Real-world treatment patterns and outcomes among patients with newly diagnosed NPM1-mutated acute myeloid leukemia in the United States Free

Approximately 30% of adults with acute myeloid leukemia (AML) have nucleophosmin 1 mutations (NPM1m). However, standard therapies do not directly target the NPM1 driver mutation. The objective of this real-world study was to describe treatment patterns and outcomes among patients with newly diagnosed (ND) NPM1m AML in the United States.

A retrospective study included adults (≥18 years) with ND NPM1m AML between 1/1/2009 and 6/30/2024. Patients were identified from COTA Healthcare, a real-world oncology database comprising approximately 200 academic and community sites in the United States. Follow-up began at AML diagnosis and continued until the earliest occurrence of relapsed/refractory (R/R) disease, loss to follow-up, or death. Patient characteristics, treatment usage, and response were reported. Treatment regimens were categorized hierarchically as intensive chemotherapy (IC), low-intensity therapy (LIT; including targeted and hypomethylating agents [HMA]), and other (i.e., unknown intensity). Use of maintenance therapy was reported separately for patients who received allogeneic hematopoietic cell transplantation (alloHCT) and for those who achieved a composite complete response (CRc) without alloHCT. Analyses included the full study population (Overall Cohort) and were stratified by FLT3 co-mutation status and by diagnosis year (Early Cohort: 2009–2018; Recent Cohort: 2019–6/30/2024) to evaluate the impact of venetoclax approval (November 2018). Proportions were compared using the 2-sample Chi-squared test. Kaplan-Meier (KM) estimates of risk of R/R disease at specific timepoints were compared using the 2-sample z-test with variances derived via the Greenwood formula.

The Overall Cohort included 742 patients with ND NPM1m AML. Median age at AML diagnosis was 62 years; 60.9% were treated in academic settings. Reported co-mutations included FLT3-ITD (44.6%), DNMT3A (28.8%), IDH2 (15.9%), FLT3-TKD (14.8%), and IDH1 (13.9%). In the frontline setting, 64.8% received IC and 21.2% received LIT. Among those who received IC, the most common regimens were IC-only (65.2%). Among those who received LIT, the most common regimens were venetoclax-based (60.5%). In the Overall Cohort, 57.2% of patients had FLT3 co-mutations (n=425) and 64.0% were diagnosed with AML between 2009 and 2018 (Early Cohort; n=475).

Among patients receiving frontline therapy, 72.9% of patients (n=541) achieved a documented CRc, 27.4% (n=203) received alloHCT, and 46.5% (n=345) achieved CRc without subsequent alloHCT. Among the 203 alloHCT recipients, 33.0% received maintenance therapy. Among the 345 with documented CRc who did not receive alloHCT, 19.4% received maintenance therapy. After a median follow-up of 16.1 months, 46.5% of patients had developed R/R disease (relapsed, 85.8%; primary refractory, 14.2%). The KM-estimated risk of R/R disease at 6 and 12 months was 20.4% and 39.7%, respectively.

The CRc rates were similar between patients with and without FLT3 co-mutation (71.8% vs 74.4%, p=0.465). Compared with those without FLT3 co-mutations, patients with FLT3 co-mutations were more likely to undergo alloHCT (33.9% vs 18.6%, p<0.001) and to receive maintenance therapy afterward (38.9% vs 18.6%, p=0.009). They also had a higher risk of R/R disease at 6 months (24.1% vs 15.4%, p=0.006) and 12 months (44.3% vs 33.2%, p=0.010).

In the Recent Cohort (n=267), frontline IC use decreased (52.1% vs 72.0% in the Early Cohort, p<0.001) and LIT use increased (38.2% vs 11.6%, p<0.001). While CRc (75.3% [Recent] and 71.6% [Early]) and alloHCT rates (27.3% [Recent] and 27.4% [Early]) were similar between cohorts, use of maintenance therapy for patients with CRc who did not receive alloHCT was higher in the Recent Cohort (30.2% vs 12.9% in the Early Cohort, p<0.001). In the Recent Cohort, among patients without alloHCT, 74.4% of patients received an HMA-containing regimen for maintenance. The KM-estimated risk of R/R disease was higher in the Early Cohort (6 months: 22.9% vs 16.0%, p=0.032; 12 months: 42.7% vs 34.5%, p=0.070).

This real-world study showed that rates of R/R disease were high in patients with ND NPM1m AML, including those without FLT3 co-mutations. Outcomes with frontline therapies remain suboptimal despite recent treatment advances such as venetoclax, highlighting the need for novel therapeutic approaches.