Socioeconomic determinants of childhood cancer outcomes in a large contemporary cohort

ABSTRACT Despites improvements in childhood cancer survival in the last several decades, marked racial, ethnic, and socioeconomic disparities in outcomes persist. Compared with non-Hispanic white children, non-Hispanic black and Hispanic children experience lower survival from many cancers, including leukemia, the most commonly diagnosed cancer in children. The underlying causes of these survival differences are poorly understood and may vary by cancer type, and both biological and socioeconomic pathways have been proposed. Recent evidence has suggested that lower socioeconomic status (SES) is associated with survival from some childhood cancers. The Children's Oncology Group (COG) is an international clinical trial cooperative group of over 200 hospitals which together treat more than 90% of all children and adolescents diagnosed with childhood cancer in the United States and Canada. In 2007 the COG opened the Childhood Cancer Registration Network (CCRN; COG protocol ACCRN07) to create a research registry. A total of over 56,000 childhood cancer cases were enrolled on ACCRN07 through the end of enrollment on December 8, 2017. All children and parents enrolled on ACCRN07 provided address information which was current at the time of diagnosis. We will work with investigators at the Minnesota Population Center to geocode all ACCRN07 patients with a valid U.S. address, contextualize with socioeconomic status data, and return small-area SES data to COG for dissemination. We will contextualize each geocoded address with Census data at the block level using seven variables from these data we will use factor analysis to derive a five-level SES indicator. We will then examine the influence of SES on risk of minimum residual disease at the end of induction therapy, relapse, other serious toxicities and adverse events, and survival in >9,500 acute lymphoblastic leukemia (ALL) patients. Over 9,500 ALL patients on ACCRN07 with a valid address will also have been treated on COG protocols. Ours will be the first study to evaluate SES as a predictor of childhood cancer outcomes on a large scale within the Children's Oncology Group, and will include detailed cytogenetic and molecular characterization of each tumor. Additionally, to our knowledge, this will be the first analysis of SES predictors of short-term treatment toxicities. We will create a highly useful resource on a large scale for a contemporary cohort of childhood cancer patients. Our findings will have translational potential in that outcomes related to SES may indicate the need to develop tailored interventions for low-resource patient populations. Additionally, this cohort's utility will extend beyond outcomes of therapy and into survivorship with linkages to the National Death Index (NDI) to obtain mortality data. Our long-term goal is to understand the factors that contribute to disparities in childhood cancer relapse, survival, and the burden of morbidity in survivors. Thus, this effort will inform targeted follow-up recommendations and risk-reducing interventions.