AUTHORS
Laura Rolke, PhD, MPH, Danielle Boselli, MS2, Jing Ai, MD, PhD3, Brittany Knick Ragon, MD4, Aleksander Chojecki, MD5, Nilay Shah, MD6, Srinivasa Reddy Sanikommu, MD7, Kelly Brook Weaver, DNP, APRN, ACNS-BC, NP-C, OCN, AOCNP8, Lawrence J Druhan, PhD9, Michael R. Grunwald, MD10, Thomas G Knight, MD11
AFFILIATIONS
1 Atrium Health Levine Cancer Institute, Wake Forest University School of Medicine, Charlotte, North Carolina, United States
2 Atrium Health Levine Cancer Institute, Wake Forest University School of Medicine, Charlotte, North Carolina, United States
3 Atrium Health Levine Cancer Institute, Wake Forest University School of Medicine, Charlotte, North Carolina, United States
4 Atrium Health Levine Cancer Institute, Wake Forest University School of Medicine, Charlotte, North Carolina, United States
5 Atrium Health Levine Cancer Institute, Wake Forest University School of Medicine, Charlotte, North Carolina, United States
6 Atrium Health Levine Cancer Institute, Wake Forest University School of Medicine, Charlotte, North Carolina, United States
7 Atrium Health Levine Cancer Institute, Wake Forest University School of Medicine, Charlotte, North Carolina, United States
8 Atrium Health Levine Cancer Institute, Wake Forest University School of Medicine, Charlotte, North Carolina, United States
9 Atrium Health Levine Cancer Institute, Wake Forest University School of Medicine, Charlotte, North Carolina, United States
10 Atrium Health Levine Cancer Institute, Wake Forest University School of Medicine, Charlotte, North Carolina, United States
11 Atrium Health Levine Cancer Institute, Wake Forest University School of Medicine, Charlotte, North Carolina, United States
ABSTRACT
Introduction: Pediatric-inspired treatment regimens have significantly improved outcomes among adolescents and young adults (AYAs) with acute lymphoblastic leukemia (ALL), but survival disparities persist. This study evaluated whether county-level social vulnerability was associated with overall survival (OS) and cancer-specific survival (CSS) among AYAs with ALL in a population-based national cancer registry.
Methods: A retrospective cohort was identified using the NCI SEER 17 Research Plus Incidence Database, which includes state and county FIPS codes and represents 26.5% of the U.S. population across 17 regional cancer registries. Eligible patients were aged 15-39 when diagnosed with ALL (AYA ICD-O-3 definition) between 2003 and 2017. Patients were excluded if they were diagnosed at death or autopsy, had no survival time, or had incomplete survival data. We constructed Kaplan–Meier survival and Cox proportional hazards models in R Studio to examine 5-year OS and CSS by social vulnerability index (SVI) quintile. Patients were linked by their county of residence to the 2014 SVI national rankings and placed into corresponding 0.2 categories (Q1 = least vulnerable to Q5 = most vulnerable). Covariates in the models were age group, sex, race/ethnicity, time period of diagnosis, and U.S. Census region.
Results: A total of 4,279 AYA patients were included in the analysis. The cohort was primarily male (n=2,757, 64.3%), Hispanic (n= 1,976, 46.2%) or non-Hispanic White (n=1,610, 37.6%), and living in the West census region (n=2,912, 68.0%). Individuals were unevenly distributed across SVI quintiles as follows: Q1 (n=218, 5.1%), Q2 (n=514, 12.0%), Q3 (n=1,293, 30.2%), Q4 (n=1,297, 30.3%), and Q5 (n=957, 22.4%).
Five-year OS decreased progressively with increasing vulnerability: Q1 – 73.3%, Q2 – 65.8%, Q3 – 59.8%, Q4 – 56.0%, and Q5 – 55.0%. A similar pattern was observed for 5-year CSS, falling from 77.0% survival in Q1 to 59.0% in Q5.
In multivariable Cox models, higher SVI was independently associated with increased risk of cancer-specific mortality at 5 years, with significantly elevated risk observed in Q3 (HR=1.48, 95% CI: 1.07-2.04, p<0.05), Q4 (HR=1.58, 95% CI: 1.14-2.18, p<0.01), and Q5 (HR=1.66, 95% CI: 1.19-2.30, p<0.01) relative to Q1. Of the covariates, older age was strongly associated with increased mortality risk (20–29 years: HR=2.19, 95% CI: 1.92-2.49, p<0.001; 30–39 years: HR=2.56, 95% CI: 2.24-2.93, p<0.001), as were non-Hispanic Black race (HR=1.53, 95% CI: 1.24-1.87, p<0.001) and Hispanic ethnicity (HR=1.43, 95% CI: 1.26-1.62, p<0.001). Encouragingly, more recent years of diagnosis were associated with improved survival, with significant reductions in mortality for patients diagnosed in 2008–2012 (HR=0.82, 95% CI: 0.72-0.92, p<0.01) and 2013–2017 (HR=0.59, 95% CI: 0.52-0.67, p<0.01).
The adverse impact of socioeconomic vulnerability was not uniform across racial/ethnic subgroups. Residing in the most vulnerable counties (Q5) was associated with higher cancer-specific mortality compared with Q1 for both non-Hispanic White (HR 1.54, 95% CI: 1.06-2.24, p<0.01) and Hispanic (HR 2.23, 95% CI: 1.10-4.52, p<0.01) patients, with pronounced effects in the latter group.
Analysis of SVI subcomponents revealed that multiple domains contributed to increased risk of mortality. In the highest quintile (Q5 vs Q1), socioeconomic status (HR 1.41, 95% CI: 1.13-1.67, p<0.01), household composition (HR 1.31, 95% CI: 1.02-1.69, p<0.05), and housing and transportation (HR 1.43, 95% CI: 1.03-1.89, p<0.05) were all significantly associated with increased risk of cause-specific death.
Conclusions: In this population-based cohort of AYAs with ALL, higher county-level social vulnerability was associated with inferior OS and CSS after adjustment for demographic factors. These findings suggest that community-level disadvantages influence cancer outcomes beyond individual characteristics. Persistent disparities among non-Hispanic Black and Hispanic AYA patients underscore the need to understand how community vulnerability intersects with race, ethnicity, access to care, and treatment delivery. The gradient relationship between SVI and survival supports its use as a population-level risk stratification tool to identify communities at increased risk for poor AYA ALL outcomes and inform interventions to improve care and survival equity.
DISCLOSURES
Thomas G Knight, MD: No relevant conflicts of interest to declare.