AUTHORS
Xin Ma, BS, Weixi Zhu, PhD2, Rasmi Girijavallabhan Nair, MBBS, MPH, PhD3, Heather Wolfe, MD4, Elif Yilmaz, MD5, Farrukh T Awan, M.D.M.S.,M.B.A.6, Yu-Lun Liu, PhD7, Praveen Ramakrishnan Geethakumari, MD, MS, MBA, FACP8
AFFILIATIONS
1 UT Southwestern Medical Center, Dallas, Texas, United States
2 UT Southwestern Medical Center, Dallas, Texas, United States
3 UT Southwestern Medical Center, Dallas, Texas, United States
4 UT Southwestern Medical Center, Dallas, Texas, United States
5 UT Southwestern Medical Center, Dallas, Texas, United States
6 UT Southwestern Medical Center, Dallas, Texas, United States
7 UT Southwestern Medical Center, Dallas, Texas, United States
8 UT Southwestern Medical Center, Dallas, Texas, United States
ABSTRACT
Background
Chimeric antigen receptor T-cell (CAR-T) therapy has transformed the treatment landscape for relapsed/refractory (R/R) B-cell lymphomas. However, current CAR-T delivery is resource-intensive, requiring referral to specialized cellular therapy centers, utilizing substantial healthcare infrastructure, thus creating barriers to equitable access beyond clinical eligibility. Although disparities in CAR-T receipt have been described, comprehensive evaluation of patient-, socioeconomic-, and healthcare system-level determinants of CAR-T utilization remain limited. We therefore evaluated factors independently associated with CAR-T utilization among patients with R/R follicular (FL) and diffuse large B-cell lymphomas (DLBCL) using a nationwide electronic health record-derived database.
Methods
A retrospective cohort study was conducted using de-identified electronic health record data from Flatiron Health. Adult patients with FL who initiated third-line or later (3L+) systemic therapy and DLBCL who initiated second-line or later (2L+) systemic therapy between January 2011 and May 2024, potentially eligible for standard-of-care CAR-T were included. The primary outcome was receipt of CAR-T. Baseline demographic characteristics, including age, sex, race, ethnicity, insurance status, neighborhood socioeconomic status (SES; 1=lowest, 5=highest), and practice type (academic, community, or hybrid academic/community), were compared between CAR-T recipients and non-recipients. Continuous variables were analyzed using Wilcoxon rank-sum tests and categorical variables using Pearson’s chi-square tests. Separate multivariable logistic regression models were fitted for FL and DLBCL, with all prespecified baseline covariates entered to identify independent predictors of CAR-T receipt.
Results
The study population included 787 patients with FL receiving 3L+ therapy, of whom 107 (13.6%) received CAR-T therapy and 680 (86.4%) did not. The DLBCL cohort included 2,381 patients receiving 2L+ therapy, of whom 272 (11.4%) received CAR-T therapy and 2,109 (88.6%) did not. In both cohorts, CAR-T recipients were significantly younger (median age; FL: 61 vs. 70 years; DLBCL: 61 vs. 68 years; both P<0.001), more frequently treated at academic centers and more likely to be from higher-SES category, than non-recipients. Commercial insurance was also more common among CAR-T recipients in univariate analyses.
On multivariable analysis, advanced age remained independently associated with lower odds of CAR-T receipt in both R/R FL (OR 0.95, 95% CI 0.93–0.97; P<0.001) and DLBCL (OR 0.97, 95% CI 0.96–0.99; P<0.001). Compared with academic centers, treatment at community practices was associated with substantially lower odds of CAR-T utilization in R/R FL (OR 0.29, 95% CI 0.18–0.48; P<0.001) and DLBCL (OR 0.42, 95% CI 0.31–0.56; P<0.001).
Compared with patients in SES category 1, R/R FL patients from SES categories 2-5 had significantly greater odds of receiving CAR-T (OR range 4.16–4.97; all P<0.05). In R/R DLBCL, patients from SES category 4 were independently associated with increased CAR-T utilization (OR 1.75, 95% CI 1.11–2.84; P=0.020). Race and sex were not independently associated with CAR-T utilization in either cohort. In R/R DLBCL, Hispanic ethnicity was independently associated with increased CAR-T utilization (OR 1.74, 95% CI 1.05–2.82; P=0.027). In R/R FL, Medicare (OR 0.33, 95% CI 0.12–0.80; P=0.022) and other insurance (OR 0.25, 95% CI 0.06–0.80; P=0.035) were associated with lower odds of CAR-T utilization compared with commercial insurance.
Conclusions
In this nationwide real-world study, CAR-T utilization was low, at 13.6% among patients with R/R FL and 11.4% among patients with R/R DLBCL. Limitations include retrospective study design, CAR-T eligibility defined per FDA approved treatment-line and unavailability of individual patient clinical assessment. Younger age, treatment at academic centers, and higher socioeconomic status were consistently associated with greater CAR-T utilization whereas older age and care in community practices were independently associated with lower utilization. These findings suggest that healthcare delivery infrastructure and socioeconomic context are major determinants of CAR-T access in routine practice. Future research and health policy efforts should focus on optimizing referral pathways and reducing healthcare barriers to improve equitable access to CAR-T.
DISCLOSURES
Xin Ma, BS: No relevant conflicts of interest to declare.