Sage Journals: Discover world-class research

Abstract

Background:

Vedolizumab is a monoclonal antibody approved for adults with moderately to severely active ulcerative colitis and an inadequate response/intolerance to conventional, or anti-tumor necrosis factor (TNF) treatment. Additional efficacy data for second-line vedolizumab after first-line anti-TNF treatment are required to determine its optimal position in the treatment sequence.

Objectives:

To evaluate efficacy and health-related quality of life (HRQoL) associated with second-line vedolizumab after first-line anti-TNF in patients with ulcerative colitis.

Design:

Meta-analysis.

Data sources and methods:

Data from randomized, placebo-controlled phase III trials (GEMINI-1 (NCT00783718) and VISIBLE-1 (NCT02611830)) were used. Patients who received second-line vedolizumab after first-line anti-TNF treatment within 5 years before trial enrollment were included. The primary outcome was clinical remission with two definitions based on the adapted Mayo score. Secondary outcomes were corticosteroid (CS)-free remission, Mayo endoscopic score (MES) ⩽1, and HRQoL (Inflammatory Bowel Disease Questionnaire and EuroQoL-5 Dimension Visual Analog Scale). Outcomes were assessed at Week 52. Efficacy of vedolizumab versus placebo was compared using fixed-effects models. Regression analyses assessed treatment effects, adjusting for disease location/duration and previous CS use.

Results:

Overall, 137 patients from GEMINI-1/VISIBLE-1 were included. Patient characteristics were similar across cohorts, although mean disease duration ranged from 6.6 to 9.9 years. Second-line vedolizumab was associated with a higher likelihood of achieving both definitions of clinical remission and all secondary outcomes (CS-free remission, MES ⩽1, and HRQoL) versus placebo. Regression analysis showed that patients treated with vedolizumab were 7.13–11.35 times more likely to achieve clinical remission (both definitions) versus placebo.

Conclusion:

Second-line vedolizumab after first-line anti-TNF treatment significantly increased the likelihood of achieving clinical remission, MES ⩽1, and improved HRQoL at Week 52 versus placebo in patients with ulcerative colitis. The small sample size, associated limitations, and heterogeneity observed in some outcomes warrant further research to strengthen these findings.

Plain language summary

A combined analysis of two studies to assess how well vedolizumab works in patients with ulcerative colitis, who first received anti-tumor necrosis factor treatment

Ulcerative colitis (UC) is a chronic condition that affects the lining of the large intestine. Many drugs are available for patients with UC to receive, including a type of drug called biologics. Anti-tumor necrosis factor (TNF) is often the first biologic given to patients with UC, known as first-line treatment. Some patients may require a different treatment if the first-line treatment doesn’t work well enough, or causes side effects. Vedolizumab is another biologic that can be used by patients with UC. More data are needed on vedolizumab as a second treatment, known as second-line, to find out the best order of biologics for patients with UC. This study assessed how well second-line vedolizumab worked, known as the efficacy, and its effect on patients’ quality of life. Data were gathered from two clinical trials of patients with UC who had previously received first-line anti-TNF treatment and went on to receive second-line vedolizumab. A process called a meta-analysis was used to combine the data from these studies and assess different outcomes. The main outcome assessed was whether patients’ disease improved after a year of treatment, known as clinical remission, compared with patients who received a placebo. The meta-analysis tested this outcome and adjusted for factors such as how long the patient had UC. A total of 137 patients were included. Patients in the vedolizumab treatment groups and placebo groups had similar characteristics. The meta-analysis showed that patients who received second-line vedolizumab were significantly more likely to reach clinical remission after 1 year of treatment than patients who received a placebo. Patients who received second-line vedolizumab also had small improvements in quality-of-life measures. These results provide evidence that can help guide decisions about the order of treatments, but further research is needed to strengthen the findings.

Graphical Abstract

Keywords

efficacy health-related quality of life inflammatory bowel disease meta-analysis treatment sequence/pattern ulcerative colitis vedolizumab

Introduction

Ulcerative colitis is a chronic inflammatory condition that affects the mucosa of the colon.¹ Treatment objectives include symptom control, clinical remission, endoscopic healing, reducing the risk of mid-term and long-term complications, improving health-related quality of life (HRQoL), and avoiding disability.^2–5 There is no consensus on first-line biologic treatment (after conventional therapies) in patients with moderately to severely active ulcerative colitis.^6,7 In recent years, multiple agents have been approved for the treatment of ulcerative colitis, many of which have efficacy data in the first-line setting (biologic-naïve) and later in the treatment sequence.^8–10 Anti-tumor necrosis factor (TNF) agents, such as adalimumab and infliximab, are frequently used in the first-line setting owing to extensive experience with these treatments in clinical practice and economic considerations.^6,10 However, efficacy data for second-line treatment after failure of one anti-TNF treatment are limited.^9,11,12

