Acute severe ulcerative colitis (ASUC) is a severe flare of ulcerative colitis (UC) that affects approximately 25% of patients with UC at some point in their lives [1]. It is defined by the Truelove and Witts criteria as 6 or more bowel actions per day with blood, along with at least one of the following: pulse rate > 90 beats/min, temperature > 37.8°C, hemoglobin < 10.5 g/dL, and erythrocyte sedimentation rate > 30 mm/hr [2].

Approximately 30% of patients are refractory to intravenous (IV) steroids and require rescue therapy with either infliximab or cyclosporin [3,4]. Approximately 50% of patients respond to infliximab or cyclosporin [3,5,6]. Patients who fail rescue therapy face the prospect of emergency colectomy. Poor prognostic factors associated with lack of response to traditional rescue therapy have been recently identified [7] and are summarized in Fig. 1.

Emerging data suggests that clinical outcomes of steroid-refractory ASUC may be improved with novel infliximab dosing strategies, use of cyclosporin as a bridge to slower acting biologic agents and Janus kinase (JAK) inhibitors as sequential therapy [8-11]. In this review, we provide an overview of the traditional rescue therapies available in ASUC and mechanisms of failure of rescue therapies. We then summarize and synthesize the emerging literature on contemporary approaches to clinical management of ASUC in the setting of failure to respond to traditional rescue therapies.

References for this review were identified through electronic searches of Ovid Medline and EMBASE (inception to December 2023), and conference proceedings identified using EMBASE (January 2020 to December 2023). The following search terms were used alone or in combination: “acute severe colitis,” “ASUC,” “ulcerative colitis,” “hospital,” “severe,” “infliximab,” “cyclosporin,” “ciclosporin,” “tofacitinib,” “upadacitinib,” “vedolizumab”, or “ustekinumab.” Only papers published in English were reviewed. The final reference list was generated based on originality and relevance to the scope of this review.

Therapies in ASUC require a rapid onset of action to target the aberrant immune response. Patients with ASUC who fail first-line therapy with steroids need rescue therapy with either infliximab or cyclosporin, both of which are equally effective based on data from 2 randomized controlled trials [5,12]. A meta-analysis of non-randomized studies indicated that infliximab was associated with better clinical response, odds ratio (OR) 2.96 (95% confidence interval [CI], 2.12–4.12) and lower 12 months colectomy rates, OR 0.42 (95% CI, 0.22–1.28) [6]. Regardless of these findings the clinical response and early colectomy rates have been similar between the 2 drugs. Long-term response rates have also remained similar; albeit with higher relapse rates in cyclosporin compared with infliximab-treated patients [3].

Whilst infliximab and cyclosporin are considered equally effective as rescue therapy in steroid-refractory ASUC, selecting the right rescue therapy in ASUC can be challenging and should be individualized based on patient factors, the presence of comorbidities, the intended choice of maintenance therapy after successful induction of remission, and physician and local hospital familiarity with the rescue therapy of choice.

Three main mechanisms of infliximab failure have been postulated in ASUC: (1) altered pharmacokinetics of infliximab in ASUC; (2) high TNF burden in ASUC leading to target-mediated drug deposition; and (3) alternate (non-TNF-mediated) immune pathways in some patients [35]. Fig. 2 provides a summary of the mechanisms of infliximab failure and how clinical management may be adapted.

Data on mechanisms of cyclosporin failure in ASUC are limited. Multidrug resistance gene-1 (MDR1) encodes for the transporter P-glycoprotein, a plasma membrane protein involved in the transport of cyclosporin out of the cell into the gut lumen [51]. In ASUC patients receiving cyclosporin, Daniel et al. [51] observed significant differences in MDR1 between responders and non-responders.

Other potential reasons for cyclosporin failure may relate to the high interindividual variability in cyclosporin metabolism and pharmacokinetics, which can in turn be partly explained by genetic polymorphisms. Cyclosporin is predominantly metabolized by cytochrome P450 3A4 (CYP3A4) in the liver and small bowel [52], and primarily excreted by the biliary system [53]. As such, inter-patient differences in metabolism via the CYP3A4 system may affect cyclosporin levels and/or dosing requirements [54]. Similarly, concomitant use of drugs, such as phenytoin and carbamazepine, that induce CYP3A4 may contribute to subtherapeutic cyclosporin levels [55] and potentially contribute to higher rates of non-response.

In view of the hepatic metabolism of cyclosporin, patients with advanced chronic liver disease may be at increased risk of supratherapeutic cyclosporin levels [56] and therefore require closer monitoring of blood cyclosporin levels and dose titration.

