Estriol-RRMS
Estriol combined with glatiramer acetate for women with relapsing-remitting multiple sclerosis: a randomised, placebo-controlled, phase 2 trial
Year of Publication: 2016
Authors: Rhonda R Voskuhl, HeJing Wang, T C Jackson Wu, ..., Robert Elashoff
Journal: The Lancet Neurology
Citation: Lancet Neurol 2016; 15: 35–46
Link: http://dx.doi.org/10.1016/S1474-4422(15)00322-1
PDF: https://escholarship.org/content/qt8zz3v374/qt8zz3v374.pdf
PDF: Clinical Question
Does oral estriol treatment, added to glatiramer acetate, reduce relapse rates in women with relapsing-remitting multiple sclerosis?
Bottom Line
Estriol 8 mg daily plus glatiramer acetate reduced the annualized confirmed relapse rate by 37% compared to placebo plus glatiramer acetate (rate ratio 0.63, p=0.077), meeting the pre-specified phase 2 significance threshold of p<0.10. Treatment was well tolerated over 24 months with no increase in serious adverse events, supporting further investigation in a phase 3 trial.
Major Points
- Relapses decrease >70% during the third trimester of pregnancy when estriol is highest; this trial tested whether exogenous estriol could replicate this protective effect
- Estriol 8 mg daily achieved serum concentrations equivalent to early second trimester of pregnancy
- Primary endpoint met at pre-specified phase 2 threshold (p<0.10): 37% relative reduction in confirmed relapse rate
- At 12 months (before glatiramer acetate reached full potency), estriol showed 51% relapse reduction (p=0.016)
- Post-hoc MRI analyses showed less cortical grey matter atrophy at 12 months in estriol group (p=0.056)
- Cognitive improvement (PASAT) correlated with higher estriol concentrations and cortical grey matter preservation
- Estriol concentrations declined by 24 months due to reduced compliance, potentially attenuating late treatment effects
- Irregular menses more common with estriol (23% vs 4%), but vaginal infections less common (1% vs 11%)
- No concerning signals for breast or uterine pathology over 24 months
Design
Study Type: Randomized, double-blind, placebo-controlled, parallel group, phase 2 trial
Randomization: 1
Blinding: Patients, treating physicians, and all investigators assessing outcomes were masked. Study statisticians and pharmacy staff were unmasked but had no patient interaction. Placebo matched estriol by appearance and taste.
Enrollment Period: June 28, 2007 to Jan 9, 2014
Follow-up Duration: 24 months
Centers: 16
Countries: USA
Sample Size: 158
Analysis: Intention-to-treat. Negative binomial regression for relapse rates adjusted for age, baseline EDSS (<2 vs ≥2), relapses in prior 12 months (≤1 vs >1), time since diagnosis (<1 vs ≥1 year), previous glatiramer acetate treatment, and previous interferon beta treatment. Kaplan-Meier and Cox proportional hazards for time-to-event. Linear mixed effects models for repeated measures. Pre-specified significance level α=0.10 for phase 2 trial.
