The fusion gene CBFB/MYH11 results from a pericentric inversion of chromosome 16, inv(16)(p13.1q22), or less commonly from a t(16;16)(p13.1;q22). This rearrangement is found in 6-8% of adult de novo AML cases and associated with favourable prognosis. Physiologically, CBFB binds to RUNX1 forming the core binding factor (CBF), which is a transcription factor essential for normal hematopoiesis and myeloid development. By disrupting transcription factor activity of CBF, the CBFB/MYH11 fusion protein causes repression of the CBF-dependent target genes resulting in a block of differentiation. Since expression of CBFB/MYH11 alone is not sufficient to cause leukemia it is likely that additional mutations are required for malignant transformation.To systematically identify non-silent mutations, which may collaborate with CBFB/MYH11 during leukemogenesis, we performed whole exome sequencing (WES) of 9 adult AML samples with inv(16) or t(16;16) and matched remission samples. Using this approach, we found 4-11 leukemia-specific sequence variants per patient (median: 6 mutations). These include mutations in genes known to cooperate with CBFB/MYH11 [e.g. NRAS (n=4), KRAS (n=4), FLT3 (n=2), KIT (n=1)] as well as in genes, which have not been described as mutated in AML so far (e.g. ZFHX4, NFE2).To test for recurrent mutations in selected genes, we designed a custom targeted resequencing assay (Haloplex, Agilent), comprising candidate genes from exome sequencing, as well as genes known to be recurrently mutated in AML (129 genes, 396.45 kbp total target sequence). We performed targeted deep amplicon sequencing on the diagnostic samples from 68 CBFB/MYH11-rearranged AML patients with a median read depth of 645. The results are summarized in Figure 1. Twenty-six genes were found mutated in at least 2 patients. The mutation frequencies of NRAS (35%), KRAS (21%), FLT3 (27%), KIT (22%) and WT1 (9%) were similar to previous reports. In addition, we found recurrent mutations with frequencies of above 5% in FAT1 (6%), PTPN11 (6%), NFE2 (6%) and ZFHX4 (6%). Recently, acquired NFE2 (Nuclear Factor, Erythroid-Derived 2) mutations were described in 2-3% of myeloproliferative neoplasms (MPN) cases. These truncating NFE2 mutations confer a proliferative advantage (Jutzi et al., 2013, JEM). ZFHX4 (Zinc Finger Homeobox 4) is required for the maintenance of tumor initiating cells in glioblastoma (Chudnovsky et al., 2014, Cell Rep). In 59% (40/68) of CBFB/MYH11-rearranged patients, more than one additional mutation was identified, but each of them at a distinct variant allele-frequency, indicating clonal heterogeneity.Recurrent mutations described in RUNX1/RUNX1T1-rearranged leukemia (Krauth et al., 2014, Leukemia, n= 139 patients, Micol et al., 2014, Blood, n= 110 patients) affecting genes such as IDH1/2 (4%), JAK2 (3%), ASXL1 (12%) and the recently described ASXL2 (23%) mutation were not found in our patients with CBFB/MYH11-rearranged CBF leukemia or considerably underrepresented (IDH1/2 0%, JAK2 0%, ASXL1 3%, ASXL2 0%). Commonly mutated genes in CBF leukemia like RAS-genes and FLT3 could be detected at higher frequency in our CBFB/MYH11 cohort (RAS-genes: 56% vs. 17%, FLT3: 27 vs. 13%), whereas KIT mutations occur in similar frequency (22% vs. 17-32%).Furthermore, we performed custom targeted resequencing in corresponding relapse samples from 7 patients with inv(16) in order to assess the clonal evolution. Surprisingly, all mutations in kinase genes [NRAS (n=1), KIT (n=2), FLT3 (n=3)] detected in 6 out of 7 cases at diagnosis were lost at relapse. Of note, one of these patients acquired mutations in WT1 and ZFHX4 and another patient acquired a CBL mutation at relapse.Taken together, our findings suggest that the mutatome of CBFB/MYH11-rearranged AML is genetically complex with co-existence of different subclones at diagnosis. Mutations in kinase genes such as NRAS, KIT, FLT3 seem to be unstable during disease progression and the actual driver of relapse remains elusive in many cases. [Display omitted] DisclosuresNo relevant conflicts of interest to declare.
Read full abstract