← Back
NeuroTrials.ai
Neurology Clinical Trial Database

ENDEAR

Nusinersen versus Sham Control in Infantile-Onset Spinal Muscular Atrophy

Share Share Copied! Compare

Year of Publication: 2017

Authors: Finkel RS, Mercuri E, Darras BT, ..., Tizzano E

Journal: New England Journal of Medicine

Citation: N Engl J Med. 2017 Nov 2;377(18):1723-1732

Link: https://pubmed.ncbi.nlm.nih.gov/29091570/

Clinical Question

Does intrathecal nusinersen improve motor function, event-free survival, and overall survival compared to sham procedure in infants with SMA Type 1?

Bottom Line

Nusinersen significantly improved motor milestone achievement (51% vs 0%), reduced risk of death or permanent ventilation (HR 0.53), and reduced mortality (HR 0.37) in infantile-onset SMA. Trial stopped early due to overwhelming efficacy.

Major Points

  • First randomized controlled trial to demonstrate efficacy of any therapy in infantile-onset SMA Type 1
  • Motor milestone response (HINE Section 2): 51% nusinersen vs 0% sham at final analysis (P<0.001); 41% vs 0% at interim
  • Event-free survival (time to death or permanent ventilation) significantly favored nusinersen: HR 0.53 (P=0.005)
  • Overall survival significantly improved: HR for death 0.37 (P=0.004)
  • Trial stopped early at prespecified interim analysis due to overwhelming efficacy -- independent data monitoring committee recommendation
  • Earlier treatment initiation (shorter disease duration at screening) was associated with greater motor milestone response
  • Led to FDA approval of nusinersen (Spinraza) in December 2016 as first therapy for SMA

Design

Study Type: Phase 3, randomized, double-blind, sham-controlled, multinational trial

Randomization: 1

Blinding: Double-blind; sham control used skin prick to mimic lumbar puncture

Enrollment Period: August 2014 to November 2016

Follow-up Duration: Terminated early at interim analysis; planned endpoint assessment at Day 183 (motor milestones) and Day 394 (event-free survival)

Centers: 31

Countries: United States, Canada, Germany, Italy, France, Spain, United Kingdom, Belgium, Sweden, Turkey, Japan, South Korea, Australia

Sample Size: 122

Analysis: Hierarchical testing strategy to control type I error at 0.05 level; 2:1 randomization (nusinersen:sham); interim analysis prespecified at ~80 infants enrolled for >=6 months

Inclusion Criteria

  • Genetic diagnosis of 5q SMA (homozygous SMN1 deletion, mutation, or compound heterozygote)
  • Two copies of SMN2 gene
  • Symptom onset at or before 6 months (180 days) of age
  • Age <=7 months (210 days) at screening
  • Adequate nutrition and hydration (with or without gastrostomy)
  • Body weight >=3rd percentile for age
  • Gestational age 37-42 weeks at birth
  • Standard of care immunizations met (influenza, pneumococcal, RSV prophylaxis)

Exclusion Criteria

  • Hypoxemia (oxygen saturation <96% awake or asleep without ventilatory support)
  • Untreated or inadequately treated active infection requiring systemic antiviral or antimicrobial therapy
  • History of brain or spinal cord disease interfering with lumbar puncture, CSF circulation, or safety assessments

Arms

FieldNusinersenControl
InterventionIntrathecal nusinersen 12 mg administered by lumbar puncture: 4 loading doses (Days 1, 15, 29, 64), then maintenance doses (Day 183, 302, 394)Small needle prick to skin overlying lumbar spine on same schedule to maintain blinding; no intrathecal injection
Duration13 months (Day 1 to Day 394)13 months

Outcomes

OutcomeTypeControlInterventionHR / OR / RRP-value
Motor-milestone response defined as improvement in >=1 HINE Section 2 category (head control, sitting, rolling, crawling, standing, walking) and no worsening in any categoryPrimary0% (0/27 at interim; 0/37 at final)41% (21/51 at interim); 51% (37/73 at final)3<0.001
Event-free survival (time to death or permanent ventilation >=16 h/day for >21 days or tracheostomy)SecondaryMedian time to event: 22.6 weeks (sham)Median not reached (nusinersen)0.530.0046 (95% CI 0.32-0.89)
Overall survival (time to death from any cause)Secondary16/41 (39%) died13/80 (16%) died0.3720.0041 (95% CI 0.18-0.77)
CHOP INTEND >=4 point improvement from baselineSecondary1/37 (3%)52/73 (71%)<0.0001
CMAP responders (peroneal nerve amplitude >=1 mV)Secondary2/37 (5%)26/73 (36%)0.0004
Specific motor milestones achieved (nusinersen group)Secondary0% for all milestonesFull head control: 22%; Rolling: 34%; Independent sitting: 8%
Any AEAdverse40/41 (97.6%)77/80 (96.3%)
Severe AEAdverse33/41 (80.5%)45/80 (56.3%)
Any SAEAdverse39/41 (95.1%)61/80 (76.3%)
Discontinued due to AEAdverse16/41 (39.0%)13/80 (16.3%)
PyrexiaAdverse22/41 (54%)39/80 (49%)
ConstipationAdverse9/41 (22%)24/80 (30%)
Upper respiratory tract infectionAdverse9/41 (22%)20/80 (25%)
Respiratory distressAdverse14/41 (34%)19/80 (24%)
PneumoniaAdverse6/41 (15%)17/80 (21%)
Respiratory failureAdverse14/41 (34%)17/80 (21%)
SAE - Respiratory failureAdverse16/41 (39%)20/80 (25%)
SAE - Respiratory distressAdverse8/41 (20%)21/80 (26%)
SAE - PneumoniaAdverse5/41 (12%)19/80 (24%)
SAE - Cardio-respiratory arrestAdverse5/41 (12%)5/80 (6%)
Death during studyAdverse16/41 (39%)13/80 (16%)HR 0.37, P=0.004

Subgroup Analysis

Disease duration at screening was the strongest predictor of treatment response. Below median (<=13.1 weeks): HR 0.240 (95% CI 0.10-0.58, p=0.0003); event rate at Day 364: sham 77.3% vs nusinersen 27.1%. Above median (>13.1 weeks): HR 0.844 (95% CI 0.43-1.67, p=0.40); event rate: sham 72.5% vs nusinersen 62.5%. This dramatic difference underscores the importance of early treatment initiation.

Criticisms

  • Trial stopped early at interim analysis, limiting collection of long-term efficacy and safety data
  • Sham control raises ethical questions, though necessary for regulatory rigor in a disease with known fatal natural history
  • Timing sensitivity suggests early treatment is critical -- patients with longer disease duration at screening showed less benefit, raising questions about treatment window
  • Intrathecal administration requires repeated lumbar punctures in infants, which is technically challenging and carries procedural risks
  • Limited to infants with 2 SMN2 copies; applicability to other SMN2 copy numbers not established by this trial
  • High cost of therapy (~$750,000/year maintenance) raises access and health economic concerns
  • Industry-funded (Biogen/Ionis) with potential conflict of interest

Funding

Biogen and Ionis Pharmaceuticals

Based on: ENDEAR (New England Journal of Medicine, 2017)

Authors: Finkel RS, Mercuri E, Darras BT, ..., Tizzano E

Citation: N Engl J Med. 2017 Nov 2;377(18):1723-1732

Content summarized and formatted by NeuroTrials.ai.