The feasibility of collecting social determinants of health in the national multi-center GRASP sickle cell gene therapy trial

Abstract

Abstract Introduction: Gene therapy offers a transformative, potentially curative opportunity for patients with sickle cell disease (SCD), a highly morbid and life-threatening condition that affects 100,000 Americans. Optimizing access to and outcomes from SCD gene therapy requires evaluation of adverse social determinants of health (SDOH) including poverty. In oncology populations, poverty is independently associated with reduced access to centralized, resource-intensive, curative treatments akin to gene therapy—including stem cell transplant and CAR T-cell therapy. Further, poverty-exposed children with cancer are more likely to relapse and die, even when treated on highly standardized multi-center clinical trials. More than half of patients with SCD are exposed to poverty—including unmet basic needs defined as household material hardship ([HMH]; food, housing, utility or transportation insecurity)—and poverty is associated with more frequent disease complications, higher healthcare utilization, and inferior health-related quality of life among children with SCD (Khan, PBC 2023). Whether poverty-associated disparities in access to and outcomes from gene therapy for SCD exist is unknown. Sociodemographic data elements routinely collected in clinical trials have been limited to race, ethnicity and insurance, which serve as proxies for poverty exposure and other adverse SDOH but can neither identify mechanisms underlying disparities nor provide modifiable intervention targets. We report the feasibility of baseline patient/parent-reported poverty data collection as an embedded aim of the national, phase II Gene Therapy to Reduce All Sickle Cell Pain (GRASP) trial (NCT05353647, BMT CTN 2001). Methods: GRASP enrolled patients 13-40 years with HbSS or HbS/β0 thalassemia genotype, clinically severe SCD, and no available HLA-identical marrow donor across 9 centers from 2022-2025 as an open-label, non-randomized, phase 2, single-arm gene transfer trial to induce fetal hemoglobin. Parents/guardians of participants <18 years or participants ≥18 years completed a brief (21-item) “Household Survey” at baseline and 24-month follow-up to evaluate the associations between poverty and trial outcomes. Survey domains included primary language, health literacy, marital status, education, HMH, income, and social support. The primary exposure of interest was HMH. Paper/pencil or electronic surveys were available in English or Spanish and could be administered in any language with an interpreter. Surveys were self-completed or read aloud by research staff at the treating site, and data were centrally monitored. Participants did not receive remuneration for survey completion. The primary outcome for this secondary analysis was feasibility of baseline poverty data collection, defined as the proportion of trial participants with completed baseline surveys; acceptability was characterized by the proportion of surveys with evaluable HMH data (the primary exposure of interest). Results: Among 24 trial-evaluable participants, 24 (100%) completed the baseline Household Survey across 8 sites. Surveys were completed solely in English, and primarily via paper/pencil (92%). One hundred percent of surveys had evaluable HMH data. Surveys were completed a median of 150 days (range 31-477) from trial enrollment, and 96% within the protocol-specified window (prior to gene therapy infusion). HMH-exposure was reported by 9 (38%) participants—with food (n=8, 33%) and housing (n=6, 25%) insecurities the most frequently reported domains. Conclusion: Collection of patient/parent-reported poverty data as an embedded aim of a multi-center, SCD gene-therapy trial is feasible and acceptable. That 100% of participants were willing to share data on food, housing, utility and transportation insecurity for research purposes supports the acceptability of this data collection. While 1 in 3 trial-participants reported HMH, this prevalence is substantially lower than the published frequency of HMH in the US SCD patient population, warranting future analyses of trial access barriers such as insurance coverage, employment flexibility, and distance to enrolling sites. Evaluation of the associations between HMH and clinical and patient-reported outcomes is pending mature trial data. Integration of this approach to data collection in future trials is critical to identifying patients who may benefit from targeted interventions to optimize access to and outcomes from this novel therapy.