Between 2000 and 2005, five randomized double-blind placebo-controlled trials with IFX in patients with moderate-to-severe UC were published [27–30]. The initial three trials provided conflicting results, but were underpowered mainly due to slow patient enrollment [31]. Furthermore, comparison between the different trials is difficult due to the heterogeneous patient populations and the different endpoints used.

In a pilot study by Sands et al. 11 (of 60 planned) patients with severe, active IV steroid-refractory UC were randomized to receive a single intravenous infusion of placebo or IFX at a dose of 5, 10, or 20 mg/kg body weight [27]. Patients had to have active disease for at least 2 weeks (modified Truelove and Witts >10) and have received at least 5 days of IV CS prior to randomization. The primary endpoint used in this trial was treatment failure at week 2, defined as absence of clinical response (modified Truelove and Witts <5 and 5 point reduction compared to baseline), increase in CS dosage, addition of immunosuppressive agents, colectomy, or death. Treatment failure occurred in all three patients randomized to placebo, compared to four out of eight patients randomized to IFX. No significant adverse events were observed.

A second randomized placebo-controlled trial by Probert et al. was conducted in 43 patients with active UC (ulcerative colitis symptom score or UCSS ≥6 and Baron endoscopy score ≥2) who failed treatment with CS (≥30 mg prednisolone or equivalent) for at least 1 week [28]. Patients were randomized to receive an infusion with IFX 5 mg/kg body weight or placebo at week 0 and 2. Clinical remission was defined as a UCSS ≤2, while sigmoidoscopic remission was defined as a Baron score of 0. At week 6, no significant difference was observed in clinical (39 % vs. 30 %, p = 0.76) or sigmoidoscopic remission rates (26 % vs. 30 %, p = 0.96) between the IFX and placebo groups. Furthermore, no significant difference was observed in sigmoidoscopic remission. The two reported serious adverse events (sepsis and colectomy) occurred in the placebo group.

A third trial by Jarnerot et al. included 45 (of the 70 planned) patients with severe to moderately severe UC [29]. Patients were randomized to one single infusion of IFX at a dose of 4–5 mg/kg body weight or placebo either at day 4 after initiation of CS treatment if they fulfilled the index criteria for fulminant UC on day 3 (fulminant colitis index ≥8, n = 28) or on days 6–8 if they fulfilled index criteria for a severe or moderately severe acute attack of UC on days 5–7 (Seo index >150, n = 17). Within 90 days after randomization, colectomy occurred less frequently in patients randomized to IFX compared to placebo (29 % vs. 67 %, p = 0.017). After 3 years of follow-up, colectomy was observed in 50 % of patients randomized to IFX compared to 76 % randomized to placebo (p = 0.012) [32]. Of note, sub-analysis of the initial data showed that the difference was only significant in the patients with a severe or moderately severe attack of UC randomized on days 6–8 [29]. Interestingly, none of the eight patients achieving endoscopic remission at 3 months underwent colectomy within the 2-year follow-up period compared to 50 % of patients who did not achieve endoscopic remission (p = 0.02) [32].

The two largest studies in patients with moderate-to-severe UC, the active ulcerative colitis trials 1 and 2 (ACT1 and ACT2), were conducted in parallel and published in 2005 by Rutgeerts et al. [30]. In each study, 364 outpatients with moderate-to-severe colitis (total Mayo score ≥6 points with an endoscopic sub-score of ≥2) were included. Patients were refractory to CS and/or thiopurines (ACT1) or refractory to at least one standard therapy including 5-aminosalicylates, CS, and/or thiopurines (ACT2).

Patients were randomized to receive IV placebo or IFX 5 or 10 mg/kg body weight for 22 weeks in ACT2 or 46 weeks in ACT1. Clinical response was defined as a decrease from baseline in total Mayo score of ≥3 points and ≥30 % with an accompanying decrease in the sub-score for rectal bleeding of ≥1 point or absolute rectal bleeding sub-score ≤1 point. Clinical remission was defined as a total Mayo score of ≤2 points, with no individual sub-score exceeding 1 point. Mucosal healing was defined as an absolute endoscopic sub-score of 0 or 1 point.

As depicted in Table 15.1, both in ACT1 and ACT2, clinical response, clinical remission, and mucosal healing were significantly more frequently observed at week 8 in patients randomized to IFX (both 5 and 10 mg/kg body weight) compared to placebo. Furthermore, differences remained significant at week 30 (ACT1 and ACT2) and week 54 (ACT1). Of note, the proportion of serious adverse events was similar in all groups (Table 15.2).

