Efficacy and safety of induction therapy with calcineurin
inhibitors followed by vedolizumab maintenance in 71 patients
with severe steroid-refractory ulcerative colitis

Summary
Background: Following induction therapy with a calcineurin inhibitor (CNI) in severe
ulcerative colitis, transitioning to vedolizumab as maintenance therapy could be an
option.
Aim: To report on the largest cohort of patients successfully induced with CNIs who
were transitioned to vedolizumab maintenance therapy.
Methods: A retrospective observational study of adult patients with severe ster￾oid-refractory ulcerative colitis. Patients were included if they were induced with
a CNI followed by maintenance therapy with vedolizumab between January 2014
and December 2018. The primary endpoint was colectomy-free survival. Secondary
endpoints included survival without vedolizumab discontinuation as well as clinical,
steroid-free and biochemical remission at week 14.
Results: A total of 71 patients (59% male) were treated with vedolizumab after in￾duction therapy with CNIs for severe steroid-refractory colitis. Patients were fol￾lowed for a median time of 25 months (IQR 16-36). Colectomy-free survival rates
from vedolizumab initiation were 93% at 3 months, 67% at 1 year and 55% at 2 years.
At the end of induction with vedolizumab at week 14, 50% of patients were in clinical
remission, and 62% of patients had a normal CRP. At 1 and 2 years following vedoli￾zumab initiation, 43% and 28% of patients were still on vedolizumab respectively.
Vedolizumab was dose escalated to infusions every 4 weeks in 44% of patients. The
median time to dose escalation was 5.6 months (IQR 4.1-8.2). No serious adverse
events were recorded in our patient cohort.
Conclusions: Transitioning to vedolizumab following induction of remission with
CNIs is effective and safe.
2  |    OLLECH et al.
1 | INTRODUCTION
Ulcerative colitis is a chronic inflammatory disease of the colon with
varying degrees of disease severity as defined by several clinical,
biochemical and endoscopic parameters.1,2
The options for the medical management of patients with se￾vere steroid-refractory ulcerative colitis are limited and include
calcineurin inhibitors (CNIs) or infliximab,1-6 with infliximab being
the predominant agent used in such a setting, due to its ease of ad￾ministration and familiarity with its use by prescribing physicians.7
Patients who are successfully induced with infliximab continue to
maintenance therapy with this drug. Secondary to its increased use
in ulcerative colitis, patients failing infliximab are becoming more
prevalent.8
While highly effective at inducing remission in patients with
ulcerative colitis,3,5 protracted use of CNIs is limited by adverse
events, including infection, nephrotoxicity, hypercholesterolaemia
and hypertension. Consequently, their use in inflammatory bowel
disease has been limited to induction therapy. Patients who are suc￾cessfully induced with CNIs in the setting of severe steroid-refrac￾tory ulcerative colitis are bridged to azathioprine for maintenance
therapy.1,2 Previously, patients who had failed or were intolerant to
thiopurines were not considered candidates for calcineurin therapy
due to the absence of effective maintenance therapy.
Vedolizumab is an effective and safe medication approved
for induction and maintenance therapy in ulcerative colitis.9,10
Vedolizumab’s impressive safety data10 makes it an ideal candi￾date agent for use as maintenance therapy in combination with the
fast-acting CNIs as induction therapy. Our group previously reported
our preliminary data on the potential safety and efficacy of this se￾quential combination regimen.11 That study, however, was limited by
the small number of patients and heterogeneous population included
(11 patients with ulcerative colitis and nine patients with Crohn’s dis￾ease) and as such encouraged further larger observational cohorts
using this treatment strategy. In this study, by enrolling a significant
number of subsequent patients with steroid-refractory ulcerative
colitis, we are now able to cumulatively report on long-term safety
and efficacy in a more homogenous population.
We describe the largest cohort of patients with steroid-refrac￾tory ulcerative colitis successfully induced with CNIs who were then
transitioned to vedolizumab maintenance therapy and show that
such a management strategy is effective and safe over a long fol￾low-up period.
2 | METHODS
2.1 | Study design and patients
We performed a retrospective single-centre observational study of
adult ulcerative colitis patients followed at the University of Chicago
Inflammatory Bowel Disease Center, a large tertiary referral centre.