Vedolizumab is a monoclonal antibody that blocks α4β7 integrin, resulting in gut-selective anti-inflammatory activity.^13,14 Vedolizumab is indicated for the treatment of adult patients with moderately to severely active ulcerative colitis or Crohn’s disease¹⁵ who have had an inadequate response with, lost response to, or were intolerant to either conventional therapy or an anti-TNF.¹⁶ This use is supported by the European Crohn’s and Colitis Organisation Guidelines on Therapeutics in Ulcerative Colitis.³ The efficacy of vedolizumab in patients with ulcerative colitis has been demonstrated in several phase III studies. In the GEMINI-1 trial (NCT00783718),¹³ vedolizumab showed improved response rates versus placebo at Week 6 (induction; adjusted difference 21.7 percentage points, p < 0.001) and Week 52 (maintenance; adjusted difference 26.1 percentage points, p < 0.001 for vedolizumab every 8 weeks (Q8W) and 21.9 percentage points for vedolizumab every 4 weeks (Q4W)). Further, a similar rate of adverse events was observed with vedolizumab compared with placebo in patients with active ulcerative colitis who had a documented unsuccessful previous treatment (glucocorticoids, immunosuppressive medications, or anti-TNF treatments).¹³ The VISIBLE-1 trial (NCT02611830)¹⁷ demonstrated that subcutaneous (SC) vedolizumab is effective as maintenance therapy in patients with moderately to severely active ulcerative colitis, who had a clinical response to intravenous (IV) vedolizumab during 6-week induction treatment and an unsuccessful response to at least one previous treatment (corticosteroids (CSs), immunomodulator, or anti-TNF).¹⁷ Clinical remission was achieved by 46.2% and 42.6% of patients in the SC and IV vedolizumab cohorts, respectively, compared with 14.3% of patients in the placebo cohort.¹⁷

There is published evidence supporting the use of vedolizumab as a first-line biologic in patients with ulcerative colitis.^10,18 In contrast, the evidence for second-line vedolizumab after first-line treatment with one anti-TNF is limited.^10,12,19 For example, a double-blind, multicenter, randomized controlled trial of 151 patients with ulcerative colitis reported significantly higher rates of steroid-free clinical remission (p = 0.033) and endoscopic improvement (Mayo endoscopic score (MES) ⩽1; p = 0.027) with second-line vedolizumab versus infliximab after failure of first anti-TNF treatment.²⁰ In addition, previous trials examining the efficacy of vedolizumab have included both anti-TNF-naïve and -experienced patients (20.8%–52.2% of patients had prior anti-TNF treatment).^13,14,17,21 Further efficacy data for second-line vedolizumab after first-line treatment with one anti-TNF is needed to determine the optimal position of vedolizumab in the treatment sequence for patients with ulcerative colitis. To address this knowledge gap, we conducted a meta-analysis of efficacy and HRQoL outcomes associated with second-line vedolizumab after first-line treatment with one anti-TNF in patients with ulcerative colitis compared with placebo using data from the GEMINI-1 and VISIBLE-1 trials.

Methods

Study design and data sources

Patient-level data from two phase III, randomized, double-blind, placebo-controlled trials were pooled for this meta-analysis.^13,17 GEMINI-1 (NCT00783718),¹³ conducted in 34 countries from 2008 to 2012, assessed the efficacy and safety of induction and maintenance vedolizumab treatment in patients aged 18–80 years with moderately to severely active ulcerative colitis. During the maintenance phase, patients received vedolizumab 300 mg IV Q8W or Q4W, or placebo. The VISIBLE-1 (NCT02611830)¹⁷ trial, conducted across 29 countries from 2015 to 2018, investigated the efficacy and safety of maintenance vedolizumab for patients with moderate to severely active ulcerative colitis. During the maintenance phase, patients received either vedolizumab 108 mg SC every 2 weeks (Q2W), vedolizumab 300 mg IV Q8W, or placebo.

Eligibility for meta-analysis

The feasibility of pooling cohorts was assessed, considering the differences in the patient populations and study designs, as well as the doses and administration of vedolizumab in the GEMINI-1 and VISIBLE-1 trials (Figure S1). Patients from the GEMINI-1 trial who received vedolizumab 300 mg IV Q8W were considered eligible for inclusion; patients in the GEMINI-1 trial who received Q4W dosing were excluded because there was no Q4W dosing equivalent in the VISIBLE-1 trial. Patients from the VISIBLE-1 trial who received vedolizumab 108 mg SC Q2W were dose-equivalent to those who received vedolizumab 300 mg IV Q8W and, therefore, eligible for pooling. Non-randomized patients from both the GEMINI-1 and VISIBLE-1 trials were excluded. Patients in the VISIBLE-1 trial were not randomized until the maintenance phase; therefore, data from the induction phases of these studies were not analyzed. Data from the maintenance phases of the GEMINI-1 vedolizumab 300 mg Q8W cohort (compared with placebo) and the VISIBLE-1 vedolizumab 300 mg Q8W and 108 mg Q2W cohorts (both compared with placebo) were pooled.

Study population

Patients who met the trial-defined criteria of ulcerative colitis who received second-line vedolizumab after inadequate response, loss of response, or intolerance to first-line treatment with one anti-TNF within 5 years before trial enrollment (as per the GEMINI-1 protocol) were included in the meta-analysis. Patients were recorded as having prior anti-TNF treatment if any one of adalimumab, golimumab, or infliximab was documented in their medication history, and the end date of use was within 5 years before trial randomization. A secondary meta-analysis of patients with ulcerative colitis who had not received prior anti-TNF treatment (population with vedolizumab as a first-line biologic) was also performed in this study.

Study outcomes

The primary outcome was to assess the effect of second-line vedolizumab versus placebo on clinical remission at maintenance (Week 52). Two definitions of clinical remission, adapted from the Mayo score, were explored in this analysis. These definitions were based on the results of a post hoc analysis of the GEMINI-1 and VISIBLE-1 trials, in accordance with guidance from the US Food and Drug Administration that advised the removal of the physician’s global assessment from the full Mayo score to reduce subjectivity.^22–24 Definition 1 included a stool frequency subscore (SFS) ⩽1 with a ⩾1-point decrease in SFS from baseline, a rectal bleeding subscore (RBS) of 0, and an endoscopic subscore (ES) of ⩽1. Definition 2 included an SFS ⩽1, an RBS of 0, and an ES of ⩽1.

