Potential Pharmacologic Treatments in Spinal Cord Injury: A Narrative Review

Abstract

Spinal cord injury (SCI) remains one of the most devastating neurological conditions with poor prognosis and limited therapeutic options. Early surgical decompression has been established as the cornerstone of acute management, yet multiple pharmacologic strategies have been extensively investigated to improve neurological outcomes. High-dose methylprednisolone sodium succinate, once widely adopted, ultimately failed to demonstrate consistent efficacy, underscoring the challenges of translating preclinical promise into clinical benefit. Advances in understanding the sequential pathophysiology of SCI, including excitotoxicity, oxidative stress, inflammation, apoptosis, glial scarring, and chronic inhibitory signaling, have broadened therapeutic targets across temporal phases of injury. In this narrative review, we revisit historical pharmacologic therapies such as corticosteroids, and highlight current and emerging candidates including riluzole, minocycline, granulocyte colony-stimulating factor, and biologics targeting the Rho/ROCK pathway (Cethrin, anti-Nogo-A, and anti-repulsive guidance molecule A antibodies). Although compelling preclinical evidence exists, most clinical trials have been limited by patient heterogeneity, narrow therapeutic time windows, and outcome assessment challenges, preventing any drug from achieving guideline-level endorsement. Future success will likely depend on tailored multimodal regimens according to patient stratification, optimized therapeutic timing and delivery, and integration with surgical decompression, hemodynamic management, rehabilitative interventions, and novel bioengineered strategies. Continued efforts in these directions remain essential to establish pharmacologic therapy as a critical component of multimodal SCI care.

GRAPHICAL ABSTRACT

INTRODUCTION

Spinal cord injury (SCI) is one of the most devastating conditions in neurotrauma, leading to permanent neurological deficits, lifelong disability, and substantial socioeconomic burden worldwide.¹²,⁴⁵,⁴⁶⁾ Epidemiologic studies estimate a global incidence of 10–80 cases per million annually, and this rate has not declined in recent decades. A significant proportion of patients are young adults in their most productive years, and those with complete (American Spinal Injury Association [ASIA] Impairment Scale [AIS] A) or severe incomplete (AIS B) injuries often experience limited neurological recovery, resulting in lifelong dependency and considerable healthcare and social costs.⁴⁶,⁶⁴⁾ These factors underscore the continuing need for therapeutic strategies capable of modifying the course of injury and improving long-term outcomes.

While early surgical decompression and hemodynamic optimization remain the standard of care,⁴,²⁶⁾ neurological recovery is often limited. Recent evidence underscores that surgical decompression within 24 hours is associated with superior motor recovery, and that ultra-early decompression (within 8 hours) may further enhance neurological outcomes.⁴,²⁴⁾ This time-sensitive principle, often captured by the phrase “time is spine,” highlights the importance of rapid diagnosis, efficient transfer, and coordinated surgical decision-making to minimize secondary ischemic damage and optimize functional recovery.⁵³⁾ However, logistical barriers such as delayed transport, diagnostic evaluation, and surgical readiness still limit the routine application of ultra-early surgery in many clinical settings. Hemodynamic optimization plays a complementary role in acute SCI management. Maintaining a mean arterial pressure (MAP) between 85 and 90 mmHg during the first post-injury week is recommended to ensure adequate spinal cord perfusion. In addition, spinal cord perfusion pressure (SCPP, MAP − intrathecal pressure) has been proposed as a physiologic target, and maintaining SCPP above approximately 60 mmHg may correlate with improved neurological recovery.⁴⁷,⁵⁸,⁵⁹⁾ Despite these established interventions, no pharmacologic therapy has yet demonstrated robust efficacy in altering the natural course of SCI.

The late 20th century marked the first major pharmacologic efforts. In 1990, the landmark National Acute Spinal Cord Injury Study (NASCIS) II trial reported that high-dose methylprednisolone sodium succinate (MPSS), if given within 8 hours, produced modest motor score improvements.⁶⁾ In 1997, NASCIS III extended this approach, but also highlighted increased risks of complications.⁷⁾ Based on these trials, MPSS became standard in the 1990s. However, subsequent meta-analyses showed no sustained benefit and consistently noted adverse events such as infections and gastrointestinal bleeding.²²⁾ By the 2010s, many centers had abandoned routine steroid therapy, and international guidelines diverged, leaving the steroid controversy as a cautionary lesson in balancing efficacy and risk.²⁸⁾ Other contemporaneous attempts, such as the Sygen trial of GM-1 ganglioside³²⁾ and trials of naloxone,⁶⁾ similarly failed to demonstrate meaningful functional recovery.

In recent decades, advances in experimental neurobiology have clarified SCI as a dynamic, multiphase process involving secondary injury cascades such as excitotoxicity, oxidative stress, inflammation, apoptosis, and chronic inhibitory signaling. This understanding has given rise to the concept of phase-specific therapy, in which pharmacologic interventions are temporally aligned with distinct stages of injury progression—acute, subacute, intermediate, and chronic. To frame potential pharmacologic approaches, it is essential to understand the pathophysiology and sequential injury phases of SCI (TABLE 1).²⁾ This review aims to delineate pharmacologic strategies according to these temporal phases (TABLE 2), summarize major clinical trials of pharmacologic agents for SCI (TABLE 3), critically appraise current clinical evidence, and outline future research directions toward integrated phase-specific and multimodal therapy.

TABLE 1. Pathophysiology and sequential cascade of SCI.

