PF162 FREQUENCY AND PROGNOSTIC IMPACT OF ZEB2 H1038/Q1072 MUTATIONS IN CHILDHOOD B‐OTHER ACUTE LYMPHOBLASTIC LEUKEMIA

Abstract

Background:Novel biologically and clinically relevant genetic aberrations have been discovered recently via massive parallel sequencing‐based genome‐wide profiling. We performed whole‐exome/‐transcriptome sequencing (WES/WTS) of a consecutive cohort of relapsed pediatric B‐other acute lymphoblastic leukemia (ALL), and we found recurrent mutations of codons H1038 and Q1072 of the ZEB2 gene. ZEB2 encodes a transcriptional repressor implicated in the pathogenesis of early T‐cell acute leukemia. Although recurrence of the ZEB2 mutations (ZEB2m) has been already noticed in B‐other ALL, their role has not been addressed so far.Aims:We aimed to determine the frequency and clinical relevance of ZEB2 mutations in childhood B‐other ALL.Methods:We performed deep amplicon sequencing of ZEB2 codons H1038 and Q1072. Discovery “diagnostic” and “relapse” cohorts comprised 233 and 36 Czech children with newly manifesting and relapsed B‐other ALL, respectively, consecutively diagnosed between November 2002 and December 2017 and treated according to four successive protocols. Validation cohorts comprised German children with newly diagnosed B‐other ALL from BFM 2000 (n = 235) or BFM 2009 (n = 391) studies or with relapsed ALL (n = 102) selected based on availability of material.Results:In the discovery diagnostic cohort, ZEB2m occurred in 3.8% (9/233) of patients. Relapses were significantly more frequent in patients with ZEB2m compared to the remaining patients (Cumulative incidence of relapse (CIR) 67%, SE 18% versus 16%, SE 3%, P = 0.0001). The frequency of ZEB2m in the discovery relapse cohort was 29% (8/36); the enrichment of ZEB2m in relapses compared to initial manifestations was statistically significant (P = 0.0005). The frequency of ZEB2m in the validation diagnostic cohort was 2.7% (17/626). While ZEB2m tended to be associated with higher relapse frequency in patients from BFM 2000 study (CIR 60% SE 26% versus 18% SE 3%, P = 0.03), the relapse rate was low in those from BFM 2009 study (1/12, plus two patients suffered different events). In the validation cohort, ZEB2m frequency did not differ between the diagnostic and relapse cohorts (2.7% versus 4.9%, P = 0.3). Altogether, in 7/8 evaluable patients, diagnostic ZEB2m was preserved in relapse, generally with equal or higher variant allele frequency, suggesting a positive selection of the ZEB2m subclones. Of 20 ZEB2m cases analyzed by WTS, 5 and 2 were classified into DUX4‐rearranged and BCR‐ABL1‐like subtypes, respectively, while 13 cases did not belong to any established ALL subtype. The limited number of ZEB2m‐positive cases did not allow us to properly analyze the potential impact of additional biological/clinical factors on relapse risk neither to compare the distribution of such factors in discovery vs. validation cohorts. Similarly, we could not evaluate the impact of different therapy, ethnic composition or possible sample‐selection bias in validation cohorts. Nevertheless, we did not observe any evident association of relapse occurrence with ALL subtype, risk group, mutation type (H1038R versus Q1072K/R) or mutation load.Summary/Conclusion:In summary, we specified the frequency of ZEB2 H1038/Q1072 mutations in newly diagnosed and relapsed B‐other ALL and their impact on outcome. Although the discovery study suggested ZEB2m as a potential relapse driver, this finding was not univocally validated in independent cohorts. With respect to the relatively low frequency of ZEB2m, studies on even larger cohorts and perhaps a multivariate approach are needed to further dissect their potential prognostic impact. Supported by NV15‐30626A, Primus/MED/28.