Population-based study of the pattern of molecular markers of minimal residual disease in childhood and adult acute lymphoblastic leukemia: an assessment of the practical difficulty of representative sampling for trial purposes. Northern Region Haematology Group.

BACKGROUND

The prevalence, in unselected patients with acute lymphoblastic leukaemia (ALL), of clonal rearrangements suitable for minimal residual disease (MRD) studies has not been formally investigated.

PROCEDURE: This was a prospective, demographic study of the frequency of molecular markers of MRD in all patients with ALL presenting over 5 years within the Northern Health Region of England (population 3.1 million). Presentation marrow samples were examined to detect informative markers.

RESULTS

One hundred twenty-four children (age <15 years) developed non-Burkitt ALL. No material was available for study in 21. Eighty-six had clonal gene rearrangements (BCR/ABL, immunoglobulin heavy chain (IGH) and/or T cell receptor (TCR) gene rearrangements). All entered remission; 84 (68% of the original cohort) survived to become eligible for MRD studies. One hundred sixteen adults developed ALL, of whom 48 were not studied due to insufficient cellular material in the bone marrow aspirate or to logistical problems in central referral of samples from other hospitals. Material from elderly adults (age >55 years) was less likely to be sent for analysis, 36% vs. 59% (P = 0.024). Thirty-eight had BCR/ABL and/or IGH/TCR gene rearrangements. Thirty-one (27% of the original cohort) entered remission and became eligible for MRD studies. Informative gene rearrangements were more common in children than adults (83% vs. 63%, P < 0.003).

CONCLUSIONS

The results reveal substantial potential, unintentional, selection bias. Large-scale multicentre studies of MRD in children may well produce clinically relevant and representative data. Those who mount similar studies in adults should not assume they will be similarly representative or as successful in accrual of material.