Further studies using the ACT1 and ACT2 study population clearly showed that IFX is able to alter the course of UC with an improvement in hospitalization rates, colectomy rates, and quality of life. Sandborn et al. reported on the 1-year colectomy and hospitalization rates (Table 15.2) [33]. Complete follow-up was available in 87 % of the patients. The cumulative incidence of colectomy through week 54 was 10 % for the combined IFX group and 17 % for the placebo group (p = 0.02). Looking at the subgroups, the difference was only significant between IFX 10 mg/kg body weight and placebo (p = 0.007). However, one should take into account that patients were allowed to receive rescue commercial IFX. Commercial IFX was used by 11 % of patients in the placebo group compared with 6 % in the combined IFX group. Therefore, rescue therapy with commercial IFX was evaluated as a potential confounder to the colectomy analysis. Post hoc analysis of the time to colectomy or use of commercial IFX demonstrated that both IFX 5 mg/kg and 10 mg/kg body weight significantly reduced the incidence rates of colectomy or the use of commercial IFX compared with placebo (p = 0.001 and p < 0.001, respectively). Furthermore, a significantly reduced number of UC-related hospitalizations per 100 patient-years was observed for both IFX groups compared with placebo (21 for IFX 5 mg/kg, 19 for IFX 10 mg/kg body weight, and 40 for placebo, p = 0.02 and p = 0.007).

Further analysis of the ACT1 and ACT2 data showed a significant and rapid improvement in health-related quality of life evaluated by both the Inflammatory Bowel Disease Questionnaire (IBDQ) and the 36-item short form health survey (SF-36) [34]. Improvement in all components of the IBDQ and SF-36 were significantly greater in both IFX groups compared to the placebo group (Table 15.2). Furthermore, continued improvement in health-related quality of life was maintained throughout the study period (54 weeks in ACT1 and 30 weeks in ACT2).

A recently published sub-analysis by Colombel et al. highlighted the benefit of achieving early mucosal healing [35]. Lower endoscopy sub-scores at week 8 were associated with increased rates of symptomatic remission, steroid-free symptomatic remission, mucosal healing, and colectomy-free survival at weeks 30 and 54 (Table 15.3). The Kaplan–Meier curves for colectomy survival separated as early as 8 weeks for IFX-treated patients who achieved mucosal healing at week 8 (Mayo endoscopic sub-score 0 or 1) when compared with those who did not (p < 0.001). No significant separation was observed between patients with a sub-score of 0 and those with a sub-score of 1 (p = 0.87), suggesting that achieving complete mucosal healing (absence of all endoscopic abnormalities) may not be mandatory to alter the course of the disease.

A total of 229 of 484 IFX-treated patients from the ACT1 and ACT2 trials entered a long-term extension cohort study published by Reinisch et al. [36]. During a median follow-up of 128 weeks, 70 patients who entered the extension phase (31 %) discontinued IFX administration. In 12 patients, this was due to a lack of efficacy (5 %), while 24 patients discontinued IFX due to adverse events (10 %). During this extension phase, no new or unexpected safety signals were observed.

In patients with CD naïve to immunomodulatory and biological therapy, the SONIC trial has clearly shown a benefit of using combination therapy with both IFX and AZA in inducing CS-free clinical remission and mucosal healing [37]. In UC such a benefit is less clear. The unpublished double-blind randomized controlled UC success trial included 231 patients with moderate-to-severe UC (Mayo score ≥6) who had failed CS and were either naïve to AZA or had stopped AZA ≥3 months prior to inclusion [38]. Patients were randomized to receive AZA monotherapy, IFX monotherapy, or a combination therapy with both AZA and IFX. At week 8, nonresponders (Mayo score reduction <1 point) in the AZA arm were eligible for rescue therapy with IFX. At week 16, a significantly greater proportion of patients in the combination group achieved steroid-free remission (total Mayo score ≤2), clinical response (decrease in total Mayo score of ≥3 points and ≥30 %), and mucosal healing (Mayo endoscopic sub-score 0 or 1) compared to the AZA group. A difference between combination therapy with IFX and AZA and monotherapy with IFX was only seen for steroid-free clinical remission (40 % vs. 22 %, p = 0.017), but not for clinical response (77 % vs. 69 %, p = 0.514) or mucosal healing (63 % vs. 55 %, p = 0.295). Based on these data, an early introduction of IFX in combination with a thiopurine could be advocated in UC. However, overtreatment with possible severe adverse events warrants further characterization of a population at risk for a more complicated disease behavior. In contrast to CD, risk factors for a more complicated disease behavior are poorly defined in UC.