Lost in translation: why pharmacological trials keep failing in acute spinal cord injury

Abstract

Despite four decades of translational research, no pharmacological agent has been approved for acute traumatic spinal cord injury (SCI) based on class-I evidence. The current narrative review critically examines the seven agents evaluated in phase-II to -III clinical trials—methylprednisolone (MPSS), GM-1 monosialotetrahexosylganglioside, minocycline, riluzole, anti-Nogo-A antibody NG101, VX-210, and granulocyte colony-stimulating factor (G-CSF)—with emphasis on quantitative effect estimates, confidence intervals, and mechanisms of translational failure. Polylaminin, under phase-I evaluation approved by the Brazilian Health Regulatory Agency (Agência Nacional de Vigilância Sanitária, ANVISA, in Portuguese), is discussed in the context of disproportionate public communication relative to the stage of evidence; available human data are limited to a small, uncontrolled, unreviewed pilot study insufficient for efficacy inferences. High-dose MPSS conferred no motor recovery benefit in a meta-analysis of 1,863 participants and was significantly associated with gastrointestinal hemorrhage (odds ratio [OR] = 2.07; 95%CI: 1.02–4.20) and respiratory infections (OR = 1.73; 95%CI: 1.12–2.68). Furthermore, GM-1 ganglioside, G-CSF, and VX-210 failed to meet the prespecified primary endpoints. Minocycline produced inconclusive results in an underpowered phase-II study. Riluzole (Riluzole in Spinal Cord Injury Study [RISCIS]), which was prematurely terminated at 55% of enrollment due to the coronavirus disease 2019 (COVID-19) pandemic, showed a significant motor benefit in patients classified as grade C according to the American Spinal Injury Association (ASIA) Impairment Scale (AIS) (+8.0 points; 95%CI: 1.5–14.4; 2-tailed p = 0.034). The Nogo Inhibition in Spinal Cord Injury (NISCI) trial (2025), the most methodologically rigorous SCI trial to date, demonstrated no primary endpoint benefit but identified functional signals in motor-incomplete injuries. In total, six structural domains of translational failure are identified and methodological solutions proposed.