Phase	Time frame	Key pathophysiology	Representative mechanisms	Therapeutic implications
Primary injury	Immediate (seconds–minutes)	Mechanical disruption of neural and vascular elements	Direct contusion, compression, laceration, hemorrhage	Prevent further mechanical insult (immobilization, surgical decompression)
Acute phase	0–48 hours	Secondary injury cascade begins	Excitotoxicity (glutamate), ionic imbalance, Ca2+ overload, mitochondrial dysfunction, oxidative stress, microvascular ischemia, early inflammation (cytokine surge)	Neuroprotection (e.g., MPSS, riluzole, minocycline)
Subacute phase	2–14 days	Cell death, demyelination, immune activation	Microglial/macrophage infiltration, oligodendrocyte apoptosis, persistent ischemia, demyelination, initiation of glial scar formation	Anti-inflammatory and cytoprotective agents (e.g., G-CSF, EPO)
Intermediate phase	2 weeks–6 months	Tissue remodeling, inhibitory barrier formation	Cystic cavitation, astroglial scar maturation, extracellular matrix accumulation, myelin-associated inhibitors (Nogo-A, MAG, OMgp → Rho/ROCK)	Neuroregeneration-priming therapies (e.g., Cethrin, anti-Nogo-A, anti-RGMa antibodies)
Chronic phase	>6 months	Stabilization, maladaptive plasticity, insufficient repair	Persistent inhibitory signaling, incomplete remyelination, maladaptive synaptic plasticity, neuropathic pain	Combination strategies (cells, scaffolds, neuromodulation, biomimetic delivery systems)

SCI: spinal cord injury, MPSS: methylprednisolone sodium succinate, G-CSF: granulocyte colony-stimulating factor, EPO: erythropoietin, MAG: myelin-associated glycoprotein, OMgp: oligodendrocyte-myelin glycoprotein, RGMa: repulsive guidance molecule A.

TABLE 2. Phase-specific potential pharmacologic agents in SCI.

Phase	Pharmacologic agents	Mechanism of action	Representative clinical trial	Current status/summary
Acute (≤48 hours)	MPSS	Synthetic glucocorticoid; inhibits lipid peroxidation and secondary inflammation after SCI by suppressing NF-κB activation and apoptosis pathways.	NASCIS II (1990), NASCIS III (1997)	Historical standard; benefit limited to ≤8 hours window; no longer routine due to infection and GI complication risk.
	Riluzole	Sodium-channel blocker reducing glutamate release and excitotoxic neuronal injury.	NACTN (phase I), RISCIS (phase III)	Investigational; safe profile; modest benefit signal in AIS C subgroup; dose optimization under study.
	Minocycline	Tetracycline derivative with anti-inflammatory and anti-apoptotic effects; microglial and caspase inhibition.	MASC (phase III)	Investigational; promising motor recovery trend in cervical SCI but no confirmed efficacy.
Subacute (2–14 days)	G-CSF	Hematopoietic growth factor stimulating neutrophil proliferation and neurotrophic signaling; reduces neuronal apoptosis.	G-SPIRIT (phase III)	Investigational; safe but underpowered; signals in AIS C and elderly subgroups.
	Erythropoietin	Glycoprotein hormone with anti-apoptotic, angiogenic, and anti-inflammatory neuroprotection via JAK2/STAT5 pathway.	Preclinical	Preclinical/exploratory; translational potential but human evidence limited.
Intermediate (2 weeks–6 months)	Cethrin (VX-210)	Recombinant RhoA antagonist inhibiting Rho/ROCK pathway → prevents growth cone collapse and enhances axonal regeneration.	SPRING (phase IIb/III)	Development discontinued after futility analysis; landmark in intracellular pathway targeting.
	Anti-Nogo-A mAb (NG-101)	Neutralizes Nogo-A, a myelin-associated inhibitor; enhances axonal sprouting & plasticity.	NISCI (phase IIb)	Safe but failed primary endpoint; possible benefit in motor-incomplete (AIS C) patients.
	Anti-RGMa mAb (Elezanumab)	Human monoclonal antibody blocking RGMa; facilitates CST sprouting and neuroprotection.	ELASCI (phase II)	Ongoing international trial; results expected 2026. Supported by preclinical efficacy and early-phase safety data suggesting neuroprotective and regenerative potential of RGMa inhibition.
Chronic (>6 months)	G-CSF	Hematopoietic growth factor that may enhance neuroplasticity and functional recovery by promoting neurogenesis, angiogenesis, and remyelination.	IRCT201108297441N1 (phase III)	Demonstrated significant improvement in ASIA motor and functional scores in chronic incomplete SCI; however, population heterogeneity and uncertain translation to acute injury limit generalization.
	Symptom-modulating agents (e.g., baclofen, tizanidine, gabapentinoids, botulinum toxin, or riluzole for spasticity control)*	N/A	RILUSCI (phase Ib/IIb)	Management in the chronic phase primarily relies on symptom-modulating agents combined with cell-based, scaffold, and neuromodulation approaches.

SCI: spinal cord injury, MPSS: methylprednisolone sodium succinate, NF: nuclear factor, NASCIS: The National Acute Spinal Cord Injury Study, GI: gastrointestinal, NACTN: The North American Clinical Trials Network, RISCIS: Riluzole in Spinal Cord Injury Study, AIS: American Spinal Injury Association Impairment Scale, MASC: Minocycline in Acute Spinal Cord Injury, G-CSF: granulocyte colony-stimulating factor, G-SPIRIT: G-CSF mediated SPinal cord Injury Recovery Induction Trial, SPRING: SPinal Cord Injury Rho INhibition InvestiGation, mAb: monoclonal antibody, NISCI: Nogo Inhibition in Spinal Cord Injury, RGMa: Repulsive guidance molecule A, CST: corticospinal tract, ELASCI: ELezanumab in Acute Spinal Cord Injury, ASIA: American Spinal Injury Association, N/A: nonavailable, RILUSCI: Riluzole for Spasticity in Chronic SCI.

^*This review primarily focuses on neuroprotective or regenerative pharmacologic therapy.

TABLE 3. Summary of major clinical trials of potential pharmacologic agents in SCI.

