CD22 Exon 12 Deletion as an Independent Predictor of Poor Treatment Outcomes in B-ALL

Abstract

We previously reported a splicing defect (CD22ΔE12) associated with the deletion of exon 12 of the inhibitory co-receptor CD22 (Siglec-2) in leukemia cells from patients with CD19+ B-precursor acute lymphoblastic leukemia (B-ALL). CD22ΔE12 causes a truncating frameshift mutation and yields a dysfunctional CD22 protein that lacks most of the cytoplasmic domain required for its inhibitory function, and it is associated with aggressive in vivo growth of human B-ALL cells in mouse xenograft models. Although CD22ΔE12 with selective reduction of CD22 exon 12 (CD22E12) levels was detected in a high percentage of newly diagnosed as well as relapsed B-ALL patients, its clinical significance remains unknown. We hypothesized that B-ALL patients with very low levels of wildtype CD22 would exhibit a more aggressive disease with a worse prognosis because the missing inhibitory function of the truncated CD22 molecules could not be adequately compensated by competing wildtype CD22. Here, we demonstrate that newly diagnosed B-ALL patients with very low levels of residual wildtype CD22 (“CD22E12low”), as measured by RNAseq-based CD22E12 mRNA levels, have significantly worse leukemia-free survival (LFS) as well as overall survival (OS) than other B-ALL patients. CD22E12low status was identified as a poor prognostic indicator in both univariate and multivariate Cox proportional hazards models. CD22E12low status at presentation shows clinical potential as a poor prognostic biomarker that may guide the early allocation of risk-adjusted, patient-tailored treatment regimens and refine risk classification in high-risk B-ALL.