Abstract
In contrast to western countries, outcomes for childhood acute leukemia in Sub Sahara Africa (SSA) remain poor. A preponderance of unfavorable phenotypes of acute leukemia in children in SSA, in addition to disparities in access to cancer care may contribute to these inferior outcomes. We compared clinical phenotypes, immunophenotypes, and outcomes of acute leukemia among two cohorts of children ages 0-14 years at Mulago National Referral Hospital (MNRH) Uganda and Texas Children's Hospital (TCH), USA. The overarching goal is to quantify the disparities in survival of childhood leukemia in SSA and identify targetable causes of the disparities.
We conducted a retrospective cohort (February 2019 - December 2021) of children with acute leukemia, n=219 (MNRH) and n=279 (TCH). All new cases of acute leukemia were consecutively included. Children at both hospitals were diagnosed using similar morphology and flow cytometry criteria and treated on Children's Oncology Group-based regimens. We computed odds ratios (OR) and contingent 95% Confidence Intervals (CI) and P values to establish the association of the immunophenotypes of acute leukemia with the children's population of origin (Uganda vs USA).
The proportion of children with Acute Lymphoblastic Leukemia (ALL) and Acute Myeloid Leukemia (AML) was 62% and 38%, vs. 83% vs.18% at MNRH and TCH respectively. Among ALL, the proportion of children with high-risk disease was higher at MNRH vs TCH, 67% and 48%, OR 2.20 (95% CI 1.41-3.44), respectively. There was a higher proportion of T-cell ALL 30% vs.16%, OR 2.24 (95% CI 1.35-3.75), and white blood cell count ³ 50,000/mL, OR 1.81 (1.14-2.90), respectively at MNRH compared to TCH. Overall, 93% vs 96% of children achieved complete remission; 64% vs.77% were negative for minimal residual disease at the end of induction at MNRH and TCH respectively. The 3-year overall survival was 40% and 97% at MNRH and TCH, respectively, HR=2.46 (95% CI 1.73 -3.20).
Children in SSA have unfavorable acute leukemia phenotypes. Cytogenetic and genomic analysis is needed to determine and target the biological underpinnings of high risk leukemias in this African population.
Disclosures
No relevant conflicts of interest to declare.
Author notes
Asterisk with author names denotes non-ASH members.