The GRADUATE program was a comprehensive Phase 3 clinical trial program evaluating gantenerumab, a fully human anti-amyloid-beta (Aβ) monoclonal antibody, for the treatment of early Alzheimer's disease. The program consisted of two identical Phase 3 trials—GRADUATE I (NCT03444870) and GRADUATE II (NCT03443973)—designed to evaluate whether gantenerumab could slow cognitive and functional decline in patients with mild cognitive impairment due to AD or mild AD dementia[@roche2023][@honig2023].
Gantenerumab represented a unique approach among anti-amyloid antibodies due to its ability to bind with high affinity to aggregated forms of Aβ, including oligomers, fibrils, and plaques[@bohrmann2012]. The GRADUATE program was notable for its ambitious scale, rigorous biomarker confirmation requirements, and use of high-dose subcutaneous administration[@salloway2022].
Despite achieving substantial amyloid plaque reduction—a finding that validated the antibody's mechanism of action—the trials failed to meet their primary clinical endpoint of slowing cognitive decline[@roche2023]. This paradox has provided critical insights into the relationship between amyloid clearance and clinical benefit, informing the development of next-generation anti-amyloid therapies.
| Attribute | GRADUATE I | GRADUATE II |
|---|---|---|
| NCT Number | NCT03444870 | NCT03443973 |
| Phase | Phase 3 | Phase 3 |
| Status | Completed (Terminated) | Completed (Terminated) |
| Sponsor | Roche/Genentech | Roche/Genentech |
| Start Date | October 2018 | October 2018 |
| Completion | January 2023 | January 2023 |
| Enrollment | ~1,400 patients | ~1,400 patients |
The GRADUATE trials were randomized, double-blind, placebo-controlled, parallel-group Phase 3 studies conducted at over 250 sites globally[@roche2023]:
The trials employed an innovative titration approach designed to minimize amyloid-related imaging abnormalities (ARIA) while maximizing brain penetration[@salloway2022].
The trials enrolled patients meeting the following criteria[@salloway2022][@roche2023]:
Inclusion Criteria:
Exclusion Criteria:
Gantenerumab operates through three complementary mechanisms to clear amyloid-beta from the brain[@bohrmann2012][@abbright2019]:
The primary clearance mechanism involves Fc gamma receptor-mediated phagocytosis. After gantenerumab binds to aggregated Aβ on plaque surfaces, the IgG1 Fc domain engages Fc gamma receptors on microglia, triggering engulfment and degradation of the antibody-Aβ complexes. This mechanism leverages the brain's natural immune surveillance to remove pathological protein deposits[@bohrmann2012].
Gantenerumab binds to multiple Aβ species on plaque surfaces, potentially destabilizing the aggregated structure and facilitating disaggregation. The antibody's conformational epitope recognition allows it to target both fibrillar core regions and more diffuse plaque components[@abbright2019].
Like other anti-amyloid antibodies, gantenerumab may create a peripheral sink by binding plasma Aβ, promoting efflux from the brain. This mechanism involves the antibody binding circulating Aβ in peripheral blood, establishing a concentration gradient that drives Aβ efflux from the central nervous system[@delrieu2017].
Gantenerumab binds to a conformational epitope present on Aβ aggregates, with preference for N-terminal regions of fibrillar Aβ. This specificity distinguishes it from other anti-amyloid antibodies:
| Endpoint | Instrument | Purpose |
|---|---|---|
| Cognition | ADAS-Cog13 | Memory, language, praxis |
| Daily Function | ADCS-MCI-ADL | Activities of daily living |
| Global Change | CIBIC-Plus | Clinician's impression |
| Biomarker | Amyloid PET | Target engagement |
| Biomarker | CSF p-tau/tau | Downstream effects |
| Brain Structure | MRI volumetry | Brain atrophy rate |
The GRADUATE trials did not meet their primary endpoint[@roche2023][@honig2023]:
| Measure | GRADUATE I | GRADUATE II |
|---|---|---|
| CDR-SB treatment difference | -0.31 points | -0.19 points |
| Relative reduction vs placebo | 8% | 6% |
| Statistical significance | Not significant (p=0.22) | Not significant (p=0.40) |
For context, lecanemab achieved a 27% relative reduction in CDR-SB decline (p<0.001) in the CLARITY AD trial, and donanemab achieved a 35% reduction (p<0.001) in TRAILBLAZER-ALZ 2[@vanDyck2023][@loyle2023].
