Pilot Study of Longitudinal Geriatric Assessment and Neurocognitive Evaluation in CAR T-Cell Therapy for Older Patients with Relapsed Refractory, Large B-Cell Lymphoma

Introduction: Chimeric antigen receptor T-cell (CAR T) therapy is increasingly utilized for older patients with relapsed refractory, large b-cell lymphoma (LBCL) with promising efficacy. However, the feasibility of conducting serial geriatric assessments with appropriate management has not been examined in this vulnerable population of patients. Moreover, the short-term and long-term impact of CAR T-cell therapy on the older patient's overall physical and neurocognitive functioning remains unknown. Methods: In this prospective pilot study (NCT04300998), we enrolled older patients with LBCL receiving commercial CAR T and conducted serial geriatric assessments (GA), formal neurocognitive testing in domains sensitive to cancer therapy side effects including attention, memory, and executive functions, and quality-of-life surveys prior to and following CAR T treatment at 3, 6, and 12 months among patients who remained in remission. CAR T manufacturing, treatment, and follow-ups followed national and institutional guidelines. Cytokine release syndrome (CRS) and Immune effector cell-associated neurologic syndromes (ICANS) were graded according to the ASTCT criteria. Standard statistical analysis was performed for CAR T treatment outcomes. GA impairment in each domain (binary outcome) was tabulated as percentages. For neurocognitive testing, means across timepoints were estimated for each test score using a linear mixed model predicting the score by timepoint. A random intercept was included to account for multiple assessments per patient. Results: Between April 2020 to March 2022, 18 patients were enrolled. The median age was 69.8 years (range 63.7 to 85.0) and 8 of 18 patients, 44%, were female. Eight patients received Axicabtagene ciloleucel while 6 and 4 patients received Tisagenlecleucel and Lisocabtagene maraleucel, respectively. Fourteen out of 18 patients, 78% (95% confidence interval [CI] 52 - 94), completed all required GA visits (baseline and all follow-up timepoints while in remission), meeting the prespecified feasibility endpoint of 75%. Geriatric vulnerabilities detected by GA were highly prevalent at the baseline including functional impairment by instrumental activities of daily living (44%), mobility impairment (24%), multimorbidity (61%), polypharmacy (44%), poor psychosocial support (39%), and depression (28%). Moreover, on a panel of 7 neurocognitive tests, 52.9% of patients had at least one and 29.4% had 2 or more impairments at baseline. Wit