Inclusion Criteria
- Women aged 18–50 years
- Diagnosis of relapsing-remitting multiple sclerosis according to McDonald criteria
- Baseline EDSS score of 0–4.5
- Relapsing disease activity in the previous 24 months
Exclusion Criteria
- Progressive multiple sclerosis
- Taking glatiramer acetate for more than 2 months before randomisation
- Currently smoking
- Other concurrent disease-modifying treatments
- Concurrent hormonal treatments
Baseline Characteristics
| Characteristic | Control | Active |
|---|---|---|
| N | 76 | 82 |
| Age (years), mean (SD) | 37.1 (7.3) | 37.7 (7.6) |
| Ethnic origin - White | 62 (82%) | 65 (79%) |
| Ethnic origin - Black | 7 (9%) | 9 (11%) |
| Ethnic origin - Hispanic | 6 (8%) | 7 (9%) |
| Ethnic origin - Other | 1 (1%) | 1 (1%) |
| Time since diagnosis (years), mean (SD) | 2.9 (4.5) | 3.3 (4.6) |
| Relapses within 1 year before screening, mean (SD) | 1.5 (0.7) | 1.5 (0.7) |
| Relapses within 2 years before screening, mean (SD) | 2.3 (0.9) | 2.0 (0.7) |
| Previous glatiramer treatment - Never | 27 (36%) | 25 (30%) |
| Previous glatiramer treatment - Before screening | 6 (8%) | 17 (21%) |
| Previous glatiramer treatment - During screening | 43 (57%) | 40 (49%) |
| Previous interferon beta treatment - No | 50 (66%) | 59 (72%) |
| Previous interferon beta treatment - Yes | 26 (34%) | 23 (28%) |
| Mean EDSS score, mean (SD) | 2.1 (1.1) | 2.2 (1.2) |
| EDSS score 0 | 6 (8%) | 9 (11%) |
| EDSS score 1.0 or 1.5 | 21 (28%) | 16 (20%) |
| EDSS score 2.0 or 2.5 | 24 (32%) | 27 (33%) |
| EDSS score 3.0 or 3.5 | 22 (29%) | 25 (30%) |
| EDSS score 4.0 | 2 (3%) | 4 (5%) |
| EDSS score 5.5 | 1 (1%) | 1 (1%) |
| Number of gadolinium-enhancing lesions, mean (SD) | 0.9 (2.1) | 1.0 (2.3) |
| Active lesions on brain MRI - No | 53 (70%) | 55 (67%) |
| Active lesions on brain MRI - Yes | 22 (29%) | 26 (32%) |
| Volume of T2 lesions (cm³), mean (SD) | 7.7 (11.2) | 6.8 (8.9) |
Arms
| Field | Control | Estriol + Glatiramer Acetate |
|---|---|---|
| Intervention | Oral placebo (matched by appearance and taste) daily plus glatiramer acetate 20 mg/day subcutaneous injection plus second placebo matched to progestin for 2 weeks every 3 months starting at 6 months | Oral estriol 8 mg daily plus glatiramer acetate 20 mg/day subcutaneous injection plus norethindrone 0.7 mg daily for 2 weeks every 3 months starting at 6 months (for uterine protection) |
| Duration | 24 months with 4-week taper | 24 months with 4-week taper |
Outcomes
| Outcome | Type | Control | Intervention | HR / OR / RR | P-value |
|---|---|---|---|---|---|
| Annualized confirmed relapse rate at 24 months. Confirmed relapse defined as new or worsening neurological symptoms lasting ≥48 hours in a patient stable or improving for ≥30 days, accompanied by objective change (worsening by ≥0.5 points on EDSS or ≥1.0 points on pyramidal, cerebellar, brainstem, or visual functional system scores), not due to fatigue alone, and not associated with fever or infection. | Primary | 0.37 relapses/year (95% CI 0.25–0.53) | 0.25 relapses/year (95% CI 0.17–0.37) | 0.077 | |
| Time to first confirmed relapse at 24 months | Secondary | 42.9% (95% CI 32.1–55.5) | 33.3% (95% CI 23.8–45.4) | 0.63 (HR) | 0.096 |
| Annualized relapse event rate at 24 months | Secondary | 0.46 relapses/year (95% CI 0.32–0.65) | 0.32 relapses/year (95% CI 0.22–0.46) | 0.65 (RR) | 0.098 |
| Time to first relapse event at 24 months | Secondary | 46.9% (95% CI 35.9–59.3) | 40.5% (95% CI 30.0–53.0) | 0.70 (HR) | 0.179 |
