Real-world Description of Non-ICANS Neurologic Events Among Patients with Relapsed or Refractory Multiple Myeloma Treated with Ciltacabtagene Autoleucel Using Two Large US Databases

Introduction/Background/Significance: Ciltacabtagene autoleucel (cilta-cel) is a BCMA-directed CAR-T cell therapy approved in the US for relapsed/refractory multiple myeloma (RRMM) as early as after first relapse based on pivotal CARTITUDE-1 (≥4 prior lines of therapy [5L+]) and CARTITUDE-4 (1-3 prior lines of therapy [2L-4L]) trials demonstrating high response rates and prolonged survival; rates of all-grade parkinsonism were 6% and 1%, while rates of cranial nerve palsy were 3% and 9%, respectively. This study aimed to characterize the real-world incidence of key non-immune effector cell-associated neurotoxicity syndrome (non-ICANS) neurologic events (NEs) following cilta-cel treatment.

Materials and Methods/Case Presentation/Objective: This retrospective cohort study analyzed adults with RRMM treated with cilta-cel using two data sources: the Komodo Research Database (01/01/2016-11/30/2024) and Loopback Analytics electronic medical records database (02/28/2017-12/31/2024). Patients were followed from the date of cilta-cel infusion (index date) until the earliest event: end of follow-up, end of data availability, or death. New-onset non-ICANS NEs were identified using ICD-10-CM codes and were defined as ≥2 diagnoses of the same condition on distinct days within 30 days, with no prior history of the condition. Analyses were stratified by 2L-4L and 5L+ and conducted separately in each database.

Results/Description/Main Outcome Measures: In Komodo, 124 patients received cilta-cel at 2L-4L and 524 at 5L+. Median ages were 66 and 64 years, with 36.3% and 43.7% female, respectively. Over the median follow-up of 3.4 and 13.3 months for 2L-4L and 5L+, observed non-ICANS NE incidence were: parkinsonism (2L-4L: 0 [0%] and 5L+: 5 [1.0%]), Guillain-Barré syndrome (0 [0%] and 1 [0.2%]), and cranial nerve palsy (7 [5.6%] and 24 [4.6%]).

In Loopback, 79 patients received cilta-cel at 2L-4L and 191 at 5L+. Median ages were 67 and 65 years, with 43.0% and 44.0% female, respectively. Over the median follow-up of 3.5 and 13.2 months for 2L-4L and 5L+, non-ICANS NEs were: parkinsonism (2L-4L: 0 [0%] and 5L+: 2 [1.0%]), Guillain-Barré syndrome (0 [0%] and 1 [0.5%]), and cranial nerve palsy (4 [5.1%] and 2 [1.0%]).

Conclusions: To our knowledge, this is the first real-world study assessing non-ICANS NEs following cilta-cel treatment across two large US databases, comprising a total of 918 cilta-cel patients. Observed rates of non-ICANS NEs were not higher than those reported in pivotal trials or previously reported real-world studies. These findings underscore the importance of ongoing monitoring and proactive NE management in real-world settings. While use of secondary databases and diagnostic coding may underrepresent non-ICANS NEs, the large sample size and consistency of findings across two distinct data sources enhance the robustness of these results.