Analysis of WT1 Mutations and Combined Aberrations in Childhood Acute Myeloid Leukemia using Whole Exome Sequencing

Introduction:

Mutations in the WT1 zinc finger gene (OMIM: 607102) are associated with an unfavorable clinical outcome in pediatric AML (Ho et al. 2010). In a subgroup of WT1^mut patients additional mutations in other AML typical genes like FLT3 are also detectable. Balgobind et al. (2011) showed a worst outcome for pediatric patients with WT1-mutations combined with a FLT3-ITD (5-year overall survival=0,22, SE= 0,14). We and others identified combined mutations not only in the FLT3-TKD but also other genes like NPM1, nRAS or c-KIT. Using DNA from WT1^mut AML pediatric patients with combined mutations, we performed whole exome sequencing (WES) to identify additional genetic aberrations to gain further insight into the genetic profile of this disease.

Methods:

From a cohort of 19 WT1^mut patients, we selected 10 patients with CN and additional mutations in the FLT3-TKD, NPM1 or n-RAS gene. Concentration and quality of the purified genomic DNA (gDNA) was determined by Agilent Technologies 2100 Bioanalyzer (Agilent Technologies; Waldbronn, Germany). Fragmentation of 100ng gDNA was performed on a Covaris S2 (duty cycle 10%, intensity 4, 200 cycles per burst for 80s) to obtain fragments with an average length of 300 base pairs (bp). The DNA sequencing library was generated from 100ng of fragmented gDNA using Agilent SureSelectXT Reagent Kits v5 (50Mb) according to the manufacturer´s protocols. The final DNA sequencing library was purified, size-controlled by Agilent Technologies 2100 Bioanalyzer (High Sensitivity DNA Chip) and prepared for sequencing according to the manufacturer´s protocol (Illumina). The libraries were sequenced on Illumina HiSeq2500 using TruSeq SBS Kit v3-HS (200 cycles, paired end run) with an average of 12.5 x 10⁶ reads per single exome (mean coverage: 50x). The GATK Pipeline (GenomeAnalysisTK-1.7) was applied for read quality trimming, read alignment to reference (hg19) and quality trimmed variant calling (McKenna et al., 2010). Variant annotation was done using Annovar (Wang et al., 2010).

Results:

In 19 out of 164 patients, who were screened for WT1, mutations in exon 7 or 9 of the WT1- gene were detected. Patients with WT1^mut had an inferior 2-years event free survival compared to patients with WT^wt (0.18, SE=0.09 vs. 0.61, SE=0.04, p=0,0001). DNA from WT1^mut patients (n=10) with normal karyotype CN and additional mutations in the FLT3-TKD, NPM1- , or the n-RAS gene were elected for WES.

The Annovar annotation suggests that each sample contains 5,359 – 5,480 genes that are mutated and where the mutations are predicted to have an effect. 2,324 of these mutated genes were present in all 10 samples together. After review of the literature, ZNF717, MUC16, MAP2K3, PRIM2, PKD1L2, MUC6, CDC27, TTN and MUC5B were identified as candidates for ongoing validation analysis. With respect to functional correlation, 90 - 113 frameshift deletions, 68 - 93 frameshift insertions, 103 - 125 non-frameshift deletions, and 96 - 118 non-frameshift insertions were predicted.

Discussion:

WT1^mut AML pediatric patients with combined mutations in other AML- specific genes present a special subgroup with an adverse prognosis. Nevertheless, more information about the genetic network is necessary to understand the pathobiology and to improve treatment of the disease. For this purpose, we decided to perform whole exome sequencing in 10 samples from patients who were treated after the AML BFM 04 protocol. We identified mutations in more than 9 genes and several aberrations in functional regions of all samples. Currently, validation of these findings is being performed to evaluate the role of the mutated genes. Nevertheless, the results of this analysis will also shed also light onto the clonal evolution of the leukemic clone and the importance for the treatment and the prognosis of the patients.