Agent		Clinical trial	Year*	Trial number	Phase	Study design	Protocol summary		Primary outcome	Key results	Limitations/current status
MPSS		NASCIS II	1990 6)	N/A	Phase II	Multicenter, randomized, double-blind, placebo-controlled trial	Acute SCI; 95 % treated ≤14 hours		Change in motor and sensory scores at 6 months	MPSS ≤8 hours showed significant improvement (motor gain 16.0 vs. 11.2, p=0.03; pinprick 11.4 vs. 6.6, p=0.02; touch 8.9 vs. 4.3, p=0.03). No benefit >8 hours or for Naloxone. Mortality similar among groups.	Benefit limited to ≤8 hours subgroup; ↑ infection and GI bleeding risk. Provided basis for NASCIS III but routine MPSS use later declined.
							Three arms (total n=487):
								(1) MPSS 30 mg/kg IV bolus + 5.4 mg/kg/hr × 23 hours (n=162);
								(2) Naloxone 5.4 mg/kg bolus + 4.0 mg/kg/hr × 23 hours (n=154);
								(3) Placebo (n=171)
		NASCIS III	1997 7)	N/A	Phase III	Multicenter, randomized, double-blind, controlled trial	Acute SCI; treated ≤8 hours		Change in motor function and FIM from baseline to 6 months	In subgroup treated 3–8 hours, 48 hours MPSS had better motor recovery at 6 months vs. 24 hours (p=0.01); more likely to gain one neurologic grade (p=0.03) and FIM (p=0.08) at 6 months. Tirilazad no benefit vs. 24 hours MPSS.	Benefit limited to 3–8 hours subgroup. Increased risk of sepsis/pneumonia in 48 hours arm. Generalizability questioned. Tirilazad showed no benefit.
							All participants got MPSS bolus 30 mg/kg initially.
							Then randomized into 3 arms (total n=499):
								(1) MPSS infusion 5.4 mg/kg/hr × 24 hours (n=166);
								(2) MPSS infusion 5.4 mg/kg/hr × 48 hours (n=167);
								(3) Tirilazad mesylate 2.5 mg/kg every 6 hours for 48 hours (n=166)
Riluzole		Safety of riluzole in acute SCI (NACTN phase I)	2014 32)	NCT00876889	Phase I	Prospective, multicenter, matched-comparison group trial	Acute SCI (AIS A–C; 28 cervical, 8 thoracic) treated ≤12 hours		Safety, PK, exploratory motor recovery	No SAEs. Mild-moderate liver enzyme elevations. Cervical injury subgroup (n=24) showed greater motor gain at 90 days (31.2 vs. 15.7, p=0.021).	Small sample, non-randomized, comparison via matched registry rather than concurrent control, preliminary efficacy only. Formed the basis for RISCIS phase III trial.
								(1) All received riluzole 50 mg PO/NG BID × 14 days (n=36);
								(2) Matched comparison group from NACTN registry (n=36).
		RISCIS	2023 13,22)	NCT01597518	Phase III	International, multicenter, adaptive, randomized, double-blind, placebo-controlled trial	Acute cervical SCI (AIS A–C, C4–C8); treated ≤12 hours		Change in UEMS at 180 days	No significant difference overall (UEMS gain 1.76; 95% CI, −2.54 to 6.06; p=0.25). In the AIS C subgroup, mean motor gain was +8.0 (95% CI, 1.5 to 14.4; p<0.05). No drug-related SAEs.	Underpowered (193/351 enrolled); adaptive futility stopped early due to COVID-19; subgroup findings exploratory. A PK/PD substudy 13) identified time-dependent PK and linked riluzole exposure to improved motor recovery and lower pNF-H levels, suggesting potential for dose optimization.
							Two arms (total n=193):
								(1) Riluzole 100 mg PO BID × 24 hours, then 50 mg BID × 13 days (n=96);
								(2) Placebo (n=97)
		RILUSCI	2023† (completed)	NCT02859792	Phase Ib/IIb	Multicenter (France), adaptive, randomized, double-blind, placebo-controlled trial	Adults (18–65 years) with chronic (≥12 months) traumatic SCI (C4–T12, AIS A–D) with spasticity (MAS >1, NRS ≥4)		Improvement in MAS or NRS for spasticity	N/A (no results published). Safety, PK, and secondary effects on pain, bladder function, QoL under evaluation.	Study completed; results pending publication. Focused on spasticity control rather than neuroregeneration—first riluzole trial repurposing riluzole for symptom-level modulation.
							Two arms (planned total n=90, 1:1):
								(1) Riluzole PO daily × 2 weeks — dose via phase 1b Bayesian reassessment (25/50/75/100 mg BID);
								(2) Placebo (25 or 50 mg capsules)
Minocycline		Minocycline and perfusion pressure augmentation in acute spinal cord injury	2012 8)	NCT00559494	Phase II	Single-center, randomized, double-blind, placebo-controlled trial	Acute traumatic SCI (AIS A–C); treated ≤12 hours		Safety and dose tolerability; exploratory motor recovery at 1 year	Treatment was well tolerated with no SAEs. Transient liver enzyme elevation occurred in one patient. Mean motor score gain was +6 overall (95% CI, −3 to 14; p=0.20) and ≈ +14 in the cervical subgroup (95% CI, 0 to 28; p=0.05).	Small sample; single center; underpowered for efficacy. Formed basis for MASC Phase III trial.
							Two arms (total n=52):
								(1) Minocycline IV loading dose 800 mg, then 400 mg BID × 7 days (n=27);
								(2) Placebo (n=25)
		MASC	2018† (ongoing)	NCT01828203	Phase III	Multicenter, randomized, double-blind, placebo-controlled trial	Acute non-penetrating cervical SCI; treated ≤12 hours		Improvement in ASIA motor score between 3 months and 1 year post-injury	N/A (no results posted)	No confirmatory phase III results published—evidence remains preliminary and inconsistent; preclinical and meta-analysis data suggest potential efficacy but require further high-quality trials.
							Two arms (n target=248):
								(1) Minocycline IV BID (800 mg tapering to 400 mg) × 7 days;
								(2) Placebo
G-CSF		Neuroprotective therapy using G-CSF	2012 66)	N/A	Phase I/IIa	Open-label, single-arm, dose-escalation trial	Acute SCI; treated ≤48 hours		Safety, feasibility; change in ASIA motor score and AIS grade	No SAEs; 9/16 improved ≥ 1 AIS grade; motor and sensory gains greater in 10 µg/kg group (p<0.05).	Small, uncontrolled, open-label; early-phase exploratory data only.
							Two dose cohorts:
								(1) 5 µg/kg/day × 5 days (n=5);
								(2) 10 µg/kg/day × 5 days (n=11)
		G-SPIRIT	2021 40)	UMIN000018752	Phase III	Multicenter (Japan), randomized, double-blind, placebo-controlled trial	Acute cervical SCI (AIS B/C); treated ≤48 hours		Change in ASIA motor score at 3 months	No significant difference (p=0.40 FAS; p=0.22 PPS); trend favoring G-CSF at 6 months (p=0.062) and 12 months (p=0.073). Older (>65 years) subgroup trend (p=0.056). No G-CSF-related SAEs.	Underpowered; primary endpoint not met; signals in AIS C and elderly subgroups need confirmation.
							Two arms (total n=88):
								(1) G-CSF 400 µg/m2/day × 5 days (n=44);
								(2) Placebo (n=44)
		G-CSF for chronic incomplete SCI	2018 16)	IRCT201108297441N1	Phase III	Randomized, double-blind, placebo-controlled, parallel trial	Chronic incomplete SCI (≥6 months, AIS B–D)		Change in ASIA motor score at 6 months	G-CSF group +5.5 vs. +0.8 in placebo (p<0.001); SCIM and IANR-SCIFRS also significantly higher (p<0.05).	Conducted in chronic SCI; heterogeneous population; effect in acute injury unclear.
							Two arms (total n=120):
								(1) G-CSF 300 µg/day × 7 days (n=60);
								(2) Placebo (n=60)
Cethrin (VX-210)		Recombinant Rho protein antagonist in acute SCI	2011 25)	N/A	Phase I/IIa	Open-label, dose-escalation study	Acute cervical or thoracic SCI		Safety, feasibility	Acceptable safety profile and trend toward greater neurological improvement in cervical AIS A patients	No control group; small sample; preliminary signals only; formed basis for SPRING trial
							Extradural application during decompression surgery (n=48 across dose cohorts)
		SPRING	2021 21)	NCT02669849	Phase IIb/III	Multicenter, randomized, double-blind, placebo-controlled trial	Acute cervical SCI (AIS A–C); treated ≤72 hours		Change in UEMS at 6 months	No significant difference in UEMS gain between VX-210 and placebo at 6 months (least square mean difference, 1.0; 95% CI, −2.5 to 4.5; p>0.05). No drug-related SAEs.	Trial stopped early after futility interim analysis; underpowered; no confirmatory benefit demonstrated; further optimization of dose and timing required.
							Two arms (total n=72):
								(1) VX-210, applied extradurally in fibrin sealant during decompression surgery (n=36);
								(2) Placebo (n=36)
Anti-Nogo-A antibody
	(ATI355)	Intrathecal ATI355 safety/PK in acute SCI	2018 43)	NCT00406016	Phase I	International, multicenter, open-label, single-arm, dose-escalation study	Acute cervical/thoracic SCI;		Safety, tolerability, PK	Intrathecal delivery was generally well tolerated; no drug-related SAEs. Some exploratory neurologic improvements observed.	First-in-human feasibility; no control arm; efficacy exploratory; informed design of later NISCI trial
							Intrathecal ATI355 over multiple dose cohorts (n=52)
	(NG-101)	NISCI	2025 71)	NCT03935321	Phase IIb	International, multicenter, randomized, double-blind, placebo-controlled trial	Acute cervical SCI (AIS A–D); treated 4–28 days		Change in UEMS at 6 months	No significant difference in UEMS gain (mean difference, 1.37, 95% CI, −1.44 to 4.18); AEs and SAEs comparable between groups. Infections most common and no treatment-related fatalities.	Did not meet primary endpoint. Post-hoc subgroup signals of benefit in motor-incomplete injuries. Further optimization needed; follow-up and biomarker-guided analyses pending.
							Two arms (initial 1:1, adjusted to 3:1, final 2:1, total n=126):
								(1) NG-101 intrathecal 45 mg bolus × 6 injections every 5 days (n=78);
								(2) Placebo (n=48)
Elezanumab (anti-RGMa mAb)		ELASCI	2026† (ongoing)	NCT04295538	Phase II	International, multicenter, randomized, double-blind (quadruple masking), placebo-controlled trial	Acute traumatic cervical SCI (C4–C7, AIS A or B); treated ≤24 hours		Change in UEMS at 52 weeks	N/A (no results posted). Safety monitoring ongoing; trial active but not recruiting.	Ongoing trial; no published efficacy data yet. Distinct for its long-term repeated dosing; results expected in 2026.
							Two arms (≈ 2:1 ratio, target total n=54)
								(1) Elezanumab (ABT-555) IV every 4 weeks × 13 doses (0–48 weeks);
								(2) Placebo

