Real-world patient characteristics, outcomes, and resource utilization of chimeric antigen receptor (CAR) T cell therapy across Foundation for the Accreditation of Cellular Therapy (FACT) and non-FACT treatment centers in large B-cell lymphoma (LBCL).

Abstract

7023 Background: CAR T cell therapy has transformed the LBCL treatment landscape and is now being offered in some non-FACT accredited centers. We evaluated clinical characteristics, AEs, time to next treatment (TTNT), and health care resource utilization (HCRU) of patients with LBCL who received CAR T cell therapy from FACT versus non-FACT centers. Methods: This retrospective cohort study used the Komodo Health Claims Database (01/2015–08/2025). Adults with LBCL receiving CAR T cell therapy were identified via International Classification of Diseases, Ninth/Tenth Revision, diagnosis and procedure codes. Centers were classified as FACT or non-FACT based on public records. Cytokine release syndrome (CRS), immune effector cell–associated neurotoxicity syndrome (ICANS), and infections were assessed within 15 and 30 days after infusion. TTNT was estimated using KM methods. All-cause HCRU was reported as per patient per month (PPPM) to adjust for variation in follow-up duration; inpatient utilization was assessed by length of stay (LOS). Comparisons used chi-square test. Results: Among 6857 CAR T cell therapy−treated eligible patients with LBCL (FACT, n = 6748; non-FACT, n = 109), demographics and characteristics were comparable between cohorts; mean Charlson Comorbidity Index score was 4.6 versus 4.9, respectively. Significantly more patients treated at FACT sites traveled > 50 miles compared with those treated at non-FACT sites (43% vs 31%; P < 0.05); median distance traveled was 34 versus 17 miles, respectively. Rates of CRS (44% vs 40%), ICANS (21% vs 24%), and infections (39% vs 36%) within 15 days after infusion were similar between cohorts (FACT vs non-FACT, respectively); rates were also comparable within 30 days. Median (95% CI) TTNT was numerically lower in the FACT-treated (23 months [21–25]) versus non-FACT–treated cohort but not statistically significant (42 months [18–not estimable]; P = 0.20; estimates limited by small sample size); 12-month TTNT rates were 58% versus 67%, respectively. Excluding infusion visits, FACT-treated patients had marginally higher postinfusion HCRU (PPPM) over 1 year, including inpatient (0.93 vs 0.61; P = 0.52), outpatient (3.84 vs 3.16; P < 0.05), and ICU visits (0.13 vs 0.07; P < 0.05). Standardized 1-year inpatient LOS was 18.4 versus 15.9 days, respectively ( P = 0.58). Among patients treated in an inpatient setting, HCRU including infusion visit exhibited a similar trend. Conclusions: In this real-world LBCL cohort, clinical characteristics, safety, TTNT, and HCRU were generally comparable between FACT- and non-FACT– accredited centers. Despite the non-FACT sample size and claims-based limitations, these findings suggest patients have similar outcomes across both types of authorized treatment centers.