Despite failing to meet the primary clinical endpoint, gantenerumab produced substantial biomarker changes that validated its mechanism of action[@salloway2022][@abbright2019]:
| Trial | Centiloid Reduction (Active) | Difference from Placebo |
|---|---|---|
| GRADUATE I | -75.73 | -66.44 |
| GRADUATE II | -65.96 | -56.46 |
These reductions, while substantial, were lower than those achieved by lecanemab (~80 centiloids) and donanemab (~70 centiloids)[@vanDyck2023][@loyle2023].
Only 28% of gantenerumab-treated participants achieved amyloid-negative PET status at endpoint, compared to approximately:
This incomplete amyloid clearance may explain the limited clinical efficacy[@honig2023].
Gantenerumab demonstrated downstream effects on tau pathology and synaptic dysfunction[@klein2019]:
These biomarker changes suggest that while amyloid was cleared, it may have been insufficient to meaningfully impact downstream tau pathology and neurodegeneration[@musiek2023].
The safety profile was generally consistent with other anti-amyloid antibodies, with ARIA being the primary safety concern[@salloway2022]:
| Adverse Event | Gantenerumab | Placebo |
|---|---|---|
| ARIA-E (edema) | 24.9% | 2.5% |
| Symptomatic ARIA-E | 5.0% | 0.4% |
| ARIA-H (hemorrhage) | Associated with ARIA-E | - |
| Injection site reactions | 24.4% | 18.8% |
Most ARIA events were mild to moderate and resolved without sequelae. The rate of ARIA-E was higher than observed with lecanemab (12.6%) but comparable to donanemab (~24%)[@salloway2022].
The GRADUATE trials implemented rigorous ARIA monitoring protocols[@salloway2022][@masterman2023]:
The gantenerumab GRADUATE program, despite its negative results, provided critical insights for the field of Alzheimer's disease drug development[@isselbacher2022][@fila2023]:
The relationship between amyloid reduction and clinical benefit may require near-complete clearance. The 28% amyloid-negative rate compared to 68-80% for newer antibodies suggests that incomplete target engagement may limit efficacy[@honig2023].
Even substantial amyloid clearance may be insufficient if downstream tau pathology is too advanced. The GRADUATE results support the hypothesis that amyloid-targeting therapies may need to be combined with tau-targeting approaches for maximum clinical benefit[@musiek2023][@reim2023].
Biomarker-confirmed amyloid pathology alone may not be sufficient. Future trials may need to incorporate:
| Feature | Gantenerumab (GRADUATE) | Lecanemab (CLARITY) | Donanemab (TRAILBLAZER) |
|---|---|---|---|
| Developer | Roche/Genentech | Eisai/Biogen | Eli Lilly |
| Administration | Subcutaneous | IV infusion | IV infusion |
| Dose | 510 mg q2w | 10 mg/kg q2w | 350 mg q4w |
| Amyloid reduction | ~50% | ~80% | ~70% |
| Amyloid-negative rate | 28% | 68% | 80% |
| CDR-SB benefit | Not significant | 0.45 (27%) p<0.001 | 0.70 (35%) p<0.001 |
| ARIA-E rate | 24.9% | 12.6% | ~24% |
| Trial duration | 116 weeks | 79 weeks | 76 weeks |
| Status | Discontinued | Approved (2023) | Approved (2024) |
Roche discontinued gantenerumab development in January 2023 following the GRADUATE trial results[@roche2023]. However, the learnings from this program have informed the development of next-generation anti-amyloid therapies:
Roche's pipeline includes trontinemab, a brain-penetrant anti-Aβ antibody that may address the brain penetration limitations observed with gantenerumab. This program represents a direct application of lessons learned from the GRADUATE experience.
The GRADUATE results have influenced the broader anti-amyloid field:
The gantenerumab program, while not successful clinically, has advanced our understanding of Alzheimer's disease pathophysiology and treatment:
Combination Approaches: The field is now exploring amyloid + tau combination therapies, informed by the recognition that amyloid clearance alone may be insufficient
Earlier Intervention: Prevention trials targeting amyloid-positive but cognitively normal individuals may offer greater opportunity for clinical benefit
Enhanced Brain Penetration: Next-generation antibodies and BBB-crossing technologies may overcome the penetration limitations observed with gantenerumab
Personalized Medicine: Genetic and biomarker stratification may help identify patients most likely to benefit from anti-amyloid therapy