PDF: Vitamin E Sano
(1997)Objective
Alpha-tocopherol (vitamin E) 2000 IU/d, selegiline 10 mg/d, or both — to evaluate whether antioxidants or MAO-B inhibitors slow clinical progression in moderately severe Alzheimer's disease.
Study Summary
• Time to primary endpoint (death, institutionalization, loss of ADLs, or severe dementia) was delayed by ~7 months after baseline MMSE adjustment.
• Neither drug improved cognition on ADAS-cog or MMSE — the benefit was functional, not cognitive.
• Combination therapy offered no additional benefit over either drug alone.
• Adverse events including falls and syncope were more common with active treatment but generally tolerable.
• Established a functional rather than cognitive disease-modifying rationale for vitamin E in AD — influenced guidelines for two decades despite later controversy over high-dose safety.
Intervention
Alpha-tocopherol 2000 IU/d, selegiline 10 mg/d, both, or placebo for 2 years in patients with moderate Alzheimer's disease.
Inclusion Criteria
Adults with probable Alzheimer's disease of moderate severity (CDR 2), living at home.
Study Design
Arms: Placebo vs Selegiline 10 mg/d vs Vitamin E 2000 IU/d vs Both
Patients per Arm: 341 total across 4 arms (~85 per arm)
Outcome
• Median time to endpoint: 440 d (placebo), 585 d (selegiline), 670 d (vitamin E), 585 d (combination)
• Unadjusted survival analysis: no significant differences — baseline MMSE imbalance required adjustment
• No significant cognitive improvement on ADAS-cog or MMSE in any active arm
• Falls and syncope somewhat more common with active treatments
Clinical Question
Does vitamin E or selegiline delay clinical progression of moderate Alzheimer's disease?
Bottom Line
In moderate Alzheimer's disease, high-dose vitamin E (2000 IU/d) or selegiline (10 mg/d) each independently delayed the composite primary endpoint (death, institutionalization, loss of ADLs, or severe dementia) by ~7 months after adjustment for baseline MMSE, without meaningful cognitive improvement. Combination therapy did not provide additional benefit. This is the first major trial to suggest antioxidant therapy slows AD progression functionally.
Major Points
- Phase 3 multicenter double-blind placebo-controlled 2x2 factorial randomized trial by the Alzheimer's Disease Cooperative Study
- 341 community-dwelling patients with probable AD of moderate severity (CDR 2) randomized
- Four arms: placebo, selegiline 10 mg/d, alpha-tocopherol 2000 IU/d, or both
- Treatment duration: 2 years
- Primary endpoint: time to any of death, institutionalization, loss of ability to perform basic ADLs, or severe dementia (CDR 3)
- Despite randomization, baseline MMSE was higher in placebo group — analysis adjusted for baseline MMSE
- Adjusted for baseline MMSE: vitamin E, selegiline, and combination all significantly delayed primary endpoint (HR ~0.53-0.56)
- Median time to endpoint: 670 d (vitamin E), 585 d (selegiline), 585 d (both), 440 d (placebo)
- Unadjusted analysis: differences did not reach statistical significance
- Neither drug produced measurable improvement on ADAS-cog or MMSE — the benefit was exclusively functional
- Combination did not exceed either monotherapy — no additive benefit
- Higher falls and syncope with active treatment
- High-dose vitamin E (2000 IU/d) later raised safety concerns (HOPE trial; meta-analyses suggesting CV risk)
- Influenced AD clinical guidelines for two decades; later trials (DHA, Souvenaid, etc.) generally did not replicate functional benefit
Study Design
- Study Type
- Phase 3 multicenter randomized double-blind placebo-controlled 2x2 factorial trial
- Randomization
- Yes
- Blinding
- Double-blind
- Sample Size
- 341
- Follow-up
- 2 years
Primary Outcome
Definition: Time to composite endpoint: death, institutionalization, loss of ADL, or severe dementia (CDR 3)
| Control | Intervention | HR/OR | P-value |
|---|---|---|---|
| Median 440 days (placebo) | Median 670 (vitamin E), 585 (selegiline), 585 (both) | - (Adjusted analyses) | Vitamin E p=0.001; selegiline p=0.012; both p=0.049 (all adjusted) |
Limitations & Criticisms
- Baseline MMSE imbalance despite randomization required post-hoc adjustment
- Primary endpoint significance depended on MMSE adjustment — unadjusted analysis did not reach significance
- High-dose vitamin E (2000 IU/d) later raised safety concerns in other indications
- No cognitive improvement on standard scales — functional benefit only
- Combination therapy did not add to monotherapy despite complementary mechanisms
- No imaging or biomarker correlates available in 1997
Citation
N Engl J Med 1997;336(17):1216-1222