CONSTRUCTMust knowUC · Acute severe colitis
Infliximab versus ciclosporin for steroid-resistant acute severe ulcerative colitis (CONSTRUCT): a mixed methods, open-label, pragmatic randomised trial
Williams JG, et al.
Lancet Gastroenterol Hepatol 2016 · PMID 27595142 (PubMed, opens in a new tab) · doi 10.1016/S2468-1253(16)30003-6 (full paper, opens in a new tab)
| Outcome | Infliximab | Ciclosporin | Comparison |
|---|---|---|---|
| CUCQ area under the curve, mean (SD) (primary endpoint) | 564.0 (241.9) | 587.0 (226.2) | Adjusted difference 7.9 (95% CI −22.0 to 37.8); P=0.603 |
| Colectomy | 55/135 (41%) | 65/135 (48%) | P=0.223 |
| Time to colectomy, mean days (95% CI) | 811 (707–912) | 744 (638–850) | P=0.251 |
| Serious adverse events | 21 in 16 patients | 25 in 17 patients | No difference |
| Deaths | 3 | 0 | — |
Source: Williams 2016, abstract (analysis by treatment allocated). A chart is not used because the primary outcome is a mean score.
Structured abstract
BackgroundInfliximab and ciclosporin were both used as rescue therapy for steroid-refractory acute severe UC and seemed similarly effective, but their clinical and cost-effectiveness had not been compared in a pragmatic trial.
MethodsOpen-label, pragmatic, mixed-methods randomised trial at 52 hospitals in England, Scotland and Wales; follow-up to 3 years. 270 patients. Primary endpoint: Quality-adjusted survival: area under the curve of Crohn's and Ulcerative Colitis Questionnaire (CUCQ) scores over up to 3 years.
ResultsPrimary result. No difference: mean AUC 564.0 (SD 241.9) with infliximab vs 587.0 (SD 226.2) with ciclosporin; adjusted difference 7.9 (95% CI −22.0 to 37.8); P=0.603
Take-homeIn steroid-refractory acute severe UC, infliximab and ciclosporin are equally effective rescue therapies; choose by local experience, patient factors and the maintenance plan.
| Patients | 270 |
|---|---|
| Population | Adults admitted with acute severe UC who had not responded to IV hydrocortisone within about 5 days |
| Comparison | Infliximab 5 mg/kg IV at 0, 2 and 6 weeks vs ciclosporin 2 mg/kg/day IV for up to 7 days, then oral 5.5 mg/kg/day for 12 weeks |
| Drugs | infliximab, ciclosporin |
| Classes | Anti-TNF, Calcineurin inhibitor |
| Published | Williams JG, et al. Lancet Gastroenterol Hepatol 2016;1:15–24. |
Key points
- Colectomy overall: 41% (55/135) with infliximab vs 48% (65/135) with ciclosporin (P=0.223). In hospital: 21% vs 25%; at 12 months: 35% vs 45% (full text).
- Mean time to colectomy: 811 days (95% CI 707–912) vs 744 days (95% CI 638–850); P=0.251.
- Serious adverse events: 21 in 16 patients (infliximab) vs 25 in 17 patients (ciclosporin). Deaths: 3 vs 0.
- Health service costs over 30 months were £5632 (95% CI 2773–8305) higher with infliximab, mainly from drug cost (full text, 2012–13 prices).
Limitations
- Open-label.
- The primary outcome was quality of life, not colectomy; colectomy differences were secondary and not significant.
- UK cost data predate cheaper anti-TNF biosimilars (the authors note the cost gap is narrowing).
Commentary
Why it matters. CONSTRUCT and CYSIF (PMID 23063316) are the two randomised trials comparing the standard rescue drugs in acute severe UC. Both found no meaningful difference, so the choice rests on practical factors.
Practical points. Infliximab is simpler to give (no drug-level monitoring) and can continue as maintenance. Ciclosporin acts quickly and is short-lived, which some prefer when colectomy is near, but it needs level, blood pressure, renal and magnesium monitoring and is usually bridged to a thiopurine. CONSTRUCT centres were encouraged to give co-trimoxazole as Pneumocystis prophylaxis with ciclosporin. Patients interviewed after colectomy suggested surgery was not a bad outcome, a useful point when counselling.
Dosing. PREDICT-UC (PMID 39236736) found no advantage of a 10 mg/kg first infliximab dose over 5 mg/kg in steroid-refractory disease.
Journal club
Design
Open-label, pragmatic, mixed-methods randomised trial at 52 hospitals in England, Scotland and Wales; follow-up to 3 years. 270 patients.
Primary result. No difference: mean AUC 564.0 (SD 241.9) with infliximab vs 587.0 (SD 226.2) with ciclosporin; adjusted difference 7.9 (95% CI −22.0 to 37.8); P=0.603
Validity and applicability
- Open-label.
- The primary outcome was quality of life, not colectomy; colectomy differences were secondary and not significant.
- UK cost data predate cheaper anti-TNF biosimilars (the authors note the cost gap is narrowing).
Where it sits now
In steroid-refractory acute severe UC, infliximab and ciclosporin are equally effective rescue therapies; choose by local experience, patient factors and the maintenance plan.
- ACTIVE (2025): After a steroid-responsive episode of acute severe UC in a thiopurine- and biologic-naive patient, starting infliximab plus azathioprine reduced 1-year treatment failure compared with azathioprine alone.
- PREDICT-UC (2024): In steroid-refractory acute severe UC, a 10 mg/kg first infliximab dose or an accelerated schedule was not better than standard 5 mg/kg dosing.
- TACOS (2024): Adding tofacitinib to IV steroids improved day-7 response in acute severe UC in one small single-centre trial; promising, but not yet standard care.
- CYSIF (2012): Ciclosporin and infliximab perform similarly as rescue therapy for steroid-refractory acute severe UC, in the short and long term.
Questions for discussion
- The primary outcome was quality of life rather than colectomy. Why might the investigators have chosen it, and what does it capture that colectomy rates miss?
- The table gives unadjusted means and an adjusted difference. What would you need to know (scale direction, adjustment, which arm is subtracted) before calling either drug better?
- With similar effectiveness, which practical factors decide between infliximab and ciclosporin for a patient on the ward today?