GEMINI 1 & 2
Tolebrutinib
The parallel GEMINI trials found that tolebrutinib did not reduce annualized relapse rates more than teriflunomide in relapsing multiple sclerosis.
- Phase
- Phase III
- Population
- Relapsing multiple sclerosis (RMS)
- Controlled period
- Jun 11, 2020 — Jul 16, 2024
What was compared
Tolebrutinib
60 mg orally once daily with matching placebo
Teriflunomide
14 mg orally once daily with matching placebo
Primary endpoint
Annualized relapse rate
Event-driven follow-up in each trial
Primary endpoint results
GEMINI 1 — annualized relapse rate
Adjusted relapses per participant-year
GEMINI 2 — annualized relapse rate
Adjusted relapses per participant-year
Neither trial demonstrated superiority: rate ratio 1.06 (95% CI 0.81–1.39; P=0.67) in GEMINI 1 and 1.00 (95% CI 0.75–1.32; P=0.98) in GEMINI 2.
Selected secondary outcomes
Pooled whole-trial estimates: 78/933 participants (8.3%) with tolebrutinib versus 106/940 (11.3%) with teriflunomide during the trial; HR 0.71 (95% CI 0.53–0.95). Formal hypothesis testing was not conducted because the primary endpoints failed.
Adjusted mean counts were 0.53 versus 0.29 in GEMINI 1 (rate ratio 1.86; 95% CI 1.36–2.55) and 0.46 versus 0.22 in GEMINI 2 (rate ratio 2.12; 95% CI 1.50–2.99), tolebrutinib versus teriflunomide. Results are descriptive under the failed testing hierarchy.
Adjusted means were 5.61 versus 5.18 in GEMINI 1 (rate ratio 1.08; 95% CI 0.88–1.34) and 5.09 versus 4.37 in GEMINI 2 (rate ratio 1.17; 95% CI 0.91–1.50), tolebrutinib versus teriflunomide. Results are descriptive under the failed testing hierarchy.
Who was studied
Key inclusion criteria
- Age 18–55 years
- RMS diagnosis based on McDonald criteria 2017
- Baseline EDSS ≤ 5.5
- At least 1 relapse in the previous year, at least 2 in the previous 2 years, or at least 1 gadolinium-enhancing brain lesion in the previous year
Key exclusion criteria
- PPMS or nonrelapsing SPMS
Participants at baseline
GEMINI 1
| Characteristic | TolebrutinibN=486 | TeriflunomideN=488 |
|---|---|---|
| Age, years — mean ± SD | 36.8 ± 9.0 | 36.6 ± 9.4 |
| Female sex — n (%) | 334 (68.7%) | 325 (66.6%) |
| Time since MS diagnosis, years — mean ± SD | 4.8 ± 6.2 | 4.6 ± 6.0 |
| EDSS — mean ± SD | 2.4 ± 1.2 | 2.4 ± 1.2 |
| Relapses in previous 12 months — mean ± SD | 1.2 ± 0.6 | 1.2 ± 0.6 |
| Previous DMT — n (%) | 171 (35.2%) | 197 (40.4%) |
| Gadolinium-enhancing lesions present — n/N (%) | 168/484 (34.7%) | 186/485 (38.4%) |
SD: standard deviation; EDSS: Expanded Disability Status Scale; DMT: disease-modifying therapy.
Source: Oh et al., NEJM 2025, Table 1
GEMINI 2
| Characteristic | TolebrutinibN=447 | TeriflunomideN=452 |
|---|---|---|
| Age, years — mean ± SD | 36.6 ± 9.3 | 36.1 ± 9.3 |
| Female sex — n (%) | 300 (67.1%) | 293 (64.8%) |
| Time since MS diagnosis, years — mean ± SD | 3.8 ± 5.4 | 3.8 ± 5.4 |
| EDSS — mean ± SD | 2.4 ± 1.2 | 2.3 ± 1.2 |
| Relapses in previous 12 months — mean ± SD | 1.1 ± 0.5 | 1.2 ± 0.6 |
| Previous DMT — n (%) | 146 (32.7%) | 152 (33.6%) |
| Gadolinium-enhancing lesions present — n/N (%) | 145/447 (32.4%) | 146/448 (32.6%) |
SD: standard deviation; EDSS: Expanded Disability Status Scale; DMT: disease-modifying therapy.
Source: Oh et al., NEJM 2025, Table 1
Safety signal
- Overall adverse-event incidence was similar between treatment groups.
- Minor bleeding events were more frequent with tolebrutinib, including petechiae and heavy menstrual bleeding.
- Liver-enzyme monitoring remained important because rare marked elevations occurred with tolebrutinib.
Why it mattered
GEMINI 1 and 2 showed that BTK inhibition with tolebrutinib did not improve relapse control over teriflunomide. The pooled disability signal generated a separate hypothesis, but the failed primary endpoints prevented formal confirmation and cannot be interpreted as established superiority.
Important limitation
Because both primary endpoints were negative, the pooled disability result was not formally tested under the prespecified hierarchy and cannot overturn the primary conclusion.