All patients with severe steroid-refractory ulcerative colitis who
received a CNI (ciclosporin or tacrolimus) as induction therapy fol￾lowed by maintenance therapy with vedolizumab between January
2014 and December 2018 were included. All patients responded to
IV ciclosporin or oral tacrolimus and received at least one vedoli￾zumab infusion while on calcineurin therapy.
2.2 | Medications
All patients were treated with either IV ciclosporin or oral tacrolimus.
Ciclosporin was given as a continuous infusion at an initial dose
of 2-4 mg/kg/d, aiming for serum trough levels of 300-400 ng/mL.
In case of response (decrease of bowel movement frequency by 50%
with the absence of haematochezia), a switch to an oral formula￾tion was performed using a total daily dose equivalent to twice the
24 hours intravenous dose.
Tacrolimus was started orally at 0.1 to 0.2 mg/kg/d, targeting
a blood concentration of 10-15 ng/g. The choice of the CNI used,
ciclosporin or tacrolimus, was made according to the attending phy￾sician discretion.
In patients who responded to calcineurin induction, vedolizumab
was administered as 300 mg infusions with standard loading doses;
weeks 0, 2 and 6, and then every 8 weeks. In case of continued
clinical activity or objective evidence of continued inflammation
(laboratory or endoscopic), vedolizumab could be prescribed every
4 weeks following induction. Following the induction infusions of
vedolizumab the CNI was tapered off. In the case of vedolizumab
failure, subsequent treatments were documented, as well.
2.3 | Data collection
At inclusion, the following characteristics were recorded for each
patient: sex, age, disease duration, smoking status, disease extent
according to the Montreal classification, prior received treatments
(steroids, thiopurine, methotrexate, infliximab, adalimumab, goli￾mumab), endoscopic activity measured with the endoscopic Mayo
subscore, C-reactive protein (CRP), haemoglobin and albumin levels.
All patients included had a clinic follow-up of at least 3 months.
2.4 | Endpoints
The primary endpoint was colectomy-free survival. Secondary end￾points included survival without vedolizumab discontinuation as well
as clinical, steroid-free, biochemical remission at week 14 as well as
time to endoscopic remission and rate of endoscopic remission at
12 months. Clinical remission was defined as the absence of blood
in the stools and <3 stools per day with the lack of abdominal pain.
Biochemical remission was defined by a normal CRP level (<5 mg/L).
Endoscopic remission was defined as an Mayo endoscopic score of
0 or 1. In addition, all adverse events were described. The study was
approved by the University of Chicago Institutional Review Board.
   | OLLECH et al.  3
2.5 | Statistical analysis
Descriptive statistics for demographic and clinical characteristics
include median (IQR) for continuous variables and frequency dis￾tributions for categorical data. Kaplan-Meier curves were gener￾ated for time-to-event data, that is, time from first vedolizumab
infusion until colectomy or cessation of vedolizumab treatment.
Patients who did not have a colectomy or who were still on ved￾olizumab were censored as of the date of the last follow-up.
Cox regression models were fit to examine the effects of differ￾ent covariates on time to colectomy and vedolizumab cessation.
Covariates analysed included age, sex, smoking status, disease
extent, pre-vedolizumab treatments, clinical and biochemical re￾mission at week 14, as well as haemoglobin, CRP and albumin. Due
to the limited number of events, models for time to colectomy or
vedolizumab cessation included only one covariate at a time to
avoid overfitting.
3 | RESULTS
3.1 | Patients
A total of 71 patients (59% male) were treated with vedolizumab after
induction therapy with CNIs for severe steroid-refractory ulcera￾tive colitis. Eighty-five per cent (n = 60) of patients had previously
been exposed to TNF antagonists’ medications. Most had extensive
disease (72%); the median disease duration was 44.1 months (IQR
15.2-115). Truelove and Witts criteria for acute severe ulcerative co￾litis (ASUC) were present in 76% (n = 54) of patients. Moderate to
severe endoscopic disease was present in 97% of patients. Patients
were followed for a median time of 25 months (IQR 16-36).