Sequential therapy refers to the use of a second rescue agent after failure of a first rescue agent in a bid to induce remission and avoid emergency colectomy.

At present, the evidence supporting maintenance therapy following successful induction of remission following rescue therapy is relatively limited. Data to guide the optimal maintenance strategy following successful induction of remission with sequential therapy or third-line strategies are also lacking.

Data from a retrospective multicenter French study found lower rates of relapse after a first episode of ASUC in those treated with anti-TNF containing maintenance regimens compared to those containing 5-aminosalicylates or conventional immunosuppressants [74].

In biologic-naïve patients, 5-aminosalicylates and/or azathioprine with or without combination infliximab can be used. The use of maintenance azathioprine compared to azathioprine in combination with infliximab is currently being investigated in the ACTIVE trial (NCT02425852) among IV steroid responders and PREDICT UC (NCT02770040) among IV steroid-refractory patients.

In previous failure of infliximab, drugs that are yet to be evaluated in clinical trials include vedolizumab, ustekinumab, and JAK-inhibitors.

Vedolizumab, a selective anti-adhesion monoclonal antibody targeting α4β7 integrin has been demonstrated to be effective in moderate to severe UC [75]. Vedolizumab prevents α4β7 integrin interaction with MAdCAM-1 expressed on gastrointestinal epithelial cells, thereby inhibiting T-cell migration into intestinal tissue. Its slow onset of action limits its use in ASUC; however, it can be used in combination with a rapid-acting agent such as IV steroids or cyclosporin rescue as a bridge to vedolizumab maintenance [76].

The safety and effectiveness of vedolizumab has been evaluated in a recent systematic review by Gisbert et al. of 8 studies [34,76-83]. The majority of studies were retrospective, included patients who had failed anti-TNFs and most used cyclosporin as bridge to vedolizumab. The colectomy-free rate in patients treated with vedolizumab for ASUC was 69% (weighted mean 95% CI, 61%–76%) and the rate of serious adverse events (or adverse events leading to discontinuation) was 1.8% with no mortality due to cyclosporin or vedolizumab [34]. Since then, another successful case of cyclosporin rescue as a bridge to vedolizumab has been reported by Ashat et al. [84]. Hence, calcineurin inhibition as a bridge to vedolizumab maintenance therapy appears to be a safety and effective strategy in patients with ASUC.

Ustekinumab, a monoclonal antibody targeting the p40 subunit of IL-12 and IL-23, has also been shown to be effective in patients with moderate to severe UC. Use of cyclosporin/tacrolimus as a bridge to ustekinumab in ASUC has been evaluated in 3 small retrospective studies comprising a total of 13 patients who had previously failed anti-TNF and/or vedolizumab [34,85-87]. Colectomy was avoided in all patients with no severe adverse events or mortality observed with either agent [34].

In a recent prospective study of 11 patients with steroid-refractory ASUC (10 of whom were biologic experienced, and 8 of whom had previously failed ≥ 3 biologic therapies), ustekinumab was commenced after a mean of 3.2 weeks after initiation of cyclosporin rescue, and combination therapy continued for a mean of 11.5 weeks. Six patients (54%) were clinical responders at week 16, with 5 patients (45%) in endoscopic remission at the same time point. At long-term follow-up at 52 weeks, 5 patients (45%) were in histological remission [88].

Hence, although the data are limited calcineurin inhibition as bridge to ustekinumab maintenance therapy may be a safe and effective strategy in ASUC, even in the setting of prior biologic treatment experience.

The JAK-inhibitors tofacitinib, upadacitinib and filigotinib are effective in moderate to severe ulcerative colitis [89-91]. Potential benefits of these agents include their oral administration, relatively rapid onset of action, ease of transition from induction to maintenance therapy and lack of immunogenicity. However, data regarding the safety and efficacy of these agents to maintain remission following successful induction of remission in ASUC are awaited.

The evolving nature of therapeutic approaches and availability of new drugs for ASUC has increased the complexity behind decision making as to when it is appropriate to proceed to colectomy. The time to clinical response of individual drugs varies from patient to patient, hence decision making is a day-by-day proposition. In the absence of validated predictive indices of therapeutic response to rescue therapies, one should proceed to colectomy if there is a negative trend in: clinical indices (comprised of vital signs, rectal bleeding and abdominal tenderness), biomarkers (rising CRP), colonic dilatation on abdominal imaging or endoscopic deterioration. Monitoring of such parameters is all the more important in the setting of sequential therapy. Decision making regarding when to proceed to colectomy therefore involves timely risk assessment, assessment of treatment response and multidisciplinary team management.

