the world in which the karyotype (written description of the chromosomal findings) was not backed-up by a karyogram (a picture of the metaphase chromosomes, paired and arranged in order). Routine preparation of karyograms, considered by many as essential for the detection and comparison of subtle chromosomal changes and incontestably invaluable for the checking of analysis, was not standard practice in the UK. Secondly, in the USA ALL treatment trials, the cytogenetic results were coordinated through databases run by experienced leukaemia cytogeneticists. It seemed that the cytogenetics of ALL in the UK might be improved if procedures applied elsewhere were adopted. During this period, the LRF had also formed the view that cytogenetics was an important, but underfunded, factor in the management of patients with leukaemia. Accordingly, following discussions between the LRF and the UKCCG, the Database was established with significant funding from the LRF. The Database was set up in June 1992 in the newly created LRF Cytogenetics Group in the Department of Haematology, The Royal Free Hospital School of Medicine, London (now the Royal Free and University College Hospital Medical School). Staffing initially comprised two cytogeneticists, a data manager and a part-time secretary, and now includes six full-time posts. Patients treated in one of the UK ALL trials since 1990 are entered onto the Database. The trials are for different age groups. The infant protocols (for 0–1 years of age) are Infant6 (1992 to August 1999) and the current international protocol, Interfant, which opened in August 1999. For children (1–14 years), the trials are UKALLXI (October 1990 to March 1997) and ALL97 (April 1997 to present). The trial for adults (15–55 years) is UKALLXII (January 1993 to present). Thirty laboratories receive samples from patients with leukaemia. Each week a list of patients entered to the trials is sent from the CTSU to the data manager and from there to laboratories for the identification of samples. Cytogenetic analysis is routinely attempted on more than 98% of patients entered to the trials. Cases are prepared and analysed locally, then dispatched to the Database with details of samples and stored material, slides with representative metaphases and a description of the karyotype. The karyotypes are reviewed by the Database cytogeneticists. This includes karyogram preparation, which is often complemented by FISH studies. The extent of the review procedure is highly variable from case to case, and is dependent on the quality of preparations and the level of analysis already undertaken locally. Definitive karyotypes with copies of karyograms and FISH images are sent to the referring laboratories. Final karyotypes, accompanied by clinical and patient details, supplied by the CTSU, are entered to the Database computer. Cases are classified as ‘normal’, ‘abnormal’ or ‘fail’. Abnormal karyotypes are described by chromosome number (ploidy) and chromosomal subgroup, according to established structural changes. This chromosomal classification is required, for each case, by the CTSU. A computer program has been developed for the Database. This enables the storage, retrieval and analysis of cytogenetic and clinical data of all cases. It is based on Paradox software and runs on a computer network. A karyotype search mechanism that interprets karyotype descriptions written in the ‘International System for Human Cytogenetic Nomenclature’ (ISCN, 1995) has been incorporated. This allows the Database to be searched for chromosomal abnormalities involving any part of the karyotype and provides a much more flexible system than that previously used, which relied on coded karyotypes and classification by chromosomal subgroup. The contribution of cytogenetics to trials depends on both the quality of bone marrow samples sent for analysis by the consultant haematologists and on the skill and application of the cytogenetics laboratories. The cytogenetic findings in successive US trials, organized by the Cancer and Leukaemia Group B (CALGB) for adults and the Children's Cancer Group (CCG) and Paediatric Oncology Group (POG) for children, and in two previous UK trials are shown in Table I. These figures indicate that, whereas a successful result was obtained for between 56% and 91% of samples, this represented only 38–80% of patients entered into the trials. Similarly, the 60–76% of successfully analysed patients in whom a clone was detected represented only between 26% and 52% of trial patients. The Database results, shown in Table II, demonstrate the success of its methods. The structured collaboration between the Database and local laboratories, combined with an increased awareness among haematologists, has resulted in an improvement in results over those in comparable studies. In UKALLXI, the 82% success and abnormality rates represent 79% and 65%, respectively, of the total patients in the trial. An earlier summary of the cytogenetic findings in UKALLXI and UKALLXII have been published (Secker-Walker, 1997). An update of UKALLXI is given in Tables III and IV. The prognostic implications of a number of cytogenetic abnormalities have been known for some time. In the UK, the cytogenetic result has had a direct influence on treatment only in the current childhood trial, ALL97. Patients with known, poor-risk