Secondary outcomes were based on those used in the GEMINI-1 and VISIBLE-1 trials.^13,17 These were assessed in patients who received vedolizumab versus placebo at Week 52 and included: CS-free clinical remission (defined as patients with clinical remission who, if they had received concomitant CSs at baseline, had discontinued CSs by Week 52); an MES ⩽1; and change in HRQoL from baseline (assessed via the EuroQoL 5 Dimension Visual Analog Scale (EQ-5D-VAS) and Inflammatory Bowel Disease Questionnaire (IBDQ)). The EQ-5D-VAS is a generic measure of quality of life consisting of a vertical scale from 0 (worst imaginable health state) to 100 (best imaginable health state).²⁵ The IBDQ is a disease-specific quality-of-life measurement, comprised of four domains and 32 items, rated on a seven-point scale, where higher scores indicate better quality of life.²⁶

All primary and secondary outcomes were also evaluated for patients who received vedolizumab as a first-line biologic. A scenario analysis was conducted exclusively for patients who received second-line vedolizumab who did not respond to, lost response to, or were intolerant to first-line anti-TNF treatment. Therefore, this scenario analysis excluded patients who had previously discontinued anti-TNF treatment for any other reason.

Meta-analysis and statistical analysis

Individual patient data meta-analyses were performed using a two-stage approach. A modified Poisson regression (i.e., with robust error variance) was used to estimate the trial-specific relative risks using a generalized linear model, with a Poisson family and log link.^27,28 To address the low number of events for some outcomes, Firth’s bias-reducing adjusted score equations were used. For continuous outcomes, generalized linear models with a Gaussian family and an identity link were used to estimate between-group mean differences. Mean differences were estimated using the measurement at baseline and Week 52. Only fixed-effects models were fit to compare the efficacy of vedolizumab with placebo because only two studies were included in the meta-analysis. Inverse variance weighting was used to estimate the average weighted effect size for the trials combined, and estimates of heterogeneity (Cochrane’s Q and I²) were reported. Cochrane’s Q was obtained using the Chi-squared test: a significant Q value indicates that heterogeneity was present. I² was interpreted according to the Cochrane Handbook for Systematic Reviews of Interventions,²⁹ where an I² of 0%–40% is considered unimportant, 30%–60% may indicate moderate heterogeneity, 50%–90% may encompass substantial heterogeneity, and 75%–100% suggests considerable heterogeneity is present. A regression analysis was also conducted to assess the relative risk of clinical remission in patients who received vedolizumab versus placebo adjusting for disease duration, disease location, and CS use in second-line patients. A modified Poisson regression was fit using a generalized linear model with a Poisson family and a log link. As per the analysis above, Firth’s bias-reduced adjusted score was used for analyses if the number of events was low. Non-responder imputation was carried out for missing data for binary outcomes, as per the original trial analyses. For continuous outcomes (EQ-5D-VAS and IBDQ scores), a complete case analysis was performed. R version 4.3.0³⁰ was used for all analyses including gglot2,³¹ haven,³² brglm2,³³ and metafor³⁴ packages. The meta-analysis was performed according to a pre-defined protocol and statistical analysis plan, neither of which was registered. This study was conducted and reported in accordance with the applicable Preferred Reporting Items for Systematic Reviews and Meta-Analyses.

Results

Patient characteristics

In total, 137 patients with ulcerative colitis from the GEMINI-1 and VISIBLE-1 trials were included in the second-line vedolizumab population. Patient characteristics were generally similar across cohorts (Table 1), although some differences were observed. Mean (SD) disease duration ranged from 6.6 (4.4) years in the GEMINI-1 vedolizumab Q8W cohort to 9.9 (10.1) years in the VISIBLE-1 placebo cohort. In the GEMINI-1 trial, most patients (vedolizumab Q8W: n = 39 (88.6%); placebo: n = 35 (94.6%)) had previously received infliximab as their prior anti-TNF therapy, whereas all patients in the VISIBLE-1 trial had previously received adalimumab.

Table 1.

Baseline characteristics of patients from the GEMINI-1 and VISIBLE-1 trials who received second-line vedolizumab after first-line treatment with one anti-TNF.

Characteristic	GEMINI-1		VISIBLE-1
Characteristic	Vedolizumab Q8W (n = 44)	Placebo(n = 37)	Vedolizumab Q8W and Q2W (n = 45)	Placebo(n = 11)
Age (years)
Mean (SD)	42 (11.81)	42 (13.65)	38 (13.63)	38 (12.45)
Sex, n (%)
Male	27 (61.4)	22 (59.5)	30 (66.7)	7 (63.6)
Race, n (%)
White	41 (93.2)	34 (91.9)	39 (86.7)	9 (81.8)
Body weight (kg)
Mean (SD)	81.2 (17.64)	81.4 (24.35)	74.6 (17.31)	63.7 (13.07)
Current smoker, n (%)
Yes	3 (6.8)	3 (8.1)	3 (6.7)	0 (0.0)
Duration of disease (years)
Mean (SD)	6.6 (4.38)	9.3 (7.92)	9.0 (7.11)	9.9 (10.07)
Mayo score
Mean (SD)	8.6 (1.83)	8.0 (1.67)	9.0 (1.41)	9.4 (1.29)
Site of disease, n (%)
Rectum and sigmoid colon only	5 (11.4)	1 (2.7)	1 (2.2)	1 (9.1)
Left side of colon	12 (27.3)	14 (37.8)	21 (46.7)	4 (36.4)
Proximal to the splenic flexure	5 (11.4)	7 (18.9)	3 (6.7)	1 (9.1)
All of the colon	22 (50.0)	15 (40.5)	20 (44.4)	5 (45.5)
Prior anti-TNF therapy, n (%)
Adalimumab	2 (4.5)	2 (5.4)	45 (100)	11 (100)
Infliximab	39 (88.6)	35 (94.6)	0 (0.0)	0 (0.0)
Golimumab	3 (6.8)	0 (0.0)	0 (0.0)	0 (0.0)
Reason for discontinuation, n (%)
Intolerance	9 (20.5)	6 (16.2)	2 (4.4)	0 (0.0)
Loss of response	16 (36.4)	11 (29.7)	10 (22.2)	1 (9.1)
Inadequate response	14 (31.8)	18 (48.6)	8 (17.8)	0 (0.0)
Missing data	5 (11.4)	2 (5.4)	25 (55.6)	10 (90.9)
Glucocorticoids only, n (%)
Yes	17 (38.6)	15 (40.5)	8 (17.8)	1 (9.1)
Immunosuppressants only, n (%)
Yes	2 (4.5)	6 (16.2)	1 (2.2)	0 (0.0)
Glucocorticoids and immunosuppressants, n (%)
Yes	9 (20.5)	8 (21.6)	36 (80.0)	10 (90.9)

