OBJECTIVE To identify the incidence of PAX5 deletion in childhood B-lineage acute lymphoblastic leukemia (B-ALL) without reproducible chromosomal abnormalities and to investigate the association between PAX5 abnormalities and prognosis of ALL. METHODS Multiplex ligation-dependent probe amplification was used to determine the copy numbers of PAX5 gene in children newly diagnosed with B-ALL without reproducible chromosomal abnormalities between April 2008 and April 2013 and controls (children with non-hematologic diseases or tumors). The patients were classifiied into deletion group and non-deletion group based on the presence of PAX5 deletion. RESULTS Eighteen (21%) out of 86 children with B-ALL had PAX5 deletion. The deletion group had a significantly higher total white blood cell count at diagnosis than the non-deletion group (P=0.001). The Kaplan-Meier analysis demonstrated that the deletion group had a significantly lower disease-free survival (DFS) rate than the non-deletion group (0.69±0.12 vs 0.90±0.04; P=0.017), but there was no significant difference in the overall survival rate between the two groups (P=0.128). The Cox analysis showed that PAX5 deletion was a risk factor for DFS (P=0.03). CONCLUSIONS PAX5 deletion is an independent risk factor for DFS in B-ALL children without reproducible chromosomal abnormalities.