SCI: spinal cord injury, MPSS: methylprednisolone sodium succinate, NASCIS: The National Acute Spinal Cord Injury Study, N/A: nonavailable, IV: intravenous, FIM: functional independence measure, NACTN: The North American Clinical Trials Network, PK: pharmacokinetics, AIS: American Spinal Injury Association Impairment Scale, PO: by mouth, NG: nasogastric, BID twice daily, SAEs: serious adverse events, RISCIS: Riluzole in Spinal Cord Injury Study, UEMS: upper-extremity motor score, CI: confidence interval, COVID-19: coronavirus disease 2019, PD: pharmacodynamics, RILUSCI: Riluzole for Spasticity in Chronic SCI, MAS: Modified Ashworth Scale, NRS: numerical rating score, QoL: quality of life, MASC: Minocycline in Acute Spinal Cord Injury, ASIA: American Spinal Injury Association, G-CSF: granulocyte colony-stimulating factor, G-SPIRIT: G-CSF mediated SPinal cord Injury Recovery Induction Trial, FAS: full analysis set, PPS: per-protocol set, SCIM: Spinal Cord Independence Measure, IANR-SCIFRS: International Association of Neurorestoratology Spinal Cord Injury Functional Rating Scale, SPRING: SPinal Cord Injury Rho INhibition InvestiGation, NISCI: Nogo Inhibition in Spinal Cord Injury, AEs: adverse events, RGMa: Repulsive guidance molecule A, mAb: monoclonal antibody, ELASCI: ELezanumab in Acute Spinal Cord Injury.

