TANGO
(2020)Objective
Intravenous tocilizumab (IL-6 receptor antagonist) monthly — to evaluate efficacy vs oral azathioprine in highly relapsing neuromyelitis optica spectrum disorder.
Study Summary
• Time to first relapse hazard ratio was 0.236 (95% CI 0.107-0.522; p=0.0004), favoring tocilizumab.
• 14% of tocilizumab patients relapsed vs 47% of azathioprine patients during median ~60-week follow-up.
• EDSS and pain scores favored tocilizumab; MRI activity reduced.
• Both drugs generally well tolerated; infection rates similar; tocilizumab has expected transaminitis and lipid changes.
• Established tocilizumab as an effective off-label treatment option for NMOSD, later superseded by approved agents (eculizumab, inebilizumab, satralizumab).
Intervention
Tocilizumab 8 mg/kg IV every 4 weeks vs azathioprine 2-3 mg/kg/d oral. Open-label randomized phase 2 trial.
Inclusion Criteria
Adults ≥18 years with seropositive or seronegative NMOSD per 2015 criteria, ≥2 relapses in preceding 12 months (highly relapsing).
Study Design
Arms: Tocilizumab 8 mg/kg IV q4wk vs Azathioprine 2-3 mg/kg/d oral (1:1)
Patients per Arm: Tocilizumab 59; Azathioprine 59 (N=118)
Outcome
• Relapse rates: 14% (tocilizumab) vs 47% (azathioprine) during median follow-up ~60 weeks
• Annualized relapse rate significantly lower with tocilizumab
• EDSS progression and pain scores favored tocilizumab
• MRI activity reduced with tocilizumab
• Serious infections: similar rates
• Transaminitis and lipid elevations: more with tocilizumab (expected class effect)
Clinical Question
Does tocilizumab reduce relapses more than azathioprine in highly relapsing NMOSD?
Bottom Line
In adults with highly relapsing NMOSD, monthly IV tocilizumab 8 mg/kg reduced time to first relapse by 76% compared with oral azathioprine (2-3 mg/kg/d) over a median 60-week follow-up (HR 0.236; 95% CI 0.107-0.522; p=0.0004), with 14% vs 47% relapse rates. Established tocilizumab as an effective alternative to azathioprine in highly relapsing NMOSD.
Major Points
- Phase 2 open-label multicenter randomized trial at 6 hospitals in China
- 118 adults with highly relapsing NMOSD (≥2 relapses in preceding 12 months) randomized 1:1
- Enrollment November 2017 to August 2018
- Tocilizumab 8 mg/kg IV every 4 weeks vs oral azathioprine 2-3 mg/kg/d
- Primary endpoint: time to first relapse
- Secondary: relapse rates, EDSS progression, pain (short-form Brief Pain Inventory), MRI activity
- Median follow-up ~60 weeks
- Time to first relapse: HR 0.236 (95% CI 0.107-0.522); p=0.0004 favoring tocilizumab
- Relapse rates: 14% (tocilizumab) vs 47% (azathioprine)
- Annualized relapse rate significantly lower with tocilizumab
- EDSS progression and pain scores favored tocilizumab
- MRI activity (new/enlarging T2 lesions, gadolinium-enhancing lesions) reduced with tocilizumab
- Serious infections: similar rates between arms
- Transaminitis and lipid elevations more common with tocilizumab (expected)
- Demonstrated IL-6 blockade as effective alternative to azathioprine in highly relapsing NMOSD
- Bridged the therapeutic landscape to approved agents (eculizumab 2019, inebilizumab 2020, satralizumab 2020)
Study Design
- Study Type
- Phase 2 open-label multicenter randomized active-controlled trial
- Randomization
- Yes
- Blinding
- Open-label
- Sample Size
- 118
- Follow-up
- Median ~60 weeks
- Centers
- 6
- Countries
- China
Primary Outcome
Definition: Time to first relapse
| Control | Intervention | HR/OR | P-value |
|---|---|---|---|
| 47% relapsed (azathioprine) | 14% relapsed (tocilizumab) | 0.236 (0.107-0.522) | p=0.0004 |
Limitations & Criticisms
- Open-label design (blinding not feasible due to oral vs IV administration)
- Single-country (China) enrollment limits external generalizability
- Phase 2 scale (N=118) and median ~60-week follow-up
- Active comparator (azathioprine) has known inadequacy in highly relapsing NMOSD
- Did not directly compare with approved monoclonal antibodies (eculizumab, inebilizumab, satralizumab)
- Cost and biosimilar availability may limit uptake in resource-limited settings
Citation
Lancet Neurol 2020;19(5):391-401