chromosomal abnormalities are exempt from randomization and are treated using a high-risk protocol. These include patients who are Philadelphia chromosome (Ph) positive (with fusion of BCR and ABL), those who have an abnormality of chromosome band 11q23 (with rearrangement of the MLL gene) or those who have near-haploidy with 23–28 chromosomes. The ALL97 protocol has recently been revised to include cases with 44 chromosomes or less in the high-risk group in accordance with the CCG protocol (Heerema et al, 1999a). Rapid and accurate identification of these high-risk cases is therefore vital for treatment decisions to be taken early. The cytogenetic findings of any series of ALL patients include a proportion of cases in which cytogenetic analysis is unsuccessful (Tables I and II) or of very poor quality, with an unknown incidence of cryptic cytogenetic change. Therefore, there was concern that relying on cytogenetic analysis alone would result in some high-risk cases being missed. Accordingly, the Database, in collaboration with the UKCCG, has established a molecular cytogenetic screening programme for these abnormalities as well as for certain other important chromosomal abnormalities of prognostic significance. An interphase-FISH ‘ALL-specific, triple-test’ has been developed for the high-risk abnormalities BCR–ABL fusion and MLL rearrangements and for the possibly good-risk ETV6–AML1 fusion of the cytogenetically cryptic t(12;21)(p13;q22) (Kearney, 1999; Harrison, 2000). This screening programme is proving to be practical and successful. Of 1307 patients entered into ALL97 by the end of September 2000, 888 (68%) had been tested, of which 243 (27%) gave a positive result. Patients with BCR–ABL fusion (19 cases), with rearrangement of MLL (26 cases) and ETV6–AML1 fusion (198 patients), have been identified. The numerical chromosomal abnormalities, including near-haploidy, are unlikely to be missed on metaphase analysis. Screening for near-haploidy is therefore undertaken only on cases with a failed or normal cytogenetic result. The use of chromosome-specific, centromeric probes enables the identification of all chromosomal gains and losses, and has the benefit that it enables the detection of both near-haploidy and good-risk patients with hyperdiploidy. Near-haploidy has been detected in five cases and hypodiploidy with 30–36 chromosomes in a further five cases in ALL97. High hyperdiploidy (> 50 chromosomes) has been found in 16 out of 84 (19%) failed and normal cases. In total, the interphase-FISH screening programme has successfully detected an abnormality in 69 cases in ALL97 that would otherwise have failed to produce a cytogenetic result. The application of molecular biology techniques, such as reverse-transcriptase polymerase chain reaction (RT-PCR) and Southern blotting will detect BCR–ABL and ETV6–AML1 fusions and rearrangements of MLL. However, at the present time, these techniques are not available in all regional centres in the UK and there is no system in place for the coordination of molecular results on ALL trial patients. The high level of cooperation between the cytogenetics laboratories and the Database has been responsible for the success of the interphase-FISH screening programme. This emphasizes the value of centralized coordination for such testing. The Database is pioneering new cytogenetic techniques in ALL. These include multiple-colour FISH, multiplex-FISH (M-FISH), in which all 24 chromosomes are ‘painted’ in a different colour (Harrison et al, 2000), sequential FISH, the serial hybridization of probes and chromosome ‘paints’ to the same metaphases (Pinson et al, 2000), and colour banding, in which the chromosomal arms have bands in combinations of seven different colours (Harrison et al, 2001). The application of these procedures has revealed a number of new cryptic chromosomal abnormalities, as shown in Fig 1, which will be investigated in a larger series of patients. This could lead to the identification of new recurring cryptic abnormalities of prognostic significance in ALL, for which appropriately modified therapy could influence survival. A representative metaphase processed using M-FISH from the abnormal bone marrow of a child with ALL. The individual chromosome pairs and the sex chromosomes are hybridized with chromosome-specific probes that ‘paint’ them in 24 different colours. Chromosomal translocations can be identified by colour changes along the chromosome arms. The arrows indicate two copies of a chromosome derived from chromosomes 1 (brown), 10 (blue) and 13 (yellow) that could not be identified from G-banded chromosomal analysis. The results of the Database to date are highly encouraging. Publications and abstracts arizing from the Database are listed in Appendix 1. The close collaboration between the Database and local laboratories provides coordination, valuable support and extra analyses for ALL cytogenetic samples in the UK. Thus, the problems relating to inadequate funding and lack of organization of cytogenetics in previous ALL trials have been largely overcome. The continuation of the Database in its present with further in FISH and molecular techniques, will that the already large collection of karyotypes to and in The value of an series of patient karyotypes in ALL is in such a series will the incidence and significance