Several meta-analyses have assessed the efficacy and safety profile of sulfasalazine and the newer aminosalicylates for remission maintenance of quiescent UC [21–25]. Sulfasalazine in daily doses of 1–4 g has demonstrated a dose–response effect and can maintain remission in 71–89 % of patients [26, 27]. In the 2006 Cochrane systematic review, oral 5-ASA preparations at doses ranging from 0.8 to 4 g per day were superior to placebo in maintaining clinical and/or endoscopic remission with a pooled odds ratio of 0.47 [95 % confidence interval (CI), 0.36–0.62 and a number needed to treat (NNT) of 6] [20]. This meta-analysis demonstrated that sulfasalazine was more effective [OR 1.29 (95 % CI, 1.05–1.57)] and had a similar adverse event profile to mesalamine [OR 1.16 and 1.31, respectively] but has been criticised for including only patients who tolerated sulfasalazine. These results have been challenged by the recent large meta-analysis of Ford et al. [22] that confirmed that sulfasalazine, mesalamine, balsalazide and olsalazine were effective at preventing clinical relapse [11 trials of 6–12 months duration, 849 5-ASA treated vs. 653 placebo-treated patients, RR of 0.65, p = 0.02, NNT of 4], endoscopic relapse (3 trials, RR of 0.56, p = 0.01, NNT of 4) or both (6 trials, RR of 0.59, NNT of 4) without any significant differences between 5-ASA formulations. Nonetheless, subgroup analysis showed superiority over placebo of sulfasalazine (RR of 0.45, NNT of 3) and mesalamine (RR of 0.65, NNT of 4) but not olsalazine (RR of 0.72). In another two meta-analyses, sulfasalazine and mesalamine were equally effective to balsalazide for remission maintenance, but balsalazide was superior to sulfasalazine for withdrawals due to adverse events (RR of 0.17, p = 0.001) [23, 24]. Regarding comparisons between mesalamine formulations, Ito et al. [25] performed a 12-month non-inferiority trial and found no significant differences in the proportion of patients without bloody stools between a 2.4 g/day pH-dependent release and a 2.25 g/day time-dependent mesalamine formulation. 5-ASA formulations were shown to maintain remission irrespective of the extent of disease [26, 27].

Based on these results, sulfasalazine should be recommended as the first-choice maintenance treatment in UC due to its efficacy and lower cost, whereas mesalamine, olsalazine and balsalazide should be reserved for patients intolerant of sulfasalazine [4]. However, as the optimal maintenance dose of sulfasalazine (4 g/day) cannot be tolerated by a significant proportion of patients, most gastroenterologists incline to use the newer 5-ASA preparations because they are better tolerated at greater than equivalent doses of sulfasalazine [3, 28, 29].

The optimal maintenance mesalamine dose is unknown. An Italian study documented that 1.2 g/day oral mesalamine was equally effective to 2.4 g/day for remission maintenance [12]. In another study, the minimum effective dose of oral mesalamine was 0.8 g/day, but an incremental benefit for patients receiving 1.6 g/day was shown [30]. However, although a clear dose–response at doses over 0.8 g 5-ASA has never been convincingly demonstrated, dose-ranging studies in patients who required doses of mesalamine up to 4.8 g/day to achieve remission have not been performed [31]. In fact, there is some evidence that higher doses may be more effective. Thus, in a 1-year prospective controlled trial, per protocol analysis of data demonstrated that 3.0 g of oral mesalamine (Salofalk^®) once daily was superior to 1.5 g once daily or 0.5 g three times weekly (86 %, 67 % and 78 %, respectively, p = 0.024) [32]. In a meta-analysis of seven randomised clinical trials, 5-ASA doses greater than 2.0 g/day were more likely to reduce the risk for relapse for 6–12 months compared to doses lower than 2.0 g/day, but the quality of the trials that were subjected to analysis was not optimal [22]. Indirect evidence comes also from a post hoc analysis of two clinical trials with Multi-Matrix (MMX) mesalamine in mild-to-moderate UC [10] where 196 of 218 (89.9 %) patients who achieved remission and were maintained on the induction dose of mesalamine were still relapse free after 1 year, indicating that the higher doses that induced remission are expected to maintain also high rates of remission.