Q2W, every 2 weeks; Q8W, every 8 weeks; SD, standard deviation; TNF, tumor necrosis factor.

In total, 285 patients with vedolizumab as a first-line biologic were included in the population analysis. Overall, the patient demographics and disease characteristics were similar to those of patients who received second-line vedolizumab; the main difference was a shorter mean (SD) duration of disease and different baseline therapies (Table S1).

Primary outcome

Second-line vedolizumab after first-line treatment with one anti-TNF was associated with a higher likelihood of achieving clinical remission (according to the adapted Mayo score definitions 1 and 2) at Week 52 in patients with ulcerative colitis (Figure 1). The estimated pooled effect size was 19.24 (95% confidence interval (CI): 11.94, 31.01) for definition 1 of clinical remission and 4.92 (95% CI: 1.92, 12.62) for definition 2. Definition 1 was associated with high heterogeneity (estimated I² of 90.38), and definition 2 was associated with moderate heterogeneity (estimated I² of 39.41). Pooled treatment effects for both definitions of clinical remission (Figure 1) demonstrated congruence between the studies. The treatment effect size for GEMINI-1 using either definition of clinical remission was higher than that for VISIBLE-1 (particularly evident for definition 1: 26.07 vs 2.53). The heavier weighting of the GEMINI-1 trial when using definition 1 suggests more within-trial variance for this outcome than with the VISIBLE-1 trial. The within-trial variance between the two trials was more closely aligned when using definition 2.

Figure 1.

Clinical remission at Week 52 associated with second-line vedolizumab versus placebo after first-line treatment with one anti-TNF. Definition 1 was an SFS ⩽1 with a ⩾1-point decrease in SFS from baseline, an RBS of 0, and an ES of ⩽1. Definition 2 was an SFS ⩽1, an RBS of 0, and an ES of ⩽1. Cochran’s Q was defined as the Chi-squared statistic testing the “no heterogeneity” hypothesis; a significant Q value indicates that heterogeneity is present. I² was defined as the total proportion of variation due to heterogeneity between trials; estimates exceeding 50% indicate that substantial heterogeneity is present.

Secondary outcomes

Treatment with second-line vedolizumab was associated with a significantly higher likelihood of achieving all secondary outcomes at Week 52, compared with placebo (Figure 2 and Table S2). The estimated study treatment effect was higher for GEMINI-1 versus VISIBLE-1 for all outcomes. For HRQoL outcomes, treatment with vedolizumab was associated with a statistically significant improvement of EQ-5D-VAS and IBDQ scores at Week 52 relative to baseline (standardized mean difference (SMD) 0.56 (95% CI: 0.08, 1.04) and 0.97 (95% CI: 0.54, 1.41), respectively). The estimated SMD relative to baseline for EQ-5D-VAS and IBDQ scores were similar between GEMINI-1 (0.50 and 1.18, respectively) and VISIBLE-1 (0.62 and 0.68, respectively).

Figure 2.

CS-free remission and MES ⩽1 at Week 52 for patients with ulcerative colitis who received second-line vedolizumab versus placebo after first-line treatment with one anti-TNF. CS-free remission was defined as patients who received CSs at baseline and who achieved clinical remission and discontinued concomitant CS use by Week 52.

Regression analyses

The regression analyses estimating the treatment effect of vedolizumab on clinical remission, adjusting for disease duration, disease location, and CS use, are presented in Table 2. At Week 52, patients who received second-line vedolizumab after first-line treatment with one anti-TNF were 11.35 times (95% CI: 3.25, 39.63) more likely to achieve definition 1 of clinical remission and 7.13 times (95% CI: 2.47, 20.60) more likely to achieve definition 2, compared with those who received placebo. For both definitions, patients who received CSs and had ulcerative colitis that did not affect the whole colon were less likely to achieve clinical remission at Week 52 compared with placebo. Disease duration did not affect the likelihood of clinical remission using either definition.

Table 2.

Effect of second-line vedolizumab, disease duration, CSs, and disease location on clinical remission at Week 52 in patients with ulcerative colitis.