3.2 | Study medications
Sixty-eight per cent of patients (n = 48) received ciclosporin, and 32%
(n = 23) of patients received tacrolimus. The CNI was continued for
a median duration of 3.5 months (IQR 2.4-5.4). The majority (79%,
n = 55) were induced with CNIs as inpatients, with ciclosporin being
the drug of choice in most inpatients (88%, n = 48). Vedolizumab
was started after a median of 29 days (IQR 16-44) from the initia￾tion of the CNI. Vedolizumab was dose escalated to infusions every
4 weeks in 44% of patients. The median time to dose escalation was
5.7 months (IQR 4.1-8.2). Table 1 describes the baseline characteris￾tics of the study cohort allocated according to the CNI used.
3.3 | Efficacy
Thirty patients (42%) underwent colectomy during the follow￾up period. Colectomy-free survival rates from vedolizumab ini￾tiation were 93% at 3 months, 67% at 1 year and 55% at 2 years
(Figure 1A). Only lack of clinical remission at week 14 was asso￾ciated with colectomy (P = 0.023). There was no significant dif￾ference in colectomy rates between anti-TNF naïve patients and

TABLE 1 Baseline characteristics:
allocation according to CNI type—
ciclosporin and tacrolimus
4  |    OLLECH et al.
anti-TNF experienced patients, P = 0.79 (Figure 1B). Likewise,
choice of the CNI used to induce remission was not associated
with colectomy rates (P = 0.91). Patients meeting Truelove and
Witts criteria for ASUC had numerically higher colectomy rates
at 3 months and 1 year when compared to patients not meeting
these criteria (9.5% and 37.1% vs 6.25% and 19.65% respectively),
but this difference was not statistically significant (P = 0.8). Table 2
describes the baseline characteristics of patients with and without
ASUC.
At the end of induction with vedolizumab at week 14, 50% of
patients were in clinical remission, and 62% of patients were in bio￾chemical remission with a normal CRP. Steroid-free remission rates
were 27%, 43% and 76% at 3 months, 12 months and 24 months
respectively.
Sixty-three per cent of patients were still on vedolizumab
6 months after induction. After 1 and 2 years, 43% and 28% of pa￾tients were still on vedolizumab respectively (Figure 1C). Both lack
of clinical and biologic remission at week 14 were associated with
vedolizumab discontinuation (P = 0.0003 and P = 0.003 respec￾tively). Previous TNF antagonist therapy was not associated with
an increased rate of vedolizumab discontinuation (P = 0.86). Data
regarding endoscopic remission were available for 44 participants
(61.9% of the study cohort), Twenty-one patients (48%) reached the
endpoint of endoscopic remission during follow-up. The cumulative
rate of endoscopic remission at 12 months was 20.9%.
3.4 | Drug therapy following discontinuation of
vedolizumab
Thirty-one patients (44%) were transitioned to another drug after
failing vedolizumab maintenance therapy. Seventeen patients were
transitioned to an anti-TNF (71% infliximab); of these 17 patients, 15
(88%) had previously been on anti-TNF therapy, and 12 (71%) of these
patients averted colectomy at the end of follow-up. Eleven patients
were transitioned to tofacitinib. Eight (73%) of these patients had pre￾viously been on anti-TNF therapy, and six (55%) averted colectomy
at the end of follow-up. Three patients were transitioned to usteki￾numab. Two of these patients were previously on anti-TNF therapy,
and two patients averted colectomy at the end of follow-up. Of these
31 patients who failed vedolizumab and transitioned to another drug,
20 (65%) averted colectomy at the end of follow-up. Colectomy-free
survival times were similar after excluding patients who received
other biologics after vedolizumab failure (Figure 1A). The median time
to switch in therapy was 7.1 months (IQR 3.7-14.3) from vedolizumab
initiation.
3.5 | Adverse events
Eighteen patients (25.4%) experienced adverse events. No mortality
events were recorded during the follow-up period. In addition, all
adverse events documented occurred while on CNIs, with none of
the adverse events leading to discontinuation of the drug. The most
common adverse event was acute kidney injury in eight patients
(11.3%) (Table 3).
