Long-Term Outcome of Ciclosporin and Infliximab as Rescue Therapy in Steroid-Refractory Acute Severe Ulcerative Colitis Open Access

Acute severe ulcerative colitis (ASUC) still represents a potentially life-threatening condition with a mortality of around 1% and a risk of short-term colectomy of around 30% [1]. Intravenous steroids remain the first-line therapy for the initial treatment of ASUC [2, 3], but up to 40% of patients will not respond adequately [4, 5] and a rescue therapy will be necessary to avoid, or at least delay, short-term colectomy. Two randomized controlled trials (RCTs) showed similar efficacy and safety for either ciclosporin (CsA) or infliximab (IFX) in patients not responding to intravenous steroids [6, 7]. However, real-life data are sparse and not always in line with the abovementioned two RCTs [8‒10]. In recent years, new options such as sequential therapy (IFX after CsA or vice versa) and a therapy with high-dose tofacitinib emerged [11, 12]. Nevertheless, due to lack of evidence and safety reasons, European Crohn’s and Colitis Organisation (ECCO) and the British Society of Gastroenterology (BSG) guidelines do not recommend the use of a sequential therapy [13, 14]. However, although data about long-term outcome and switching to a second rescue therapy is limited, recent retrospective studies demonstrated an acceptable safety profile using a sequential therapy [15, 16]. In this single-center retrospective study we aimed to compare the long-term outcome and safety of patients with steroid-refractory ASUC treated with either CsA, IFX or sequential therapy at a tertiary center.

Between January 2010 and July 2021, we analyzed patients with established criteria for the diagnosis of ASUC at the USZ [17]. Eligible patients admitted for ASUC were at least 16 years old and refractory to standard high-dose steroid therapy. We excluded patients with documented refusal for use of their health-related personal data, with Crohn’s disease and contraindications to CsA or IFX. We used the Montreal classification for determining the disease extension (ulcerative proctitis, E1; left-sided UC, E2; and extensive UC, E3) [18].

We investigated patients with a single rescue treatment (CsA or IFX) as well as patients with sequential treatment (with CsA and IFX or tofacitinib). The switch between the first and the second rescue therapy needed to be within 1 month after the beginning of the first rescue therapy. It was according to the discretion of the treating physician whether a therapy with IFX at a dosage of 5–10 mg/kg body weight or CsA intravenously at a dosage of ∼2 mg/kg/day was chosen.

Patients with a severe or fulminant course of the disease and insufficient response to the rescue therapy and/or in case of patient’s preference were evaluated for colectomy. The indications for colectomy were in line with the ECCO guidelines and were discussed in an multidisciplinary approach (gastroenterologists and surgeons) with the patient [14].

We derived the inpatient and outpatient medical files from the patient’s medical record in the hospital’s respective computer system (USZ, KISIM, Cistec AG, Zurich, Switzerland, CH). A total of 274 entries from 170 patients were reviewed between January 2010 and July 2021. A total of 48 patients with ASUC received at least one rescue therapy and had signed an informed consent. Due to further treatment in an external hospital, 2 patients had to be excluded because the necessary follow-up data of at least 1 year was not fulfilled. Hence, 46 of the patients met the inclusion criteria for our study. The follow-up data were obtained until last data up to July 2022 or colectomy/death.

We collected the following patient characteristics at the start of the rescue therapy: age, sex, disease duration, disease extension, albumin, CRP, hemoglobin and previous immunotherapies (biologics and immunomodulators). Any adverse events (AEs) or serious AEs (SAEs) after the induction of the rescue therapy were assessed at 3 and 12 months.

The primary endpoint of this study was colectomy-free survival at 1, 3, and 5 years. Secondary outcomes such as predictors of colectomy, the presence of AEs at 1 year and mortality during the entire follow-up were also evaluated.

A descriptive statistical analysis of the data was performed. Data were reported as means and population percentages with standard error. A p value <0.05 was regarded as statistically significant.

Colectomy-free survival rates at 1, 3, and 5 years and changes in therapy were estimated through Kaplan-Meier method.

A logistic regression model was utilized to estimate the impact of age, sex, dose of steroids, duration of disease, albumin, number of previous medications, and the chosen therapy on colectomy-free survival at 1 year. McFadden’s R-squared (pseudo-R-squared) is calculated as a measure of the goodness of fit for the logistic regression model. It can be interpreted as the proportion of the variance in the response variable that is explained by the model, adjusted for the number of parameters in the model. Finally, it was also tested that model assumptions for logistic regression models are verified and found that the model verifies these assumptions.