| Annualized confirmed relapse rate at 12 months (exploratory) | Secondary | 0.48 relapses/year (95% CI 0.33–0.69) | 0.25 relapses/year (95% CI 0.16–0.40) | 0.49 (RR) | 0.016 |
| Fatigue (MFIS) change from baseline at 24 months | Secondary | 0.5 (SD 12.8) | -5.0 (SD 14.0) | 0.009 | |
| PASAT change from baseline at 12 months (all patients) | Secondary | 0.13 (SD 4.46) | 1.93 (SD 5.59) | 0.054 | |
| PASAT change at 12 months (baseline score <55, post-hoc) | Secondary | 1.60 (SD 5.99) | 4.70 (SD 6.56) | 0.011 | |
| Cortical grey matter % change at 12 months (post-hoc) | Secondary | -0.72% (SD 0.80) | -0.44% (SD 0.92) | 0.056 | |
| Disability progression over 24 months | Secondary | 15.8% (95% CI 8.8–27.6) | 11.4% (95% CI 5.9–21.7) | 0.81 (HR) | 0.664 |
| Any adverse event | Adverse | 67 (88%) | 76 (93%) | 0.21 | |
| Serious adverse events | Adverse | 10 (13%) | 8 (10%) | 0.54 | |
| Irregular menses or spotting | Adverse | 3 (4%) | 19 (23%) | 0.0005 | |
| Vaginal infection | Adverse | 8 (11%) | 1 (1%) | 0.012 | |
| Upper respiratory tract infection | Adverse | 26 (34%) | 22 (27%) | 0.31 | |
| Urinary tract infection | Adverse | 10 (13%) | 15 (18%) | 0.34 | |
| Glatiramer acetate injection area abnormalities | Adverse | 12 (16%) | 21 (26%) | 0.13 | |
| Fatigue | Adverse | 8 (11%) | 13 (16%) | 0.30 | |
| Depression or anxiety | Adverse | 9 (12%) | 12 (15%) | 0.56 | |
| Endometrium thickness >8 mm | Adverse | 27 (36%) | 24 (29%) | 0.46 | |
| Uterine fibroids | Adverse | 8 (11%) | 8 (10%) | 0.91 | |
| B-cell lymphoma | Adverse | 1 (1%) | 0 (0%) | 0.49 |
Subgroup Analysis
Post-hoc analysis showed that in patients with baseline PASAT score <55 (lower cognitive function), estriol treatment resulted in significantly greater PASAT improvement at 12 months (4.70 vs 1.60 points, p=0.011), whereas no significant effect was seen in patients with baseline PASAT ≥55 (p=0.694). In patients without enhancing lesions at baseline, those in the estriol group had significantly less cortical grey matter atrophy at 12 months (p=0.043), suggesting neuroprotective effects independent of anti-inflammatory effects. In patients with enhancing lesions at baseline, the estriol group had more white matter atrophy (p=0.012), consistent with pseudoatrophy from resolving inflammation.
Criticisms
- Small sample size (n=158) limited statistical power; actual relapse rates were lower than anticipated, reducing power from 80% to 74%
- Both arms received glatiramer acetate, making it difficult to isolate the independent effect of estriol
- Estriol serum concentrations declined significantly by 24 months due to compliance issues, potentially attenuating late treatment effects
- Menstrual irregularities in the estriol group (23% vs 4%) may have partially unmasked treatment allocation
- Phase 2 significance threshold of p<0.10 is less stringent than conventional p<0.05
- Long-term safety of estriol beyond 24 months remains unknown
- Results applicable only to women of reproductive age; cannot be extrapolated to men or postmenopausal women
- MRI volumetric outcomes showing grey matter preservation were post-hoc analyses
- Single-country (USA) study limits generalizability
Funding
National Institutes of Health, National Multiple Sclerosis Society, Conrad N Hilton Foundation, Jack H Skirball Foundation, Sherak Family Foundation, and the California Community Foundation. Synthetic Biologics provided estriol and placebo free of charge.
Based on: Estriol-RRMS (The Lancet Neurology, 2016)
Authors: Rhonda R Voskuhl, HeJing Wang, T C Jackson Wu, ..., Robert Elashoff
Citation: Lancet Neurol 2016; 15: 35–46
Content summarized and formatted by NeuroTrials.ai.