Abstract 1043


FLT3-ITD is supposed to function as driver mutation in acute myeloid leukemia (AML), e.g. higher WBC counts are found in FLT3-ITD mutated (FLT3+) cases. Also for NPM1 mutations (NPM1+) higher blast proportions were described. As both mutations can be concomitantly found we investigated correlations and the biological background of these aspects.


We analyzed 805 pts (m/410; f/395; median, 66.6 yrs; 20.0–93.3 yrs) with de novo normal karyotype AML for FLT3+ and NPM1+ and by bone marrow (BM) cytomorphology. Blast definition and classification was done according to FAB and WHO criteria on May Giemsa Gruenwald, MPO, and NSE stained BM aspirates. The FLT3-ITD status and the FLT3-ITD mutant/wildtype ratio were analyzed by fragment analysis (genescan); the NPM1+ status was investigated by melting-curve based PCR assay.


The overall mutation rate for FLT3+ was 219/805 (27.2%), for NPM1+ 391/805 (48.6%). Mean WBC count was higher in FLT3+ cases when compared to the FLT3- (69.0 × 109/l vs. 25.7 × 109/l; p<0.001). The 2-year survival rate was best in the NPM1+/FLT3- pts (n=240; 82.3%) when compared to all other subgroups (NPM1+/FLT3-ITD+: n=151; 55.2%; NPM1-/FLT3+: n=68; 40.5%; NPM1-/FLT3-: n=346; 62.1%; p=0.003). This demonstrates the universal validity of our data set. Frequencies of both FLT3-ITD or NPM1 mutated cases were increasing by blast percentages per decade from 20% to 100%: For the FLT3+: blasts 20–29%: 15/108 (12.2% of cases); 30–39%: 7/91 (7.7%); 40–49: 11/69 (15.9%); 50–59%: 14/80 (17.5%); 60–69%: 25/102 (24.5%); 70–79%: 34/114 (29.8%); 80–89%: 55/123 (44.7%); 90–100%: 58/103 (56.3%) (p<0.001). For NPM1+: blasts 20–29%: 38/123 (30.9% of cases); 30–39%: 31/91 (34.1%); 40–49: 20/69 (29.0%); 50–59%: 36/80 (45.0%); 60–69%: 51/102 (50.0%); 70–79%: 62/114 (54.4%); 80–89%: 83/123 (67.5%); 90–100%: 70/103 (68.0%) (p<0.001). Subsequently, we separated the whole cohort according to the threshold 40% BM blasts: 2 yrs survival rate was significantly lower in pts with ≥40% of blasts (n=591; 63.4%) when compared to those with <40% (n=214; 77.5%; p=0.004). Patients with ≥40% of blasts showed significant higher rates of NPM1+/FLT3+ (blasts <40%: 13/214; 6.1% vs. ≥40%: 138/591; 23.4%; p<0.001) and NPM1-/FLT3+ (blasts <40%: 9/214; 4.2% vs. ≥40%: 59/591; 10%; p=0.009). NPM1-/FLT3- cases were less frequent in cases with ≥40% of blasts (blasts <40%: 136/214 NPM1-/FLT3-; 63.6% vs. blasts ≥40%: 210/591; 35.5%; p<0.001). Focusing on the blast levels in the different molecular subgroups, cases with 90–100% blasts were most frequent in the NPM1+/FLT3+ (double mutated) subgroup (45/151; 29.8%). In the NPM1+/FLT3- and the NPM1-/FLT3+ subgroups, cases with 80–89% of blasts were most frequent (43/240; 17.9%; and 15/68; 22.1%, respectively). In contrast, the NPM1-/FLT3- pts were most frequently from the 20–29% blast category (79/346; 22.8%) (p<0.001 for comparison of all subgroups). A higher FLT3-ITD-mutant/wildtype ratio was correlated with a higher proportion of BM blasts (Spearman; p<0.001). In univariate analysis for prognostically relevant parameters, OS was significantly related to age (p<0.001; RR=1.36 per 10 years of increase), WBC (p<0.001; RR=1.06 per 10×109/l), blasts <40% (p=0.005; RR=2.29), the NPM1+/FLT3- subgroup (p≤0.001; RR=0.40), and a FLT3mut/wildtype ratio ≤0.5 (p=0.024; RR=1.78). No significance was found for gender, Hb, thrombocytes, and FAB subtypes. In multivariate analysis, age (p<0.001), WBC (p=0.003), blasts <40% (p=0.008) and the NPM1+/FLT3- status (p=0.002) retained prognostic significance.


The frequencies of the FLT3+ and the NPM1+ cases are continuously increasing in parallel to increasing blast percentages in normal karyotype AML. Thresholds of 40% of blasts are suitable to discriminate different prognostic groups which can be related to significantly higher frequencies of NPM1+/FLT3+ and NPM1-/FLT3+ in patients with higher blast proportions. Therefore, both molecular markers apparently contribute to blast proliferation in normal karyotype AML. Combinations of these markers are also relevant since blasts were highest in the double NPM1/FLT3-mutated cases and lowest in the NPM1-/FLT3- cases in our study.


Haferlach: MLL Munich Leukemia Laboratory: Employment, Equity Ownership. Haferlach:MLL Munich Leukemia Laboratory: Employment, Equity Ownership, Research Funding. Kern:MLL Munich Leukemia Laboratory: Employment, Equity Ownership. Schnittger:MLL Munich Leukemia Laboratory: Employment, Equity Ownership.

Author notes


Asterisk with author names denotes non-ASH members.