^*Year indicates the publication year of the trial results.

^†Trial is ongoing or completed with no publication for the results yet. Year indicates the expected completion or completion year of the study.

PATHOPHYSIOLOGY AND SEQUENTIAL PHASES OF SCI

SCI begins with a primary mechanical insult that directly disrupts neural and vascular structures, followed by a secondary cascade that evolves through distinct temporal phases. In the acute phase (<48 hours), excitotoxic glutamate release, ionic imbalance, calcium overload, mitochondrial dysfunction, oxidative stress, microvascular ischemia, and cytokine surges drive early neuronal and glial death.⁵,³⁴,⁵⁶,⁵⁷⁾ The subacute phase (2–14 days) is marked by sustained inflammation, apoptosis of neurons and oligodendrocytes, demyelination, and initiation of glial scar formation.¹⁵,²⁷,⁶³⁾ In the intermediate phase (2 weeks–6 months), cystic cavitation and a dense astroglial scar mature, reinforced by extracellular matrix and myelin-associated inhibitors such as Nogo-A, myelin-associated glycoprotein (MAG), and oligodendrocyte-myelin glycoprotein (OMgp) that converge on the Rho/ROCK pathway to block axonal regeneration.¹⁹,⁶⁰⁾ In the chronic phase (>6 months), maladaptive plasticity, incomplete remyelination, and persistent inhibitory signaling perpetuate neurological dysfunction, while endogenous repair attempts (e.g., oligodendrocyte precursors, neural stem cells) remain insufficient for meaningful recovery.⁷²⁾ These mechanisms and phases not only explain the progressive nature of SCI but also define therapeutic windows. Neuroprotective approaches are most relevant in the acute to subacute phases, whereas neuroregenerative strategies target barriers in the intermediate to chronic stages.¹,³⁹⁾ Accordingly, this review is organized around these phase-specific therapeutic windows (TABLE 2).

PHASE-SPECIFIC PHARMACOLOGIC THERAPIES

Acute phase (0–48 hours): Neuroprotection

MPSS

Historically, corticosteroids were the first widely tested pharmacologic agents in acute SCI. The NASCIS II and III trials suggested modest neurological benefit of high-dose MPSS when administered within 8 hours of injury, but subsequent analyses revealed limited efficacy and significant complications, including infections and gastrointestinal bleeding.⁶,⁷,²²⁾ Consequently, its routine use has been largely abandoned, although MPSS remains an important historical milestone that established the modern framework for SCI pharmacotherapy.²⁸⁾

Riluzole

Riluzole is a benzothiazole derivative originally approved for amyotrophic lateral sclerosis. It acts as a sodium channel blocker and inhibits pathologic glutamate release, thereby attenuating excitotoxicity—a key driver of early neuronal and oligodendrocyte death in acute SCI. Preclinical studies demonstrated reduced lesion expansion, enhanced neuronal survival, and improved functional outcomes, providing a strong rationale for clinical translation.⁵⁵⁾

The phase I trial (NCT00876889) conducted by the North American Clinical Trials Network (NACTN), established the safety, pharmacokinetics, and feasibility of riluzole administration within 12 hours of acute SCI.²⁷,³³⁾ However, the following multicenter Riluzole in Spinal Cord Injury Study (RISCIS) phase III trial (NCT01597518) was terminated early in 2021 due to the coronavirus disease 2019 pandemic and enrollment challenges rather than drug-related adverse events.²³⁾ As a result, only 193 patients (55% of the planned 351) were randomized, limiting statistical power. The trial did not meet its primary endpoint, but exploratory subgroup analyses suggested possible neurological benefit, particularly in AIS C patients. In addition, a pharmacokinetic substudy (NCT01597518, ancillary arm) demonstrated variability in drug exposure and hinted at a relationship between systemic exposure and motor recovery.¹³⁾ These findings reinforce the importance of optimized dosing strategies.

Although no large-scale confirmatory trial is currently active, riluzole remains a prototype of mechanism-driven neuroprotective pharmacotherapy in SCI, representing one of the few agents that have advanced from strong pathophysiologic rationale through multicenter randomized testing. More recently, the Riluzole in the Treatment of Spasticity in the Traumatic Chronic Spinal Cord Injury Condition (RILUSCI) trial (NCT02859792) has repurposed riluzole for chronic SCI-related spasticity, with an adaptive, placebo-controlled design to assess its effects on spasticity, pain, and bladder function.¹⁴⁾ The study was completed in 2023, but its results have not yet been published.

