AUTHORS
Callie Fry, Kate Feldman, BA2, Edie Weller, PhD3, Yael Flamand, MS4, Rahela Aziz-Bose, MD, MPH5, Puja J. Umaretiya, MD, MS6, Colleen A. Kelly, MD7, Haley Newman, MD, MSCE8, Daniel J. Zheng, MD, MHS, MSHP9, Ariana Valenzuela, MSGH10, Victoria B. Koch, BS11, Melissa A. Burns, MD12, Peter D. Cole, MD13, Hital Patel, BS14, Lisa M. Gennarini, M.15, Bruno Michon, MD16, Thai Hoa Tran, MD17, Jennifer J. G. Welch, MD18, Lewis B. Silverman, MD19, Kara M Kelly, MD20, Elizabeth Bouchard, PhD21, Justine M. Kahn, MD, MS22, Kira Bona, MD, MPH23
AFFILIATIONS
1 Dana-Farber/Boston Children’s Cancer and Blood Disorders Center, Harvard Medical School, Boston, Massachusetts, United States
2 Biostatistics and Research Design Center, Boston Children’s Hospital, Boston, Massachusetts, United States
3 Biostatistics and Research Design Center, Boston Children’s Hospital and Department of Pediatrics, Harvard Medical School, Boston, Massachusetts, United States
4 Department of Data Science, Dana Farber Cancer Institute, Boston, Massachusetts, United States
5 Dana-Farber/Boston Children’s Cancer and Blood Disorders Center, Harvard Medical School, Boston, Massachusetts, United States
6 Division of Pediatric Hematology/Oncology, Department of Pediatrics, University of Texas Southwestern Medical Center, Dallas, Texas, United States
7 Dana-Farber/Boston Children’s Cancer and Blood Disorders Center, Harvard Medical School, Boston, Massachusetts, United States
8 Division of Oncology, Children's Hospital of Philadelphia, Philadelphia, Pennsylvania, United States
9 Division of Oncology, Children's Hospital of Philadelphia, Philadelphia, Pennsylvania, United States
10 Division of Pediatric Intensive Care, Children’s Hospital Colorado, Aurora, Colorado, United States
11 Dana-Farber/Boston Children’s Cancer and Blood Disorders Center, Harvard Medical School, Bosotn, Massachusetts, United States
12 Dana-Farber/Boston Children’s Cancer and Blood Disorders Center, Harvard Medical School, Boston, Massachusetts, United States
13 Rutgers Cancer Institute, New Brunswick, New Jersey, United States
14 Department of Cancer Prevention and Control, Roswell Park Comprehensive Cancer Center, Buffalo, New York, United States
15 Montefiore Medical Center, Bronx, United States
16 Centre Hospitalier Universitaire de Quebec, Quebec, Canada
17 Division of Pediatric Hematology/Oncology, Charles-Bruneau Cancer Center, CHU Ste-Justine, University of Montreal, Montreal, Canada
18 Division of Pediatric Hematology/Oncology, Hasbro Children’s, Brown University Health, Providence, Rhode Island, United States
19 Division of Pediatric Hematology, Oncology and Stem Cell Transplantation, Columbia University Irving Medical Center, New York, New York, United States
20 Department of Pediatrics, Roswell Park Comprehensive Cancer Center, University at Buffalo Jacobs School of Medicine and Biomedical Sciences, Buffalo, New York, United States
21 Roswell Park Comprehensive Cancer Center, University at Buffalo Jacobs School of Medicine and Biomedical Sciences, Buffalo, New York, United States
22 Division of Pediatric Hematology, Oncology and Stem Cell Transplantation, Columbia University Irving Medical Center, New York, New York, United States
23 Dana-Farber/Boston Children’s Cancer and Blood Disorders Center, Harvard Medical School, Boston, Massachusetts, United States
ABSTRACT
Introduction: Children with acute lymphoblastic leukemia (ALL) exposed to poverty are more likely to relapse and die despite clinical trial participation (Winestone, PBC, 2023). Mechanisms underlying observed disparities are poorly understood, though hypothesized to include differential care delivery such as chemotherapy delays. Poverty may drive chemotherapy delays through mechanisms such as missed clinic visits or higher rates of treatment-related infectious toxicities that postpone cycle starts. We aimed to evaluate whether household material (HMH: food, housing, or utility insecurity), a concrete measure of family-reported poverty exposure, was associated with the frequency of chemotherapy cycle delays in a frontline pediatric ALL trial.
Methods: This secondary analysis used prospectively collected data from the randomized, phase III Dana-Farber ALL Consortium trial 16-001 (NCT03020030) that enrolled children with de novo ALL at 8 US and Canadian centers (2017-2022). Two embedded, opt-in, correlative studies among English-, Spanish-, and French-speaking families of patients <18 years evaluated (1) parent-reported sociodemographic characteristics via survey during the first cycle of chemotherapy and (2) chemotherapy cycle start dates documented via the medication event monitoring system (MEMS) caps during Consolidation II and Maintenance phases of therapy (total of 6 monitored 21-day cycles, 3 during each phase).
The analytic cohort was restricted to patients with available HMH data and documented cycle start dates for at least 2 consecutive chemotherapy cycles. The primary exposure was severe HMH, defined as ≥2 baseline resource insecurities vs 0-1. The primary outcome was ≥1 chemotherapy cycle delay, utilizing trial-defined delays (>3 days in Consolidation II, >7 days in Maintenance). Sociodemographic and clinical characteristics were summarized using descriptive statistics. Fisher’s exact and Wilcoxon rank-sum tests assessed univariate associations between covariates and cycle delay including child sex, age, race/ethnicity (Hispanic, non-Hispanic Black, non-Hispanic Other, non-Hispanic White), insurance (US public, US private, Canadian), treatment arm (low-risk vs intermediate/high/very high), language (dominant [US English, Canadian English/French] vs other); parent age, education (high-school education vs less), household structure (single vs dual parent), low-income household (<200% of the federal poverty level [FPL] vs other). Univariate logistic regression evaluated the association between HMH and ≥1 chemotherapy cycle delay. Statistical significance was set at α=0.05 (two-sided).
Results: The analytic cohort included 189 children (72% US, 28% Canadian) with a median age of 5 years (Q1: 4, Q3: 9) and median caregiver age of 38 years (Q1: 34, Q3: 43). Fifteen percent of children lived with severe HMH, 31.7% in a low-income household, and 24.3% had US public insurance. Children were identified as 16.9% Hispanic and 3.2% non-Hispanic Black; 18% spoke a non-dominant language, 19.6% lived in a single-parent household, and 20.1% had a parent with high-school education or less.
Overall, 32.3% of children experienced ≥1 chemotherapy cycle delay. Of these, 23.3% had a cycle delay during Consolidation II only, 7.4% during Maintenance only, and 1.6% during both. The frequency of ≥1 chemotherapy cycle delay was 39.3% among children with HMH exposure versus 31.1% among unexposed. On univariate analysis, HMH was not significantly associated with experiencing >1 cycle delay (odds ratio 0.7, 95% CI [0.3,1.59], p=0.39). Similarly, no other sociodemographic characteristics were associated with experiencing ≥1 cycle delay.
Conclusion: Among children with ALL treated on a multi-center frontline trial, 1 in 3 experienced >1 chemotherapy cycle delay, with no variation by parent-reported poverty exposure (HMH). These null findings suggest that cycle delays may not be the primary care-delivery mechanism driving socioeconomic disparities in ALL outcomes, and that other pathways—including treatment resistance and treatment adherence—warrant continued investigation.
DISCLOSURES
Callie Fry: No relevant conflicts of interest to declare.