Prognostic Impact of NPM1/FLT3-ITD genotypes from Randomized Patients with Acute Myeloid Leukemia (AML) Treated within the International Ratify Study

Background: Patients (pts) with AML harboring FLT3 -ITD mutations have poor outcomes. The higher relapse rate and inferior overall survival (OS) largely depend on the ITD mutant to wild-type (wt) allelic ratio. Furthermore, there is a strong gene-gene interaction between NPM1 mutation (NPM1^mut) and FLT3 -ITD. Consequently, the 2017 European LeukemiaNet (ELN) has included defined NPM1/FLT3- ITD genotypes for risk categorization. In the recent phase III RATIFY trial [NCT00651261; Stone et al., NEJM 2017] the addition of the multi-targeted kinase inhibitor midostaurin (M) to standard chemotherapy significantly prolonged OS and event-free survival (EFS) in pts with newly diagnosed AML and FLT3 mutation. Here, we evaluated the prognostic impact in a post-hoc analysis of the NPM1/FLT3 -ITD genotypes as defined by the 2017 ELN from randomized pts treated within the RATIFY trial.

Study Design: In total, 428 of 717 pts were included who gave informed consent for biomarker analyses and who could be categorized to one of the 4 ELN NPM1 / FLT3 -ITD subgroups: NPM1^mut/ FLT3 -ITD^low (n=85), NPM1^mut/ FLT3 -ITD^high (n=159), NPM1^wt/ FLT3 -ITD^low (n=75), and NPM1^wt/ FLT3 -ITD^high (n=109) (allelic ratio: high, ≥0.5; low, <0.5). Median age of these pts was 47 years (yrs) (range, 18 to 59 yrs); median follow-up time was 59 months (mo) (range, 42 to 81 mo). Pts were randomly assigned to receive either M or placebo (PBO) together with standard induction and consolidation therapy followed by twelve 28-day cycles of maintenance therapy with either M or PBO.

Results: With regard to pts and disease characteristics, there was a higher proportion of female pts in the subsets with concurrent NPM1^mut, whereas higher white blood cell counts (WBC) and bone marrow blasts were found in the subsets with FLT3 -ITD^high. Rates of complete remission (CR) positively correlated with NPM1^mut status, but not with FLT3 -ITD allelic ratio (p=.016). CR according to randomization only showed a trend in favor of M in NPM1^wt/ FLT3 -ITD^low pts (p=.058). OS was significantly different among the 4 groups (p=.001); median survival times were mo not reached (n.r.), 27 mo, 20 mo, and 17 mo for NPM1^mut/ FLT3 -ITD^low, NPM1^mut/ FLT3 -ITD^high, NPM1^wt/ FLT3 -ITD^low, and NPM1^wt/ FLT3 -ITD^high subgroups, respectively. This effect was even more pronounced when censoring pts at the time of allogeneic hematopoietic cell transplantation (alloHCT). OS according to randomization showed some differential effects of M among the 4 genotypes (Figure 1) with a significant beneficial effect of M in the NPM1^wt/ FLT3 -ITD^high group (median OS 26 mo vs 14 mo for PBO; p=.025), whereas for the other genotypes beneficial effects of M did not reach statistical significance. OS analysis according to randomization censored at the time of alloHCT revealed an advantage of M for the NPM1^mut/ FLT3 -ITD^low (median n.r. in both arms; p=.038) and NPM1^mut/ FLT3 -ITD^high (median, n.r. vs 18 mo for PBO; p=.032) subsets, while the significant effect present in the non-censored analysis for the NPM1^wt/ FLT3 -ITD^high group was lost (median OS, 19 mo vs 17 mo for PBO; p=0.2). Similar to OS, non-censored EFS significantly differed among the 4 groups (p=.001); median EFS times were 16 mo, 8 mo, 4 mo, and 4 mo for the NPM1^mut/ FLT3 -ITD^low, NPM1^mut/ FLT3 -ITD^high, NPM1^wt/ FLT3 -ITD^low, and NPM1^wt/ FLT3 -ITD^high groups, respectively. EFS analysis according to randomization showed a beneficial effect of M in the NPM1^wt/ FLT3 -ITD^high group (median EFS, 8 mo vs 3 mo for PBO; p=.016). Multivariable analysis including the 4 NPM1/FLT3 -ITD genotypes, treatment arm, sex, age, WBC and alloHCT revealed NPM1/FLT3 -ITD genotypes (OS: p<.0001; EFS: p=.002), treatment arm with M in favor to PBO (OS: p=.011; EFS: p=.003), and WBC (OS: p=.015; EFS: p<.0001) as prognostic variables for OS and EFS while alloHCT was only significant for OS (p<.0001); and sex only for EFS (p=.017).

Conclusions: Data from this large randomized trial suggest the high prognostic value of the NPM1/FLT3 -ITD genotypes considering the ITD mutant to wt allelic ratio. The study was not powered to show differential effects of M among genotypes; however, a beneficial effect of M on OS and EFS appeared most pronounced in the NPM1^wt/ FLT3 -ITD^high group. Multivariate analysis revealed NPM1/FLT3 -ITD genotypes, treatment arm with M in favor to PBO, WBC, and alloHCT as independent prognostic factors for OS.

KD and CT contributed equally

View largeDownload slide

View largeDownload slide

Close modal