Variable	Clinical remission definition 1		Clinical remission definition 2
Variable	RR	95% CI	RR	95% CI
Intercept	0.05	0.01, 0.18	0.07	0.02, 0.25
Treatment: vedolizumab	11.35	3.25, 39.63	7.13	2.47, 20.60
Disease duration	1.01	0.97, 1.05	1.01	0.97, 1.05
CS use: yes	0.82	0.49, 1.36	0.77	0.47, 1.25
Disease location: left-sided colitis	0.83	0.41, 1.66	0.95	0.48, 1.88
Disease location: pancolitis	0.53	0.25, 1.15	0.59	0.28, 1.24
Disease location: proctosigmoiditis	0.80	0.27, 2.36	0.81	0.28, 2.39

The reference population was patients on placebo, with no CS use, and with ulcerative colitis in all of the colon. The intercept is the estimated RR of the reference population. Missing data has been imputed as non-response for proportion-based endpoints. For continuous outcomes, the last post-baseline observation has been carried forward. Definition 1 was an SFS ⩽1 with a ⩾1-point decrease in SFS from baseline, an RBS of 0, and an ES of ⩽1. Definition 2 was an SFS ⩽1, an RBS of 0, and an ES of ⩽1. Treatment and CS use were defined as binary variables (1 for vedolizumab, 1 for CS use). Disease duration was defined as a continuous variable. Disease location was defined as a categorical variable with four levels.

CI, confidence interval; CS, corticosteroid; ES, endoscopic subscore; RBS, rectal bleeding subscore; RR, relative risk; SFS, stool frequency subscore.

Population with vedolizumab as a first-line biologic

For the primary outcome, treatment with vedolizumab as a first-line biologic was associated with a higher likelihood of achieving remission at Week 52 (Table S3). The pooled effect sizes for definitions 1 and 2 were 2.61 (95% CI: 1.72, 3.96) and 2.59 (95% CI: 1.74, 3.84), respectively. Study-specific treatment effects were similar for GEMINI-1 and VISIBLE-1 for both definitions of remission (Figure S2). Treatment with vedolizumab as a first-line biologic was also associated with a significantly higher likelihood of achieving all secondary outcomes at Week 52 (Figure S3) and, in contrast with the second-line population, the definition of clinical remission did not affect the pooled effect size. Heterogeneity was low for both definitions of remission and for secondary binary outcomes, indicating that the rates of achieving outcomes did not differ largely between the trials. As with the second-line population, treatment with vedolizumab as a first-line biologic was associated with a small but statistically significant improvement in EQ-5D-VAS and IBDQ scores at Week 52, relative to baseline (0.56 (95% CI: 0.27, 0.85; Q: 0.05; I²: 0.00) and 0.63 (95% CI: 0.35, 0.91; Q: 2.73; I²: 63.36), respectively).

Scenario analysis

Overall, 130 patients with ulcerative colitis who did not respond to, lost response to, or were intolerant to first-line anti-TNF treatment were included in the second-line vedolizumab scenario analysis. Only seven patients were excluded from the overall population; patient characteristics of this subgroup were therefore similar to the overall second-line population, with no large differences observed. Trends observed for primary and secondary outcomes were similar to those observed in the overall second-line population. For both definitions of clinical remission, treatment with vedolizumab was associated with a higher likelihood of achieving remission at Week 52 versus placebo (pooled effect size: 17.53 (95% CI: 10.63, 28.92) and 4.75 (1.85, 12.22) for definitions 1 and 2, respectively). Treatment with vedolizumab was associated with a significantly higher likelihood of achieving all secondary binary outcomes at Week 52 versus placebo (pooled effect size: 6.19 (95% CI: 3.06, 12.56) for CS-free remission; 4.12 (1.74, 9.77) for MES ⩽1). The estimated SMD for change in HRQoL outcomes was 0.56 (95% CI: 0.07, 1.05) for EQ-5D-VAS and 0.95 (95% CI: 0.51, 1.39) for IBDQ, indicating a small improvement from baseline.

Discussion

This meta-analysis of the GEMINI-1 and VISIBLE-1 trials demonstrates the efficacy of second-line vedolizumab after first-line treatment with one anti-TNF in patients with moderately to severely active ulcerative colitis compared with placebo. This analysis was specific to patients who received second-line vedolizumab after first-line anti-TNF treatment, and a larger sample size could be analyzed through the combination of GEMINI-1 and VISIBLE-1 study data. Within the context of a growing number of approved agents for patients with ulcerative colitis,^9,19 this meta-analysis provides efficacy data, thus contributing evidence for the optimal treatment sequence for patients with ulcerative colitis. Second-line vedolizumab after first-line treatment with one anti-TNF significantly increased the likelihood of achieving both definitions of clinical remission at Week 52 versus placebo. This is in line with reported vedolizumab efficacy among anti-TNF-naïve or broader anti-TNF-experienced populations. For example, a phase III randomized study reported significantly higher rates of clinical remission in the maintenance phase for patients with ulcerative colitis who received vedolizumab versus placebo.¹⁴ However, only a subset of patients had received prior anti-TNF, and the clinical remission rate was only numerically higher versus placebo in this subgroup.¹⁴ Another study, in which a proportion of patients (21%) had previous anti-TNF treatment, reported vedolizumab to be superior to adalimumab with respect to clinical remission and endoscopic improvement.²¹ A randomized controlled trial of 151 patients demonstrated that a significantly higher proportion of patients receiving second-line vedolizumab after first-line anti-TNF achieved steroid-free clinical remission and endoscopic improvement (MES ⩽1) at 14 weeks versus second-line infliximab.²⁰ In a meta-analysis of 588 patients with ulcerative colitis and failure with prior anti-TNF treatment, no statistically significant differences in rates of hospitalization, surgical intervention, or CS-free remission were found between vedolizumab and ustekinumab.¹² However, the treatment persistence of second-line vedolizumab was reported to be superior to ustekinumab in a cost-comparison analysis.³⁵ There are few real-world studies that have focused specifically on second-line treatment after one anti-TNF in patients with ulcerative colitis.^10,11,19 A real-world study of patients with ulcerative colitis demonstrated that second-line vedolizumab had superior effectiveness compared with second-line infliximab or adalimumab after 3 years of follow-up (after failure with first-line adalimumab or infliximab, respectively).¹⁰ This meta-analysis contributes to the limited evidence for patients with ulcerative colitis receiving second-line vedolizumab after treatment with one anti-TNF. Given that many patients now receive sequential biologic treatments, understanding vedolizumab efficacy by prior biologic exposure addresses a key evidence gap explored in this meta-analysis.