Secondary outcomes such as mortality during the entire follow-up between patient groups were compared using Fisher’s exact test (for count data). Fisher’s exact test is used to analyze contingency tables with small sample sizes or sparse data. Specifically, it is used to determine whether there is a significant association between two categorical variables like mortality and the patient groups. All statistical analyses were performed using R 4.2.3 GUI 1.79 High Sierra build (Vienna; Austria).

This study was performed in accordance with the Declaration of Helsinki. This human study was approved by Ethics Committee of the Canton of Zurich – approval: BASEC-Nr. 2021-01953. Since 2016 we have requested patients with general consent for further use of the data at our hospital and the written informed consent was given as part of the hospital stay.

We retrospectively analyzed a total of 46 patients who were initially treated with either CsA (n = 31) or IFX (n = 15) due to steroid-refractory ASUC at the University Hospital of Zurich. A total of 13% (6/46) patients received a treatment with a second rescue therapy after CsA. Only 1 patient received tofacitinib (n = 1), whereas all the remaining patients received IFX (n = 5) within 1 month after starting a therapy with CsA. Four patients were nonresponders to CsA, 1 patient had an intolerance and one an allergic reaction to CsA. We did not identify any IFX patient who received a second rescue therapy within 1 month.

The mean age of the sample was 36.04 ± 13.96 years (54.3% men). In total, 29 out of 46 patients (63%) were biologic-experienced and 70% (32/46) had previously used immunomodulators. All biologic-experienced patients except 3 patients (1x adalimumab, 2x vedolizumab) had received IFX in the past. The majority suffered from extensive UC (56.5%) and from left-sided UC (41.3%). Only 1 patient was diagnosed with ulcerative proctitis (2.2%). The mean disease duration was 7.2 years and in total we had a median follow-up period of 5.1 years. Detailed clinical characteristics at the time of inclusion are listed in online supplementary Table 1 (for all online suppl. material, see https://doi.org/10.1159/000546511).

All enrolled patients completed at least 12 months of follow-up. In sum, 78% (36/46), 67% (24/36), and 48% (14/29) were colectomy-free at 1, 3, and 5 years, respectively.

Colectomy-free survival rates at 1, 3, and 5 years according to treatment given at inclusion were 68% (17/25), 52.3% (11/21), and 36.8% (7/19) in patients who received CsA without any second rescue therapy and 86.7% (13/15), 81.8% (9/11), and 62.5% (5/8) in those who received only IFX, respectively.

In the group with sequential therapy, 100% (6/6) and 66.7% (4/6) were colectomy-free after 1 and 3 years. Between patients receiving IFX, CsA or a second rescue therapy there was an overall significant difference between these three groups in colectomy-free survival (p = 0.026) (online suppl. Fig. 1). To analyze a potential difference between each individual therapy to another, we performed a post hoc analysis. This post hoc analysis demonstrated a nonsignificant tendency to a better outcome with sequential therapy in comparison to CsA regarding the colectomy-free survival until the end of follow-up (p = 0.087). There was neither a significant difference in the rate of colectomy-free survival between patients receiving IFX or CsA (p = 0.149) nor between IFX and second rescue therapy (p = 0.212).

Patients with a higher number of previous immunosuppressive therapies were less likely to be colectomy-free at 1 year (p = 0.049) (online suppl. Table 2). Other variables such as age at hospitalization, sex, dose of steroids, disease duration, and albumin did not correlate with a higher risk of 1-year colectomy.

In our analysis, we distinguished between early (within 3 months) and late (between 3 and 12 months) AEs. There were two early SAEs under IFX reported. One patient (58-year-old male) developed a moderate to severe infusion reaction with dyspnea, saturation drop and hypotension. Another patient (36 aged male) suffered from toxic megacolon while under treatment with IFX, which might also indicate a severe course of the disease and was treated conservatively.

The cumulative rate of Clostridioides difficile and CMV infections in patients first treated with CsA (32%) was a bit higher compared to IFX (27%). Patients treated with CsA showed a higher rate of renal side effects compared to IFX. One patient had a moderate allergic reaction after the first use of CsA, which led to CsA discontinuation. Besides that, we did not observe any other SAEs in the CsA group and no death was recorded within 1 year. The other AEs were mild and moderate. After switching rescue therapy from CsA to IFX as a second rescue therapy within 1 month, except 1 patient with recurrent symptomatic epilepsy, we did not find any SAEs during 1 year of follow-up.

