Association of Race, Ethnicity, and Social Deprivation on the Incidence and Severity of Adverse Events in Induction in Children with Acute Lymphoblastic Leukemia

Background: Anti-cancer therapies used to treat pediatric acute lymphoblastic leukemia (ALL) lead to high cure rates but can cause significant adverse events (AEs). Prior studies show that underserved children, especially those from racial and ethnic minority groups, have increased acuity of illness at presentation, higher rates of relapse, and decreased overall survival (OS). These disparities are multifactorial, but the association between race, ethnicity, and social deprivation with the development of AEs has not been comprehensively evaluated. This study aims to compare the incidence and severity of clinically-significant AEs during induction therapy for de novo pediatric ALL by race and ethnicity, social deprivation level, and the combined effect of these factors. Methods: This single institution retrospective cohort study included children aged 0-21 years diagnosed and treated for ALL at Children's Healthcare of Atlanta (CHOA) between January 1, 2010 and September 1, 2022. Demographic data were extracted from the electronic health record (EHR) and CHOA Cancer Registry. Incidence and severity, defined as the highest grade of each AE per the National Cancer Institute Common Terminology Criteria for Adverse Events (CTCAE), were manually abstracted from the EHR following a detailed chart abstraction guide. The following clinically-significant AEs were included: acute respiratory distress syndrome (ARDS), constipation, hyperglycemia, ileus, infection/sepsis, neuropathy, pancreatitis, seizure, stroke, thromboembolic events, and typhlitis. Addresses at diagnosis from eligible patients were used to map the Social Deprivation Index (SDI) for each year during 2010-2022 to the zip code tabulation areas of patient residence. The SDI was categorized into quintiles with quintile 1 representing the areas of lowest social deprivation and quintile 5 representing the areas of highest social deprivation. For each AE, the association of (a) race and ethnicity and (b) SDI quintiles with the incidence and severity was analyzed using multivariable logistic regression, adjusting for age at diagnosis and sex. Results: Of the 814 patients in the cohort, 447 (54.8%) patients were male, median age at diagnosis was 7 years (IQR 3,11 years),181 (22.2%) were non-Hispanic Black, and 386 (47.4%) were non-Hispanic White. Considering all CTCAE grades, constipation (n=462, 56.8%) and hyperglycemia (n=800, 98.3%) were the most common AEs; ARDS, Ileus, pancreatitis, seizure, stroke, thromboembolic were rare (n=6-42, 1-5%). Non-Hispanic Black patients had significantly lower odds of ileus (0.13, 95% CI 0.02, 0.99), infection (0.63, 95% CI 0.43, 0.93), neuropathy (0.45, 95% CI 0.30, 0.70), and sepsis (0.49, 95% CI 0.25, 0.94) compared to non-Hispanic white patients. Non-Hispanic Black patients had a 2.16 times higher odds (95% CI 1.19, 3.92) of grade 3+ hyperglycemia compared to non-Hispanic white patients. Living in the highest quintile (SDI 5) conferred an independent risk (OR 2.61 (95% CI 1.09,6.68)) of grade 3+ hyperglycemia compared to those in the first quintile for SDI, after controlling for race, ethnicity, age, and sex. There was no statistically significant interaction between SDI quintile, race, and ethnicity. Conclusions: In this study, non-Hispanic Black race and living in areas with the highest social deprivation were independent risks for severe hyperglycemia during induction. This finding highlights groups targetable for early hyperglycemic prevention and intervention. In this analysis limited to induction therapy for ALL, there was no evidence of interaction between race, ethnicity, and SDI on the incidence and severity of AEs. This suggests these factors may be independent in relation to development of AEs and that higher rates of relapse and decreased OS previously described in racial and ethnic minority groups may not be due to AEs experienced during ALL induction. Abstraction of ten additional clinically-significant AEs is ongoing. Future directions include analyses of all ALL chemotherapy courses, to help further elucidate any association between race, ethnicity, and social deprivation.