This meta-analysis also demonstrated a statistically significant improvement of HRQoL scores versus baseline at Week 52 after second-line vedolizumab. The absence of disability and normalized HRQoL is recommended as a long-term treatment target for patients with ulcerative colitis.^3,4 However, few studies have reported the effect of second-line vedolizumab on quality-of-life measures. GEMINI-1 reported improvements in EQ-5D, IBDQ, and Short Form-36 scores at Week 52 versus placebo; however, differences were not significant in patients who had received prior anti-TNF treatment.³⁶ For patients receiving vedolizumab in VISIBLE-1, the least squares mean differences for EQ-5D VAS and IBDQ scores were significantly improved at Week 52 versus placebo (p < 0.001).¹⁷ Although statistically significant, the pooled SMDs in EQ-5D VAS and IBDQ scores were small. The individual trial estimates for the mean difference of IBDQ scores, and EQ-5D VAS scores (for VISIBLE-1 only), reached the values for minimal clinically important difference (MCID) reported by Feagan et al.³⁶ (⩾16 and ⩾10 points from baseline, respectively), suggesting their relevance from a patient perspective. However, while this MCID is commonly used for IBDQ,³⁷ minimal change thresholds for EQ-5D VAS vary across populations.³⁸ Consequently, the clinical relevance of the improvement in HRQoL scores from baseline should be interpreted with caution.

The meta-analysis also confirms the efficacy of vedolizumab as a first-line biologic for patients with ulcerative colitis, which aligns with previous studies.^14,18,21 Compared with the second-line vedolizumab population, patients in the first-line population generally had a shorter duration of disease and a lower proportion of patients with disease throughout the colon.

In the second-line population, differences were observed in the heterogeneity and pooled treatment effect size between clinical remission definitions 1 and 2. High heterogeneity was indicated with I² estimates for definition 1 compared with moderate heterogeneity for definition 2. This suggests variation in the proportion of patients who may have achieved clinical remission by either definition in both trials. In addition to small sample size, study design variability, and different outcome measures, the difference in prior therapies in the study populations (mostly infliximab in GEMINI-1 vs all adalimumab in VISIBLE-1) could also contribute to heterogeneity. Overall, CIs associated with the pooled treatment effect size were wide, showing that the estimated relative risk was associated with large uncertainty. This could be due to the low sample size and low number of events observed in the treatment arms included in the study. Moreover, pooled treatment effect size for clinical remission was 10-times greater for definition 1 than 2. However, this should be interpreted with caution because no patients in the placebo arm of the GEMINI-1 study achieved definition 1, the GEMINI-1 study was more heavily weighted than the VISIBLE-1 study, and definition 1 was more stringent than definition 2. Broadly, the trends seen in the primary and secondary outcomes were similar to those seen in the scenario analysis, as expected given the similar populations.

There are strengths and weaknesses associated with this study. This meta-analysis allowed pooling of data from two study populations, thereby increasing the statistical power of the analysis. The regression analysis provided information on whether the achievement of clinical remission was influenced by patient characteristics. In addition, efficacy was assessed according to two definitions of clinical remission that reflected current guidelines and may more accurately represent what is observed in current clinical practice compared with the definitions used in the GEMINI-1 and VISIBLE-1 trials.^23,24 Further, these results contribute valuable insights into the positioning of vedolizumab within the treatment sequence for patients with ulcerative colitis, addressing an evidence gap regarding second-line therapy after treatment with one previous anti-TNF. The limitations associated with this analysis include the small sample size, low number of events, and heterogeneity observed for some outcomes. Owing to these factors, a direct comparison analysis between first-line and second-line populations was not possible, and comparison of these populations should be undertaken with critical examination. Only two studies were included in the meta-analysis at each time point, owing to the availability of individual patient data, increasing the uncertainty surrounding the estimation of the between-study heterogeneity, which could have also led to biased estimates of the pooled treatment effect and possible selection bias. Because the meta-analyses conducted at Week 52 were limited to those patients randomized to the maintenance phase in the GEMINI-1 and VISIBLE-1 trials, this could have led to selection bias (i.e., only patients who responded in the induction phase were included in the maintenance phase). Thus, it may be expected that vedolizumab is efficacious at Week 52 in patients who initially responded to treatment. However, this is difficult to account for as it is an artifact of the GEMINI-1 and VISIBLE-1 trial designs. The generalizability of this meta-analysis is limited by the inclusion of only two trials and their associated inclusion criteria, such as prior anti-TNF exposure being limited to within 5 years of study enrollment. The current analyses do not evaluate effect modifiers or characterize non-responders as a distinct group and, therefore, provide limited insight into individualized treatment selection. Finally, the data may not reflect current routine clinical practices or real-world data. Further research, particularly real-world evidence and randomized head-to-head trials, and systematic literature reviews with meta-analyses of a larger number of studies are warranted to further contextualize our findings and identify potential gaps or inconsistencies in the literature.

