Mesenchymal Stromal Cell Implants for Chronic Motor Deficits After Traumatic Brain Injury: Post Hoc Analysis of a Randomized Trial

Abstract

Background and objectives: Traumatic brain injury (TBI) is frequently characterized by chronic motor deficits. Therefore, this clinical trial assessed whether intracranial implantation of allogeneic modified mesenchymal stromal (SB623) cells can improve chronic motor deficits after TBI.

Methods: Post hoc analysis of the double-blind, randomized, prospective, surgical sham-controlled, phase 2, STEMTRA clinical trial (June 2016 and March 2019) with 48 weeks of follow-up was conducted. In this international, multicenter clinical trial, eligible participants had moderate-to-severe TBI, were ≥12 months postinjury, and had chronic motor deficits. Participants were randomized in a 1:1:1:1 ratio to stereotactic surgical intracranial implantation of SB623 cells (2.5 × 10⁶, 5.0 × 10⁶, 10 × 10⁶) or surgical sham-controlled procedure. The prespecified primary efficacy end point was significantly greater change from baseline of the Fugl-Meyer Motor Scale (FMMS) score, a measure of motor status, for the SB623 pooled vs control arm at 24 weeks.

Results: A total of 211 participants were screened, 148 were excluded, and 63 underwent randomization, of which 61 (97%; mean age, 34 [SD, 12] years; 43 men [70.5%]) completed the trial. Single participants in the SB623 2.5 × 10⁶ and 5.0 × 10⁶ cell dose groups discontinued before surgery. Safety and efficacy (modified intent-to-treat) were assessed in participants who underwent surgery (N = 61; SB623 = 46, controls = 15). The primary efficacy end point (FMMS) was achieved (least squares mean [SE] SB623: +8.3 [1.4]; 95% CI 5.5-11.2 vs control: +2.3 [2.5]; 95% CI -2.7 to 7.3; p = 0.04), with faster improvement of the FMMS score in SB623-treated groups than in controls at 24 weeks and sustained improvement at 48 weeks. At 48 weeks, improvement of function and activities of daily living (ADL) was greater, but not significantly different in SB623-treated groups vs controls. The incidence of adverse events was equivalent in SB623-treated groups and controls. There were no deaths or withdrawals due to adverse events.

Discussion: Intraparenchymal implantation of SB623 cells was safe and significantly improved motor status at 24 weeks in participants with chronic motor deficits after TBI, with continued improvement of function and ADL at 48 weeks. Cell therapy can modify chronic neurologic deficits after TBI.

Trial registration information: ClinicalTrials.gov Identifier: NCT02416492. Submitted to registry: April 15, 2015. First participant enrolled: July 6, 2016. Available at: classic.clinicaltrials.gov/ct2/show/NCT02416492.

Classification of evidence: This study provides Class I evidence that intracranial implantation of allogeneic stem (SB623) cells in adults with motor deficits from chronic TBI improves motor function at 24 weeks.

Figure 1. CONSORT Flow Diagram

63 participants were randomized to SB623 cell treatment or sham surgery. However, safe stereotactic implantation trajectories could not be determined for 1 participant in the SB623 2.5 × 10⁶ cell dose group and 1 participant in the SB623 5.0 × 10⁶ cell dose group, resulting in both participants discontinuing from the trial before surgery. Both the modified intent-to-treat (mITT) and safety populations (n = 61) contained participants who were randomized and underwent SB623 cell treatment or sham surgery. All 61 participants completed 48-week evaluations.

Figure 2. SB623 Efficacy End Point Measures

(A) FMMS mean change from baseline for (1) the SB623 pooled arm at 24 weeks (●), (2) SB623 5 × 10⁶ cell dose at 24 weeks (▲), and (3) SB623 5 × 10⁶ cell dose at 48 weeks (▼). FMMS baseline mean (SD) scores were 52.2 (19.3) for SB623 pooled, 51.3 (22.0) for the 5 × 10⁶ cell dose, and 52.3 (15.1) for sham surgery control. The graphs show data from the modified intent-to-treat population, which included 61 participants who underwent surgery. (B) DRS baseline mean (SD) scores were 4.8 (3.0) for SB623 pooled and 3.7 (2.0) for sham surgery control. The graph shows data from the modified intent-to-treat population, which included 61 participants who underwent surgery. (C) ARAT baseline mean (SD) scores were 19.1 (19.5) for SB623 pooled and 20.1 (17.2) for sham surgery control. The graph shows data from the modified intent-to-treat population, which included 61 participants who underwent surgery. (D) NeuroQOL upper baseline mean (SD) scores were 32.5 (12.9) for SB623 pooled and 32.2 (9.2) for sham surgery control. The graph shows data from the modified intent-to-treat population, which included 61 participants who underwent surgery. (E) Gate velocity baseline mean (SD) scores were 0.67 (0.49) m/s for SB623 pooled and 0.81 (0.58) m/s for sham surgery control. The graph shows data from the lower extremity deficit population (N = 56). (F) NeuroQOL lower baseline mean (SD) scores were 41.5 (10.4) for SB623 pooled and 44.3 (9.6) for sham surgery control. The graph shows data from the modified intent-to-treat population, which included 61 participants who underwent surgery. ARAT = Action Research Arm Test; DRS = Disability Rating Scale; FMMS = Fugl-Meyer Motor Scale.