Introduction

T-cell acute lymphoblastic leukemia (T-ALL) in adults represents a disease with a rather unfavourable prognosis. Despite the fact that treatment stratification and minimal residual disease (MRD) monitoring have improved survival, there is still need to improve outcome by the development of novel targeted therapies. Therefore, molecular alterations are in the focus of on-going research. Until recently only few candidates were identified as recurringly mutated genes including NOTCH1, FBXW7, PTEN. The development of next generation sequencing (NGS) significantly enlarged this spectrum and identified alterations in additional genes (BCL11B, PHF6, DNM2, CNOT3, KRAS, NRAS, DNMT3A). Whereas a number of putative driver mutations have been characterized, the spectrum of recurring alterations in larger cohorts and their relevance in different leukemic subgroups remains unexplored. To unravel relevant recurring alterations in a large cohort of adult T-ALL and to explore potential target genes for novel therapeutic strategies, we performed targeted high throughput NGS of 88 candidates in 81 T-ALL samples.

Patients and methods

We investigated 67 adult T-ALL patients enrolled in the trial 07/2003 of the German Acute Lymphoblastic Leukemia Multicenter Study Group (GMALL). In addition, 14 patients with early T-precursor ALL (ETP-ALL) from other GMALL trials were analysed. Customized biotinylated RNA oligo pools (SureSelect, Agilent) were used to select the targeted regions. We performed 76-bp paired-end sequencing on an Illumina Genome Analyzer IIx platform and reads were mapped to NCBI hg19 RefSeq. For a variant call we required at least a read depth of 20, an allele frequency of 20% and an average base calling quality of Q13. Polymorphisms annotated in dbSNP 135 were excluded. The targeted region comprised 88 genes known to be frequently mutated in ALL, acute myeloid leukemia, myelodysplastic syndrome as well as genes associated with epigenetic regulation, splicing, DNA mismatch repair, and the NOTCHpathway.

Results

We obtained an average of 1.2 Mbp sequence for each sample, resulting in an average coverage of 120 reads for the target region. 79% of the targeted region was covered with a minimum of 20 reads. After exclusion of polymorphism annotated in dbSNP135, 473 single nucleotide variations (SNV) and small indels were identified, 294 of those resulted in changes on the protein level. On average three (3.1) genes per patient were mutated, and 66 (77%) of the 88 genes were mutated in at least one patient. As expected, the highest mutation rate with 53% was found for NOTCH1, with a higher frequency in thymic T-ALL (67.5%) than in early T-ALL (33.3%). Mutation frequencies of FBXW7 (10%), WT1 (10%), JAK3 (12%), and BCL11B (10%) were in the range of reported frequencies. Recently identified novel alterations in DNM2 (17%), PHF6 (11%), DNMT3A (5%) or RELN (5%) were confirmed in our cohort. Interestingly, genes that had not been described in T-ALL included recurring mutations in the histone methyl-transferase MLL2 (11%), frequently mutated in B-cell lymphomas. Like in lymphoma and in the Kabuki syndrome, MLL2 mutations were distributed over the entire gene without any obvious hot spot region. Also the protocadherins FAT1 (15%) and FAT3 (12%) were recurringly altered. FAT1and its inactivation by mutations were recently linked to activation of the WNT pathway in solid tumours.

Affected pathways significantly differed in leukemic subgroups: whereas mutations involving the NOTCH pathway were predominately enriched in the thymic subgroup (75%) and less relevant in early T-ALL (33%, P=0.004), chromatin modifiying genes (17% vs. 5%, P=0.22) and signalling genes (42% vs. 15%, P=0.09) were more frequently mutated in early T-ALL. Spliceosome mutations described in myeloid and mature lymphoid malignancies were present only in a minority (7.4%) of T-ALL.

Conclusion

Adult T-ALL reveals a highly heterogeneous spectrum of candidate gene mutations. Here we provide an original and comprehensive overview of recurring mutations that unravel preferentially pathways altered in specific leukemic subgroups. In addition, we identified novel candidate genes with potential therapeutic implications (FAT1, EZH2, DNMT3A). These mutations have to be validated in a larger cohort with a focus on clinical implications accompanied by functional assays regarding their use as therapeutic targets.

Disclosures:

Krebs:Illumina: Honoraria. Greif:Illumina: Honoraria.

Author notes

*

Asterisk with author names denotes non-ASH members.

Sign in via your Institution