Minocycline

Minocycline is a semisynthetic tetracycline derivative with pleiotropic effects beyond its antimicrobial activity. It penetrates the central nervous system effectively and exerts anti-inflammatory, anti-apoptotic, and anti-excitotoxic actions. In experimental SCI models, minocycline reduced microglial activation, attenuated apoptosis of oligodendrocytes, and improved functional recovery.³⁴,⁶⁵⁾

Based on these preclinical findings, the Canadian single center phase II randomized trial (NCT00559494) evaluated intravenous minocycline administered within 12 hours of acute SCI.⁸⁾ The study demonstrated safety and reported improved motor recovery at one year compared with placebo, particularly in cervical injuries. Building on those results, a phase III multicenter trial—Minocycline in Acute Spinal Cord Injury (MASC, NCT01828203)—was initiated, aiming to enroll up to 248 patients to test IV minocycline versus placebo within 12 hours of cervical SCI. However, the status of this trial is listed as unknown, and no results have been posted publicly yet. Despite encouraging preclinical and early clinical signals, definitive proof of clinical efficacy remains lacking. Nevertheless, minocycline continues to represent one of the most extensively studied and biologically plausible repurposed neuroprotective agents in SCI.

Other pharmacologic candidates

Naloxone, an opioid receptor antagonist thought to modulate lipid peroxidation and free radical formation, was included in the NASCIS II trial alongside MPSS, but did not improve neurological outcomes compared with placebo.⁶⁾ Tirilazad mesylate, a non-glucocorticoid 21-aminosteroid and free radical scavenger, was evaluated in the NASCIS III trial, but similarly failed to demonstrate significant benefit while adding complexity to treatment protocols.⁷⁾

Taken together, these findings delineate the current landscape of acute-phase pharmacotherapy in SCI, summarizing both its achievements and limitations. Corticosteroids remain a controversial milestone, while riluzole and minocycline represent the most advanced mechanism-driven candidates. In contrast, other agents such as naloxone and tirilazad underscore the persistent challenges of translating pathophysiological rationale into reproducible clinical benefit. Despite decades of research, no pharmacologic agent has yet achieved universal acceptance for acute SCI, emphasizing the continued need for more precise, mechanism-specific strategies.

Subacute phase (2–14 days): Anti-inflammation and cytoprotection

Granulocyte colony-stimulating factor (G-CSF)

G-CSF is a hematopoietic growth factor with pleiotropic effects, including mobilization of stem cells, inhibition of apoptosis, promotion of angiogenesis, and modulation of neuroinflammation. In preclinical SCI models, G-CSF reduced neuronal death, enhanced vascular remodeling, and improved functional recovery.¹⁾

An earlier open-label, dose-escalation phase I/IIa study (n=16) tested subcutaneous G-CSF at 5 or 10 µg/kg/day for 5 days within 48 hours after injury, demonstrating favorable safety and preliminary neurological improvement, with 9 of 16 patients showing ≥1 AIS grade gain and greater motor recovery in the high-dose group (p<0.05).⁶⁶⁾

Building on these findings, a Japanese multicenter phase III randomized controlled G-SPIRIT trial (UMIN000018752) tested G-CSF in acute SCI.⁴⁰,⁴¹⁾ In this study, patients with acute SCI (within 48 hours of injury) were randomized to receive G-CSF (400 µg/m² daily for 5 days) or placebo. The trial aimed to enroll approximately 180 participants but ultimately randomized 88 patients due to recruitment challenges. The safety profile was acceptable, with no unexpected serious adverse events. In terms of efficacy, the trial did not meet its primary outcome measure of ASIA motor score improvement at 3 months. However, post hoc subgroup analyses, especially among patients classified as AIS C at baseline, hinted at greater improvements in motor score and functional outcomes in the G-CSF arm compared to placebo. Other secondary endpoints, such as walking capacity or bladder/bowel function, showed trends favoring G-CSF but lacked statistical robustness.

Although this trial did not provide definitive evidence for G-CSF efficacy in SCI, it remains one of the largest and most rigorously conducted human studies of a growth factor therapy in this field. The results highlight both the potential and limitations of translating promising preclinical biology into clinical gains. For G-CSF to be adopted as a viable therapy, further rigorously designed trials with adequate power, stratification by injury severity, and possibly biomarker-driven patient selection will be essential.

Erythropoietin (EPO)

EPO, primarily known for its erythropoietic function, also exerts neuroprotective effects via anti-apoptotic, anti-oxidative, and anti-inflammatory mechanisms.⁵¹⁾ Preclinical studies demonstrated reduced lesion size, enhanced neuronal survival, and improved motor outcomes following systemic administration.³,⁴²⁾ Currently, no large randomized controlled trials have validated EPO in SCI, and its role remains investigational.

Taken together, these findings delineate the current landscape of subacute-phase pharmacotherapy in SCI. Both G-CSF and EPO share convergent cytoprotective and anti-inflammatory mechanisms aimed at limiting delayed neuronal loss and promoting microenvironmental repair. Although each has shown encouraging biological plausibility and acceptable safety in early human studies, neither has yet produced definitive clinical efficacy in large randomized trials. These agents collectively highlight the promise—and the translational gap—of subacute neuroprotection, underscoring the need for adequately powered, biomarker-driven studies to validate their therapeutic potential.⁴⁸,⁶¹⁾

Intermediate phase (2 weeks–6 months): Regeneration priming

Cethrin (VX-210, formerly BA-210)

Cethrin is a recombinant Rho protein antagonist, a signaling molecule activated after SCI that contributes to growth cone collapse, axonal regeneration failure, and neuronal apoptosis. By blocking the Rho/ROCK pathway, Cethrin aims to create a permissive environment for axonal regeneration and functional recovery.³⁰,⁵²⁾ Preclinical studies demonstrated enhanced axonal growth and improved locomotor outcomes following Cethrin administration.⁶⁰⁾

A phase I/IIa open-label dose-escalation study tested Cethrin applied extradurally during decompression surgery in patients with acute cervical or thoracic SCI.²⁵⁾ Results showed an acceptable safety profile and suggested neurological improvement, particularly in patients with cervical AIS A injuries, prompting further development.