Once-daily dose of mesalamine is equally effective and safe, better tolerated and preferred by patients over divided daily doses [12, 13, 33–35]. This effect appears to be similar across all studies irrespective of the mesalamine formulation. A single tablet of 1.2 g MMX mesalamine was as effective as twice-daily dosing (2.4 g) in maintaining clinical remission (88.9 % vs. 93.2 %) and combined clinical and endoscopic remission (64.4 % vs. 68.5 %) for 12 months; both dosing regimens demonstrated a similar adverse event profile and very high adherence rates [13]. In another single-blinded study, 2 g delayed-release 5-ASA granules (Pentasa^®) once daily were superior to 1 g twice daily in maintaining remission of quiescent UC for 1 year (70.9 % vs. 58.9 %, respectively, p = 0.024) [33]. In a very large study, 1.6–2.4 g mesalamine (Asacol^®) once daily was not inferior to twice-daily dosing in maintaining remission of quiescent UC for 1 year (85.4 % vs. 85.4 %, respectively) [34]. Additionally, mesalamine granules once daily at a dose of 1.5 g daily was more effective than placebo to maintain remission over a 6-month period. Finally, 2.4 g MMX mesalamine once daily was equally effective to 2.4 g/day pH-modified release mesalamine (Asacol^®) in divided doses at preventing clinical and/or endoscopic relapse of quiescent UC [35].

The efficacy and safety of various formulations (suppositories, foam and liquid or gel enemas) and dosing regimens of rectally administered mesalamine have been assessed in controlled clinical trials and case series for maintenance of remission of UP, “distal” colitis (“proctosigmoiditis”) and L-UC [36–41]. 5-ASA is delivered to the upper rectum by suppositories and to the rectosigmoid area by foam enemas. Liquid enemas may deliver the active compound up to the splenic flexure, but the actual area of distribution varies between individuals, depending on the length of the sigmoid colon, the volume and the viscosity of the enema.

The advantage of the topical (rectal) therapy is that delivering the active compound directly to the affected area in less frequent dosing schedules increases its effectiveness and reduces systemic availability and adverse effects. In controlled clinical trials, the overall efficacy of topical mesalamine to prevent clinical and/or endoscopic relapse of UP and L-UC for 1 year ranges between 52 and 80 % [4]. Corresponding figures for placebo-treated patients range between 11 % and 53 % and are statistically lower than active treatment in almost all clinical trials. Topical mesalamine administered intermittently three times or even twice weekly is at least equally effective to oral sulfasalazine and mesalamine [18, 38, 42, 43]. However, unlike older mesalamine formulations, oral MMX mesalamine which aims at delivering 5-ASA more evenly throughout the colon has not been tested against topical therapy in quiescent UC. The frequency of topical administration depends largely on patient tolerance, frequency of prior flares and treatment regimen used to induce remission of the most recent flare and may range from 1 g per day to 1–4 g every third day [18, 44]. Topical steroids including the newer formulations with low systemic bioavailability have not shown maintenance efficacy [3, 4, 18, 45].

The combination of oral and topical 5-ASA is superior to oral or topical therapy alone at preventing relapse of quiescent UC [46, 47]. There are no additional safety signals compared to oral or topical therapy, but the long-term tolerance of treatment is debatable. Topical therapy is usually administered intermittently. Combined treatment should be considered as an escalation of therapy for patients who have relapsed despite optimal oral or topical 5-ASA monotherapy [3, 4, 18, 44].

Two large studies, ACT 1 and ACT 2, have assessed the efficacy of IFX in patients with moderate-to-severe UC refractory to steroids, immunomodulators (ACT 1 and 2) and/or 5-ASA (ACT 2) [7]. Remission rates at week 30 (ACT 1 and 2) and at week 54 (ACT 1) were assessed only in patients who had achieved clinical response or remission at week 8 after induction with IFX (5 or 10 mg/kg at weeks 0, 2 and 6) or placebo and continued IFX schedule therapy (5 or 10 mg/kg) or placebo every 8 weeks. Remission rates were significantly higher for IFX groups in ACT 2 at week 30 (26 % vs. 36 % vs. 11 % for 5 mg/kg and 10 mg/kg IFX or placebo, respectively) and in ACT 1 at week 54 (34 % for combined IFX groups vs. 17 % for placebo). In ACT 1, complete mucosal healing defined as a Mayo score 0 or 1 at week 54 was superior for IFX than placebo-treated patients (55 %, 57 % and 22 %, for 5 mg/kg, 10 mg/kg IFX and placebo, respectively). Although the gain in steroid-free remission at week 54 (ACT-1) was only 14 % for the 5 mg/kg arm and 9 % for the 10 mg/kg IFX over placebo (10 %), IFX resulted in a significant reduction in hospitalisations and colectomies [78]. The degree of mucosal healing at week 8 did not predict subsequent outcome of UC but was associated with a significant reduction in the likelihood for colectomy at 1 year [79]. Extended treatment with IFX for up to three additional years was effective in maintaining remission and well tolerated without any additional safety signals [80].

