Acute Myeloid Leukemia Clinical Presentation in Standard Practice: Analysis of 20-Year Real-World Data

Introduction: Despite the recent advances in the understanding of acute myeloid leukemia (AML) biology and the extraordinary progress in the therapeutics options for these patients, the long-term survival rates remain as low as 30-35% when considering patients older than 60 years old, and 40-45% up to the age of 55 years. Similarly, most reports on the prognostic stratification by molecular and genetic abnormalities come from large centers with selected patient populations, frequently treated in clinical trials. These populations may not be representative of the general population in terms of race, ethnicity, sex, and other important characteristics, or the actual practice that is affected by access, insurance coverage and other factors. To address this gap, we conducted this study that aimed to evaluate how patients with AML present at diagnosis and how they are evaluated in standard practice.

Methods: This study utilized data from the Syapse Learning Health Network (LHN) patient population. The Syapse Network contains patient populations from community health systems with varying breadth and depth of clinical information. The health system relationships were used to retrieve real-world data from a retrospective cohort study of patients with AML. This study utilized automated ingestion of data from multiple sources, and secondary data manually curated from patient electronic medical records by oncology data specialists. Differences between groups were compared by chi-square or Fisher's exact test for categorical variables and, for continuous variables, the non-parametric Mann-Whitney U test was used. The Kaplan-Meier method was used to estimate the probability of survival. The log-rank test was applied to assess the significance of differences among actuarial survival curves with a 95% confidence interval. A significance set for p< 0.05 was adopted.

Results: 901 patients with newly diagnosed AML were analyzed. The median age at diagnosis was 71 years-old (range, 18-90 years) and 52.8% of patients were male. Nearly 90% of the patients were identified as White (88.2%) and not Hispanic or Latino (91.8%). Most patients had primary AML (pAML, n=675, 75%) while 226 (25%) were diagnosed with secondary AML (sAML), which included therapy-related (n=43) and AML progressing from myelodysplastic syndrome (MDS) or myeloproliferative neoplasms (MPN, n=183). The great majority of AML patients died within the study period (2002-2024), 80.2%.

While comparing the characteristics of pAML patients with sAML, it was observed that individuals with pAML developed the malignancy at earlier ages (pAML median, 70; range, 18-90 vs sAML median, 72; range, 30-89; p<0.001). Additionally, there was a higher rate of living pAML (23%) patients at the end of the study period than of sAML (10.2%, p<0.001). There was no statistically significant difference between pAML and sAML in terms of sex (p=0.872), race (p=1.000), and ethnicity (p=0.909). The overall survival of pAML patients was longer than individuals with sAML (median, 26.6 vs 15.4 months, p<0.001).

We further analyzed patients with primary and secondary AML according to the Eastern Cooperative Oncology Group (ECOG) performance score. AML patients were divided as: low ECOG (ECOG score 0 or 1) and high ECOG (ECOG score 2 or above). One hundred and ninety-seven (29.1%) patients with pAML and 100 (44.2%) of sAML had the ECOG score available. Most patients with pAML (72.1%) as well as with sAML (73%) fell in the low ECOG group. pAML patients with lower ECOG score were younger (median, 70.5 vs 82.0 years old, p=0.0181). We did not detect statistically significant differences regarding sex, race, and ethnicity (p>0.05). The overall survival of pAML patients with lower ECOG score was longer than individuals with higher performance status (29.5 vs 3.5 months, p<0.001). Among the 100 patients with sAML, there was no statistically significant difference between patients with low and high ECOG score regarding sociodemographic features (p>0.05). However, sAML patients with low ECOG score tended to present better overall survival (median, 16 vs 6.2 months, p=0.07).

Conclusion: Our study provides valuable information regarding patients with AML diagnosed and managed within U.S. community health systems, and assists in identifying high-risk patients and in prompting clinicians and policymakers to implement focused interventions.

Tameishi:Syapse Holdings, Inc.: Current Employment. Brandt:Syapse Holdings, Inc.: Current Employment. Brown:Syapse Holdings, Inc.: Current Employment, Current holder of stock options in a privately-held company; Sygnomics: Current holder of stock options in a privately-held company, Other: Serve as scientific/clinical advisor with equity. Cortes:Sun Pharma: Consultancy, Research Funding; Biopath Holdings: Consultancy, Research Funding; Pfizer: Consultancy, Research Funding; Rigel: Consultancy; Nerviano: Consultancy; Takeda: Consultancy; Abbvie: Research Funding; Novartis: Consultancy, Research Funding.