Genomic imbalances analysis provides new insight into prognostic factors in adult and pediatric T-ALL

• Chromothripsis, genomic complexity, gain at 6q27 (MLLT4) and del(16)(p13)/CREBBP are associated with a poor outcome in T-ALL.

• Gain at 6q27 affecting MLLT4 is a newly-identified recurrent somatic CNV observed in 3% of T-ALL.

Given the poor outcome of refractory and relapsing T-ALL, identifying prognostic markers is still challenging. Using SNP-array analysis, we provide a comprehensive analysis of genomic imbalances in a cohort of 317 newly-diagnosed T-ALL patients including 135 children and 182 adults with respect to clinical and biological features and outcomes. SNP-array results identified at least one somatic genomic imbalance in virtually all T-ALL patients (~96%). Del(9)(p21) (~70%) and UPD(9)p21)/CDKN2A/B (~28%) were the most frequent genomic imbalances. Unexpectedly del(13q14)/RB1/DLEU1 (~14%) was the second more frequent CNV followed by del(6)(q15)/CASP8AP2 (~11%), del(1)(p33)/SIL-TAL1 (~11%), del(12)(p13)ETV6/CDKN1B (~9%), del(18)(p11)/PTPN2 (~9%), del(1)(p36)/RPL22 (~9%), and del(17)(q11)/NF1/SUZ12 (~8%). SNP-array also revealed distinct profiles of genomic imbalances according to age, immunophenotype, and oncogenetic subgroups. In particular, adult T-ALL patients demonstrated a significantly higher incidence of del(1)(p36)/RPL22, and del(13)(q14)/RB1/DLEU1, and lower incidence of del(9)(p21) and UPD(9p21)/CDKN2A/B. We determined a threshold of 15 genomic imbalances to stratify patients into high- and low-risk groups of relapse. Survival analysis also revealed the poor outcome, despite the low number of affected cases, conferred by the presence of chromothripsis (n=6, ~2%), del(16)(p13)/CREBBP (n=15, ~5%) as well as the newly identified recurrent gain at 6q27 involving MLLT4 (n=10, ~3%). Genomic complexity, del(16)(p13)/CREBBP and gain at 6q27 involving MLLT4 maintained their significance in multivariate analysis for survival outcome. Our study thus demonstrated that whole genome analysis of imbalances provides new insights to refine risk stratification in T-ALL.