Abstract
Background: Acute Myeloid Leukemia (AML) is an aggressive malignancy accounting for 1% of adult cancers and 2% of cancer deaths in the United States. Disease incidence is greatest in elderly patients, with a median age of diagnosis of 69. AML is largely characterized by recurrent driver mutations in myeloid cells. A prospective, randomized-controlled trial by Papaemmanuil et al in 2016 examined 1,540 patients with AML, identifying 5,234 driver mutations in 76 genomic regions of AML cells, with the vast majority (73%) being point mutations. According to the data from this study, the 5 most frequent genes mutated in AML within the examined cohorts were, in order of frequency: FLT3, NPM1, DNMT3A, NRAS, TET2. They also showed that overall survival correlated with number of mutations. Importantly, the demographic data in this cohort was limited to age and gender and did not include racial and ethnic background, although patient samples were obtained from clinical trials of the German-Austrian AML Study Group. A larger study by Hogg et al in 2023 included a genetic analysis of 3,064 patients with AML. Similar to the German-Austrian cohort, this American study did not specify patient race or ethnicity. Race and ethnicity are, however, now understood to play integral roles in outcomes of AML. A 2024 study of 162 adult AML patients by Dong et al demonstrated clear racial disparities in AML, including higher rates of intermediate and high-risk disease but also improved overall survival in Hispanic patients. Additionally, a 2025 systematic analysis by Loeb et al demonstrated racial disparities in clinical trials for AML, with enrollment of 80.8% white participants, only 4.7% black participants, and only 3.4% Hispanic participants. To further explore the significance of demographic factors on gene mutations in AML, we evaluated the incidence of specific mutations in patients diagnosed with myeloid malignancy and examine their association with trends in demographic data including age, gender, race, and ethnicity at a diverse, urban, safety net hospital
Methods: Using Slicer Dicer and chart review in Epic Charting Software, retrospective data was collected from patients diagnosed with acute myeloid leukemia from January 2020 to June 2025 at the University of Illinois Cancer Center (Chicago, IL). Data was collected on race, ethnicity, age, gender, zip code, and neogenomic test results. The data was analyzed using Microsoft Excel and R statistical software.
Results: A total of 119 patients met inclusion criteria for the study. The cohort consisted of 53% White, 34% Black, 9% Asian and Pacific Islander, and 4% Other races. Patients were 27% Hispanic and 73% non-Hispanic (NH). The combined racial and ethnic demographic of the cohort was: 33% NH Black, 32% NH White, 27% Hispanic, 7% NH Asian, and 1% NH Other. Patient sex was 45% female and 55% male. Average age at diagnosis was 57. Average age at diagnosis by race and ethnicity were: NH Black (60, n=40), NH White (58, n=38), Hispanic (53, n=32), NH Asian (50, n=8), and NH Other (66, n=1) although these differences were not statistically significant. The five most common mutations overall were TET2 (n= 29), FLT3 (n=28), DNMT3A (n=23), ASXL1 (n=21), and IDH2 (n=20). In Hispanic patients, they were FLT3 (n=8), DNMT3A (n=6), IDH2 (n=6), TET2 (n=6), and NPM1 (n=5). In NH White patients, they were TET2 (n=14), ASXL1 (n=10), DNMT3A (n=10), NPM1 (n=9), and FLT3 (n=7). In NH Black patients they were ASXL1 (n=8), IDH2 (n=8), TP53 (n=8), TET2 (n=8), DNMT3A (n=7).
Discussion: Our cohort represents a diverse AML patient population, reflected in racial and ethnic demographics. Notable differences in demographics were also observed compared to prior seminal studies on AML genetics. In particular, the average age of diagnosis at 57 was notably lower than previously published data. This younger age was relatively maintained across all patient subgroups, pointing toward AML in younger populations at our urban hospital. The most common gene mutations also appear to be conserved between racial and ethnic backgrounds, implicating drivers involved in epigenetics, proliferation, and differentiation. Notably, NH Black patients were more likely to have high-risk TP53 mutations in our cohort. This work demonstrates racial and ethnic heterogeneity in the genetic landscape of AML. Further work is needed to explore how race and ethnicity impact acquisition of AML driver mutations and affect overall outcomes in AML.
