No CARs for miles: Geographic disparities in access to multiple myeloma CAR-T therapy

Abstract

Abstract Introduction Chimeric antigen receptor T-cell (CAR-T) therapy has revolutionized treatment for relapsed/refractory multiple myeloma (MM), promoting deeper, more durable responses than chemotherapy. A fundamental barrier to CAR T-cell access is the distance to a CAR-T center. Others have shown geographic disparities in clinical trial access, but how geography affects access to MM CAR-T centers is unclear. Here, we present a unique geospatial analysis of CAR-T access in the US, assessing county-level proximity to CAR-T centers, population drive-time coverage, and correlations with social determinants of health. We then interrogate how geography interacts with social determinants of health by comparing socioeconomic, household, minority, housing, and transportation vulnerabilities that are summarized in the CDC's Social Vulnerability Index (SVI). Methods Geospatial drive-time analysis was conducted using the locations of authorized CAR-T therapy centers in the USA (76 ide-cel and 55 cilta-cel centers, obtained April 2025) via the ARC Geographic Information System (ArcGIS, Enzi- Redwood, CA, USA). Drive-time zones (0–1 hr, 1–2 hrs, 2–5 hrs, >5 hrs) from each county centroid to the nearest CAR-T center were calculated, and counties were categorized by drive-time group. SVI and mortality data from the CDC were linked at the county level, with SVI scores compared across zones using one-way ANOVA. Pearson correlation and linear regression were used to assess associations between distance to CAR-T centers and MM mortality. Analyses were conducted in GraphPad, with significance defined as p<0.05. Results Among 3141 U.S. counties analyzed, 705 (22.4%) were within 1 hour, 1268 (40.4%) were within 1-2 hours, 1014 (32.3%) were within 2-5 hours and 154 (4.9%) were 5 hours or greater from a MM CAR-T center, equating to 59.2%, 22.7%, 16.8% and 1.4% of the population living in the 0-1, 1-2, 2-5, and >5 hour driving time area, respectively. Thirty-seven percent of counties (18.2% of population) are 2 hours or greater from the nearest CAR-T center. Examining MM deaths, 19.1% occur in counties >2 hours from the nearest CAR-T center with an overrepresentation of minority population. Regional disparities in CAR-T center localization were evident. CAR-T centers were more likely to be academic hospitals located on the East or West Coast, and more likely to be in urban areas. Vast areas of the Great Plains are markedly distant from CAR-T centers, and key predominantly minority regions with higher SVI such as the Rio Grande region of Texas and the Deep South are at least moderately distant from CAR-T centers. Using SVI quartiles to examine the most vulnerable counties, 43.0% of counties with the highest SVI quartile (Q4) existed >2 hours from the nearest CAR-T center. Counties 1-2 and 2-5 hours from CAR-T centers exhibited higher median SVIs (0.45-0.5 and 0.55-0.6, respectively) than those within 0-1 hour (0.3-0.4) (p<0.0001). Conclusion CAR-T therapy has fundamentally changed MM care, but access to CAR-T centers limits broad application of this intervention to specialized centers. Our analysis is the first of its kind to show that geographic access to CAR-T cell centers represents a unique health care disparity in MM care, with a significant portion of the US population living several hours away from the nearest CAR-T centers. Furthermore, the patients who live further away from CAR-T centers may be less able to relocate to a CAR-T center. As hospitals require specialized expertise and resources to become a CAR-T center, these geographic disparities are unlikely to be solved with current CAR-T manufacturing strategies. Thus, novel CAR-T manufacturing strategies and pipelines remain an important unmet clinical need.