Abstract
Abstract 1576
Poster Board I-602
In cytogenetically normal acute myeloid leukemias (CN-AML), mutations of nucleophosmin (NPM1) gene and internal tandem deletion of Fms-related tyrosine kinase 3 (FLT3-ITD) are used to stratify them into prognostically distinct groups: CN-AML patients with mutated NPM1 (NPM1-mt) and wild type FLT3 (FLT3-wt) form a group with favorable prognosis, whereas those with wild type NPM1 (NPM1-wt) and FLT3-ITD make up the less favorable prognosis group. FLT3 mutations in tyrosine kinase domain (FLT3-TKD) are much less common and their prognostic significance is uncertain. Molecular testing for these abnormalities is expensive, time consuming and not widely available.
We have retrospectively correlated flow cytometric profile of patients with CN-AML with abnormalities of NPM1 and FLT3, in 79 unselected CN-AML patients between the ages of 19 and 60, and who also had blast cell DNA available for determination of FLT-3 and NPM1 mutation status. The FLT3-ITD was detected by DNA PCR and the NPM1 mutation and FLT-3 point mutation by direct sequencing of exons 12 and exon 20, respectively. The 4-color flow cytometry was performed using some or all of the following antibodies: HLA-DR, CD2, CD3, CD4, CD5, CD7, CD8, CD10, CD11b, CD11c, CD13, CD14, CD15, CD19, CD20, CD22, CD33, CD34, CD45, CD56, CD61, CD64, CD71, CD79a, CD117, TDT, and MPO. All the plots were reviewed by two investigators (BD, MM) and listmode files reanalyzed as necessary. The antigen expression was recorded as negative/positive, dim/moderate/bright and partial/complete and correlated with NPM1-mt, FLT3-ITD and FLT3-TKD using 2×2 contingency tables and Fisher's exact test.
42 (53%), 34(43%) and 5 (6%) of 79 patients were NPM1-wt, FLT3-ITD and FLT3-TKD respectively. NPM-wt patients were more likely to be CD34+ (93%, p<0.0001, positive predictive value 89%, see figure); moreover, all patients with aberrant expression of CD2 (n=6) were NPM-wt (p=0.028). FLT3-ITD patients were more likely to be CD56+ (62%, p=0.02). 15/79 (19%) patients had unfavorable combination of NPM1-wt / FLT3-ITD, while 18 (23%) had favorable combination of NPM1-mt / FLT3-wt. The other two combinations NPM1-mt / FLT3-ITD and NPM1-wt / FLT3-wt had 19 (24%) and 27 (34%) patients respectively. Those with favorable combination were more likely to be CD56- (93%, p=0.0062, see figure), than all others. The patients with FLT3-TKD tended to be CD19+ (3/5) compared to those without TKD (15/71); however, due to low frequency of this mutation, the data is not statistically significant (p=0.0831). The remaining antigens, the intensity of staining and duality of populations (only a subpopulation positive) did not correlate NPM1 or FLT3 gene aberrations, nor with any of their combinations.
In CN-AML patients, flow cytometric demonstration of CD2+ and CD34- can predict NPM1-mt, while CD56+ can predict FLT3-ITD. CD56 can also identify the patients with favorable combination of NPM1-mt/FLT3-wt.
No relevant conflicts of interest to declare.
This icon denotes an abstract that is clinically relevant.
