Racial and Ethnic Disparities in Pediatric Acute Myeloid Leukemia Outcome Pubblico

Abstract

In the U.S., Black and Hispanic children with cancer are less likely to survive than White children. In acute myeloid leukemia (AML), Black race and Hispanic ethnicity have, likewise, been associated with a poor prognosis. It remains unclear, however, whether the association between race/ethnicity and outcome, could be explained by differences in other relevant variables, such as socioeconomic status (SES), age, obesity, and disease characteristics, or whether the association is simply due to worse outcome following stem cell transplant (SCT), often included in AML treatment, and known to be inferior for Black and Hispanic patients. In order to examine these and other potential explanatory variables, we did a secondary analysis of the data collected during the Children's Oncology Group phase 3 clinical trial AAML0531. The primary outcome was event-free survival (EFS), censored at the time of per-protocol SCT. We hypothesized an association between race/ethnicity and response to chemotherapy and survival; and moreover, that it would be explained by racial/ethnic differences in patient characteristics such as weight and SES and disease characteristics such as the presence of risk-stratifying cytogenetic/molecular abnormalities. A total of 1022 patients were included in the analysis, but only 914 could be classified into one of 4 meaningful racial/ethnic groups (White, Black, Asian, Hispanic-White). Kaplan-Meier survival analysis showed significantly worse 5-year EFS for Black patients compared to White (39.07% ± 4.89 vs 50.84% ± 2.23; p=0.035); however, there was no difference in remission or relapse rate between groups. In the multivariate analysis, controlling for age, cytogenetic/molecular disease characteristics, weight, and SES did not affect the association between Black race and poor EFS (HR=1.49, 95% CI: 1.10-2.00; p=0.009). Surprisingly, Black patients had significantly worse treatment related mortality (TRM) compared to White (15.21% ± 3.54 vs 5.66% ± 1.01; p=0.0006), although no consistent cause of death was identified. Therefore, we conclude that the poor outcome for Black children with AML treated on this regimen is due to excess TRM and not due to baseline differences in disease characteristics, response to therapy, or complications from SCT. Future research should be aimed at determining the underlying reason for this observation.

Table of Contents

Introduction 1-3

Background 4-8

Methods 9-17

Results 18-22

Discussion 23-26

Conclusion 27

References 28-29

Tables and Figures 30-51

• Table 1. Race and ethnicity of patients enrolled on study AAML0531 30
• Table 2a. Patient characteristics by race/ethnicity 31
• Table 2b. Disease characteristics by race/ethnicity 32
• Table 2c. Treatment characteristics by race/ethnicity 33
• Tables 3a and b. MRD positivity following Induction 1, comparing patients of different races/ethnicities 34
  • 3a. Odds ratio of being MRD positive at the end of Induction 1, compared to patients of White race 34
  • 3b. Proportion of patients who were MRD positive at the end of Induction 1, by race/ethnicity 34
• Table 4. Outcomes by race/ethnicity following 2 cycles of induction chemotherapy 35
• Tables 5a and b.Occurrence of early death, comparing patients of different races/ethnicities 36
  • 5a. Odds ratio of early death, compared to patients of White race 36
  • 5b. Proportion of patients who had early death, by race/ethnicity 36
• Figure 1. Event-free survival, censored at the time of per-protocol transplant in CR-1 37
• Figure 2. Event-free survival, censored at the time of per-protocol transplant in CR-1, comparing racial/ethnic groups 38
• Figure 3. Overall survival, censored at the time of per-protocol transplant in CR-1. 39
• Figure 4. Overall survival, censored at the time of per-protocol transplant in CR-1, comparing racial/ethnic groups. 40
• Table 6. EFS outcomes by race/ethnicity, excluding patients who had per-protocol transplant in CR-1. 41
• Table 7. EFS outcomes by race/ethnicity, NOT excluding patients who had per-protocol transplant in CR-1. 42
• Figure 5. Cumulative incidence of relapse, censored at the time of per-protocol transplant in CR-1. 43
• Figure 6.Treatment-related mortality, censored at the time of per-protocol transplant in CR-1 44
• Table 8. Cause of death when death was the only event (descriptive). 45
• Table 9. Univariate predictors of EFS parameters (induction failure, relapse, death) 46
• Tables 10a and b.Cox Proportional Hazards Model of the effect of the interaction between race/ethnicity and t(8;21) on EFS. 47
  • 10a. Hazard ratios for EFS parameters (induction failure, relapse, death), comparing patients with t(8;21) to those without, controlling for race 47
  • 10b. Hazard ratios for EFS parameters (induction failure, relapse, death), comparing patients of different racial/ethnic groups to White patients, controlling for the presence of t(8;21) 47
• Tables 11a and b.Cox Proportional Hazards Model of the effect of the interaction between race/ethnicity and Inversion 16 on EFS. 48
  • 11a. Hazard ratios for EFS parameters (induction failure, relapse, death), comparing patients with Inversion 16 to those without, controlling for race 48
  • 11b. Hazard ratios for EFS parameters (induction failure, relapse, death), comparing patients of different racial/ethnic groups to White patients, controlling for the presence of Inversion 16 48
• Tables 12a and b.Cox Proportional Hazards Model of the effect of the interaction between race/ethnicity and Risk Group on EFS. 49
  • 12a. Hazard ratios for EFS parameters (induction failure, relapse, death), comparing High Risk to Low Risk patients, controlling for race/ethnicity 49
  • 12b. Hazard ratios for EFS parameters (induction failure, relapse, death), comparing patients of different racial/ethnic groups, controlling for Risk Group. 49
• Tables 13a and b.Cox Proportional Hazards Model of the effect of the interaction between race/ethnicity and payment type on EFS. 50
  • 13a. Hazard ratios for EFS parameters (induction failure, relapse, death), comparing patients with Medicaid/no means of payment to those with Private Insurance/Self Pay, controlling for race/ethnicity 50
  • 13b. Hazard ratios for EFS parameters (induction failure, relapse, death), comparing patients of different racial/ethnic groups to White patients, controlling for the payment type 50
• Table 14. Multivariate Cox Proportional Hazards Model: Race/Ethnicity as a predictor of EFS, controlling for age, weight, payment type, and risk classification 51

About this Master's Thesis

Rights statement

• Permission granted by the author to include this thesis or dissertation in this repository. All rights reserved by the author. Please contact the author for information regarding the reproduction and use of this thesis or dissertation.

School	Laney Graduate School
Department	Clinical Research Curriculum Program
Degree	MS
Submission	Master's Thesis
Language	English
Research Field	Health Sciences, Oncology
Parola chiave	Outcome Ethnicity AML Pediatric Acute Myeloid Leukemia Race
Committee Chair / Thesis Advisor	Woods, William G, Emory University
Committee Members	Qayed, Muna , Emory University Klein, Mitchel, Emory University

Ultima modifica

Primary PDF

Thumbnail	Title	Date Uploaded	Actions
	Racial and Ethnic Disparities in Pediatric Acute Myeloid Leukemia Outcome ()	2018-08-28 15:10:52 -0400	Press to Select an action Download

Supplemental Files

Thumbnail	Title	Date Uploaded	Actions