Clinical significance of cytogenetic abnormalities in adult acute lymphoblastic leukemia.

ACUTE LYMPHOBLASTIC leukemias (ALL) are characterized by clonal proliferation, accumulation, and tissue infiltration of neoplastic cells. They are mainly regarded as childhood diseases, with an early incidence peak at 2 to 5 years of age, where they represent about 80% of the childhood leukemias in the United States, and occur with an incidence of up to 30 cases per 1 million population per year.1 The age-adjusted incidence of ALL in adults (usually defined as 15 years of age and older) amounts to about one third of that in children.1 However, ALL has a bimodal distribution, with a second peak around age 50 and a low but steady rise in incidence with increasing age.2 Improvements in cytogenetic techniques have yielded significant insight as to the importance of cytogenetic abnormalities in the pathophysiology and prognosis of hematologic malignancies. Heim and Mitelman3 reported an overall increase in the number of reported cases of cancer with cytogenetic alterations from 3,844 in 1983 to more than 22,000 in 1994. Of all neoplasms, leukemias have been by far the most intensively investigated and account for more than 60% of all listed chromosomal aberrations, including more than 3,000 cases of ALL. The majority of cases of ALL demonstrate an abnormal karyotype, either in chromosome number (ploidy) or as structural changes such as translocations, inversions, or deletions. These changes were detected in only half of ALL patients in the first banding studies.3 The scantiness of information gained from chromosomal findings in ALL has been, in large part, due to technical difficulties. Chromosome studies in ALL exhibit poor morphology; chromosomes tend to spread poorly, and appear blurred and fuzzy with indistinct margins, making banding studies challenging or even impossible.4,5 Improvements in spreading and banding techniques have resulted in higher rates of detection, and most studies now report chromosomal changes in 60% to 85% of ALL cases.6-10 The Third International Workshop on Chromosomes in Leukemia (TIWCL) found the majority of cytogenetic changes in cases of Bprecursor ALL, with only 39% occurring in T-cell ALL.6,9 Williams et al4 used a direct technique of bone marrow (BM) chromosomal analysis developed particularly for studies in ALL, which paid attention to sampling and processing steps using specific flaming techniques and modified G-banding procedures. They identified clonal karyotypic abnormalities in 94% to 98% of cases of ALL. Such improved techniques also detected nonrandomly occurring cytogenetic abnormalities in cases with hyperdiploid chromosome numbers (.50) that had previously been classified as normal in karyotype.11 These results showed a high prevalence of clonal chromosomal abnormalities in ALL, as was shown for acute nonlymphoblastic leukemias by Yunis,12 who used high-resolution banding techniques. These and similar studies underscore the significant yield achieved with thorough cytogenetic studies in ALL. That cytogenetic abnormalities confer important prognostic information in ALL was first reported by Secker-Walker et al13 in 1978 in a series of childhood ALL. The investigators reported better clinical outcomes in cases with hyperdiploid karyotypes than in those with hypodiploidy or pseudodiploidy, and these findings were confirmed in the follow-up study14 and by other researchers.8,15-17 The TIWCL examined 330 newly diagnosed ALL patients (172 adults and 157 children) and found that chromosomal abnormalities distinguished high-risk from lowrisk patients. Complete remission (CR) rates, remission durations, as well as disease-free-survivals (DFS) were significantly affected by the karyotypic abnormalities.6 Among adult patients the highest likelihood of cure (21% to 30%) was projected in patients with chromosome numbers of .50, or 47 to 50, with 6q2, or with a normal karyotype.18 As with children, karyotypes in adults were significant independent predictors of remission duration or DFS, after considering covariates such as age, leukocyte count at presentation, or French-AmericanBritish (FAB) morphology.19 Secker-Walker et al10 reported the prognostic effect of certain structural rearrangements, such as t(9;22), to be independent of other single variables. Most studies on karyotypic abnormalities and their clinical significance have been performed in childhood ALL.15,20-24 Adult ALL may show nonrandom chromosomal abnormalities similar to those found in childhood ALL, but their distribution and, possibly, their biological significance are different. Few