The rearrangement patterns of Ig and T-cell receptor (TcR) genes were studied by Southern blot analysis in 30 precursor B-cell acute lymphoblastic leukemias (B-ALLs) and 10 T-ALLs at diagnosis and subsequent relapse. Eight precursor B-ALLs appeared to contain biclonal/oligoclonal Ig heavy-chain (IgH) gene rearrangements at diagnosis. Differences in rearrangement patterns between diagnosis and relapse were found in 67% (20 cases) of precursor B-ALLs (including all eight biclonal/oligoclonal cases) and 50% (five cases) of T-ALLs. In precursor B-ALLs, especially changes in IgH and/or TcR-delta gene rearrangements were found (17 cases), but also changes in TcR-beta, TcR-gamma, Ig kappa, and/or Ig lambda genes (11 cases) occurred. The changes in T-ALLs concerned the TcR-beta, TcR-gamma, TcR-delta, and/or IgH genes. Two precursor B-ALLs showed completely different Ig and TcR gene rearrangement patterns at relapse, suggesting the absence of a clonal relation between the leukemic cells at diagnosis and relapse and the development of a secondary leukemia. The clonal evolution in the other 23 ALL patients was based on continuing rearrangement processes and selection of subclones. The development of changes in Ig and TcR gene rearrangement patterns was related to remission duration, suggesting an increasing chance of continuing rearrangement processes with time. These immunogenotypic changes at relapse occurred in a hierarchical order, with changes in IgH and TcR-delta genes occurring after only 6 months of remission duration, whereas changes in other Ig and TcR genes were generally detectable after 1 to 2 years of remission duration. The heterogeneity reported here in Ig and/or TcR gene rearrangement patterns at diagnosis and relapse might hamper polymerase chain reaction (PCR)-mediated detection of minimal residual disease (MRD) using junctional regions of rearranged Ig or TcR genes as PCR targets. However, our data also indicate that in 75% to 90% of ALLs, at least one major rearranged IgH, TcR-gamma, or TcR-delta band (allele) remained stable at relapse. We conclude that two or more junctional regions of different genes (IgH, TcR-gamma, and/or TcR-delta) should be monitored during follow-up of ALL patients for MRD detection by use of PCR techniques. Especially in biclonal/oligoclonal precursor B-ALL cases, the monitoring should not be restricted to rearranged IgH genes, but TcR-gamma and/or TcR-delta genes should be monitored as well, because of the extensive changes in IgH gene rearrangement patterns in this ALL subgroup.