This led to the phase IIb/III SPinal Cord Injury Rho INhibition InvestiGation (SPRING) trial (NCT02669849), which investigated a single extradural application of VX-210 during decompression within 72 hours of acute cervical SCI.²¹⁾ However, the trial was terminated prematurely in 2020 after interim analyses indicated it was unlikely to meet the primary efficacy endpoint (improvement in upper extremity motor score).²⁰⁾ Although development of VX-210 has been discontinued, Cethrin remains a landmark example of translating intracellular pathway modulation into clinical trials for SCI.

Anti-Nogo-A antibodies

Nogo-A is a myelin-associated inhibitory protein that restricts axonal regeneration and plasticity after SCI by binding to the Nogo receptor complex and activating downstream RhoA/ROCK signaling. Neutralizing this pathway using anti-Nogo-A monoclonal antibodies aims to enhance sprouting of spared axons and promote functional recovery.¹⁹,⁵⁷,⁶⁰⁾

Based on this rationale, an early first-in-human, international multicenter phase I dose-escalation study (NCT00406016) evaluated the intrathecal administration of anti-Nogo-A antibody (ATI355), demonstrating an acceptable safety profile without drug-related serious adverse events and exploratory neurological improvements in some patients.⁴³⁾

These encouraging findings led to the international Nogo Inhibition in Spinal Cord Injury (NISCI) trial (NCT03935321), a phase IIb, randomized, double-blind, placebo-controlled study. Patients (n=126, randomized; 78 NG-101, 48 placebo) with acute cervical SCI (AIS A–D), enrolled within 4–28 days post-injury, received six intrathecal bolus injections of 45 mg NG-101 every 5 days over 4 weeks. The primary endpoint was change in upper-extremity motor score (UEMS) at 6 months. The study initially used a 1:1 randomization but later shifted to 3:1 (final 2:1 NG-101:placebo) to enhance enrollment and exposure.

Published in The Lancet Neurology (2025),⁷¹⁾ the trial confirmed safety and feasibility of intrathecal antibody delivery but did not yield a statistically significant difference in UEMS gain at 6 months (mean difference, 1.37; 95% confidence interval, −1.44 to 4.18). However, post-hoc analyses suggested potential benefit in motor-incomplete (AIS C) patients, consistent with earlier preclinical and early-phase findings.

Anti-repulsive guidance molecule A (RGMa) antibodies (Elezanumab)

RGMa is a myelin-associated inhibitory protein that limits axonal regeneration and plasticity by activating the Nogo receptor–Rho/ROCK signaling pathway. Elezanumab (ABT-555) is a fully human monoclonal antibody targeting RGMa, designed to enhance neuroprotection, plasticity, and axonal regrowth after SCI.³⁶⁾ In rodent and non-human primate SCI models, Elezanumab improved neuromotor recovery, preserved white matter integrity, and promoted corticospinal tract sprouting.³⁶,⁵⁴⁾ Importantly, intravenous administration yielded superior efficacy compared to intrathecal delivery. A first-in-human single ascending dose and multiple ascending dose study in healthy volunteers and multiple sclerosis patients demonstrated that Elezanumab was well tolerated, with dose-proportional pharmacokinetics, CSF penetration, and reduction in free RGMa levels as evidence of target engagement.³⁷⁾ Building on these findings, the international ELezanumab in Acute Spinal Cord Injury (ELASCI) trial (NCT04295538) was initiated as a phase II, randomized, double-blind, placebo-controlled study evaluating Elezanumab in patients with acute traumatic cervical SCI (C4–C7, AIS A–B) treated within 24 hours of injury. The primary endpoint is the change in UEMS at 52 weeks, along with safety and biomarker assessments. The study remains active but not recruiting, with results not yet published. The trial is expected to be completed in 2026.

Collectively, these approaches represent the frontier of regeneration-priming pharmacotherapy in spinal cord injury. Agents such as Cethrin, anti-Nogo-A, and anti-RGMa antibodies share a mechanistic focus on overcoming intrinsic and extrinsic growth inhibition within the injured spinal cord through modulation of the Rho/ROCK signaling axis. Although none have yet demonstrated definitive clinical efficacy, they collectively validate the translational feasibility of biologic and intracellular-pathway-targeted interventions. Ongoing and future trials integrating precise patient stratification, biomarker-guided endpoints, and multimodal rehabilitation frameworks will be essential to realize the regenerative potential of these molecular strategies.

Chronic phase (>6 months): Symptom modulation and combinatorial strategies

G-CSF

G-CSF, previously described as a potential agent for the subacute phase, has also been investigated as a neurorestorative therapy in chronic, incomplete SCI. A randomized, double-blind, placebo-controlled phase III trial (IRCT201108297441N1) reported significant improvements in ASIA motor scores and functional outcomes following short-term subcutaneous administration of G-CSF (300 µg/day × 7 days).¹⁶⁾ However, the heterogeneous study population and post-rehabilitation design limit extrapolation to acute or complete injuries, and confirmatory trials are warranted.

Riluzole (repurposing for spasticity)

To date, no pharmacologic agent has demonstrated proven neuroprotective or regenerative efficacy in the chronic phase of SCI. Therapeutic efforts at this stage therefore focus primarily on symptom modulation and functional maintenance, aiming to alleviate spasticity, neuropathic pain, and bladder dysfunction. Drugs such as baclofen, tizanidine, gabapentinoids, and botulinum toxin remain standard of care, while riluzole has recently been repurposed for spasticity control in the RILUSCI trial (NCT02859792).¹⁴⁾ This adaptive, randomized, placebo-controlled phase Ib/IIb trial is designed to assess safety and pharmacokinetics, with the protocol indicating a favorable safety profile. Ongoing analyses are evaluating efficacy endpoints including changes in spasticity, pain, and bladder function. Final results are not yet published.

