Wilms Tumor gene 1 (WT1) is overexpressed in the vast majority of patients (pts) in Acute Myeloid Leukemia.(1) Recently the decrease of WT1 expression was suggested to reflect cytoreduction after chemotherapy in AML.(2) However, despite its potential role in minimal residual disease (MRD) assessment, at the moment there is no consensus on its clinical application.


233 clinically and molecularly well characterized consecutive AML pts were treated with intensive induction chemotherapy according to age, performance status and co-morbidities in a single hematology center. WT1 expression was analyzed by Real Time Polymerase Chain Reaction (RT-PCR), using the standardized European LeukemiaNet (ELN) method, at baseline and after induction chemotherapy. Further, WT1 expression was compared with several demographic (sex, age), clinical (organomegaly and leucocytosis), biological (genetic and genomic aberrations, immunophenotype) and therapy related variables.

Aim of this retrospective analysis was to investigate the role of WT1 expression in risk stratification of de novo AML, its potentially added value to already established risk factors and its possible role for response evaluation.


WT1 expression at diagnosis was determined in the bone marrow of all pts with a mean value of 9379 copies (range 1-83200 copies). There was no statistical significance in the difference of WT1 expression when pts were divided on the basis of demographic or clinical variables. Interestingly with increasing age of pts at diagnosis WT1 expression decreased significantly, for instance the patient group 19-30 years (yrs) showed higher copies than the 41-50 yrs- or the 51-60 yrs group, p=0.019 and p=0.004, respectively. Organomegaly or amount of blasts in peripheral blood didn’t show any correlation, whereas a high number of blasts in the bone marrow expressing AML specific antigens displayed significant correlation with high WT1 expression (p=0.024).

There was no significant correlation between WT1 expression and cytogenetic abnormalities, whereas there was a significant correlation between WT1 expression and the presence or absence of FLT3. Patients with FLT3 mutations (n=38) had significantly higher WT1 expression levels than pts with FLT3 wild type (wt) (n=170) at baseline, 58709,4 vs 8710,3 copies (p=0.0008). Significant correlation was also obtained when subdividing the pts on the basis of NPM1 mutations. Patients presenting a NPM1 mutation showed higher WT1 expression than NPM1 wt pts (p=0.018). Even the combination of FLT3 and NPM1 yielded statistically significant correlation with WT1 expression. Patients with FLT3 and NPM1 mutations had higher WT1 copies than FLT3 and NPM1 wt pts (p=0.028).

Bone marrow analysis of WT1 expression was repeated after induction chemotherapy in 133 pts (57%) with a mean value of 891 copies (range 0-23261 copies).

The achievement of complete remission (CR) according to ELN criteria (3) in 110 pts was related to significant decrease of WT1 expression (p<0,0001), whereas 23 non-responders didn’t show significant diminution in WT1 expression (p=0,1) after induction chemotherapy.

Moreover, 17 CR pts with high WT1 expression (>350 copies) after first induction chemotherapy showed reduced disease-free survival (DFS) compared to 89 pts in CR with low WT1 expression (<350 copies, median 16 months vs 40 months, p=0.01).

In conclusion, in the present cohort high WT1 expression correlates significantly with high disease burden as expressed by a high number of AML blasts in the bone marrow at diagnosis as well as acknowledged bad prognosis markers such as FLT3 mutation at diagnosis. Additionally, significant reduction of the WT1 expression level is detected in the patient group that shows complete response after intensive induction chemotherapy. The subgroup of pts with CR after induction chemotherapy who failed to reduce significantly WT1 expression had reduced DFS, probably because of higher burden of residual disease.

These results may support the value of the quantitative assessment of WT1 expression in the bone marrow as an additional potential marker for therapy response evaluation in AML and warrant a prospective study in order understand better its value.

(1) Cilloni D, Saglio G. Acta Haematol 2004;112:79-84

(2) Grimwalde H, Vyas P, Freeman S. Curr Opin Oncol 2010;22:656-663

(3) Dohner H, Estey EH, Amadori S et al. Blood 2010;115:453-474


No relevant conflicts of interest to declare.

Author notes


Asterisk with author names denotes non-ASH members.