Analysis of ASXL1, IDH1, IDH2, c-CBL, and WT1 Mutations in De Novo Acute Myeloid Leukaemia

Abstract 1450

The clinical relevance and prognostic implications of some recently identified mutations in acute myeloid leukemia (AML) is not yet well established. Among them, we have selected to be analyzed those affecting the following genes: Additional Sex Combs-Like 1 (ASXL1), Isocitrate Dehydrogenase (IDH1 and IDH2), Casitas B-lineage Lymphoma (c-CBL), and Wilms Tumor 1 (WT1). They have been previously reported with a variable incidence: ASXL1 mutations in 10.8% patients with normal karyotype (NK), IDH1 and IDH2 mutations in 8 – 33% of de novo AML, c-CBL mutations in 2% of de novo AML, and WT1 mutations in 5–12% of de novo AML patients.

In order to know the incidence and prognostic impact of these mutations and their possible cooperative role in leukemogenesis, we have screened for ASXL1, IDH1, IDH2, c-CBL, WT1, FLT3, NPM1 and CEBPa, mutations in a cohort of de novo AML patients from a single centre.

We studied 174 de novo AML patients [98M/76F; median age: 62 yr. (range: 16 – 88); favourable (n= 13), intermediate (n= 86) and high (n= 51) cytogenetic risk classification by the MRC group]. DNA was isolated from bone marrow samples obtained at diagnosis.

In order to determine cooperating mutations, we developed a new combination of high-resolution melting (HRM) assays on a LightCycler® 480 and lastly direct sequencing, to detect somatic mutations for ASXL1 (exon 12), IDH1 (exon 4), IDH2 (exon 4), WT1 (exons 7, 8 and 9) and c-CBL (exons 8 and 9). All mutations reported in this study were confirmed al least twice. FLT3 (ITD and D835Y), NPM1 (exon 12) and CEBPa were performed as described previously by standard methods. Sequence analysis was checked by its corresponding GeneBank Accession Number.

The number of patients found to carry mutations in our series was: 16 patients with ASXL1 mutations (9.2%), 16 patients with IDH mutations (2.9% had a IDH1^R132, 12.6% the SNP rs11554137 and 6.3% IDH2^R140), 5 patients with WT1 mutations (2.9%), 37 patients with FLT3 mutations (21.3%), 44 patients with NPM1 mutations (25,3%) and 8 patients with CEBPa mutations (4.6%). No mutations where found in c-CBL.

We could not found a pattern of cooperating mutations in the studied group of genes. WT1, FLT3 and NPM1 were associated with leukocyte count >30 × 10⁹/L at diagnosis (80% vs. 31% for WT1, P =0,022; 68% vs. 22% for FLT3, P= 0.001; and 50% vs. 24% for NPM1, P= 0.002; in mutated vs. wild-type patients, respectively). WT1 was also associated with a platelet count > 50 × 10⁹/L at diagnosis (100% vs. 57% in mutated vs. wild-type patients, respectively; P =0,048). Besides, FLT3 and NPM1 mutations were more frequent in the intermediate cytogenetic risk group (82% and 74%; P =0.004 and P =0.047; respectively). ASXL1 and IDH mutations were not correlated with any of the clinical and biological features studied.

In univariate analysis, only age and cytogenetics had an impact on overall survival (OS, median of 12mo vs. 3mo, for patients < and ≥65 yr., P <0.001 and 24mo, 11mo and 3mo for favourable, intermediate and high risk, P =0.005). Mutational status of ASXL1, IDH1, IDH2, WT1, FLT3, NPM1 and CEBPa did not impact on outcome in the whole series. However, when the analysis was restricted to patients with intermediate cytogenetic risk, patients with FLT3 mutations had a shorter OS (19mo vs. 8mo, wild-type vs. mutated patients; P =0.047) and those with WT1 mutations showed a trend towards an inferior OS (11mo vs. 1mo, wild-type vs. mutated patients; P = 0.066).

In multivariate analysis in patients with intermediate cytogenetic risk, the age [HR (95% CI) = 3.3 (1.9 − 5.9) P <0.001], and FLT3 status [HR (95% CI) = 2.2 (1.2–3.9) P =0.008] retained an independent adverse significance for OS. In terms of relapse free survival any of the variables showed a significant implication.

To sum up, the incidence found for the studied genes was lower than the previously reported: ASXL1, 9.2%; IDH1^R132, 2.9%; IDH2^R140, 6.3%; WT1, 2.9%; and c-CBL, 0%. We were unable to find a pattern of cooperating mutations in the studied group of genes or any impact of these mutations on the outcome.