Additionally, we also evaluated AEs and SAEs from patients who switched to secondary rescue therapy over 1 month after initial rescue therapy. Beside some infections (CMV, C. difficile, HSV), we observed 1 patient with pancreatitis who received a secondary rescue therapy with IFX, which was most likely associated to the concomitant mesalazine therapy and had a benign course of disease after stopping it.

During the entire follow-up, 4 patients deceased, with 3 patients dying more than 6 years after rescue therapy and as far as known not directly associated with ASUC. One patient (75 aged male) in the IFX group died shortly after hospital discharge. The cause of death was not clearly determined but might be related to a severe course of ASUC.

A woman aged 35 years who was treated with CsA as rescue therapy died from a metastatic rectal cancer 6 years after enrolment in the trial. A male aged 87 years, who got colectomy after rescue therapy with CsA, died from respiratory failure due to end-stage interstitial lung disease and a cerebrovascular infarction 9 years after inclusion. A multimorbid male (diabetes type 1, chronic kidney failure, ischemic cardiopathy, periphery artery disease), aged 64 years and exposed to IFX died from unknown reasons nearly 7 years after enrolment.

Mortality was not significantly different between the groups during the follow-up period (p = 0.7996). All the AEs and SAEs are presented in online supplementary Table 3.

This real-life data of patients with ASUC evaluated the long-term outcome of CsA and IFX as rescue therapy in a single-center study. We confirmed an equal efficacy and safety of both regimens over a period of 5 years. Despite the low numbers of patients with sequential therapy, we could demonstrate that a second rescue therapy may be an option in selected patients without increasing their risk of complications. Patients having had more immunosuppressive therapies in the past, had a lower chance of being colectomy-free after 1 year.

The American College of Gastroenterology (ACG) und the European Crohn’s and Colitis Organization (ECCO) recommend in patients with ASUC the use of rescue therapy with IFX or CsA after inadequate response to intravenous corticosteroids. To date, there is currently no general recommendation in favor of one of those two rescue therapies. The optimal selection should be individualized considering possible contraindications and adapted to the local expertise with each drug [19, 20].

Our real-world data confirmed that colectomy-free survival is not influenced by the initial monotherapy (CsA or IFX) in patients with steroid-refractory ASUC (p = 0.149). These results are consistent with the findings of the prospective study from Laharie et al. [21] and the CONSTRUCT trial [6, 22]. The more favorable outcome after 5 years of treatment observed in other studies [21, 23] may be explained by the high rate of biologic-experienced and immunomodulator experienced patients in our study.

We showed that patients with a higher number of previous immunosuppressive therapies were less likely to be colectomy-free at 1 year (p = 0.049). This finding is consistent with the trial of Nasuno et al. [24], showing that patients with previous immunosuppressive therapies generally tend to be more difficult to treat and therefore may be at higher risk for colectomy. Interestingly, neither sex, disease duration nor albumin at hospital admission were significant variables in predicting colectomy at 1 year. Grazie et al. [10] did not find an increased risk for early colectomy associated with albumin, CRP or hemoglobin either. Nasuno et al. [24] reported that female gender was a prognostic factor for good outcomes in patients treated with IFX. In opposite, other studies [23, 25] did not demonstrate a significant effect of sex on colectomy rate. The role of sex on the clinical course of ulcerative colitis remains controversial after all and further long-term studies are needed.

There is still a lack of long-term data about the outcome of sequential rescue therapy in ASUC. Protic et al. [15] demonstrated in a multicenter study with sequential rescue therapies a decrease of the overall colectomy rate by 18% at 1 year. Furthermore, Narula et al. [26] showed in their systematic review a short-term treatment response of sequential therapy of 62.4% without any major safety concerns. Since we could not demonstrate any trend against more complications, our data support the use of sequential therapy in selected patients.

All patients treated with a second rescue therapy were colectomy-free after 1 year (6/6). At least 4 patients remained colectomy-free during the follow-up for at least 3 years. Since there were 2 patients with a lost to follow-up, we are not aware of the outcome of the remaining 2 patients. However, since our clinic is a tertiary clinic, most patients with ASUC and need of colectomy in our area will be transferred to our hospital. Therefore, we hypothesize that these two lost to follow-up patients have not had a colectomy in the meantime.