The treatment landscape of ASUC is rapidly evolving with new rescue therapy options, although most of these are still under evaluation. Whilst the traditional therapies of infliximab and cyclosporin remain the mainstays of treating steroid-refractory disease, these therapies are not always effective. Hence, attention has shifted towards investigating mechanisms of drug failure, predicting non-response to therapy and devising strategies to overcome non-response.

Non-response or incomplete response to infliximab may be driven by increased infliximab clearance, tissue mediated drug deposition or alternate non-TNF mediated pathways [35]. While incomplete response may be overcome by dose optimization via accelerated or intensified dosing, the optimal infliximab dosing strategy is yet to be defined and is the subject of ongoing trials (NCT03937609 and NCT02770040). Non-response to infliximab in ASUC that cannot be overcome by dose optimization suggests the need for sequential therapy to address alternate non-TNF mediated pathways. While cyclosporin has represented a feasible alternative in such patients, preliminary data suggest that JAK inhibitors may be safe and effective as sequential therapy and therefore require further investigation in relation to their safety, efficacy, role and positioning as rescue therapy agents.

Despite its more challenging route of administration cyclosporin is enjoying a renaissance in some centers familiar with its use. Not only is cyclosporin being used among those who are thiopurine and infliximab naïve but it is also finding a role as a bridging therapy induction agent to allow slower acting biologic therapies such as vedolizumab and ustekinumab to take effect as maintenance agents.

A unique advantage of JAK inhibitors is their capacity to act as both a rescue therapy option as sequential therapy as well as a maintenance agent following successful induction; however, the optimal dosing regimen for both induction and maintenance also requires further investigation.

Whilst advanced therapies have traditionally been used in the setting of steroid non-response (failure) in ASUC, the treatment paradigm may shift in the future with the potential use of JAK inhibition either as an alternative or adjunct to IV steroids. In the recent TACOS study, patients with ASUC were randomized to a 7-day course of tofacitinib 10 mg TDS or placebo TDS while continuing IV hydrocortisone, and treatment response was assessed at day 7 [92]. Patients who responded to tofacitinib received tofacitinib 10 mg BD maintenance therapy whereas steroid monotherapy responders received oral 5-aminosalicylates and thiopurine therapy. Patients who received tofacitinib were more likely to be treatment responders compared to those who received placebo (83.01% vs. 58.82%; OR, 3.42; 95% CI, 1.37–8.48; P= 0.007). Similarly, patients in the tofacitinib group were less likely to require rescue therapy at 7 days (OR, 0.27; 95% CI, 0.09–0.78; P= 0.01) and at 90 days compared to those in the placebo group (cumulative probability 0.13 vs. 0.38 respectively, P= 0.003) [92]. Although most adverse events were mild, there was 1 case of dural venous thrombosis in the tofacitinib arm [92]. Whilst the use of tofacitinib as an adjunct to steroids seems promising, such an approach may risk over-treating patients who will respond to conventional therapy. Similar to calcineurin-inhibitor treatment, JAK inhibitors may be used as induction therapy to bridge to selective therapies with more favorable safety profiles.

Although infliximab and cyclosporin have traditionally been used as first-line rescue therapies, 2 small recent case series have examined the use of tofacitinib as first-line rescue therapy in steroid-refractory patients with ASUC, in predominantly biologic-naïve cohorts [93,94]. Short-term clinical response rates in both series were promising and in the order of 75% to 88% (Table 2). However, the serious adverse events included a case of herpes zoster and 2 deaths, including 1 occurring in a patient who developed sepsis after colectomy, and 1 who died of bacterial pneumonia 1 month after commencement of tofacitinib [93,94]. These preliminary data suggest that although effective, first-line tofacitinib rescue may be associated with significant risks and therefore warrants further evaluation of safety and efficacy within controlled clinical trials.

In the setting of a broadening armamentarium of traditional versus novel rescue therapies and dosing regimens it is difficult to establish the optimal timing, dosing, and duration of each successive rescue therapy agent. Novel strategies and predictive indices of rescue therapy response are currently under development in ASUC to help with decision-making regarding each of the rescue therapies and to determine the specific parameters that define therapeutic success or failure. Whilst randomized controlled trials are ideally required to discern the positioning of advanced therapies in ASUC, such trials are often not practical to conduct in the setting of failure of first-line rescue therapy. Hence, data to guide management following failure of first-line rescue are more likely to come from both real-world and modeling studies. Irrespective of the advances in medical therapy, knowing when to proceed to colectomy requires timely risk assessment, assessment of treatment response and multidisciplinary management. Colectomy constitutes an appropriate intervention for treatment refractory disease and should therefore not be considered a failure of contemporary ASUC management.