| Abstract: | Crosslineage T-cell receptor delta (TCR delta) rearrangements are widely used as tumor markers for the follow up of minimal residual disease in childhood B-precursor acute lymphoblastic leukemia (ALL) by polymerase chain reaction (PCR). The major drawback of this approach is the risk of false-negative results due to clonal evolution. We investigated the stability of V delta 2D delta 3 rearrangements in a group of 56 childhood B-precursor ALL patients by PCR and Southern blot analysis. At the PCR level, V delta 2D delta 3-to-J alpha rearranged subclones (one pathway for secondary TCR delta recombination) were demonstrated in 85.2% of V delta 2D delta 3-positive patients tested, which showed that small subclones are present in the large majority of patients despite apparently monoclonal TCR delta Southern blot patterns. Sequence analysis of V delta 2D delta 3J alpha rearrangements showed a biased J alpha gene usage, with HAPO5 and J alpha F in 26 of 32 and 6 of 32 clones, respectively. Comparison of V delta 2D delta 3 rearrangement status between diagnosis and first relapse showed differences in seven of eight patients studied. In contrast, from first relapse onward, no clonal changes were observed in six patients studied. To investigate the occurrence of crosslineage TCR delta rearrangements in normal B and T cells, fluorescence-activated cell sorter-sorted peripheral blood CD19+/CD3- and CD19-/CD3+ cell populations from three healthy donors were analyzed. V delta 2D delta 3 rearrangements were detected at low frequencies in both B and T cells, which suggests that V delta 2-to-D delta 3 joining also occurs during normal B-cell differentiation. A model for crosslineage TCR delta rearrangements in B-precursor ALL is deduced that explains the observed clonal changes between diagnosis and relapse and is compatible with multistep leukemogenesis of B-precursor ALL. |
