A single infusion of mivocabtagene autoleucel was associated with a median 45.6% reduction in walking time at 16 weeks among patients with treatment-refractory stiff person syndrome (SPS), and 80.8% achieved clinically meaningful mobility gains. All 26 treated patients remained free of immunomodulatory or immunosuppressive therapy during follow-up, after discontinuing these treatments before infusion as required by the study protocol.
In the nonrandomized, open-label phase 2 KYSA-8 trial, investigators evaluated the safety and efficacy of mivocabtagene autoleucel, an autologous CD19-directed chimeric antigen receptor (CAR) T-cell therapy, at 3 US centers.
From September 2024 to May 2025, 27 patients were enrolled, and 26 received treatment. Eligible patients were aged 18 to 75 years, had antibody-positive SPS, and had experienced an inadequate response to at least 1 immunomodulatory treatment. The median age was 56 years, 16 patients (61.5%) were female, and median follow-up was 6.5 months.
Before infusion, patients received lymphodepleting chemotherapy with fludarabine and cyclophosphamide for 3 days, followed by a single intravenous infusion of 100 million autologous CD19-directed CAR T cells. The primary efficacy endpoint was change from baseline to week 16 in the timed 25-foot walk test, with safety evaluated as a coprimary endpoint. Secondary measures included the modified Rankin Scale, Distribution of Stiffness Index, Heightened Sensitivity Scale, and Hauser Ambulation Index.
At week 16, the mean improvement in the timed 25-foot walk was 7.7 seconds. Overall, 21 of 26 patients (80.8%) achieved an improvement of at least 20%, a threshold considered clinically meaningful in prior studies of multiple sclerosis. Additionally, 8 patients (30.8%) completed the test in less than 5 seconds, compared with none at baseline.
Among 12 patients who required a walking aid at baseline, 8 (66.7%) no longer needed assistance at week 16. Improvements persisted among 16 patients evaluated at week 24, with a median 44.4% reduction in walking time.
Secondary measures of disability, stiffness, sensory sensitivity, and ambulation also improved. At week 16, mean scores decreased by 0.8 points on the modified Rankin Scale, 1.5 points on the Distribution of Stiffness Index, 3.2 points on the Heightened Sensitivity Scale, and 1.6 points on the Hauser Ambulation Index. All 4 changes were statistically significant. Four patients (15.4%) achieved a Distribution of Stiffness Index score of 0, indicating no stiffness in the regions assessed by that scale.
Exploratory assessments showed a mean increase of 125 meters in 6-minute walking distance and improvements across all 8 domains of the 36-Item Short Form Health Survey. Overall, 25 of 26 patients (96.2%) improved on at least 1 primary or secondary efficacy measure.
The authors also observed changes in immune biomarkers following treatment. Peripheral B-cell depletion was complete by day 5. Among 20 evaluable patients with elevated glutamic acid decarboxylase 65 antibody levels, median antibody titers declined by 56% at week 16. By that time, 14 patients (53.8%) had experienced B-cell reconstitution. Changes in B-cell subsets and regulatory T cells were suggestive of an immune reset, although its long-term durability remains uncertain.
Treatment-related adverse events occurred in all 26 patients. Cytokine release syndrome developed in 24 patients (92.3%), with all cases classified as grade 1 or 2. Management of cytokine release syndrome included tocilizumab and corticosteroids in 21 patients each. Immune effector cell–associated neurotoxicity syndrome occurred in 3 patients (11.5%), with all cases classified as grade 1. Both complications resolved without sequelae.
Grade 3 or 4 neutropenia occurred in 5 patients (19.2%), and 11 patients (42.3%) experienced at least 1 grade 3 or higher treatment-related adverse event. No serious infections associated with neutropenia were reported during follow-up. Three patients (11.5%) experienced treatment-related serious adverse events, including a grade 2 peritonsillar abscess and grade 4 localized neck edema that required elective intubation and resolved within 9 days. No deaths or treatment discontinuations resulting from adverse events were reported.
The trial was small, nonrandomized, and lacked a control group, limiting the ability to distinguish treatment effects from other influences. Follow-up was relatively short, and only 16 patients had reached the 24-week assessment. The findings may not generalize to patients with less severe or antibody-negative SPS.
These limitations preclude conclusions about comparative efficacy or the durability of the observed benefits, but the authors noted that long-term follow-up is in progress. They also highlighted a practical consideration for delivering CAR T-cell therapy in this population.
"A future consideration will therefore include access to treating physicians with experience managing CAR T-cell therapy and potential associated events," wrote lead study author Amanda L. Piquet, MD, of the University of Colorado Anschutz, Aurora, and colleagues.
Kyverna Therapeutics funded the study and participated in its design, conduct, analysis, and manuscript preparation, but not direct data collection. Several authors reported employment, consulting fees, research support, stock ownership, or patent interests involving Kyverna. Additional industry relationships were disclosed.
Source: JAMA Neurology
