Background: Acute myeloid leukemia (AML) is associated with high early mortality, especially among older adults and those with limited access to care. This study utilizes a national cancer registry to identify demographic and disease-related predictors of early death in AML patients.

Methods: A retrospective cohort analysis was conducted using data from the SEER 8 database (Nov 2024 submission) covering cases diagnosed between 1975 and 2022. Patients with a diagnosis of non-M3 AML, defined by ICD-O-3 histology codes were included. Cases identified solely through autopsy or death certificate were excluded. Two primary endpoints were evaluated: early mortality within one month of diagnosis (EM1) and within two months of diagnosis (EM2). Binary logistic regression was performed to identify predictors of early mortality, and odds ratios (ORs) with 95% confidence intervals (CIs) were reported.

Results: The cohort included 37,835 patients diagnosed with non-M3 AML. The median age at diagnosis was 70 years, with 54.8% male and 83.3% identified as Caucasian. Most patients (67.3%) had AML not otherwise specified. Chemotherapy was administered to 66.7% of patients.

EM1 occurred in 12,620 (33.4%) patients. Among patients under 60 years of age, 15.2% died within the first month, compared to 32.2% of those aged 60 to 75 and 52.8% of patients over 75 (p < 0.001). Receiving chemotherapy was strongly associated with significantly lower EM1 (20.6% vs. 61.2%, p < 0.001). Patients residing in metropolitan areas had lower EM1 rates than those in non-metropolitan areas (31.6% vs. 35.7%, p < 0.001). Similarly, married individuals had a lower EM1 as compared to unmarried individuals (32.0% vs. 35.9%, p < 0.001). EM1 was also inversely associated with household income: 35.5% of patients with household incomes below $65,000, compared to 32.2% in the $65,000–90,000 group and 31.6% among those with incomes above $90,000 (p < 0.001).

In adjusted logistic regression analysis, age was a strong predictor of early mortality. Compared to patients under 60, those aged 60 to 75 had significantly higher odds of early death (OR 2.35, 95% CI, 2.17–2.56; p < 0.001), as did those over 75 (OR 3.76, 95% CI, 3.46–4.09; p < 0.001). Later years of diagnosis were modestly protective (OR 0.993; 95% CI, 0.989–0.996; p < 0.001). Compared to AML with minimal differentiation, several subtypes were associated with significantly different risks of early mortality: AML with maturation (OR 0.72; 95% CI, 0.54–0.96; p = 0.023), AML with myelodysplasia-related changes (OR 0.65; 95% CI, 0.51–0.83; p = 0.001), AML with t(8;21)(q22;q22); RUNX1-RUNX1T1 (OR 0.44; 95% CI, 0.30–0.65; p < 0.001), therapy-related AML (OR 0.39; 95% CI, 0.29–0.51; p < 0.001), and acute monoblastic and monocytic leukemia (OR 1.29; 95% CI, 1.01–1.66; p = 0.045). Receiving chemotherapy significantly reduced the odds of EM1 (OR 0.19; 95% CI, 0.18–0.20; p < 0.001), and being married was also associated with a lower risk (OR 0.89; 95% CI, 0.83–0.94; p < 0.001).

EM2 occurred in 15,014 patients. EM2 was significantly associated with older age, lack of chemotherapy, unmarried status, lower household income, non-metropolitan residence, and earlier year of diagnosis (all p < 0.001), while sex was not a significant factor. In multivariable analysis, age remained a strong predictor of EM2, with ORs of 2.62 (95% CI, 2.43–2.83; p < 0.001) for ages 60 to 75 and 4.78 (95% CI, 4.42–5.18; p < 0.001) for those over 75, compared to patients under 60. Chemotherapy was highly protective (OR 0.12; 95% CI, 0.12–0.13; p < 0.001), and being married also conferred a modest reduction in risk (OR 0.88; 95% CI, 0.83–0.93; p < 0.001). Compared to AML with minimal differentiation, the following subtypes were associated with significantly lower odds of EM2: AML with maturation (OR 0.64; 95% CI, 0.49–0.84; p = 0.001), AML with myelodysplasia-related changes (OR 0.66; 95% CI, 0.52–0.84; p = 0.001), AML with t(8;21)(q22;q22); RUNX1-RUNX1T1 (OR 0.43; 95% CI, 0.30–0.62; p < 0.001), and therapy-related AML (OR 0.42; 95% CI, 0.33–0.55; p < 0.001).

Conclusion: Early mortality in AML remains substantial, particularly among older patients and those who did not receive treatment. Age, marital status, and AML subtype are independent predictors of early death, while race, income, sex, and rural-urban residence are not significant. These findings highlight the importance of timely treatment in high-risk groups to improve early outcomes.

This content is only available as a PDF.
2025
Sign in via your Institution