Due to the small study population in this group, our findings should be interpreted with caution. Nevertheless, these results can be considered positive and support the use of sequential rescue therapy in individual cases. Since patients who need a sequential therapy certainly belong to a difficult-to-treat population, we would expect a higher colectomy rate simply due to this selection bias. Nevertheless, our data tends to indicate the opposite. However, there is still a need for high-quality evidence with long-term data; sequential therapy should only be given at experienced referral centers with careful monitoring.

Regarding the safety profiles both drugs were comparable and most AEs were mild and transient. Except one death, the other three deaths were observed at least 6 years after initiation of the rescue therapy and not directly associated with ASUC. During the entire follow-up, mortality was not significantly different between the groups (p = 0.7996).

Our study has several strengths. The primary outcome (colectomy-free interval) was clearly defined at the beginning of this study and was not influenced by observer variation. The follow-up duration (median of 5.1 years) was longer compared to other studies, which is important for the evaluation of long-term outcomes for both rescue therapies.

The retrospective and non-randomized study design offers certain limitations. The patients were included in the study over a longer period (up to 11 years), which makes a direct comparison of the groups challenging specifically also in view of the rapidly evolving field including emergence of new therapeutic options, such as for instance tofacitinib. Due to our small sample size, the post hoc analysis should be interpreted with caution.

Since these are real-life data from a tertiary center, we do have some patients with lost to follow-up. However, although our clinic is a tertiary clinic and therefore most patients with ASUC and need of colectomy in our area will be transferred to our hospital, patients who were lost to follow-up were not counted in the colectomy-free group.

Since in our primary analysis none of the patients received CsA after IFX, we cannot generalize the safety of a sequential therapy to this specific group of patients. Furthermore, all our patients with sequential therapy had a favorable disease course making a prediction of who might benefit the most of a sequential therapy, not possible. Since nearly all biologic-experienced patients received IFX in the past, we could not compare anti-TNF versus non-anti-TNF experienced patients regarding colectomy-free survival.

In conclusion, our data demonstrate that colectomy-free survival is not dependent from initial monotherapy (CsA or IFX) in patients with steroid-refractory ASUC and the mortality did not significantly differ during the entire follow-up. In selected patients, a sequential therapy may be used without a higher risk of complications.

This study was performed in accordance with the Declaration of Helsinki. This human study was approved by Ethics Committee of the Canton of Zurich – Approval No: BASEC-Nr. 2021-01953. Since 2016 we have requested patients with general consent for further use of the data at our hospital and the written informed consent was given as part of the hospital stay.

P.S.: consulting fees from Falk Pharma, Takeda, Sanofi, Abbvie, Janssen-Cilag, Ferring, Pfizer, Galapagos, Lilly, unrestricted grant from Falk Pharma. L.B.: consulting fees, honoraria or travel support from Abbvie, Amgen, Bristol Myers Squibb, Esocap, Falk, Janssen, Lilly, Pfizer, Sanofi, and Takeda. G.R.: grants, consulting fees or honoraria from Abbvie, Ardeypharm, Arena, Augurix, Bristol Myers Squibb, Boehringer, Calypso, Celgene, Falk, Ferring, Fisher, Flamentera, Genentech, Gilead, Janssen, Lilly, MSD, Novartis, Pfizer, Phadia, Roche, UCB, Takeda, Tillots, Vifor, Vital Solutions, USB and Zeller. G.R. was a member of the journal’s Editorial Board at the time of submission. A.K.: honoraria or travel support from Amgen, Janssen-Cilag and Takeda and participation on a data safety board or advisory board for Takeda. G.N.: lecturing fees, consulting fees, and advisory fees from AbbVie, Merck Sharp and Dohme, Takeda, Janssen-Cilag, Pfizer, Gilead, Galapagos, Ferring, Vifor, Astro Pharma, Falk, Bristol Myers Squibb, Arena Pharmaceuticals, and Amgen.

This study was not supported by any sponsor or funder.

F.G. and P.S.: organization, data interpretation, and wrote the main manuscript text. P.S. and G.R. conceived and designed the project. I.H.: Statistical data analysis, prepared online supplementary Figure 1 and online supplementary Tables 1 and 2. G.R., L.B., G.N., and A.K. discussed the results and reviewed the manuscript. All authors read and approved the final manuscript.

The data that support the findings of this study are not publicly available due to their containing information that could compromise the privacy of research participants, but are available from the corresponding authors (F.G. and P.S.) upon reasonable request.