of rare or complex abnormalities be is no that the level of chromosomal analysis is for the of high-risk patients to the treatment at targeting and specific genetic abnormalities, is now being For the application of such therapy, the comprehensive and accurate description of the karyotype will be of paramount the is available for analysis, which will lead to a of the genetic changes that to leukaemogenesis. The to the Leukaemia Research Fund and the Leukaemia Fund for and & Secker-Walker, for hidden hyperdiploidy in acute lymphoblastic a Leukaemia Research Cancer Cytogenetics Group Chromosomes and & Secker-Walker, in acute lymphoblastic is a significant Chromosomes and Harrison, & Secker-Walker, The importance of cytogenetics and associated molecular techniques in the management of patients with leukaemia. Clinical Harrison, & Secker-Walker, on of the 11q23 Workshop 11q23 chromosomal & Secker-Walker, on of the 11q23 Workshop with a of cytogenetic and clinical & Secker-Walker, on of the 11q23 Workshop with a and clinical study of cases. & Secker-Walker, on of the 11q23 Workshop chromosomes of in & Secker-Walker, on of the 11q23 Workshop The in acute a cytogenetic and clinical profile of patients. & Secker-Walker, on of the 11q23 Workshop The in acute a laboratory and clinical study of cases. & Secker-Walker, on of the 11q23 Workshop The and a cytogenetic and clinical profile of patients. Secker-Walker, on of the 11q23 Workshop on the Workshop on & Secker-Walker, on of the 11q23 Workshop with a laboratory and clinical study of cases. Secker-Walker, Harrison, & of the in the of patients with of the of chromosome A combined and Southern blotting analysis. Harrison, & Secker-Walker, of of cells in Philadelphia positive and high hyperdiploid acute lymphoblastic leukaemia. Harrison, The management of patients with the role of cytogenetics in this molecular of Haematology, Secker-Walker, & Harrison, as a to in acute lymphoblastic leukaemia. Harrison, & of in a and analysis. Cancer & Harrison, An abnormal karyotype involving chromosomes in a child with acute lymphoblastic leukaemia. Secker-Walker, & Cytogenetic classification of hyperdiploidy in childhood acute lymphoblastic an LRF UKCCG of Secker-Walker, & The acute lymphoblastic leukaemia (ALL) karyotype database of Haematology, & Secker-Walker, chromosome in hyperdiploid A Leukaemia Research Fund Kingdom Cancer Cytogenetics Group Study & Secker-Walker, hidden hyperdiploidy in ALL. of Haematology, & Secker-Walker, in acute lymphoblastic leukaemia. of Haematology, & Secker-Walker, or of chromosome in acute lymphoblastic leukaemia. Secker-Walker, & of and poor chromosomal subgroups among the first cases in the LRF UKCCG karyotype database in acute lymphoblastic leukaemia of Haematology, & Secker-Walker, of the of chromosome 1 in acute lymphoblastic leukaemia in the LRF UKCCG karyotype of Haematology, & Secker-Walker, in acute lymphoblastic leukaemia in the LRF UKCCG karyotype a case for of Haematology, Harrison, & Secker-Walker, was identified by FISH in clones with in the LRF UKCCG Leukaemia Karyotype Database. of Medical 1, & Secker-Walker, A cytogenetic database on the Cytogenetics and & Secker-Walker, Abnormal in acute lymphoblastic leukaemia in the LRF UK Cancer Cytogenetics Group karyotype Cytogenetics and & Secker-Walker, High hyperdiploidy and of the of chromosome 1 in acute lymphoblastic leukaemia in the LRF UKCCG Karyotype Database. of Medical 1, & Secker-Walker, of and high hyperdiploidy in the LRF UKCCG Karyotype Database in acute lymphoblastic leukaemia. of Medical 1, Harrison, & Secker-Walker, The success of the Leukaemia Research Fund Kingdom Cancer Cytogenetics Group (UKCCG) Karyotype Database. of Medical Harrison, & Secker-Walker, of hidden high hyperdiploidy in acute lymphoblastic leukaemia. Secker-Walker, & Harrison, of the occur in the of patients with of the of chromosome a combined and Southern blotting analysis. of Haematology, Secker-Walker, & Harrison, rearrangements of the of chromosome in acute lymphoblastic leukaemia. of Haematology, Harrison, & in ALL of Haematology, & Harrison, in ALL of Haematology, & Harrison, in the Kingdom Karyotype Database for of Haematology, Harrison, of FISH in and of Medical & Harrison, or of of Medical 1, Greaves, & Harrison, Cytogenetic to the Greaves, & Harrison, Cytogenetic to the of Haematology, Harrison, & acute leukaemia in children treated for acute lymphoblastic leukaemia. Harrison, & colour fluorescence in situ hybridization all the problems of cytogenetic analysis in of Haematology, & Harrison, of the system in the detection of chromosome abnormalities in haematological Cytogenetics and & Harrison, The at diagnosis and Cytogenetics and & Harrison, The at diagnosis and of Haematology, 1, & Harrison, hyperdiploidy revealed by the Cytogenetics and & Harrison, in acute lymphoblastic leukaemia. Cytogenetics and Harrison, & The value of colour FISH in the cytogenetic analysis of leukaemia. Cytogenetics and & Harrison, FISH into in acute lymphoblastic leukaemia. of Medical Harrison, & and significance of chromosomal abnormalities in childhood acute lymphoblastic leukaemia in the of Haematology, 1, & Harrison, FISH as a prognostic in adults with acute lymphoblastic leukaemia. of Haematology, 1, & Harrison, FISH in adult acute lymphoblastic leukaemia. of Haematology, 1,