Integrated or combinatorial therapeutic strategies

Stem cell-based therapy

Stem cell transplantation (e.g., neural stem/progenitor cells, mesenchymal stem cells, induced pluripotent stem cell-derived neural lineages) remains a leading candidate. These cells may support remyelination, provide trophic support, and integrate into host circuits.⁶⁸,⁶⁹⁾ Clinical trials using mesenchymal stem cells, neural stem/progenitor cells, or induced pluripotent stem cell-derived neural lineages have shown acceptable safety and modest improvements in ASIA motor and sensory scores, with occasional AIS grade conversion and partial bladder recovery.¹⁰,¹⁸,⁶⁷⁾ However, the overall evidence remains limited by small sample sizes, heterogeneous methodologies, and short-term follow-up.⁶²⁾

Tissue-engineering approaches

Biomaterial scaffolds and hydrogels (e.g., collagen, fibrin, decellularized matrices) serve as structural frameworks, guide axonal growth, and deliver growth factors or cells into the lesion cavity.³¹⁾ Early human pilot studies report safety and modest functional signals, though consistent efficacy has not yet been demonstrated.

Emerging molecular and genetic strategies

Recent preclinical studies explore modulation of ferroptosis, autophagy pathways, exosome-based therapies, optogenetics, and neuromodulation to support circuit rewiring.⁹,¹¹,²⁹,⁵⁰⁾ These approaches remain largely in the animal/model stage but hold promise for future translation, particularly when combined with cell and scaffold-based systems. Nanomaterial-based delivery systems, such as graphene quantum dots carried by macrophages, also illustrate the potential of biomimetic platforms for bioimaging and photodynamic therapy.³⁸,⁴⁹⁾

Rehabilitation and neuromodulation

Rehabilitation remains a cornerstone of functional recovery after SCI and is increasingly being combined with pharmacologic and biologic interventions.¹⁷⁾ Intensive task-specific and activity-based rehabilitation has been shown to induce neuroplastic remodeling and strengthen spared neural circuits, particularly when initiated early and sustained over time. Neuromodulatory strategies—including epidural electrical stimulation, transcranial magnetic stimulation, and transcutaneous spinal stimulation—have further expanded therapeutic potential by facilitating voluntary motor output and autonomic recovery in chronic SCI.³⁵,⁷⁰⁾

Together, these converging therapeutic modalities underscore the paradigm shift from isolated monotherapies toward integrated, systems-level restoration in SCI. While pharmacologic agents may optimize the biochemical milieu for regeneration, concurrent application of stem cell transplantation, biomaterial scaffolds, and neuromodulation can provide structural and electrophysiologic reinforcement. Coordinating these strategies with precision rehabilitation may enable true functional rewiring rather than symptomatic compensation.

DISCUSSION

SCI remains one of the most devastating conditions in neurotrauma, with poor prognosis despite advances in surgical and rehabilitative care. Among available interventions, early surgical decompression has consistently emerged as the most critical determinant of neurological recovery, supported by both experimental and clinical evidence.⁴⁾ Nevertheless, for decades, there has been intense effort to identify pharmacologic strategies capable of improving outcomes beyond surgical treatment.

MPSS became the most representative pharmacologic candidate following the NASCIS trials.⁶,⁷⁾ However, enthusiasm diminished as subsequent studies and meta-analyses failed to confirm robust benefit while consistently highlighting increased risks of infection and gastrointestinal complications.²⁸⁾ MPSS thus remains a historical milestone, but not a universally accepted standard of care.

Despite these setbacks, advances in understanding SCI pathophysiology—including recognition of the sequential cascade of excitotoxicity, oxidative stress, inflammation, apoptosis, glial scarring, and chronic inhibitory signaling—have revealed multiple therapeutic targets across different temporal windows. This has led to development of novel pharmacologic approaches such as riluzole, minocycline, G-CSF, Cethrin, anti-Nogo-A, and anti-RGMa antibodies. While each demonstrated compelling preclinical efficacy, none has yet delivered reproducible benefit in large-scale randomized controlled trials. The failures highlight a set of recurring barriers: the heterogeneity of injury severity and location, variability in treatment environments, the narrowness of therapeutic time windows, and the improbability that a single agent could dramatically alter outcomes in such a complex disease.

Future directions must therefore emphasize precision and integration. First, identifying the right therapeutic targets is essential, coupled with refined timing of intervention and patient stratification based on biomarkers or injury characteristics.⁴⁴⁾ Second, multimodal approaches—combining multiple drugs targeting complementary mechanisms, or integrating pharmacologic agents with surgical decompression, hemodynamic management, and rehabilitative interventions—are likely to be more effective than monotherapy. Third, novel delivery systems such as nanoparticles, scaffolds, and bioengineered carriers may overcome pharmacokinetic limitations and improve local efficacy. Importantly, non-pharmacologic innovations such as cell transplantation, biomaterials, and neuromodulation should be viewed not as alternatives but as potential synergistic partners with drugs.

In summary, while no pharmacologic agent has yet become standard therapy for SCI, the scientific rationale remains strong, and the translational setbacks should not be interpreted as futility. Rather, they underscore the need for more nuanced trial designs, combinatorial strategies, and integration with emerging technologies. Continued persistence in these directions is crucial, as the potential for pharmacologic intervention in SCI remains real and should not be abandoned.

CONCLUSION

Despite decades of effort, no pharmacologic therapy has yet become standard care for SCI, with early milestones such as MPSS ultimately failing to demonstrate consistent benefit. Advances in understanding SCI pathophysiology have identified time-dependent therapeutic windows, inspiring neuroprotective, neuroregenerative, and biologic strategies. While trials of agents like riluzole, minocycline, G-CSF, Cethrin, anti-Nogo-A, and anti-RGMa antibodies have yet to show definitive efficacy, they represent a maturing, mechanism-based approach. Future success will likely depend on biomarker-guided patient stratification, precise timing, and combination with surgical, rehabilitative, and bioengineered interventions, underscoring that pharmacological treatments will form one component of a multimodal treatment paradigm for SCI.