Increased Mortality In High-Risk Black Children with Acute Lymphoblastic Leukemia at the University of Alabama at Birmingham.

Approximately 2500 children per year in the United States are diagnosed with Acute Lymphoblastic Leukemia (ALL). Although the incidence of ALL in black children is lower than in other ethnic groups, black children diagnosed with ALL have historically been more likely to have high-risk disease and have had an overall survival of 64% v. 78% for white patients. The disparity in survival persists when controlling for high-risk disease. Published studies have only evaluated those patients registered on clinical trials, and very little is known about the impact of socioeconomic parameters on this disparity.

All children diagnosed and treated for ALL at the University of Alabama at Birmingham for the time period 1999–2009 were included in the study (n= 237). Disease characteristics and demographic variables were collected on all patients and related to patient outcomes. Kaplan-Meier curves were constructed for high v. low risk, and black v. white patients and hazard ratios were used to determine the effects of various parameters on survival.

75.4% of the sample was white, 18.6% black and 5.9% Hispanic. There were no differences in the age or proportion presenting with high versus standard risk disease (43%). Black children were significantly less likely to have private insurance (22% versus 69%, p <0.001) and less likely to have a primary pediatrician at the time of diagnosis (p = 0.015). Black patients lived within zip codes with less of the total population in the work force (59.4% v. 62.4%, p = 0.04), lower median household income ($33131 versus $39951, p = 0.009) and were also more likely to live in an apartment (15.9% v. 2.6%, p=0.0006). Black patients were significantly more likely to relapse (29.6% of blacks versus 13.6% of whites, p = 0.02). Overall survival was significantly decreased for black patients with a hazard ratio of 2.2 (p=0.04). Controlling for disease characteristics at diagnosis (white blood cell count < or > 50,000; high or low risk leukemia blast genetics; central nervous system disease; age < 1 or >10 years of age) adjusted the hazard ratio to 1.85. The increased risk in black patients was primarily due to those patients classified as high-risk (white blood cell count >50,000, high risk genetics, or age >10 on presentation), with a 5 year survival of 77% in high-risk whites and Hispanics and only 45% in high-risk blacks (log-rank test for equality of Kaplan-Meier curves p < 0.0001). Controlling for known socioeconomic factors (public versus private insurance; zip code level demographic information and pediatrician at diagnosis) in addition to disease factors increased the hazard ratio to 6.9 (p = 0.06).

The reasons for the disparities in outcomes for childhood ALL are not well understood. Black race in this study was associated with increased mortality and the risk was primarily in those diagnosed with high-risk disease. In contrast to other studies showing a larger proportion high-risk disease in black patients, 43% of both black and whites in this study were classified as having high-risk disease. This survival disparity is primarily due to the increased relapse rate observed in high-risk black patients. Controlling for disease characteristics, the observed increase in the hazard ratio for black patients persists. Although black patients have demographics consistent with lower socioeconomic status, controlling for socioeconomic parameters did not change the observed increased hazard ratio. This study lends support to the hypothesis that genetic and/or host factors contribute to the worse outcomes observed in black patients. Further studies with more patient-specific socioeconomic information could further clarify this relationship.

No relevant conflicts of interest to declare.