Conclusion

This meta-analysis demonstrates that second-line vedolizumab after first-line treatment with one anti-TNF is an effective therapeutic option for patients with moderately to severely active ulcerative colitis. Clinical outcomes and HRQoL were significantly improved at Week 52 compared with placebo. These findings should be interpreted in the context of the small number of studies in the analysis and the potential selection bias arising from the inclusion of patients who had already responded to induction therapy, which may limit generalizability and overstate long-term benefit relative to an unselected population. While not definitive for all patients who have received treatment with one prior anti-TNF, these results provide supportive evidence for the positioning of vedolizumab within the treatment sequence for patients with ulcerative colitis and help to address the evidence gap regarding treatment sequencing decisions.

Supplemental Material

sj-docx-1-tag-10.1177_17562848261431467 – Supplemental material for Efficacy and health-related quality of life associated with second-line vedolizumab after first-line treatment with one anti-TNF in patients with ulcerative colitis: a meta-analysis of the GEMINI-1 and VISIBLE-1 trials

Supplemental material, sj-docx-1-tag-10.1177_17562848261431467 for Efficacy and health-related quality of life associated with second-line vedolizumab after first-line treatment with one anti-TNF in patients with ulcerative colitis: a meta-analysis of the GEMINI-1 and VISIBLE-1 trials by Raja Atreya, Julien Kirchgesner, Zoe Cheah, Patricia Biedermann, Lytske Bakker, Marlies Meyer and Tim Raine in Therapeutic Advances in Gastroenterology

Supplemental Material

sj-docx-2-tag-10.1177_17562848261431467 – Supplemental material for Efficacy and health-related quality of life associated with second-line vedolizumab after first-line treatment with one anti-TNF in patients with ulcerative colitis: a meta-analysis of the GEMINI-1 and VISIBLE-1 trials

Supplemental material, sj-docx-2-tag-10.1177_17562848261431467 for Efficacy and health-related quality of life associated with second-line vedolizumab after first-line treatment with one anti-TNF in patients with ulcerative colitis: a meta-analysis of the GEMINI-1 and VISIBLE-1 trials by Raja Atreya, Julien Kirchgesner, Zoe Cheah, Patricia Biedermann, Lytske Bakker, Marlies Meyer and Tim Raine in Therapeutic Advances in Gastroenterology

Footnotes

Acknowledgements

At the direction of the authors, medical writing support was provided by Harriet Smith, MBBCh, of Oxford PharmaGenesis Ltd, Oxford, UK and funded by Takeda Pharmaceuticals International AG, Zurich, Switzerland. We would also like to acknowledge Paul Talsma for his careful review and valuable input on the statistical analyses. Following the International Committee of Medical Journal Editors (ICMJE) recommendations and Good Publication Practice, all authors had full access to the study data and take responsibility and accountability for the components of the work to which they contributed for this publication. The analysis plan and research questions were defined by Takeda in consultation with authors, who further contributed and refined the final data analysis performed by OPEN Health. Takeda reviewed the scientific content for medical accuracy and intellectual property clearance. Authors were responsible for all decisions on scientific content, interpretation of data, conclusions, and approved the final submitted manuscript, in accordance with ICMJE authorship criteria.

Declarations

ORCID iDs

Raja Atreya

Patricia Biedermann

Supplemental material

Supplemental material for this article is available online.

References

Ordás

Eckmann

Talamini

, et al. Ulcerative colitis. Lancet 2012; 380: 1606–1619.

Cai

Wang

Treatment of inflammatory bowel disease: a comprehensive review. Front Med (Lausanne) 2021; 8: 765474.

Raine

Bonovas

Burisch

, et al. ECCO guidelines on therapeutics in ulcerative colitis: medical treatment. J Crohns Colitis 2022; 16: 2–17.

Turner

Ricciuto

Lewis

, et al. STRIDE-II: an update on the selecting therapeutic targets in inflammatory bowel disease (STRIDE) initiative of the International Organization for the Study of IBD (IOIBD): determining therapeutic goals for treat-to-target strategies in IBD. Gastroenterology 2021; 160: 1570–1583.

Le Berre

Peyrin-Biroulet

Sandborn

, et al. Selecting end points for disease-modification trials in inflammatory bowel disease: the SPIRIT consensus from the IOIBD. Gastroenterology 2021; 160: 1452–1460.e21.

Bressler

Is there an optimal sequence of biologic therapies for inflammatory bowel disease?

Therap Adv Gastroenterol 2023; 16: 17562848231159452.

Lee

The role and prospect of tofacitinib in patients with ulcerative colitis. Intest Res 2023; 21: 168–169.

Singh

Murad

Fumery

, et al. First- and second-line pharmacotherapies for patients with moderate to severely active ulcerative colitis: an updated network meta-analysis. Clin Gastroenterol Hepatol 2020; 18: 2179–2191.e6.

Hyun

Zhang

H-S

, et al. Comparative effectiveness of second-line biological therapies for ulcerative colitis and Crohn’s disease in patients with prior failure of anti-tumour necrosis factor treatment. BMC Gastroenterol 2022; 22: 143.

10.

Kapizioni

Desoki

Lam

, et al. Biologic therapy for inflammatory bowel disease: real-world comparative effectiveness and impact of drug sequencing in 13 222 patients within the UK IBD BioResource. J Crohns Colitis 2024; 18: 790–800.

11.

Jairath

Zoughlami

Mahmoud

, et al. P0595 Comparative effectiveness between ustekinumab and vedolizumab after anti-TNF failure in ulcerative colitis: real-world experience. J Crohns Colitis 2025; 19: i1185.

12.

