Evaluation Of Different NPM1 Mutations In AML Patients According To Clinical, Cytogenetic and Molecular Features and Impact On Outcome

![Graphic][1] Background NPM1 mutations ( NPM1 mut) are known to be prognostically favorable in the absence of FLT3 -ITD. About 80% of patients (pts) have previously been shown to carry a distinct NPM1 mut (“type A”). The remaining NPM1 mut pts show various rare mutations (“rare types”) with different individual frequencies. Whether these “rare types” differ from “type A” with respect to clinical outcome and accompanying genetic abnormalities is unknown. Aim To evaluate the incidence of “rare types” of NPM1 mut in AML and to investigate whether they show differences in cytogenetics, concomitant molecular mutations, biological or prognostic profiles compared to pts with the common “type A”. Patients and Methods A cohort of 2,859 newly diagnosed adult AML pts were screened for NPM1 mut by melting curve analysis and subsequent Sanger sequencing of the positive cases. In 877 (30.7%) pts NPM1 mut were detected. All 877 NPM1 mut pts were also analyzed for FLT3 -ITD. The following markers were screened in the respective subcohorts: FLT3- TKD (n=869), MLL -PTD (n=872), WT1 (n=785), CEBPA (n=690), IDH2 R172 (n=590), IDH1 R132 (n=582), IDH2 R140 (n=577), RUNX1 (n=531), DNMT3A (n=529), ASXL1 (n=432), NRAS (n=390), KRAS (n=231), and TET2 (n=216). Results In total, 660/877 (75.3%) mutations were “type A” mutations (c.860\_863dupTCTG), whereas 217/877 (24.7%) represented “rare types”. The most common of the latter was type B (c.863\_864insCATG; n=77; 8.8% of all NPM1 mut) followed by type D (c.863\_864insCCTG; n=56; 6.4%), and type I (c.863\_864insCTTG; n=15; 1.7%). The remaining 41 types were present in less than 1% each. Pts with “rare types” were slightly younger than pts with “type A” (mean 61.4 yrs vs 63.8 yrs; p=0.035) and showed a trend for lower WBC counts (50.6 vs 61.2x109/L; p=0.051). Neither gender, hemoglobin level, platelet count, the history of the disease, the percentage of bone marrow blasts, normal vs aberrant karyotypes, nor distribution between the intermediate and adverse cytogenetic risk scores according to the refined MRC criteria varied between “rare types” and “type A”. “Rare types” more often harbored WT1 mutations than “type A” pts (10.6% vs 3.3%; p<0.001) and less frequently had DNMT3A mutations (33.0% vs 58.0%; p<0.001). No associations were seen for any of the other molecular mutations. Focusing on FLT3- ITD, no differences within the FLT3 -ITD/wt ratio was seen between “rare types” and “type A”. Focusing on the prognostic relevance, we limited our cohort to 603 intermediate-risk karyotypes pts with available information on outcome who have been treated in curative intention: “Rare types” (n=153) had a trend towards a favorable overall survival (OS) compared to “type A” (n=450; median 76.0 vs 62.2 months (m); p=0.083). Same holds true for event-free survival (EFS) (20.3 vs 14.7 m; p=0.084). Restricting the analysis to pts with FLT3 -ITD, “rare types” (n=63) showed significantly better OS than “type A” pts (n=182; 76.0 vs 20.7 m; p=0.049) and as well better EFS (22.8 vs 8.0 m; p=0.021). Even more striking: dividing the cohort to the 4 possible combinations of these two markers we found 1) 63 pts presenting with “rare types” and FLT3 -ITD, 2) 90 with “rare types” and no FLT3 -ITD, 3) 131 pts with “type A” and FLT3 -ITD, and 4) 319 pts with “type A” and no FLT3 -ITD. Only those pts harboring “type A” and FLT3 -ITD showed adverse OS (18.1 m) compared to all other subgroups (p=0.010 vs “rare type” and FLT3 -ITD; p<0.001 vs. the two other subgroups). Furthermore, the three other subgroups did not differ amongst each other in OS (“rare type” and FLT3 -ITD: 76.0 m; “rare type” and no FLT3 -ITD: not reached; “type A” and no FLT3 -ITD: 73.4 m; n.s.; [Figure][2]). This effect is also seen for EFS where “type A” and FLT3 -ITD pts showed adverse EFS (6.9 m) compared to all other subgroups (p=0.006 vs “rare types” and FLT3 -ITD; p<0.001 vs. the two other subgroups). Again, the three other subgroups did not differ amongst each other (“rare types” and FLT3 -ITD: 22.8 m; “rare types” and no FLT3 -ITD: 18.1 m; “type A” and no FLT3 -ITD: 18.7 m; n.s.). ![Figure 1][1] Figure 1 Conclusions 1) Rare NPM1 mut types harbored more frequently WT1 mut and less frequently DNMT3A mut. 2) The adverse effect of FLT3 -ITD mutations within NPM1 mut pts is not seen within the “rare types”, but only in those pts showing NPM1 mut “type A”. 3) Therefore, NPM1 mut pts showing “rare types” might not be stratified by their FLT3- ITD status, but by their prognostically favorable NPM1 mut only. Disclosures: Alpermann: MLL Munich Leukemia Laboratory: Employment. Haferlach: MLL Munich Leukemia Laboratory: Employment, Equity Ownership. Dicker: MLL Munich Leukemia Laboratory: Employment. Eder: MLL Munich Leukemia Laboratory: Employment. Kohlmann: MLL Munich Leukemia Laboratory: Employment. Kern: MLL Munich Leukemia Laboratory: Employment, Equity Ownership. Haferlach: MLL Munich Leukemia Laboratory: Employment, Equity Ownership. Schnittger: MLL Munich Leukemia Laboratory: Employment, Equity Ownership. [1]: pending:yes [2]: #F1