Numerous uncontrolled case series have been reported on patients with UC of varied degrees of severity and refractoriness to prior therapies who were treated with IFX alone or combined with AZA for variable periods of time. Although these data are difficult to subject to group analysis, a nonsystematic review [81] and a systematic review [82] of the literature have confirmed the efficacy and safety of IFX to induce and maintain clinical response, remission, mucosal healing and wane steroids in a considerable proportion of patients with active moderate-to-severe UC.

Colectomy-free survival after IFX therapy was investigated in four studies. Gustavsson et al. [83] reported on a 3-year follow-up of patients who received a single infusion of 4–5 mg/kg IFX as salvage therapy for severe intravenous steroid-refractory UC. Overall, 12/24 (50 %) of IFX-treated patients versus 16/21 (76 %) of placebo-treated patients underwent colectomy (p = 0.012). Lack of mucosal healing at 3 months after the single infusion of IFX predicted subsequent colectomy. In a cohort of 121 patients from Leuven [84] and in a French cohort of 119 patients [85], who were treated with IFX for a median of 33 (IQR 17.0–49.8) months and 18 (IQR 8–32) months, respectively, predictors for colectomy were the lack of short-term response to IFX, baseline levels of CRP ≥5 mg/l [84] or ≥10 mg/l [85], prior treatment with intravenous steroids [84] and/or ciclosporin [84, 85] and an indication of IFX for severe UC [85]. In a Canadian cohort, 46/115 (40 %) of UC patients treated with IFX and/or immunomodulators came to colectomy after a median of 5.3 months. Patients with a detectable serum IFX trough level but not antibodies to IFX had higher rates of remission (69 % vs. 15 %; p < 0.001) and endoscopic improvement (76 % vs. 28 %, p < 0.001). In contrast, an undetectable serum IFX was highly predictive of increased risk for colectomy (55 % vs. 7 %, OR 9.3; 95 % CI 2.9–29.9; p < 0.001). Concomitant use of immunomodulators did not influence the clinical outcomes [86].

In a recent trial involving patients with steroid-refractory UC who were naïve to immunomodulators and anti-TNF therapy, patients were assigned randomly to receive intravenous infusions of infliximab (5 mg/kg at weeks 0, 2, 6 and 14) plus daily oral placebo capsules, oral azathioprine 2.5 mg/kg daily plus placebo infusions on the infliximab schedule or combination therapy with the two drugs. Corticosteroid-free clinical remission was evaluated at week 16 as a primary end point. Corticosteroid-free remission at week 16 was achieved by 39.7 % (31 of 78) of patients receiving infliximab/azathioprine, compared with 22.1 % (17 of 77) receiving infliximab alone (p = 0.017) and 23.7 % (18 of 76) receiving azathioprine alone (p = 0.032). Mucosal healing at week 16 occurred in 62.8 % (49 of 78) of patients receiving infliximab/azathioprine, compared with 54.6 % (42 of 77) receiving infliximab (p = 0.295) and 36.8 % (28 of 76) receiving azathioprine (p = 0.001). Thus, in antitumour necrosis factor-alpha-naive patients with moderate-to-severe UC treated with infliximab plus azathioprine were more likely to achieve corticosteroid-free remission at 16 weeks than those receiving either monotherapy. Combination therapy treatment was associated with better mucosal healing than azathioprine monotherapy.

Whether UC patients in remission on combined AZA and IFX should discontinue AZA is currently unknown. In Crohn’s disease, long-term combination therapy is associated with a higher IFX trough level [87] and lower rate of loss of response to IFX [88] than IFX monotherapy [88]. Whether the full dose of AZA is required to achieve these goals awaits clarification by future trials. However, the benefits of any long-term combination therapy should be balanced against the increased risk for infections and/or malignancies especially in young male patients [84, 88–91]. On the other hand, bridging with IFX to AZA monotherapy in AZA-naïve patients may be easier to achieve in UC than in Crohn’s disease. In any case, the decision to stop IFX and continue on AZA should be discussed only for patients in long-standing sustained steroid-free clinical, serologic (normal CRP), endoscopic (mucosal healing) and/or histologic remission who have a detectable IFX trough level. Again, safety issues, patient preferences, adherence, cost and also the risk of no response or allergic reaction upon re-treatment with IFX should be considered in decision-making.

