Abstract
-
Background/Aims
- Vedolizumab and infliximab are widely used biologic agents for ulcerative colitis (UC), yet their comparative long-term effectiveness and safety in biologic-naive patients remain uncertain. This meta-analysis evaluated long-term efficacy and safety outcomes of vedolizumab versus infliximab in this population.
-
Methods
- A systematic search of MEDLINE, Scopus, the Cochrane Library, and ClinicalTrials.gov was conducted. Studies enrolling biologic-naive UC patients with a minimum follow-up of 12 months and reporting comparative outcomes between vedolizumab and infliximab were included. Random-effects meta-analyses were performed to pool odds ratios with 95% confidence intervals.
-
Results
- Twenty-five studies met the inclusion criteria. Vedolizumab was associated with significantly greater treatment persistence than infliximab at 12, 24, 36, and 48 months, while no significant difference was observed at 60 months. Patients receiving vedolizumab were more likely to achieve clinical remission and endoscopic remission at 12 months. Rates of steroid-free remission at 12 months were comparable between groups. Vedolizumab was linked to lower hospitalization rates at 12 and 36 months, with no difference at 24 months, and showed similar risks of inflammatory bowel disease-related surgery. The need for dose optimization was lower with vedolizumab at 12 months. Regarding safety, vedolizumab was associated with a lower incidence of serious adverse events compared with infliximab.
-
Conclusions
- In biologic-naive UC patients with long-term follow-up, vedolizumab demonstrates favorable treatment persistence, remission, and safety profiles compared with infliximab. Given the predominance of observational evidence, these findings should be interpreted cautiously and warrant confirmation in prospective comparative studies.
-
Keywords: Ulcerative colitis; Biologic-naive; Vedolizumab; Infliximab; Meta-analysis
Graphical abstract
INTRODUCTION
Ulcerative colitis (UC) is a sustained inflammatory condition of the colonic mucosa whose relapsing course often obliges clinicians to move beyond conventional drugs to biological therapies in order to achieve durable symptom control and mucosal healing [1].
Two biologic agents commonly used as first-line options in moderate-to-severe UC are infliximab, a systemic anti-tumor necrosis factor (anti-TNF) antibody, and vedolizumab, a monoclonal antibody that selectively blocks the α4β7 integrin and limits gut-directed lymphocyte trafficking [2,3]. Both agents demonstrated clear efficacy in randomized, placebo-controlled trials for induction and maintenance of remission–infliximab in the ACT trials and vedolizumab in the GEMINI-1 program–but their mechanisms, onset of action, and safety considerations differ in ways that may affect long-term patient trajectories [2,3].
Randomized controlled trials (RCTs) have been critical to establishing short-term and maintenance benefits, yet most pivotal trials emphasize induction endpoints (measured in weeks) and are not designed to directly compare different biologic classes over extended periods in real-world settings [4]. In consequence, there remains a paucity of randomized, head-to-head evidence that evaluates how initial choice between infliximab and vedolizumab shapes outcomes beyond the early maintenance phase. This gap leaves clinicians to infer comparative durability and safety from heterogeneous postmarketing cohorts and observational series.
Longitudinal outcomes, including how long a patient remains on their first biologic (treatment persistence), the probability of achieving steroid-free clinical remission, and the attainment of endoscopic remission, are central to both individual patient prognosis and health-system planning because they reflect sustained disease control, steroid minimization, and mucosal healing, all of which are linked to fewer hospitalizations and surgeries [5]. Endoscopic remission in particular has been associated with better long-term trajectories, making multi-year follow-up necessary to judge whether early gains translate into meaningful, lasting benefit [5].
Real-world comparative studies that follow biologic-naive UC populations for 1 year or longer have produced mixed findings: some cohorts report greater drug survival, higher rates of steroid-free remission, and better mucosal outcomes with vedolizumab, whereas other extended series find similar effectiveness and safety between the 2 agents over multi-year observation [6,7]. These divergent signals likely reflect differences in patient selection, definitions of outcomes, and follow-up duration, complicating simple translation into practice.
Because treatment persistence and steroid-free/endoscopic remission are outcomes that accrue over months to years, head-to-head randomized long-term comparisons are scarce, and observational datasets show inconsistent results, a focused synthesis of long-term comparative data is needed. Restricting the analysis to biologic-naive cohorts with at least 1 year of follow-up concentrates on the point at which initial biologic choice is most likely to influence sustained control, cumulative steroid exposure, and safety signals relevant to chronic care.
