Real-world incidence and management of non-icans neurologic events following ciltacabtagene autoleucel in multiple myeloma

Abstract Introduction: Ciltacabtagene autoleucel (cilta-cel) is an established effective treatment strategy for relapsed/refractory multiple myeloma in fifth line or later (5L+) and was recently approved for patients who have received ≥1 prior line of therapy. Although rare, non-immune effector cell-associated neurotoxicity syndrome (ICANS) neurologic events (NEs), including cranial nerve palsy (CNP), parkinsonism, and Guillain-Barré syndrome, may occur post-infusion. This study evaluated the incidence, clinical characteristics, and management strategies of non-ICANS NEs to inform patient care. Methods: This retrospective cohort analysis utilized Loopback Analytics electronic medical records (Feb 2017–May 2025) supplemented with physician chart notes from academic and community centers across the US. Adults treated with cilta-cel in second to fourth line (2L–4L) and 5L+ settings were included if they had ≥1 absolute lymphocyte count (ALC) test within 30 days pre- and post-infusion. Incident non-ICANS NEs were defined as ≥2 ICD-10 diagnoses of the same condition on distinct days within 30 days of each other. Non-ICANS NEs were further verified through manual chart review using unstructured data elements from patient charts to ensure clinical accuracy. Pre-lymphodepletion ALC (x 103/μL; i.e., closest value to cilta-cel infusion),post-infusion peak ALC (x 103/μL), and management strategies for non-ICANS NEs were described along with cilta-cel response and mortality. Interim results are presented; data collection is ongoing with updated analyses planned for additional patients. Results: Among 174 patients (2L–4L cilta-cel: 73; 5L+ cilta-cel: 101), median age was 64 years (range: 37-83 years). With a median follow-up of 6.1 months (2L–4L; range: 0.2-13.3) and 17.2 months (5L+; range: 1.1-33.0), non-ICANS NEs remained rare. CNP occurred in 4 (5.5%) patients in the 2L–4L cohort and 3 (3.0%) in the 5L+ cohort. Parkinsonism and Guillain-Barré syndrome were each reported in 1 patient, both in the 5L+ cohort, with no cases observed in 2L–4L. Among patients without non-ICANS NEs, the median post-infusion peak ALC was 2.1 in the 2L–4L cohort and 2.0 in the 5L+ cohort. In the 2L–4L cohort, 4 CNP cases occurred at a median of 29 days post-infusion. Median pre-lymphodepletion ALC was 0.10, while post-infusion peak ALC was 7.6 (range: 1.2-16.8; median day 11 post-infusion). Management included prednisone and valacyclovir. Over a median follow-up of 6.7 months, 3 (75.0%) patients showed improvement in CNP (median 15 days), and 1 (25.0%) patient experienced full resolution by day 62. In the 5L+ cohort, 3 CNP cases occurred at a median of 25 days post-infusion. Median pre-lymphodepletion ALC was 0.10, and post-infusion peak ALC was 8.8 (range: 8.3-26.9; median day 11 post-infusion). Management reported for one 5L+ CNP patient resulting in full resolution of CNP included prednisone, dexamethasone, and valacyclovir; data on CNP improvement wasn't documented for the remaining two patients in the medical charts. Over a median follow-up of 7.9 months, all 3 patients achieved a complete response to cilta-cel. The one case of parkinsonism in the 5L+ cohort occurred on day 46 post-infusion. Pre-lymphodepletion ALC was 0.60, with a post-infusion peak ALC of 14.5 on day 15. Management for parkinsonism included cyclophosphamide. The patient achieved complete response to cilta-cel within 15.5 months of follow-up. Data on neurologic symptom improvement wasn't documented for this patient in the medical charts. Finally, the case of Guillain-Barré syndrome in the 5L+ cohort occurred on day 105 post-infusion. Pre-lymphodepletion ALC was 0.04, with a post-infusion peak ALC of 8.8 on day 10. During the 15.7 month follow-up period, management included intravenous immunoglobulin and plasma exchange, and the patient achieved complete response post-cilta-cel. Data on neurologic symptom improvement wasn't documented for this patient in the medical charts. Conclusion: In this real-world cohort, CNP, parkinsonism, and Guillain-Barré syndrome cases were infrequent following cilta-cel infusion. Affected patients had higher post-infusion peak ALCs compared to those without non-ICANS NEs, suggesting peak ALC may serve as a potential biomarker for identifying patients at risk for NEs and guiding prophylactic and therapeutic strategies. Despite experiencing non-ICANS NEs, all patients with available response assessments responded to cilta-cel, and no deaths were reported.