Although the number of available therapies for acute myeloid leukemia (AML) has increased dramatically in the past five years, significant disparities in outcomes among different racial and ethnic groups persist. Studies have demonstrated that Black patients with AML have worse survival and may be less likely to receive intensive chemotherapy and undergo allogeneic stem cell transplantation than white patients.1–3 However, while differences in access to treatment certainly may be contributing to inequalities in outcomes, access to care alone does not appear to be sufficient to explain the disparities in survival that have been observed. A recent analysis of 1,339 patients treated uniformly with intensive induction chemotherapy on Alliance clinical trials found that Black patients had worse survival than white patients (29% vs. 42%, respectively, were alive at 3 years; p=0.02).2
Structural racism is the concept that economic, political, and social systems, including laws and governmental regulations, have caused and maintained racial inequities.4,5 To assess the role of structural racism in the disparities in survival across racial and ethnic groups in patients with AML, Dr. Ivy Abraham and colleagues recently conducted a retrospective analysis of data collected from six institutions in the Chicago area. In addition to clinical characteristics and outcomes, data collected included demographic information (race and ethnicity were self-reported) and health care access as indicated by insurance status and enrollment onto clinical trials. Geocoding software was used to map each patient to a residential census tract, and data from the American Community Survey for years 2006 to 2010 were then used to measure the following proxies for structural racism: census tract disadvantage (including factors such as the proportion of families with incomes below the poverty line and the proportion of unemployed adults), tract affluence (defined by measures such as the proportion of adults with at least a college education and the proportion of adults employed in professional or managerial occupations), and segregation (tract proportion non-Hispanic Black [NHB] and tract proportion non-Hispanic white [NHW]).
This analysis included 822 patients with AML, 40 percent of whom were part of a minority population (n=126 NHB, n=117 Hispanic, and n=82 other racial and ethnic groups). A discrete survival analysis for hazard of leukemia death after adjusting for age, gender, and study site showed that NHB patients had worse survival compared to NHW patients (HR, 1.59; 95% CI, 1.15-2.22), and there was also a trend toward worse survival in Hispanic patients compared to NHW patients (HR, 1.25; 95% CI, 0.88-1.79). There were significant differences in census tract measures between groups, with NHB and Hispanic patients more likely to live in more disadvantaged and less affluent census tracts. A mediation analysis was then conducted to understand the relative impact of six variables (structural racism, tumor biology, health care access, comorbidities, treatment, and intensive care unit admissions) on outcomes. This analysis showed that among these six variables, structural racism was the biggest contributor to the disparities in leukemia deaths observed between NHB and NHW patients and between Hispanic and NHW patients.
Although several studies have reported on disparities in AML outcomes among racial and ethnic groups, this study is unique in that the authors attempted to understand why these disparities exist. The use of census tract measures as a proxy for structural racism is a novel approach to understanding AML outcomes, but it should be noted that it is unknown whether the scores the authors created to measure census tract disadvantage, tract affluence, and segregation are indeed accurate measures of the impact of structural racism. Additionally, this analysis was limited to patients treated at academic institutions in a single urban center; therefore, patients who were not referred for treatment or who were unable to travel to these centers for treatment were not included in the study. The median duration of follow up for patients included in this analysis was short (only 1.29 years from date of diagnosis), and it will be important to also assess which variables contribute to disparities in long-term outcomes.
In summary, this study provides important evidence that structural racism contributes to disparities in AML outcomes. Further research will be needed to validate measures of structural racism and to test interventions to reduce disparities in AML outcomes.
Dr. McMahon indicated no relevant conflicts of interest.