No Difference in End-induction Minimal Residual Disease or Survival For Children Living in Hispanic/Latinx Enclaves Following Diagnosis of Acute Lymphoblastic Leukemia
Introduction
Hispanic/Latinx children with acute lymphoblastic leukemia (ALL), the most prevalent pediatric cancer, experience poorer survival than their non-Hispanic white peers. Research into the causes of this disparity is imperative to reduce the burden of childhood cancer in Texas. Hispanic/Latinx enclaves are neighborhoods characterized by high proportions of recent immigrants, ethnic-specific businesses, and households without an adult who speaks English fluently. Despite the socioeconomic challenges often faced by enclave residents, studies involving adults with cancer suggest that residing in an enclave may be associated with survival benefit due to stronger social ties or differences in sociocultural and lifestyle factors. In this study, we sought to determine whether residence in an enclave was associated with treatment outcomes in children with ALL, adjusting for patient-level clinical characteristics and neighborhood-level socioeconomic status (SES).
Methods
We obtained demographic, clinical, and residential data for N=1,348 patients with ALL at Texas Children’s Hospital (Houston, TX), Cook Children’s (Fort Worth, TX), and Vannie Cook Children’s Cancer and Hematology Clinic (McAllen, TX). We computed Hispanic enclave index (HEI) and Yost socioeconomic status index scores for Texas census tracts using data from the 2010 decennial census and the 2008-2016 American Community Surveys. We defined enclaves as tracts in the highest quartile of HEI scores relative to the statewide distribution, and low SES neighborhoods as those with Yost scores in the lowest quartile. We classified children as living in the following types of neighborhoods based on their address at diagnosis: low income/non-enclave (referent), low income/enclave, high income/non-enclave, and high income/enclave. We compared end-induction minimal residual disease (MRD) positivity, overall survival (OS), and relapse-free survival (RFS) among the groups using the Kaplan-Meier method and generalized linear mixed models, adjusting for sex, age at diagnosis, cytogenetics (favorable, unfavorable, not prognostic), risk group (standard, high) and ALL immunophenotype.
Results
In univariate analyses, the proportions of children who experienced MRD positivity, relapse, and death were similar across neighborhoods, and neighborhood characteristics were not associated with these outcomes in generalized linear mixed models. The median follow-up time was 6.6 years; in Kaplan-Meier analyses of subjects with ≥3 years of follow-up, neither OS nor RFS differed by neighborhood.
Conclusion
We did not identify disparities in MRD positivity, relapse, or mortality following ALL diagnosis among children living in enclaves when compared to those living in non-enclaves after controlling for individual clinical characteristics and neighborhood SES. The mechanisms postulated to impact survival in adults residing in enclaves include stronger social ties and better community support during hardships. These may not hold the same significance for the pediatric population, or community and social support may be similar for residents of enclaves and non-enclaves. Alternatively, societal and governmental interventions unique to the pediatric population, such as Children’s Medicaid, may negate these. Finally, we may have been unable to detect any late differences in OS and RFS because of our relatively short median follow-up period (6.6 years). Nonetheless, additional research is warranted to determine whether residence in an enclave is associated with the long-term burden of late effects among childhood cancer survivors, given that a majority develop chronic conditions related to their therapy.