Sex, Race and Geographic Disparities in Early Age at Onset Hematologic Malignancies

Introduction: Hematologic malignancies are a large heterogeneous group of cancers caused by genetic errors occurring across different stages of hematopoiesis. As the exact etiological drivers remain largely unknown, specific prevention efforts targeting these malignancies are essentially nonexistent. A recent shift towards utilization of population-level surveillance data to investigate early-onset cancer (defined as under age 50) and geospatial incidence patterns can guide interventions that benefit specific populations experiencing high burden, an approach known as precision public health. The objective of this study is to characterize geographic, racial and gender disparities in early onset hematologic malignancies (EOHM) using surveillance data to begin to inform whether environment or where a person lives could potentially play a role in EOHM, as well as future interventions.

Methods: The data for this analysis are from the US Cancer Statistics Public Use Research Database, provided by the National Program of Cancer Registries and the Surveillance Epidemiology and End Results programs. All individuals with hematologic cancer diagnosed in 2016-2020 were included, with hematologic malignancies in the following sites: Hodgkin Lymphoma (HL), Lymphocytic Leukemia, Myeloid and Monocytic Leukemia, Myeloma, Non-Hodgkin Lymphoma (NHL), Other Leukemia. Five race/ethnic groups were defined: non-Hispanic White (NHW), non-Hispanic Black (NHB), non-Hispanic American Indian and Alaskan Native, non-Hispanic Asian/Pacific Islander and Hispanic (any race). Sex is defined as male or female. As a first step, EOHM incidence rate ratios (IRR) and their confidence intervals were calculated for gender and racial/ethnic subpopulations within early onset and late onset disease. Early onset and late onset disease was also compared amongst each subpopulation to identify which groups experience higher relative risk of early onset disease compared to late onset. The second and third steps characterized the geographic distribution of EOHM by state, identifying states with higher rates of EOHM than expected compared to the rest of the US by race and gender respectively. The fourth step evaluated EOHM subtypes to see which subtypes are likely driving disparities in the highest burden states. The result is a handful of states, sub-populations, and blood cancer subtypes that should be focused on for future public health approaches to address EOHM and related disparities.

Results: Overall, males and NHB individuals had the highest incidence rates of EOHM. However, the EOHM relative risk for females was higher for early compared to late age at onset than males. All non-White racial populations had increased relative rates of EOHM compared to late age at onset disease, whereas the NHW population had a slightly higher relative rates of late onset disease. NHB individuals recorded the highest relative risk of EOHM compared to late onset. Geographic variation and clustering of states with higher EOHM rates than expected were identified. The six states with the highest burden of overlapping gender and racial disparities included New York, Florida, New Jersey, Georgia, Connecticut, and Maryland. From those, Florida and New York were targeted as particular states of interest because Florida had high IRRs across all EOHM subtypes and New York had the high IRR across five of six subtypes (HL, NHL, Myeloma, Lymphocytic Leukemia and Other Leukemia).

Conclusion: This study is unique in its public health approach to hematologic malignancy, which is more commonly studied at the molecular level. While the development of blood cancers is viewed as sporadic, the disparate patterns in race, sex, and location observed in this study seem to suggest environment could be playing a role. Thus, future studies investigating exposures and disparities in these States appear warranted, particularly in Florida and New York. In a time when specific prevention efforts targeting these malignancies are nonexistent, this analysis can be utilized to generate hypotheses about drivers of EOHM and guide prevention and screening interventions in states most impacted by both EOHM and disparities.