Developing a Fit-for-Purpose Composite Symptom Score for Representing Symptom Burden for Managing CRS and Icans after CAR T-Cell Therapy

Introduction: With the clinical benefits of chimeric antigen receptor (CAR)- T-cell therapy are increasingly recognized, both acute and late-onset severe symptomatic toxicities, CRS and ICANS, that require initial hospitalization followed by careful monitoring while patients discharged. A more relevant and effective monitoring and managing the toxicities of CAR T-cell therapy has been proposed by CMS with Patient-Reported Outcomes (PROs). This study examined a brief set of Patient-Reported symptoms and functioning that highly representing significant CRS or ICANS during the first year of CAR T-cell therapy to provide a fit-for-purpose and easier to use symptom assessment tool for future patient care for monitoring development and recovery of these toxicities, and for applied in clinical trials for effective managing these toxicities.

Methods: This cross-sectional study surveyed relapsed/refractory lymphoma patients who scheduled or received standard of care CAR T-cell therapy at MD Anderson Cancer Center. PROs with symptoms and symptom interference to functioning were assessed using psychometric validated tool, the CAR module of the MD Anderson Symptom Inventory (MDASI-CAR), which included 13 core symptoms, 10 relevant symptoms to CAR T-cell therapy and 6 symptom interference items for functioning measure. All MDASI items are rated on a 0-10 scale to describe symptom severity during the past 24 hours. Patients also rated health status on PROMIS 29, EQ-5D and health related QoL on a 0-10 scale of single item on the same time of symptom assessment.

RESULTS: A total of 78 patients were included during anytime of their first year of treatment; 68 (87%) received Axicabtane cilocleucel (Yescarta); 25 (32%) developed grade 2-4 CRS and 27 (35%) had grade 2-4 neurotoxicity (ICANS). Total of 45 patients contributed PROs after 30 days of therapy. Using MDASI-CAR, we observed that multiple symptoms were significantly more severe post 30 days of therapy among patients with grade 2-4 CRS or ICANS, compared to those with mild AEs (both CRS and ICANS on 0-1); these included fatigue, poor appetite, poor balance or falling, dizziness, difficulty speaking (finding words) and symptom interference with walking. The effect size (ES) for the composite score of these items was 0.59 between grade 2-4 vs. 0-1 of CRS (95% CI: (-0.01, 1.20)), and 0.83 between grade 2-4 vs. 0-1 of ICANS (95% CI: (0.20, 1.45)). The Cronbach alpha was 0.72 of this composite score, indicative of a satisfied reliability. The significance test by ECOG-PS (0-1 vs. 2-5) indicated an excellent known group validity of this composite score (p < 0.0001). Physical function domain on PROMIS-29 also showed large effect sizes by group of AEs (CRS ES = 0.74; ICANS ES = 1.15), strongly supporting the inclusion of the “walking” item with a cluster of symptoms in the composite score.

CONCLUSION: This study identified a fit-for-purpose significant AE related symptom burden in patients with refractory/relapsed lymphoma who receive CAR T-cell therapy. These self-reported unique profiles of symptom burden that a brief set of PROs well representing significant CRS or ICANs could be applied in intensive patient care under ePRO-based pathway without much survey burden for repeated symptom monitoring during the first year. Additional longitudinal study with larger sample from real world observations of CAR T-cell therapy is needed to optimize the composite score and implement in the routine symptom monitoring in CAR T-cell patient care.