Novel PDGFRB fusions in childhood B- and T-acute lymphoblastic leukemia

Acute lymphoblastic leukemia (ALL) is among the most curable childhood cancers with a 90% overall survival in the good-risk category.1 Though the remission and cure rates among newly diagnosed patients have improved, the prognosis for those with relapsed disease remains poor. Better risk stratification integrating cytogenetics and genomics has enabled improvements in therapy and prediction of relapse. Recently the clinical utility of comprehensive genomic profiling (CGP) has been recognized, in particular, for risk stratification of about 25–30% of childhood ALL patients who do not appear to harbor any of the established chromosomal abnormalities (Supplementary Information).2 By way of illustration, a novel subgroup of BCR-ABL1-like ALL has been described, in which patients lack the BCR-ABL1 fusion but exhibit a gene expression profile similar to that of BCR-ABL1-positive ALL.3 Furthermore, some of these patients have been demonstrated to harbor tyrosine kinase-activating fusion genes, notably including PDGFRB and JAK2 fusions. The EBF1-PDGFRB gene fusion has been well studied and found to be responsive to the ABL1 tyrosine kinase inhibitor (TKI), imatinib mesylate.4, 5, 6 Here, we have identified three novel PDGFRB fusions (PDGFRB fused to each of SATB1, AGGF1 and DOCK2) in children with either precursor B-cell (B-ALL) or T-cell ALL (T-ALL). We report the clinical utility of CGP to characterize these previously undescribed activating fusion genes, in conjunction with childhood B- or T-ALL, and describe responses to TKIs.