Rapid CAR T Expansion Tied to Higher Risk of Neurotoxicity in Multiple Myeloma

New findings presented at the 2025 International Myeloma Society Annual Meeting suggest that rapid CAR T-cell expansion is associated with an increased risk of delayed neurotoxicity (DNT) in patients with relapsed/refractory multiple myeloma (RRMM) receiving ciltacabtagene autoleucel (cilta-cel; Carvykti, Janssen Biotech, Inc/Legend Biotech). The data provide potential tools for early risk stratification and could guide preemptive intervention strategies.

Cilta-cel, a BCMA-targeted CAR T-cell therapy, was approved by the FDA in 2022 for adult patients with RRMM after four or more prior lines of therapy, with an expanded label in 2024 allowing use after at least one prior line. These approvals were based on the CARTITUDE-1 trial (NCT03548207), which demonstrated durable responses and progression-free survival benefits. In August 2025, cilta-cel made headlines when 33% of patients remained alive and progression-free at five years without maintenance therapy.

However, cilta-cel’s clinical utility has been tempered by unique toxicities, particularly delayed neurotoxicity. To better understand this risk, investigators analyzed CAR T expansion using flow cytometry and evaluated absolute lymphocyte count (ALC) as a surrogate marker. Two patient cohorts were included: 256 patients treated between 2022 and 2024 across three institutions (ALC cohort) and 54 patients in a CAR expansion cohort.

In the ALC cohort, patients had a median age of 64 years, with 38% classified as penta-refractory. At a median follow-up of 14.7 months, progression-free survival reached 28.7 months. DNT occurred in 29 patients, including 20 cases of cranial nerve palsy and 8 of parkinsonism. In the CAR expansion cohort, baseline characteristics were similar, with 16% experiencing DNT.

The analysis showed that patients who developed DNT had significantly higher peak CAR T-cell levels (P = .04), strongly correlated with peak ALC (rho = 0.84, P < .001). Median peak ALC in patients with DNT was 5780/μL versus 2200/μL in those without (P < .001), with particularly elevated counts among patients with parkinsonism (13,335/μL vs 2270/μL; P < .001). An early ALC rise between days 7 and 12 was also predictive of DNT (5360/μL vs 1040/μL; P < .001).

Researchers identified two clinically relevant ALC thresholds to stratify DNT risk:

• ≥3000/μL between days 7–21, or

• ≥2500/μL between days 7–21 with at least a twofold rise from baseline.

These criteria captured 83% of DNT cases—including nearly all parkinsonism events—while excluding 41% of patients without DNT.

The study highlights ALC as a practical, reliable biomarker for CAR T expansion and a tool for early identification of patients at risk for neurotoxicity. According to investigators, applying these thresholds may enable preemptive interventions, improving the safety of cilta-cel without diminishing its efficacy.