Rabeeah

Hallak

Abu-Rumaileh

, et al. Comparative efficacy of vedolizumab versus ustekinumab in UC patients after anti-TNF therapy failure: a systematic review and meta-analysis. Gastroenterology 2025; 168: S11–S12.

13.

Feagan

Rutgeerts

Sands

, et al. Vedolizumab as induction and maintenance therapy for ulcerative colitis. N Engl J Med 2013; 369: 699–710.

14.

Motoya

Watanabe

Ogata

, et al. Vedolizumab in Japanese patients with ulcerative colitis: a phase 3, randomized, double-blind, placebo-controlled study. PLoS One 2019; 14: e0212989.

15.

US Food and Drug Administration. ENTYVIO^® prescribing information, https://www.accessdata.fda.gov/drugsatfda_docs/label/2024/761359s000lbl.pdf (2025, accessed 2 January 2025).

16.

European Medicines Agency. Entyvio summary of product characteristics, https://www.ema.europa.eu/en/documents/product-information/entyvio-epar-product-information_en.pdf (2025, accessed 2 January 2025).

17.

Sandborn

Baert

Danese

, et al. Efficacy and safety of vedolizumab subcutaneous formulation in a randomized trial of patients with ulcerative colitis. Gastroenterology 2020; 158: 562–572.e12.

18.

Attauabi

Madsen

Bendtsen

, et al. Vedolizumab as the first line of biologic therapy for ulcerative colitis and Crohn’s disease—a systematic review with meta-analysis. Dig Liver Dis 2022; 54: 1168–1178.

19.

Rundquist

Sachs

Eriksson

, et al. Drug survival of anti-TNF agents compared with vedolizumab as a second-line biological treatment in inflammatory bowel disease: results from nationwide Swedish registers. Aliment Pharmacol Ther 2021; 53: 471–483.

20.

Bouguen

Nachury

Nancey

, et al. OP38 Comparative efficacy of infliximab and vedolizumab after failure of a first anti-TNF in patients with ulcerative colitis: a double-blind randomized controlled trial (EFFICACI). J Crohns Colitis 2025; 19: i74.

21.

Sands

Peyrin-Biroulet

Loftus

Jr , et al. Vedolizumab versus adalimumab for moderate-to-severe ulcerative colitis. N Engl J Med 2019; 381: 1215–1226.

22.

Sandborn

Sands

Uddin

, et al. Clinical trial design in ulcerative colitis: interpreting evolving endpoints based on post hoc analyses of the vedolizumab phase 3 trials GEMINI 1 and VISIBLE 1. Crohns Colitis 360 2023; 6: otad076.

23.

US Food and Drug Administration. Ulcerative colitis: developing drugs for treatment guidance for industry, https://www.fda.gov/media/158016/download (2022, accessed 13 March 2025).

24.

US Food and Drug Administration. Ulcerative colitis: clinical trial endpoints guidance for industry, https://www.fda.gov/files/drugs/published/Ulcerative-Colitis--Clinical-Trial-Endpoints-Guidance-for-Industry.pdf (2022, accessed 20 March 2025).

25.

Rabin

de Charro

EQ-SD: a measure of health status from the EuroQol Group. Ann Med 2001; 33: 337–343.

26.

Guyatt

Mitchell

Irvine

, et al. A new measure of health status for clinical trials in inflammatory bowel disease. Gastroenterology 1989; 96: 804–810.

27.

Zou

A modified Poisson regression approach to prospective studies with binary data. Am J Epidemiol 2004; 159: 702–706.

28.

Firth

Bias reduction of maximum likelihood estimates. Biometrika 1993; 80: 27–38.

29.

Deeks

Higgins

JPT

Altman

, et al. Analysing data and undertaking meta-analyses. In: Higgins

Thomas

Chandler

, et al. (ed.) Cochrane handbook for systematic reviews of interventions. Cochrane Training, 2023. https://training.cochrane.org/handbook/current (accessed 16 December 2024).

30.

R Core Team. R: a language and environment for statistical computing. R Foundation for Statistical Computing: Vienna, Austria, https://www.R-project.org/ (2023, accessed 4 June 2025).

31.

Wickham

. ggplot2. WIREs Comp Stat 2011; 3: 180–185.

32.

Wickham

Smith

Fay

haven: import and export “SPSS”, “Stata” and “SAS” files. R package version 2.5.4, https://github.com/tidyverse/haven, https://github.com/WizardMac/ReadStat, https://haven.tidyverse.org (2023, accessed 30 January 2025).

33.

Kosmidis

brglm2: Bias reduction in generalized linear models. R package version 06.

34.

Viechtbauer

Conducting meta-analyses in R with the metafor package. J Stat Softw 2010; 36: 1–48.

35.

Martin

Zhou

Mac

, et al. P0732 A cost comparison of treatment sequencing for ulcerative colitis: vedolizumab vs ustekinumab as a second-line biologic in anti-TNF-exposed patients in Canada. J Crohns Colitis 2025; 19: i1422.

36.

Feagan

Patel

Colombel

J-F

, et al. Effects of vedolizumab on health-related quality of life in patients with ulcerative colitis: results from the randomised GEMINI 1 trial. Aliment Pharmacol Ther 2017; 45: 264–275.

37.

Katsoula

Paschos

Malandris

, et al. Systematic review and network meta-analysis: evaluating the impact of advanced therapies for moderate-to-severe ulcerative colitis on health-related quality of life. J Crohns Colitis 2025; 19: jjaf210.

38.

Al Sayah

Jin

Short

, et al. A systematic literature review of important and meaningful differences in the EQ-5D Index and Visual Analog Scale Scores. Value Health 2025; 28: 470–476.