Cytogenetic studies of 103 patients with acute myelogenous leukemia in relapse.

In order to investigate the cytogenetic patterns in relapsed acute myelogenous leukemia (AML), a clinical and cytogenetic follow-up of patients newly diagnosed for the Fourth International Workshop on Chromosomes in Leukemia (4IWCL) was evaluated at the 6IWCL. Information was received on 103 patients in relapse who were then classified into seven groups according to the diagnostic karyotype. These groups were: normal, t(8;21), t(15;17), +8, a single specific abnormality either numerical or structural other than those already listed, a single nonrandom or miscellaneous abnormality again either numerical or structural, and complex abnormalities. The patient's age, diagnostic FAB type, the number of relapses, the total survival time, and the karyotype in relapse were considered in each of these cytogenetic groups. The remission and survival rates were comparable in all groups except the +8 group, where patients relapsed earlier and had a shorter survival time. Multiple relapses occurred most frequently in the t(8;21) group, whereas none of the patients with t(15;17) relapsed more than once, although the total survival time was similar to the two groups. Thirty-nine percent of the patients relapsed with the same karyotype as at diagnosis. A more complex karyotype showing evolution was found in 53%, and 8% showed either a less-complicated karyotype or appeared to have reverted to normal. Numerical abnormalities in relapse frequently involved trisomy of chromosomes 8 and/or 21. There was a nonrandom development of 9q- with relapse in patients with t(8;21). A pericentric inversion of chromosome 4, and abnormality infrequently reported at diagnosis, was found in relapse in association with t(15;17), t(8;21), and +8 karyotypes. Changes considered to be typically secondary in nature involving 5q, 7q, and 12p were seen in only seven cases. Twenty-one patients who had an apparently normal karyotype at diagnosis remained normal in relapse, indicating that absence of clonal chromosome abnormality is a real observation in AML rather than a failure of detection.