OBJECTIVES: Chimeric antigen receptor (CAR) T cell therapy is an effective treatment option for R/R DLBCL. We evaluated how patients with R/R DLBCL value treatment attributes associated with CAR T cell therapy.

METHODS: A survey-based best-best discrete choice experiment (BB-DCE) was administered to adults with self-reported R/R DLBCL in Europe, Japan, and the US. Nine experimentally designed BB-DCE tasks consisted of 3 hypothetical treatment profiles (standard of care [SOC], CAR T cell novel therapy, and non-CAR T cell novel therapy). Treatment attributes included 2-year progression-free survival (PFS), acute treatment reaction (cytokine release syndrome, neurological events), serious infections, chronic adverse effects, dosing schedule, and administration location. A hurdle latent class logit (HLCL) model was used to account for differences in choice behavior. Preference weights were used to calculate relative attribute importance (RAI) and attribute tradeoffs for each class.

RESULTS: Two hundred ten patients completed the survey. Mean age was 59 years; 53% were male and 46% were currently receiving treatment. A 2-class HLCL model had the best fit. Most patients (class 1 [C1]; 71.9%) considered treatment benefits and risks when making choices. In C1, choice behavior was driven by changes in treatment benefit of PFS (RAI=37.4) and patients were willing to accept a 39.7% increase in risk of acute reaction to improve probability of remaining progression free at 2 years from 5% to 25%. A subset of patients (class 2 [C2]; 28.1%) only considered risk of treatment reaction (RAI=35.7) and treatment location when making choices and were more likely to prefer SOC.

CONCLUSIONS: Most patients were focused on treatment benefit and improved PFS; however, nearly one third were risk averse and preferred treatment aligned with SOC. Recognizing that patients have different priorities highlights the importance of shared clinical decision-making to ensure that patients receive treatments in line with their own priorities.