Race and Ethnicity and Overall Survival in Pediatric Acute Myeloid Leukemia
Daniel J. Zheng, Long Khuong, Catherine Aftandilian, Kira Bona, Emi H. Caywood, Caitlin W. Elgarten, Brian T. Fisher, Cody-Aaron L. Gathers, Taumoha Ghosh, M. Monica Gramatges, Gary Hettinger, Meret R. Henry, Yuan-Shung V. Huang, Yimei Li, Kelly Maloney, Amir Mian, Tamara P. Miller, Arunkumar Modi, Rajen Mody, Regina M. Myers, Haley Newman, Jose Ortiz, Alix E. Seif, Caroline Smith, Jamie Stokke, Naomi Winick, Jennifer J. Wilkes, Victor Wong, Richard Aplenc, Kelly D. GetzImportance
Hispanic and non-Hispanic Black (hereafter, Black) children with acute myeloid leukemia (AML) have historically experienced worse survival outcomes compared with non-Hispanic White (hereafter, White) children. Understanding specific intermediate pathways underlying these outcome disparities is a critical step to directing interventional efforts and resources.
Objective
To compare survival outcomes by race and ethnicity in a large contemporary cohort of pediatric patients with AML and to examine presentation acuity, frontline organ toxic effects, and relapse as potential mediators of any observed outcome disparities.
Design, Setting, and Participants
This retrospective cohort study was analyzed from April 2025 to January 2026 and included 13 US children’s hospitals across 10 states participating in the Real-World Epidemiology of Acute Leukemia-AML cohort. Participants included children treated for de novo AML from January 2011 through May 2024.
Exposure
Race and ethnicity (Black, Hispanic, and White).
Main Outcomes and Measures
The main outcome was overall survival (OS), defined as the time from AML diagnosis until death from any cause.
Results
The study cohort included 773 children (median [IQR] age, 9 [2-14] years; 402 males [52.0%]), of whom 125 (16.2%) were Black, 237 (30.7%) were Hispanic, and 411 (53.2%) were White (median [IQR] follow-up, 51.2 [25.7-86.9] months), treated for de novo AML. Among the patients, Black (5-year OS, 61.1% [95% CI, 52.6%-71.0%]; adjusted hazard ratio [AHR], 1.55 [95% CI, 1.26-1.91]) and Hispanic (5-year OS, 65.4% [95% CI, 58.9%-72.5%]; AHR, 1.32 [95% CI, 1.03-1.68]) children had worse OS compared with White children (5-year OS, 70.0% [95% CI, 65.2%-75.1%]). Compared with White patients, Black patients had increased risk of higher cardiovascular (adjusted risk ratio [ARR], 2.09 [95% CI, 1.23-3.56]), respiratory (ARR, 1.36 [95% CI, 1.05-1.77]), and renal (ARR, 3.08 [95% CI, 1.02-9.32]) acuity at presentation. There were no statistically significant increases in severe frontline organ toxic effects or relapse rates by race and ethnicity. Cardiovascular acuity (AHR, 1.83 [95% CI, 1.51-2.22]) and respiratory acuity (AHR, 1.43 [95% CI, 1.15-1.77]) at presentation were independently associated with all-cause mortality. Cardiovascular acuity (proportion mediated, 14%) and respiratory acuity (proportion mediated, 8%) at presentation partially mediated the total absolute survival difference between Black and White children.
Conclusions and Relevance
In this cohort study of pediatric patients with AML, differential cardiovascular and respiratory acuity at presentation were partial mediators of outcome disparities for Black children with AML. Other intermediates along the care continuum should be identified to account for the remainder of the observed outcomes and to prioritize focus areas for interventional efforts to mitigate the survival differences.