Accordingly, this meta-analysis aims to pool studies that directly compare vedolizumab and infliximab in biologic-naive UC patients with ≥12 months of follow-up, and to estimate relative effects on treatment persistence, clinical and steroid-free remission, endoscopic remission, and adverse events (AEs). By synthesizing long-term evidence and harmonizing outcome definitions where possible, the study seeks to provide clinicians and policy-makers with clearer, practice-relevant estimates about which first-line biologic better supports durable disease control and acceptable safety in routine care.
METHODS
1. Eligibility Criteria
This meta-analysis was performed in accordance with the PRISMA guidelines [8]. The protocol of this study has been registered in PROSPERO (CRD420261328450). Study selection was conducted using predefined eligibility criteria based on the PICOS framework (Population, Intervention, Comparator, Outcomes, and Study design). This approach was designed to identify studies that directly compared the long-term effectiveness and safety of vedolizumab and infliximab as initial biologic therapy for UC, while minimizing heterogeneity and ensuring comparability of included populations and outcomes.
1) Inclusion Criteria
(1) Population (P)
Adult patients diagnosed with UC based on accepted clinical, endoscopic, and/or histopathological criteria. Only study populations that were biologic-naive at the time of treatment initiation were considered and extracted. Studies that included mixed cohorts of UC together with Crohn’s disease (CD) or other types of inflammatory bowel disease (IBD) were included only when outcome data specific to UC could be separately identified and extracted.
(2) Intervention (I)
Vedolizumab administered as a first-line biologic therapy for UC. All approved dosing regimens and routes of administration were accepted, provided that vedolizumab represented the initial biologic agent initiated in the studied cohort.
(3) Control (C)
Infliximab used as first-line biologic treatment for UC. Studies must directly compare outcomes between patients treated with vedolizumab and those treated with infliximab within the same study period.
(4) Outcome (O)
Eligible studies must report at least 1 of the following predefined outcomes assessed at long-term follow-up: treatment persistence or drug survival, defined as continued use of the index biologic without discontinuation or switching; clinical remission, defined as partial Mayo score of 1 or less with a rectal bleeding subscore of 0; corticosteroid-free clinical remission, indicating remission achieved without ongoing steroid therapy; endoscopic remission, defined as Mayo endoscopic subscore (MES) of 0; hospitalization, defined as hospital admission with a primary diagnosis of UC or IBD-related complications; IBD-related surgery, defined as major surgical intervention related to IBD, including procedures involving intestinal incision, resection, or anastomosis, as well as operations affecting the rectum, rectosigmoid region, perirectal structures, or anal area; therapy optimization, defined as dose escalation or interval reduction; total AEs, defined as overall number or proportion of any undesirable medical occurrences reported during a specified treatment period, regardless of severity or causal relationship to the study medication; or serious AEs (SAEs), defined as events requiring hospitalization, therapy discontinuation, significant disability, or resulting in death.
(5) Study design (S)
Comparative studies with a minimum follow-up duration of 12 months to ensure assessment of long-term effectiveness and safety. Eligible designs included RCTs, prospective cohort studies, and retrospective observational studies that provide direct head-to-head comparisons between vedolizumab and infliximab.
2) Exclusion Criteria
The following studies were excluded: (1) pediatric-only cohorts or mixed populations without disaggregated results; (2) studies focused exclusively on patients with prior exposure to biologic therapy; (3) studies evaluating vedolizumab used in combination with other biologic agents; (4) studies comparing vedolizumab with anti-TNF agents as a class, without separately reporting infliximab-specific outcomes; or (5) case reports, case series without a comparator group, narrative reviews, editorials, conference abstracts without full data, and non-human studies.
2. Search Strategy and Study Selection
A comprehensive literature search was performed from February 25, 2026 to March 4, 2026 to identify all relevant studies that compared vedolizumab and infliximab as first-line biologic therapy in patients with UC. The search was conducted across multiple electronic databases, including MEDLINE (via PubMed), Scopus, the Cochrane Library, and ClinicalTrials. gov. These databases were selected to capture peer-reviewed publications as well as registered clinical studies with potentially unpublished results. The search strategy was developed using combinations of keywords and controlled vocabulary terms related to “ulcerative colitis,” “vedolizumab,” “infliximab,” and “biologic-naive.” Boolean operators and database-specific filters were applied to optimize sensitivity and specificity of the search. Details regarding search strategy for each database can be seen in Supplementary Table 1. No restrictions on publication year were applied in order to retrieve all potentially eligible long-term evidence.