Adalimumab (ADA) has been approved for the treatment of UC. Initially, preliminary evidence from small trials and retrospective case series suggests that ADA may be effective and safe in achieving and maintaining long-term clinical response or remission, mucosal healing, steroid sparing, improving quality of life and reducing colectomies in patients who have failed almost all prior therapies [92–97]. Concomitant use of AZA was the only independent factor predicting response to ADA in one trial [95]. Recently, a large controlled trial was reported in 494 patients with moderate-to-severe UC who had failed treatment with steroids, immunomodulators and/or IFX (40.3 % of the ITT population). Patients were randomised 1:1 to placebo or ADA (n = 248, 160/80 mg sc at week 0 and 2 and then 40 mg every other week). Dose escalation of ADA was allowed. At weeks 8, 52 and both 8 and 52, significantly more patients on ADA achieved clinical remission and response. Significantly more patients with prior IFX failure achieved clinical remission or response on ADA than placebo at week 52 and sustained clinical response at both weeks 8 and 52. The therapeutic gain in steroid-free remission rate at week 52 was only 7.6 % in favour of ADA but was still significantly statistically superior to placebo (5.7 %, p = 0.035). Adverse events were not significantly different between ADA and placebo [98].

Golimumab is a subcutaneously administered fully human anti-TNF antibody, previously approved for the treatment of rheumatoid arthritis, ankylosing spondylitis and psoriatic arthritis. The treatment of ulcerative colitis recently gained regulatory approval for induction and maintenance of remission in patients who have moderate-to-severe ulcerative colitis based on the data in two registration trials. There has been one randomised placebo-controlled trial [Program of Ulcerative Colitis Research Studies Utilizing an Investigational Treatment-Subcutaneous (PURSUIT-SC)] that assessed induction therapy with subcutaneous golimumab in anti-TNF-α-naive patients with moderate-to-severe ulcerative colitis (Mayo score 6–12 points with an endoscopic subscore >2 points) not responding to conventional therapy with oral mesalamine, oral corticosteroids, and AZA/6-mercaptopurine or unable to taper corticosteroids without recurrence of ulcerative colitis activity. The subsequent PURSUIT-M randomised placebo-controlled trial assessed the efficacy and safety of golimumab in maintaining clinical response in patients who responded to induction treatment with golimumab in the preceding PURSUIT-SC trial. Golimumab was shown to be more efficacious than placebo in inducing clinical response, remission and mucosal healing and improving quality of life.

Based upon the date from these two clinical trials, Golimumab is initially administered subcutaneously at the dose of 200 mg at week 0 followed by 100 mg at week 2 and after that every 4 weeks.

Vedolizumab is the most recent agent for treatment of ulcerative colitis (and Crohn’s disease) to gain regulatory approval. The efficacy of treatment with vedolizumab in ulcerative colitis was assessed. 374 patients were randomised to either drug or placebo as part of induction. The response for week 6 was measured by the Mayo score and documented mucosal healing.

During induction, 47.1 % of the patients on vedolizumab versus 25.5 % of the patients on placebo achieved remission (p < 0.001). A second cohort of patients received open-label vedolizumab, and responders from both cohorts were included in the maintenance trial that evaluated clinical remission at week 52. Patients were randomised to receive the drug every 4 or 8 weeks or a placebo. A total of 41.8 % of patients maintained remission when receiving medication every 8 weeks compared to 44.8 % who received the drug every 4 weeks; patients who received placebo had maintenance of remission at a rate of 15.9 %. There was a statistically significant difference in maintenance of remission between patients who received the drug every 8 weeks versus placebo (p < 0.001) and those receiving the drug every 4 weeks versus placebo (p < 0.001). Based on this data, the drug was approved for use in adults with moderate-to-severe ulcerative colitis when one or more standard therapies (corticosteroids, immunomodulators or tumour necrosis factor blocker medications) have not resulted in an adequate response.