← Back
NeuroTrials.ai
Neurology Clinical Trial Database

PRISMS-4

PRISMS-4: Long-term efficacy of interferon-β-1a in relapsing MS

Share Share Copied! Compare

Year of Publication: 2001

Authors: PRISMS Study Group, University of British Columbia MS/MRI Analysis Group

Journal: Neurology

Citation: Neurology 2001;56:1628-1636

Link: https://www.thelancet.com/journals/lance...3334-0/abstract

Clinical Question

Does the efficacy of IFN-β-1a persist through 4 years of treatment, is there a dose-response relationship, and does early treatment provide benefit over delayed treatment initiation?

Bottom Line

Clinical and MRI benefits of IFN-β-1a continued for both doses up to 4 years with evidence of dose-response. Outcomes were consistently better for patients treated for 4 years than for those who crossed over from placebo at year 3. Neutralizing antibody development reduced efficacy.

Major Points

  • 90% of original patients (506/560) entered the extension study
  • 4-year relapse rate: Rx44 0.72, Rx22 0.80 vs crossover 1.02 (p<0.001)
  • Time to sustained disability progression prolonged by 18 months with Rx44 vs crossover (p=0.047)
  • T2 burden decreased 6.2% with Rx44 vs increases of 3.4-9.7% in other groups
  • Crossover patients showed ~50% relapse reduction after switching to active treatment
  • Dose effect approached significance for relapses (p=0.069) and was significant for several MRI outcomes
  • Neutralizing antibodies: 23.7% Rx22 vs 14.3% Rx44 (p=0.02); associated with reduced efficacy
  • No new safety issues identified; adverse events decreased in years 3-4

Design

Study Type: Extension study of Phase 3 RCT with dose-blinded assessment

Randomization: 1

Blinding: Dose-blinded; original placebo patients re-randomized to one of two IFN doses; assessing neurologists blinded; MRI analyzed centrally by blinded researchers

Enrollment Period: Extension of PRISMS-2 (years 3 and 4)

Follow-up Duration: 4 years total (2-year extension after 2-year core study)

Centers: 22

Countries: Europe, Canada, Australia

Sample Size: 506

Analysis: Intent-to-treat; Poisson regression for relapse count; Cox proportional hazards for time-to-event; Kaplan-Meier for progression-free proportions; ANOVA on ranked data for MRI

Inclusion Criteria

  • Completion of PRISMS-2
  • Medical and neurologic condition deemed appropriate for continuing blinded treatment

Exclusion Criteria

  • Patients could be withdrawn for WHO grade IV toxicity
  • Pregnancy
  • Protocol violations
  • Unacceptable adverse events
  • Non-compliance

Arms

FieldControlRx22Rx44
InterventionPlacebo for years 1-2, then randomized to IFN-β-1a 22 μg or 44 μg TIW for years 3-4IFN-β-1a 22 μg subcutaneously three times weekly continuously for 4 yearsIFN-β-1a 44 μg subcutaneously three times weekly continuously for 4 years
Duration2 years placebo + 2 years active treatment4 years4 years

Outcomes

OutcomeTypeControlInterventionHR / OR / RRP-value
Relapse count per patient over 4 yearsPrimary1.02/year (crossover combined)RR 0.76 (Rx22 vs crossover), 0.66 (Rx44 vs crossover)<0.001 for both; Rx44 vs Rx22: p=0.069
Time to first confirmed EDSS progression (40th percentile)Secondary24.2 months (crossover)RR 0.68 (Rx22), 0.62 (Rx44)0.047 (Rx44 vs crossover); 0.289 (Rx22 vs crossover)
Proportion relapse-free at 4 yearsSecondary6.7% (crossover)<0.001 (Rx44), 0.016 (Rx22)
Median time to second relapse (months)SecondaryReference (crossover)0.006 (Rx22), <0.001 (Rx44); Rx44 vs Rx22: 0.046
Median IDSS (EDSS step-years) over 4 yearsSecondary333.5 (crossover)0.034 (Rx44 vs crossover)
Change in T2 burden of disease over 4 yearsSecondary+7.2% (placebo/22), +9.7% (placebo/44)0.003 (Rx44 vs placebo/44); 0.009 (Rx44 vs Rx22)
Median new T2 lesions per patient per scan (4 years)Secondary2.0 (placebo/22), 2.7 (placebo/44)<0.001 for all comparisons
Injection-site inflammation (years 1-4)Adverse
Flu-like symptoms (years 1-4)Adverse
Depression (years 1-4)Adverse
Lymphopenia (years 1-4)Adverse0.07
Elevated ALT (years 1-4)Adverse0.07

Subgroup Analysis

Neutralizing antibody impact: NAb-positive patients had higher relapse rates in years 3-4 (Rx44: 0.81 vs 0.50, p=0.002). MRI effects more pronounced: median T2 lesions 1.4 (NAb+) vs 0.3 (NAb-) for Rx44 (p<0.001). T2 burden increased 17.6% in NAb+ vs decreased 8.5% in NAb- patients (Rx44).

Criticisms

  • Loss of placebo control after year 2 limits definitive efficacy conclusions for extension period
  • Higher dropout rate in placebo group during years 1-2 may have selected for responders
  • Crossover design makes true treatment effect assessment difficult as control data blends placebo and treatment years
  • Cannot definitively attribute late improvements to treatment vs natural disease stabilization
  • Dose effect on primary outcome (relapses) only approached significance (p=0.069)

Funding

Serono International SA (Geneva, Switzerland) and monitored by Serono Clinical Research Associates

Based on: PRISMS-4 (Neurology, 2001)

Authors: PRISMS Study Group, University of British Columbia MS/MRI Analysis Group

Citation: Neurology 2001;56:1628-1636

Content summarized and formatted by NeuroTrials.ai.