Abstract
Introduction The continuously evolving therapeutic landscape of acute myeloid leukemia (AML) in the intensive and non-intensive settings, as well as the routine incorporation of targeted therapies have resulted in markedly improved remission rates and overall clinical outcomes. Irrespective of treatment modality, blood product transfusions retain a pivotal role in the supportive care of AML patients, especially during the induction phase. Whether transfusion burden during induction is significantly impacted by treatment intensity and/or disease characteristics is currently undefined.
Methods Retrospective chart review of electronic medical records was conducted for newly diagnosed AML patients who received induction treatment from 2021-2025 at the Weill Cornell Medicine and The New York Presbyterian Hospital. Patients provided informed consent for data collection and analysis and were stratified according to treatment intensity and disease risk using the ELN 2022 for intensively treated patients and the ELN 2024 for non-intensive treated patients. Records were reviewed for packed red blood cell (PRBC) and platelet (plt) transfusions, early death rate, and clinically significant bleeding during induction. Patients were transfused according to standard institutional practices, with routine transfusions administered for hemoglobin <7.5g/dL, plt≤10x10(3)/μL, symptomatic anemia, and clinically significant bleeding.
Results The analysis comprised 208 patients, with 108 intensive and 100 non-intensive inductions. Median age was 66 years (range 20-89), and intensively treated patients were significantly younger (56 years, range 20-75), than non-intensively treated patients (75 years, range 49-89; p<0.001). Using the ELN 2022 and ELN 2024 risk models for intensively and non-intensively treated patients, respectively, patients receiving non-intensive treatment were more likely to be designated as favorable risk compared with intensively treated patients (55.1% vs. 29.6%; p<0.001). The frequency of mutations in genes related to epigenetic/chromatin modifiers (63.2%), NPM1 (21.7%), signaling pathways (53.2%), transcription factors (41%), cohesin complex (9.9%), and tumor suppressor genes (26%) were not significantly different between treatment groups whereas patients in the non-intensive group were more likely to harbor RNA splicing factor mutations compared with their intensive treatment counterparts (40.8% vs. 19.8%; p=0.001). Plt counts at days 8, 15, and 22 from initial induction were significantly lower in intensively treated patients (median 20K vs. 28K; p=0.009, 13K vs. 20K; p<0.001, and 25K vs. 38K; p=0.014, respectively). Patients receiving intensive induction were more likely to experience clinically significant bleeding events (16.7% vs. 5%; p=0.008), received significantly more PRBC transfusions (median 7 units vs. 5 units; p=0.003) during the initial 30 days of treatment, and were more likely to receive more than 10 PRBC units (20.6% vs. 8%; p=0.021) during that timeframe than non-intensively treated patients. Intensively treated patients also incurred a significantly increased plt transfusion burden at 30 days (median 7 units vs. 4 units; p<0.001) and at 60 days (median 9 units vs. 4 units; p<0.001). Patients in the intensive subset were significantly more likely to require over 10 platelet transfusions compared with the non-intensive subset (34.6% vs. 16.1%; p<0.001). In multivariate analysis the platelet transfusion burden at 30 days (Odds ratio [OR]=1.12 [95% confidence interval {CI}, 1.05-1.19; p<0.001) and 60 days (OR=1.05 [95% CI], 1.01-1.10; p=0.013) were both significantly associated with treatment intensity. Early death rates at 30 days following induction were low in both groups and not significantly impacted by treatment intensity (1.9% in intensive vs. 3% in non-intensive; p=0.67) nor at 60 days (2.9% in intensive vs. 8.2% in non-intensive; p=0.12).Conclusion In this single-center, real-world cohort, AML patients undergoing intensive induction had significantly increased bleeding events and transfusion requirements versus those receiving non-intensive induction. Since it is increasingly recognized that certain AML patient subsets have comparable rates of remission with intensive and non-intensive regimens, prospective trials comparing these approaches should also compare bleeding events and transfusion burden.
