Nonresponse to Bridging Therapy and Peak ALC After Cilta-Cel Is Associated With Neurotoxicity, NRM in Myeloma

High lymphocyte count and nonresponse to bridging therapy following treatment with ciltacabtagene autoleucel (cilta-cel; Carvykti) was associated with a 10-fold increase in the risk of parkinsonism and higher nonrelapse mortality (NRM) rates in patients with multiple myeloma, according to findings from a real-world retrospective study presented at the 2026 Transplantation & Cellular Therapies Meetings.¹

Data from the large-scale, multicenter study, which was designed by the US MM Immunotherapy Consortium, provide critical insights into the predictors of these adverse events related to chimeric antigen receptor (CAR) T-cell therapy. By analyzing 761 patients treated across 15 centers, researchers identified potentially modifiable biomarkers and clinical factors that could redefine risk-mitigation strategies.

“We have 1 in 3 patients who are alive and progression free at 5 years. So how can we get more of those patients there with a great quality of life?” said Surbhi Sidana, MD, associate professor of medicine at Stanford University, in her presentation of the data.

What was the link between parkinsonism and lymphocyte count?

Delayed neurotoxicity (DNT) occurred in 10% (n = 75) of the cohort, manifesting primarily as cranial nerve palsies in 35 patients (4.6%) and parkinsonism in 22 patients (2.9%). Although the incidence of parkinsonism remains relatively low, its impact on quality of life and clinical management is profound.

Previous research showed that higher peak CAR T-cell expansion is correlated to DNT.² “But flow cytometry is not readily available at bedside, and even if it is available at an institution, the results don’t come for another 24 hours. So, what can we use that’s a good surrogate marker at the bedside?” Sidana asked.¹

The study confirmed a correlation between absolute lymphocyte count (ALC) and the development of DNT, particularly parkinsonism. Patients experiencing symptoms exhibited significantly higher peak ALC levels compared to those who did not (median, 5880/uL vs 1170/uL, respectively). Specific thresholds emerged as potent predictors, with a peak ALC greater than 3000/uL being associated with a 12% absolute risk of parkinsonism compared with 1% for those below the threshold (P <.001). Peak ALC of greater than 3000/uL was identified as an independent risk factor in multivariate analysis with an OR of 12.7 (P <.001).

The clear difference in risk suggests that peak ALC may serve as a highly practical biomarker for early intervention. “If an intervention is effective enough to reduce the risk of parkinsonism to baseline…the number needed to treat for such a perfect intervention is only 9 patients,” Sidana said.

Did response to bridging therapy correlate with parkinsonism?

In this population, 86% (n = 650) of patients received bridging therapy and 33% (n = 195) had a partial response or better. The data revealed that 21 of 22 parkinsonism cases (95%) occurred in patients with disease that failed to respond to bridging therapy, even though 91% achieved a response and 68% had a complete response or better to the CAR T-cell infusion itself, which Sidana said was comparable to previous trial data.

Nonresponse to bridging therapy was an independent predictor for parkinsonism (OR, 9.9; P =.03) and a significant driver of NRM (HR, 2.41; P =.046).

“Nonresponse to bridging therapy was associated with a 10-times risk of parkinsonism and a much higher NRM with cilta-cel. Effective tumor debulking with bridging is very critical to decrease the risk of these toxicities and NRM,” Sidana said.

What was the effect on NRM?

The NRM estimates stood at 9% at 1 year and 10% at 2 years. The etiology of NRM was dominated by infectious complications in 56% (n = 35) followed by immune-mediated acute adverse events in 22% (n = 14) and delayed adverse events including delayed neurotoxicity and colitis in 9.5% (n = 6).

In addition to bridging therapy failure, independent predictors for NRM included an ECOG performance status of 2 or higher (OR, 3.97; 95% CI, 1.98-7.77; P <.001), high-risk cytogenetics (OR, 2.39; 95% CI, 1.27-4.55; P =.007), and age of 70 or older (OR, 2.65; 95% CI, 1.40-5.01; P =.003).

How are these findings relevant for addressing neurotoxicity?

These findings emphasize that effective bridging therapy is not only valuable for reaching CAR T, but for ensuring the best outcomes afterward.

The identification of peak ALC as a biomarker offers a window for preemptive management. In patients trending toward high ALC expansion, clinicians may need to maintain a lower threshold for neuro-monitoring or investigate the early use of corticosteroids or other immunosuppressive agents.

Based on these findings, investigators have launched a multicenter observational study in which patients with high CAR T expansion will receive preemptive dexamethasone to reduce the risk of parkinsonism. Sidana acknowledged that dexamethasone may not be sufficient, and other interventions have significant adverse event profiles.

“We know we've identified a good biomarker—it's not perfect. But we have not yet identified a great intervention,” she concluded.

References

1. Sidana S, Reid B, Dima D, et al. Enhancing the safety of ciltacabtagene autoleucel in relapsed multiple myeloma (mm):identification of potentially modifiable risk-factors associated with delayed neurotoxicity and non-relapse mortality. Tandem Meetings: Transplantation & Cellular Therapy Meetings of ASTCT and CIBMTR, February 4-7, 2026; Salt Lake City, UT. Abstract 13.
2. Hosoya H, Velayati A, Dima D, et al. Rapid peak CAR-T expansion is associated with delayed neurotoxicity following ciltacabtagene autoleucel in multiple myeloma. Blood. 146(suppl 1):96. doi: 10.1182/blood-2025-96