4 | DISCUSSION
In this study, we have shown that patients with severe steroid-re￾fractory ulcerative colitis induced with CNIs and then transitioned
to maintenance therapy with vedolizumab avoided colectomy in 67%
of cases after 12 months and in 55% of cases after 2 years. Such a
FIGURE 1 A, Kaplan Meier curve of colectomy free survival
for all patients and after excluding patients who received other
biologics after vedolizumab failure. B, Kaplan Meier of colectomy
free survival for patients exposed and not exposed to TNF
antagonists. C, Kaplan Meier curve of time to vedolizumab
discontinuation

   | OLLECH et al.  5
strategy was well tolerated in our cohort, with relatively few adverse
events.
The strength of this study includes its large cohort of patients,
larger than all previously reported studies combined, as well as the
long median follow-up of over 2 years.
There have been two previous studies11,12 and a case series of
two patients13 that demonstrated the feasibility of inducing remis￾sion with ciclosporin followed by maintenance of remission with
vedolizumab. However, both studies had limitations, including small
patient numbers and limited follow-up time. The earlier study from
our group by Christensen et al11 reported on nine patients with
Crohn's disease and 11 patients with ulcerative colitis treated with
combination therapy of vedolizumab with either ciclosporin or tac￾rolimus, and found that by week 14 of treatment, four (44%) patients
with Crohn's disease and six (55%) patients with ulcerative colitis
were in clinical remission.
The second study by Pellet G et al12 was a multicentre retrospec￾tive study from France. In this study, information was collected on
39 patients with steroid-refractory UC. The authors reported that
after a median follow-up period of 11 months, 11 patients (28%) un￾derwent colectomy and that patients meeting Truelove and Witts
criteria for ASUC had higher colectomy rates when compared to
patients who did not meet these criteria. While both studies re￾ported on the feasibility of combination treatment with CNIs and
vedolizumab to induce and maintain remission, larger observational
studies are needed to better explore the efficacy and safety of such
a strategy before advocating for its use in everyday practice.
In our large patient cohort with prolonged follow-up time, we
have shown the safety and efficacy of such a treatment strategy. We
chose colectomy-free survival as our primary endpoint as colectomy
is an unambiguous event in a retrospective cohort series. Indeed, our
study showed that in this very sick cohort of patients, a substantial
proportion of patients were able to avoid colectomy in the short and
long term by inducing patients with CNIs and transitioning to vedol￾izumab as maintenance therapy. Unlike the study by Pellet G et al12
in our patient cohort, there was no significant difference between
patients who met Truelove and Witts criteria for ASUC and those
that did not in regards to colectomy free survival (P = 0.8). However,
patients with ASUC did have numerically higher colectomy rates at
3 months and 1 year when compared to those without ASUC (9.5%
and 37.1% vs 6.25% and 19.65% respectively). The study was likely
underpowered to detect a difference, especially as the group with￾out ASUC was relatively small at only 17 patients. It is also possible
that as this study only included patients who had already responded
to ciclosporin induction that colectomy rates between groups might
have been attenuated.
We also looked at the time to vedolizumab discontinuation, and
we have shown that despite a significant number of patients failing

6  |    OLLECH et al.
vedolizumab over time, patients were able to avoid colectomy in the
short and long term. In our cohort, after 2 years of therapy, 72% of
patients had stopped vedolizumab therapy. Of note, in the pivotal
GEMINI-1 clinical trial,9
in patients who responded to vedolizumab
induction, mucosal healing at 1 year was 50%, and steroid-free
remission was 31% at 1 year. Likewise, in the recently published
VARSITY trial,14 steroid-free remission at 1 year was 12.6%. Similar
data for other biologic therapy over time exists as well; for example,
in the ACT-1 study of infliximab for moderate to severe UC,15 remis￾sion rates at 1 year were only 20%.
At the University of Chicago, patients were treated in a ‘treat
to target’ management strategy consistent with the ‘Selecting
Therapeutic Targets in Inflammatory Bowel Disease’ Consensus
statement.16 Over time, even if patients averted colectomy, patients
who still had clinical disease with endoscopic activity or were corti￾costeroid dependent were switched to other therapies. It is plausible
that over a long follow-up period, a substantial number of patients
will not reach management targets and will be switched to other
therapies as these become available and treatment targets become
more stringent. Patients who discontinued vedolizumab were able
to successfully transition and regain response with other therapies,
mainly tofacitinib and infliximab.
