Acuity of Illness at Initial Presentation Mediates Early Mortality in Adolescent and Young Adult Leukemia

Background: Age-related disparities in outcomes for pediatric and adolescent leukemia are multifactorial. While the prevalence of pre-existing medical morbidities is low in children and adolescents, cancer-associated organ dysfunction leading to increased acuity of illness at initial presentation has been noted in up to 24% of those with pediatric leukemia. We have noted an association between adolescent and young adult (AYA) age group and acuity of illness at initial presentation (Jain et al, Blood, 2023). The current study aims to examine (1) the association between acuity of illness and overall survival, and (2) whether acuity of illness mediates age-related differences in overall survival. Methods: We conducted a retrospective analysis of 688 patients aged 1-21 years who presented to Children's Healthcare of Atlanta between 2010-2018 with any newly diagnosed leukemia, including B-cell acute lymphoblastic leukemia (ALL), T-cell ALL, acute myeloid leukemia, and other leukemias (chronic myeloid leukemia, juvenile myelomonocytic leukemia, Burkitt's leukemia). We excluded patients who lacked information in the first 72 hours following initial presentation to the hospital due to incomplete records or initial treatment at another facility. Based on manual chart review, high acuity of illness (yes/no) was defined as any ICU admission or ICU-level resource utilization by organ system (i.e., cardiovascular, respiratory, renal, hematologic, and neurologic), regardless of physical location in the hospital. We assessed overall survival with Kaplan-Meier analysis, with censoring defined by date at last follow-up, 5 years post-diagnosis, or December 31, 2020, whichever came first; the analysis was stratified by acuity of illness (high versus low) at initial presentation. Associations between age at presentation and survival were examined using the Cox proportional hazards model, controlling for sex assigned at birth, race/ethnicity, insurance status at presentation, and type of leukemia. Additionally, mediation analysis was conducted to test the extent to which high acuity of illness explained age-related differences in overall survival (Yoshida and Li, 2022). Results: In this cohort of 688 patients with leukemia (median age 6 years, interquartile range 3-12 years), high acuity of illness at initial presentation was seen in 24.7% of patients and was more likely in older patients (adjusted odds ratio for 10-21 years versus 1-9 years: 1.94, 95% confidence interval [CI]): 1.32-2.85). High (versus low) acuity of illness was significantly associated with an increased risk of death at 6 months (adjusted hazard ratio [aHR]: 3.70, 95% CI: 1.72-7.99) and 1 year (HR: 2.10, 95% CI: 1.14-3.92) post-diagnosis. Older age group was associated with an increased risk of death at 6 months (aHR for 10-21 years versus 1-9 years: 3.08, 95% CI 1.27-7.87), 1 year, (aHR: 2.43, 95% CI 1.25-4.74), 2 years (aHR: 2.46, 95% CI 1.44-4.18), and 5 years (aHR: 2.47, 95% CI 1.53-3.99) post-diagnosis. At 6 months post-diagnosis, high acuity of illness explained a significant proportion of the association between older age group and overall survival (23%, 95% CI 2-44%), although this mediation effect was statistically nonsignificant at 1 year (15%, 95% CI -4-35%), 2 years (3%, 95% CI -11-16%), and 5 years (5%, -8-17%). Conclusion: Among pediatric and AYA patients with leukemia, older age was associated with early mortality, and high acuity of illness at initial presentation mediated part of age-related differences in mortality. Early mortality differences by age group may be due to increased morbidity and toxic death in remission from ICU-level care at diagnosis. While the causality between age and acuity of illness could be multifactorial based on differences ranging from biology to access to care, this research provides new insights towards informing strategies to narrow disparities in leukemia survival.