Disaggregated outcomes of Asian Americans, Native Hawaiian and Pacific Islander populations with Acute Myeloid Leukemia

Abstract Introduction: Asian American, Native Hawaiian, and Pacific Islander (AANHPI) individuals represent one of the most diverse and rapidly growing racial group in the United States. Yet they are frequently aggregated in cancer research making it difficult to investigate critical differences in their health outcomes. In acute myeloid leukemia (AML), a heterogenous and high-risk hematologic malignancy, survival disparities by disaggregated AANHPI ethnicity are not fully understood. Existing studies have primarily focused on White, Black, and Hispanic comparisons, often overlooking within-group heterogeneity among AANHPI patients. To address this gap, we examined overall survival (OS) among AANHPI patients with AML in the National Cancer Database (NCDB), with disaggregation by ethnic subgroup to evaluate whether survival differs meaningfully across populations traditionally combined in aggregate. Methods: We used the NCDB to identify all adult patients (age ≥18) diagnosed with AML from 2004 to 2021. The NCDB was utilized because of its unique capacity to disaggregate Asian American, Native Hawaiian, and Pacific Islander (AANHPI) populations, giving detailed analysis across specific ethnic subgroups. Patients were categorized into disaggregated Asian American groups and, separately, into Native Hawaiian and other Pacific Islander groups. Non-Hispanic White patients served as the reference population. Cox proportional hazards regression was used to evaluate OS, first in univariable models with race/ethnicity as the main predictor, and then in multivariable models adjusting for age, sex, year of diagnosis, receipt of chemotherapy, area-level socioeconomic status, educational attainment, insurance status, and comorbidity index. Results: Among 145,249 adults with AML, there are 4,357 (3%) individuals from the AAHNPI subgroups who showed significantly different OS compared to non-Hispanic White patients. Chinese (HR 1.24, 95% CI 1.12–1.38, P<0.001), Filipino (HR 1.16, 95% CI 1.01–1.32, P=0.030), Native Hawaiian (HR 1.28, 95% CI 1.01–1.62, P=0.045), Korean (HR 1.34, 95% CI 1.14–1.58, P<0.001), and South Asian (HR 1.24, 95% CI 1.12–1.37, P<0.001) patients experienced significantly worse survival. By comparison, other ethnic subgroups including Japanese, Vietnamese, Laotian, Hmong, Cambodian, Thai, and Pacific Islander (non-Hawaiian) patients, did not demonstrate statistically significant differences in survival compared to the White population. In multivariable analysis adjusting for age, sex, year of diagnosis, chemotherapy receipt, socioeconomic status, education, insurance status, and comorbidity index, several AANHPI ethnic subgroups continued to show significantly poorer OS compared to White patients. Korean (HR 1.39, 95% CI 1.17–1.66, P<0.001), Native Hawaiian (HR 1.36, 95% CI 1.04–1.77, P=0.023), Chinese (HR 1.17, 95% CI 1.05–1.31, P=0.006), and South Asian (HR 1.15, 95% CI 1.03–1.28, P=0.010) patients experienced significantly worse survival. Other groups, including Japanese, Filipino, Vietnamese, Laotian, Hmong, Cambodian, Thai, and non-Hawaiian Pacific Islanders, demonstrated no statistically significant difference in OS compared White patients. Conclusions: These findings highlight the importance of disaggregating AANHPI populations in interpreting outcomes of hematologic cancer . Aggregated analyses may obscure meaningful differences in survival, as demonstrated by the significantly worse outcomes among specific ethnic patient subgroups (Korean, Native Hawaiian, Chinese, and South Asian) with AML. The finding that these disparities persisted despite adjustment for socioeconomic and educational factors, suggesting that other barriers—including language differences, variable health literacy, cultural barriers, biological factors, clinical trial underrepresentation and financial hardship related to intensive therapies such as hematopoietic stem cell transplantation—may contribute to these outcomes and require further investigation. Expanded inclusion of disaggregated AANHPI data in national cancer registries and targeted efforts to improve equity in access and care delivery for high-risk ethnic subgroups are warranted.