All records identified from the database searches were exported to reference management software, and duplicate entries were removed. Two independent reviewers screened titles and abstracts to assess initial relevance according to the predefined eligibility criteria. Articles that appeared to meet the inclusion requirements, or for which eligibility could not be determined based on the abstract alone, were retrieved in full text for detailed evaluation. Disagreements at any stage of screening were resolved through discussion, and when necessary, by consultation with a third reviewer.
3. Data Extraction
Information from all eligible studies was collected in a structured and systematic manner. Two investigators independently reviewed each included article and extracted relevant data using a standardized extraction template developed specifically for this review. This process was undertaken to ensure accuracy, consistency, and completeness of the information obtained from the primary sources.
For every study, general characteristics were recorded, including the name of the first author, year of publication, country of origin, study design, sample size, and duration of follow-up. Details regarding patient populations were also gathered, such as age distribution, sex, disease extent, MES, and confirmation that participants were biologic-naive at the time of treatment initiation. Where numerical results were presented in different formats, such as proportions, hazard ratios, or risk estimates, the necessary data were converted to a common effect measure whenever feasible. In cases in which essential information was incomplete or unclear, attempts were made to derive the required values from supplementary materials or to contact study authors for clarification.
Any inconsistencies or disagreements between reviewers during the extraction process were resolved through discussion and re-examination of the source articles. If consensus could not be reached, a third reviewer adjudicated the issue. This approach was intended to minimize errors and to ensure that the data included in the analysis accurately represented the findings of the original studies.
4. Risk of Bias Assessment
The methodological quality of all studies included in the meta-analysis was systematically evaluated to determine the potential influence of bias on the pooled results. Because the majority of eligible evidence was expected to originate from observational research, the Newcastle–Ottawa Scale (NOS) was selected as the primary tool for quality appraisal [9]. This instrument provides a structured method to assess non-randomized studies based on key domains related to study design and conduct [9].
Each study was independently reviewed by 2 investigators using the NOS criteria, which examine 3 principal aspects: selection of study participants, comparability of treatment groups, and adequacy of outcome assessment and follow-up [9]. Points were assigned according to predefined standards within these categories, resulting in a maximum possible score of 9 [9]. Consistent with commonly accepted practice, studies achieving a total score of 7 or higher were classified as high-quality research, whereas those with lower scores were considered to have moderate or greater risk of bias [9].
5. Statistical Analysis
Quantitative synthesis of eligible studies was undertaken to compare long-term outcomes between vedolizumab and infliximab in biologic-naive patients with UC. For all dichotomous endpoints, effect estimates were expressed as odds ratios (ORs) with corresponding 95% confidence intervals (CIs). Pooled estimates were calculated using the Mantel–Haenszel method within a random-effects framework to account for anticipated clinical and methodological variability among studies.
Between-study heterogeneity was evaluated using the I-square (I 2) statistic [10]. Values exceeding 50% were interpreted as indicating meaningful heterogeneity, prompting further exploration of potential sources of inconsistency [10]. In addition, the chi-square test for heterogeneity was reviewed, with a P-value below 0.10 considered suggestive of significant variation beyond chance. For outcomes demonstrating substantial heterogeneity, sensitivity analyses were conducted using a leave-one-out approach. This method involved sequentially omitting each study from the analysis to assess the stability of the pooled results and to determine whether any individual study exerted a disproportionate influence on the overall effect estimate. For outcomes with a very small number of contributing studies, additional sensitivity analyses were conducted using a fixed-effect model to evaluate the robustness of pooled estimates generated under the random-effects model.
To further investigate factors that might modify treatment effects, meta-regression analyses were performed when sufficient data were available. A restricted maximum-likelihood random-effects model was applied to explore associations between study-level characteristics and reported outcomes. Variables examined included disease extent, baseline MES, patient age, sex distribution, duration of UC, smoking status, and use of concomitant corticosteroids. Meta-regression covariates were extracted and analyzed at the study level as pooled cohort characteristics. For age and other continuous variables reported separately for the vedolizumab and infliximab groups, pooled study-level means were calculated as weighted averages using group sample sizes: x¯pooled=n1x¯1+n2x¯2n1+n2, where n1 and n2 denote the group sample sizes and x¯1 and x¯2 denote the corresponding group means. These analyses were intended to identify clinical or methodological features that could help explain observed differences across studies. For meta-regression, only studies with available data for the covariate of interest were included in each specific model. Studies with missing values for a given variable (reported as “NR” in Table 1) were not entered into that particular meta-regression analysis and were automatically excluded by the software for that covariate only. Statistical significance was defined a priori as a two-sided P-value < 0.05.