Interestingly, many patients transitioned to anti-TNFs after fail￾ing vedolizumab maintenance therapy were previously treated with
an anti-TNF. Due to the retrospective nature of our study, we were
unable to document the reason why anti-TNF therapy was stopped
for a large proportion of these patients. Some patients had a dis￾tant history of anti-TNF use, and, as many of these patients were
referred from outside institutions, no further data were available to
us in order to clarify the reason behind anti-TNF discontinuation.
Likewise, due to the small number of patients who transitioned to
further therapy with anti-TNFs and the lack of historical data for
many of these patients, these results, while interesting, should be
exploratory. Further studies should be performed looking into the
possibility of ‘circling back’ to prior therapies in order to establish the
efficacy of such a treatment strategy.
Inducing remission with CNIs and transitioning to vedolizumab
maintenance therapy is becoming more relevant as more patients
with ulcerative colitis now have a history of previous exposure and
failure to anti-TNFs and azathioprine.8
Likewise, since the publica￾tion of the CYSIF and CONSTRUCT trials,17,18 which demonstrated
similar short-term response rates between infliximab and ciclo￾sporin in the setting of acute severe ulcerative colitis, infliximab
has emerged as the predominant agent used in such patients,7
and
more patients are presenting to tertiary care centres after failing
infliximab.8
A potential reason for nonresponse to infliximab is
secondary loss of serum proteins due to monoclonal antibodies
loss through an inflamed gut19 and in such patients, the use of
a nonprotein-based therapy for induction of remission, such as a
CNI, could be useful.
The current study adds confidence to using the approach of
inducing remission with CNIs—which are potent and fast-acting
drugs with proven efficacy in treating patients with severe ulcer￾ative colitis3,5-8,17,18,20 and then transitioning to the slower acting
steroid-sparing agentvedolizumab, for maintenance therapy.9,10 We
have shown that overlapping these two drugs is effective and safe.
This study adds to the armamentarium of therapeutic interventions
available to physicians treating patients with severe steroid-refrac￾tory ulcerative colitis.
There are several limitations to our study, mainly linked to its
retrospective single-centre setting with the inherent risk of bias,
and incomplete data for some patients. Treatments were not stan￾dardised, and the side effects of drugs may have been underesti￾mated. It should also be emphasised that patients were treated by
expert physicians in a large referral centre.
In conclusion, induction of remission with CNIs with a transi￾tion to vedolizumab is effective and safe and leads to avoidance
of colectomy in a substantial subgroup of patients over a long
follow-up period. Such a treatment strategy might be considered
in patients with steroid-refractory colitis, especially if they had
previously failed either anti-TNFs or thiopurines. Such a strategy
enables the introduction of safe protein-based therapies such as
vedolizumab following stabilisation and induction of remission
with CNIs.
ACKNOWLEDGEMENTS

swelling
Tacrolimus + Vedolizumab 1 (1.4)
   | OLLECH et al.  7
with critical revisions by SD, IN, BC, and AS. All authors contributed
to the acquisition of the data. JEO, VR, PHS, RDC, NP, and DTR con￾tributed to the analysis and interpretation of data.
All authors approved the final version of the manuscript.
ORCID
Britt Christensen https://orcid.org/0000-0002-8746-4275
Atsushi Sakuraba https://orcid.org/0000-0003-2519-6129
David T. Rubin https://orcid.org/0000-0001-5647-1723
REFERENCES
1. Rubin DT, Ananthakrishnan AN, Siegel CA, Sauer BG, Long MD.
ACG clinical guideline: ulcerative colitis in adults. Am J Gastroenterol.
2019;114:384-413.
2. Harbord M, Eliakim R, Bettenworth D, et al. Third European evi￾dence-based consensus on diagnosis and management of ul￾cerative colitis. Part 2: current management. J Crohns Colitis.
2017;11:769-784.
3. Lichtiger S, Present DH, Kornbluth A, et al. Cyclosporine in se￾vere ulcerative colitis refractory to steroid therapy. N Engl J Med.
1994;330:1841-1845.
4. Järnerot G, Hertervig E, Friis-Liby I, et al. Infliximab as rescue ther￾apy in severe to moderately severe ulcerative colitis: a randomized,
placebo-controlled study. Gastroenterology. 2005;128:1805-1811.
