B-cell lineage acute lymphoblastic leukemia (B-ALL) is thought to arise from the clonal expansion of a single transformed B cell. Minimal residual disease is defined as the existence of leukemic cells unable to be detected by the conventional assessment methods (morphology and immunology). Using these methods it is possible to detect residual disease of > 1 leukemic in 10-100 normal bone marrow cells. The study of a large number of children for a long follow up period might prove that the detection of MRD is a strong prognostic factor of suspending leukemia relapse and strongly influence the treatment protocols. The use of the polymerase chain reaction (PCR) to exploit the diversity of the complementarity determining region along with the specific usage of the immunoglobulin variable heavy chain family (VH), have significantly improved the sensitivity of the method. Amplification of the CDRIII by PCR employing VH and JH consensus primers results in the production of diverse CDIII regions. Direct sequencing of this product creates difficulty in identifying the specific leukemic clone due to increased background. The use of a VH-family specific oligonucleotides can identify the molecular pattern of the individual clone, being at the same time easier to sequence. PCR (with VH1-VH6 and JH consensus primers) was used to study 8 children with B-ALL from the day of diagnosis and for a follow up period of 4 years. Direct sequencing of the 9 PCR products verified that the amplified fragments originated from the third hypervariable region of the rearranged immunoglobulin heavy chain (IgH). Hybridisation of the PCR products of the follow up samples to clone-specific oligonucleotides revealed detectable residual disease in 1/8 cases studied. The ability to identify the molecular pattern can be further used to investigate whether the choice of a specific VH family correlates to clinical parameters of the disease, ie presenting features, response to therapy and outcome.