Lower Socioeconomic Status Associated With Worse Post-Transplant Survival in Children With Cancer

According to an analysis of data from the Center for International Blood and Marrow Transplant Research (CIBMTR), children with cancer who have Medicaid insurance or live in a neighborhood with high poverty levels have worse survival outcomes following allogeneic hematopoietic cell transplantation (alloHCT). However, the authors, led by Kira Bona, MD, MPH, from Dana-Farber Cancer Institute and Boston Children’s Cancer and Blood Disorders Center, found no significant associations between these socioeconomic factors and outcomes in children who underwent alloHCT for nonmalignant disease.

To determine the affects of various socioeconomic factors on post-alloHCT outcomes, Dr. Bona and researchers assembled two pediatric cohorts of patients who received a first alloHCT between 2006 and 2015 and who were enrolled in the CIBMTR database. The cohorts included 2,053 children with malignant disease and 1,696 children with nonmalignant disease.

Neighborhood poverty exposure was defined as living in a high-poverty ZIP code (≥20% of persons below 100% federal poverty level). Researchers then evaluated overall survival (OS), as well as relapse and transplant-related mortality (TRM) in patients with malignant disease, acute and chronic graft-versus-host disease (GVHD), and infection in the first 100 days post-alloHCT.

In the malignant cohort, 299 children (15%) lived in high-poverty neighborhoods and more than one-third (n=711; 35%) were insured by Medicaid. Proportions of neighborhood high-poverty exposure and Medicaid coverage were similar in the nonmalignant cohort (n=228 [13%] and n=597 [35%], respectively). Across both cohorts, children living in high-poverty neighborhoods were more commonly Black or Hispanic.

In the malignant and nonmalignant cohorts, post-alloHCT outcomes were:

  • 5-year OS: 52% and 75%, respectively
  • 100-day probability of infection: 82% and 76%
  • 100-day probability of grade 2-4 acute GVHD: 38% and 25%
  • 5-year probability of chronic GVHD: 32% and 25%

In the malignant cohort, the 5-year incidence of TRM was 21% and the 5-year incidence of relapse was 33%.

The most common cause of death was recurrence of neoplasm in the malignant cohort (23%) and organ failure in the nonmalignant cohort (7%).

The authors determined that children living in high-poverty neighborhoods had significantly higher prevalence of TRM compared with those living in low-poverty neighborhoods (25% vs. 20%, respectively), but did not have inferior OS (53% vs. 52%). Patients who lived in high-poverty neighborhoods, they reported, were 34% more likely to experience TRM (TABLE). There also were no differences in GVHD or infection based on neighborhood poverty exposure.

“Efforts to address insurance-associated outcome disparities … will require a systematic consideration of nonbiological outcome predictors.”

—Kira Bona, MD, MPH

In the nonmalignant cohort, there also were no significant differences in post-transplant outcomes such as OS, acute GVHD, chronic GVHD, or infection through day 100 between children living in high-poverty and low-poverty neighborhoods.

When the researchers examined other sociodemographic factors, including race, insurance, and ethnicity, they found that five-year OS was significantly inferior in children with malignant diseases who were insured by Medicaid rather than private insurance (48% vs. 55%; hazard ratio [HR] for death = 1.23; p=0.004). Children insured by Medicaid also were more likely to experience TRM (24% vs. 19%; HR=1.28; p=0.006). No such associations were identified in the nonmalignant cohort.

The mechanisms underlying this relationship are “presumably complex and multifactorial,” the authors noted. “U.S. children living in poverty experience higher rates of food insecurity, comorbid illness, injury, infectious disease and hospitalization,” they explained. “It is plausible that inferior underlying health status prior to a diagnosis of cancer makes children more vulnerable to end-organ dysfunction from pre-HCT therapy – and therefore, TRM.”

Dr. Bona and colleagues identified several important limitations, including that their analysis could not evaluate the impact of neighborhood poverty (or insurance) on OS for children who were never successfully transplanted, and poverty and insurance may present barriers to getting to transplant. Data on other social determinants of health, including literacy, education, housing conditions, and social support, also were not accounted for in this report.

“That a child’s insurance status is independently associated with mortality following alloHCT for malignant disease is striking,” the researchers wrote. “Efforts to address insurance-associated outcome disparities in pediatric HCT will require a systematic consideration of nonbiological outcome predictors in future pediatric HCT data collection and therapeutic trials.”

Study authors report no relevant conflicts of interest.


Bona K, Brazauskas R, He N, et al. Neighborhood-poverty and pediatric allogeneic hematopoietic cell transplantation outcomes: a CIBMTR analysis. Blood. 2020 October 26. [Epub ahead of print]