OCTAVE Induction 1 and 2, OCTAVE SustainMust knowUC · Induction and maintenance
Tofacitinib as Induction and Maintenance Therapy for Ulcerative Colitis
Sandborn WJ, et al.
N Engl J Med 2017 · PMID 28467869 (PubMed, opens in a new tab) · doi 10.1056/NEJMoa1606910 (full paper, opens in a new tab)
| Endpoint | Tofacitinib 10 mg | Tofacitinib 5 mg | Placebo | Difference |
|---|---|---|---|---|
| OCTAVE Induction 1, Remission, week 8 (primary endpoint) | 18.5% | not in this comparison | 8.2% | P=0.007 |
| OCTAVE Induction 2, Remission, week 8 (primary endpoint) | 16.6% | not in this comparison | 3.6% | P<0.001 |
| OCTAVE Sustain, Remission, week 52 (primary endpoint) | 40.6% | 34.3% | 11.1% | P<0.001 for both doses vs placebo |
Induction remission was modest in absolute terms; the larger gain was in keeping induction responders in remission.
Source: Sandborn 2017, abstract. The abstract gives percentages only, not counts. In Sustain, placebo means withdrawal of tofacitinib after response to induction.
Structured abstract
BackgroundAdvanced therapy for UC meant injected or infused biologics. A phase 2 trial had suggested that an oral JAK inhibitor, tofacitinib, could induce remission.
MethodsThree phase 3, randomised, double-blind, placebo-controlled trials: two 8-week induction trials and one 52-week randomised-withdrawal maintenance trial. 1,139 patients. Primary endpoint: Remission at week 8 (induction) and week 52 (maintenance).
ResultsPrimary result. Induction 1: 18.5% vs 8.2% (P=0.007). Induction 2: 16.6% vs 3.6% (P<0.001). Sustain: 34.3% (5 mg) and 40.6% (10 mg) vs 11.1% (P<0.001 for both).
Take-homeTofacitinib is an effective oral induction and maintenance therapy for UC, but its infection (especially zoster), cardiovascular and cancer signals shape who should receive it.
| Patients | 1,139 |
|---|---|
| Population | Moderately to severely active UC despite conventional therapy or a TNF antagonist. Induction 1: 598; Induction 2: 541. Sustain: 593 induction responders. |
| Comparison | Tofacitinib 10 mg twice daily vs placebo (induction); tofacitinib 5 mg or 10 mg twice daily vs placebo (maintenance) |
| Drugs | tofacitinib |
| Classes | JAK inhibitor |
| Published | Sandborn WJ, et al. N Engl J Med 2017;376:1723–36. |
Key points
- Overall and serious infections were more frequent with tofacitinib during induction; in maintenance, overall infections and herpes zoster were more frequent than with placebo.
- Across all three trials, adjudicated non-melanoma skin cancer occurred in 5 tofacitinib vs 1 placebo patient, and adjudicated cardiovascular events in 5 vs 0. Lipid levels rose.
- ORAL Surveillance (rheumatoid arthritis, age ≥50 with a cardiovascular risk factor; PMID 35081280): combined tofacitinib doses vs a TNF inhibitor, hazard ratio 1.33 (95% CI 0.91–1.94) for major adverse cardiovascular events and 1.48 (95% CI 1.04–2.09) for cancer; non-inferiority was not shown.
- Tofacitinib was the first oral JAK inhibitor for UC.
Limitations
- Absolute induction remission rates were modest (16.6–18.5%).
- Maintenance used randomised withdrawal of induction responders.
- The key long-term safety data (ORAL Surveillance) come from an older rheumatoid arthritis population, which may not translate directly to UC.
Commentary
Why it matters. OCTAVE brought the first oral small molecule with a new mechanism (JAK inhibition) into UC, with onset fast enough to be useful in active disease.
How later evidence changed its use. ORAL Surveillance (PMID 35081280) found higher rates of major cardiovascular events and cancer with tofacitinib than with a TNF inhibitor in older patients with rheumatoid arthritis and cardiovascular risk. Regulators then added warnings for the JAK inhibitor class. In practice, assess age, smoking, cardiovascular and thrombotic risk and cancer history before choosing a JAK inhibitor, use the lowest effective maintenance dose, and offer recombinant zoster vaccination.
Related. Upadacitinib (U-ACHIEVE and U-ACCOMPLISH, PMID 35644166) is a more JAK1-selective option with higher placebo-adjusted remission rates in its own phase 3 trials (a cross-trial comparison, not head-to-head).
Journal club
Design
Three phase 3, randomised, double-blind, placebo-controlled trials: two 8-week induction trials and one 52-week randomised-withdrawal maintenance trial. 1,139 patients.
Primary result. Induction 1: 18.5% vs 8.2% (P=0.007). Induction 2: 16.6% vs 3.6% (P<0.001). Sustain: 34.3% (5 mg) and 40.6% (10 mg) vs 11.1% (P<0.001 for both).
Validity and applicability
- Absolute induction remission rates were modest (16.6–18.5%).
- Maintenance used randomised withdrawal of induction responders.
- The key long-term safety data (ORAL Surveillance) come from an older rheumatoid arthritis population, which may not translate directly to UC.
Where it sits now
Tofacitinib is an effective oral induction and maintenance therapy for UC, but its infection (especially zoster), cardiovascular and cancer signals shape who should receive it.
- ASTRO (2026): Guselkumab can be started with subcutaneous 400 mg injections instead of IV infusions, with clear benefit over placebo by week 12.
- RELIEVE UCCD (UC) (2026): Duvakitug, an anti-TL1A antibody, showed a dose-related remission signal in UC in phase 2b; phase 3 trials will decide its place.
- QUASAR (2025): Guselkumab induces and maintains remission in UC, and the follow-up ASTRO trial showed induction can be given subcutaneously.
- TUSCANY-2 (2025): Afimkibart missed its phase 2b primary endpoint (total Mayo remission), although remission by the modified Mayo score favoured it; anti-TL1A therapy still needs phase 3 proof.
Questions for discussion
- Induction remission rates were modest in absolute terms. How would you explain the expected benefit to a patient?
- ORAL Surveillance studied older patients with rheumatoid arthritis and cardiovascular risk. How far should it shape JAK inhibitor use in a young patient with UC?
- When would you choose 10 mg rather than 5 mg twice daily for maintenance, given the safety signals?