Ikzf1 deletions in b-all: from genetic basis to diagnostic enhancement

Because IKZF1 deletions are associated with an increased risk of relapse in patients with B-cell precursor acute lymphoblastic leukemia (B-ALL), the development of methods for rapid and accurate detection of this alteration has a great clinical impact. Although multiplex ligation-dependent probe amplification (MLPA) has been widely used for the evaluation of IKZF1 deletions, multiplex (M)-PCR is a valuable tool for the validation of those IKZF1 deletions. However, this method is still limited to detect recurrent intragenic deletions (Δ2-3, Δ2-7, Δ2-8, Δ4-7, Δ4-8). Therefore, here we evaluate the breakpoint map of IKZF1 deletions with a special focus on non-recurrent deletions to update methodologies for the diagnosis of this genetic alteration. Moreover, we explore the genetic features underlying those IKZF1 deletions to infer its molecular pathogenesis. This study included four international cohorts of pediatric and adult patients with B-ALL. IKZF1 deletions were identified by MLPA (P335 and P202 assays) and M-PCR. Non-recurrent deletions were verified by next-generation sequencing using the Illumina DNA Prep with Enrichment assay. We also revised previously published data regarding IKZF1 deletions in B-ALL for compiling breakpoint sequence information of these deletions. Sequence data from our cohorts and literature were used in further analyses. After identifying the breakpoint clusters, we performed an agnostic motif search using MEME and annotated the presence of filler DNA or microhomologies at the junctions of IKZF1 deletions. A total of 1,472 B-ALL samples were screened, and IKZF1 deletions were found in 16% of these patients. Although non-recurrent deletions were associated with 1% of the total cohort, it represents 8% of all IKZF1 deletions. The analysis of breakpoint sequences at DNA level revealed that non-recurrent IKZF1 deletions comprises miscellaneous alterations, but the majority of them encompases Δ1 and Δ1-3. Considering that IKZF1 exon 1 deletions are more difficult to reliably detect by MLPA due to the high GC content (>50%) of this region, we mapped several exon 1 deletions at DNA level and identified one commonly deleted region which is ideal for designing MLPA probes. Next, we searched for motifs, additional nucleotides/microhomologies associated with IKZF1 deletion sequences. We verified an enrichment of the 5’-CACACAGTG-3’sequence within 21 out of 24 breakpoint clusters of IKZF1 deletion (E-value: 2.9e-008). Additionally, the filler DNA within the junctions of IKZF1 deletions presented a higher GC nucleotide content at these clusters as compared to the deletions outside of them (71% vs. 56%). This study provides the genetic landscape of IKZF1 deletions in B-ALL. Although rare types of deletion may be identified at the routine diagnosis, they should be considered after using adapted MLPA protocols, including additional probes within the common deleted region nearby IKZF1 exon 1. We also verified the occurrence of the heptamer sequence and an enrichment of GC nucleotides exclusively within breakpoint clusters, suggesting that RAG recombination and TdT activity may promote most IKZF1 deletions, although rare types of alterations may be associated with another molecular mechanism of leukemogenesis.