5. Wu B, Tong J, Ran Z. Tacrolimus therapy in steroid-refractory ul￾cerative colitis: a review. Inflamm Bowel Dis. 2019. https://doi.
org/10.1093/ibd/izz068 [Epub ahead of print].
6. Hoffmann P, Wehling C, Krisam J, Pfeiffenberger J, Belling N, Gauss
A. Performance of tacrolimus in hospitalized patients with ste￾roid-refractory acute severe ulcerative colitis. World J Gastroenterol.
2019;25:1603-1617.
7. Sjöberg M, Walch A, Meshkat M, et al. Infliximab or cyclosporine as
rescue therapy in hospitalized patients with steroid-refractory ul￾cerative colitis: a retrospective observational study. Inflamm Bowel
Dis. 2012;18:212-218.
8. Weisshof R, Ollech JE, El Jurdi K, et al. Ciclosporin therapy after
infliximab failure in hospitalized patients with acute severe colitis is
effective and safe. J Crohns Colitis. 2019;13:1105-1110.
9. Feagan BG, Rutgeerts P, Sands BE, et al. Vedolizumab as induc￾tion and maintenance therapy for ulcerative colitis. N Engl J Med.
2013;369:699-710.
10. Colombel J-F, Sands BE, Rutgeerts P, et al. The safety of vedolizumab
for ulcerative colitis and Crohn's disease. Gut. 2017;66:839-851.
11. Christensen B, Gibson PR, Micic D, et al. Safety and efficacy of
combination treatment with calcineurin inhibitors and vedolizumab
in patients with refractory inflammatory bowel disease. Clin
Gastroenterol Hepatol. 2019;17:486-493.
12. Pellet G, Stefanescu C, Carbonnel F, et al. Efficacy and safety of
induction therapy with calcineurin inhibitors in combination with
vedolizumab in patients with refractory ulcerative colitis. Clin
Gastroenterol Hepatol. 2019;17:494-501.
13. Szanto K, Molnar T, Farkas K. New promising combo therapy in
inflammatory bowel diseases refractory to anti-TNF agents: cyclo￾sporine plus vedolizumab. J Crohns Colitis. 2018;12:629.
14. Sands BE, Peyrin-Biroulet L, Loftus EV, et al. Vedolizumab versus
adalimumab for moderate-to-severe ulcerative colitis. N Engl J Med.
2019;381:1215-1226.
15. Rutgeerts P, Sandborn WJ, Feagan BG, et al. Infliximab for induc￾tion and maintenance therapy for ulcerative colitis. N Engl J Med.
2005;353:2462-2476.
16. Peyrin-Biroulet L, Sandborn W, Sands BE, et al. Selecting ther￾apeutic targets in inflammatory bowel disease (STRIDE): deter￾mining therapeutic goals for treat-to-target. Am J Gastroenterol.
2015;110:1324-1338.
17. Williams JG, Alam MF, Alrubaiy L, et al. Infliximab versus ciclosporin
for steroid-resistant acute severe ulcerative colitis (CONSTRUCT):
a mixed methods, open-label, pragmatic randomised trial. Lancet
Gastroenterol Hepatol. 2016;1:15-24.
18. Laharie D, Bourreille A, Branche J, et al. Ciclosporin versus inflix￾imab in patients with severe ulcerative colitis refractory to intra￾venous steroids: a parallel, open-label randomised controlled trial.
Lancet. 2012;380:1909-1915.
19. Brandse JF, van den Brink GR, Wildenberg ME, et al. Loss of in￾fliximab into feces is associated with lack of response to ther￾apy in patients with severe ulcerative colitis. Gastroenterology.
2015;149:350-355; e2
20. Ordás I, Domènech E, Mañosa M, et al. Long-term efficacy and
safety of cyclosporine in a cohort of steroid-refractory acute
severe ulcerative colitis patients Vedolizumab from the ENEIDA registry
(1989–2013): a Nationwide Multicenter Study. Am J Gastroenterol.
2017;112:1709-1718.
How to cite this article: Ollech JE, Dwadasi S, Rai V, et al.
Efficacy and safety of induction therapy with calcineurin
inhibitors followed by vedolizumab maintenance in 71
patients with severe steroid-refractory ulcerative colitis.