All statistical computations were carried out using Review Manager version 5.4 and Comprehensive Meta-Analysis version 4 software. Results were considered statistically significant when the two-sided P-value was less than 0.05, unless otherwise specified.
RESULTS
1. Study Selection and Characteristics
A comprehensive search of 4 electronic databases using predetermined keywords identified a total of 769 potential records. After removal of duplicate entries, screening of titles and abstracts resulted in the exclusion of 680 publications. Eighty-nine articles were considered potentially relevant and were therefore reviewed in full text. Application of the predefined eligibility criteria led to the exclusion of 64 studies for the following reasons: 18 did not report the outcomes of interest, 15 were review articles, 11 included participants with prior biologic exposure, 9 lacked an appropriate comparison group, 5 did not provide results specific to UC, 2 did not present separate data for infliximab-treated patients, 2 involved exclusively pediatric populations, 1 used an indirect comparison design, and 1 had a follow-up period shorter than 1 year. Consequently, 25 studies met all inclusion criteria and were incorporated into the final meta-analysis (Fig. 1). Among the studies included in the analysis, 2 employed prospective cohort design, whereas the other 23 were based on retrospective cohort data. The geographical distribution of the evidence was diverse: 6 studies originated from the United States, 4 studies from Germany, 2 studies each from Italy, Australia, and China, and 1 study each from Saudi Arabia, Israel, Brazil, Denmark, the United Kingdom, South Korea, Japan, Kuwait, and Slovenia. Reported periods of observation ranged from a minimum of 12 months to a maximum of 60 months. Additional information concerning baseline patient characteristics, disease extent, and MES across the included studies is provided in Table 1 [6,7,11-33].
2. Quality of Study Assessment
Methodological quality of the observational studies was evaluated through application of the NOS. Assessment across the selection, comparability, and outcome domains indicated that every included study met criteria for high methodological standard, achieving total scores between 7 and 8, as summarized in Table 2 [6,7,11-33].
3. Outcomes
1) Treatment Persistence
Data on treatment persistence were available from 20 studies at 12 months, 10 studies at 24 months, 7 studies at 36 months, 6 studies at 48 months, and 3 studies at 60 months of followup. Combined analyses demonstrated that, among biologic-naive patients with UC, use of vedolizumab was associated with a greater probability of remaining on therapy (treatment persistence) compared with infliximab at follow-up intervals of 12 months (OR, 1.73; 95% CI, 1.50–1.99; P<0.001, I 2 =8%, random-effect model), 24 months (OR, 1.78; 95% CI, 1.35–2.34; P<0.001, I 2 =46%, random-effect model), 36 months (OR, 1.67; 95% CI, 1.23–2.26, P=0.001, I2 =0%, random-effect model), and 48 months (OR, 1.84; 95% CI, 1.37–2.47; P<0.001, I 2 =0%, random-effect model) (Fig. 2, Table 3). In contrast, no statistically significant difference in persistence between the 2 therapies was observed at the 60-month time point (OR, 2.19; 95% CI, 0.82–5.84; P=0.120, I 2 =0%, random-effect model) (Table 3). Sensitivity analyses performed at 60-month time point using a fixed-effect model showed consistent results (OR, 2.13; 95% CI, 0.82–5.52; P=0.120, I 2 =0%, fixed-effect model), indicating that these findings remain robust. Nonetheless, the analysis of treatment persistence at 60 months was based on only 3 studies and a very small vedolizumab sample, which substantially limits statistical power and precision. Therefore, the absence of a significant difference should be interpreted as inconclusive rather than as evidence of comparable long-term persistence between treatment groups.
2) Clinical Remission
Eight studies, comprising a total of 1,375 participants, presented outcomes on clinical remission after 12 months of therapy. Pooled results from these investigations indicated that, in biologic-naive individuals with UC, vedolizumab treatment was linked to a greater likelihood of achieving clinical remission at the one-year follow-up compared with infliximab (OR, 1.96; 95% CI, 1.37–2.81; P<0.001, I 2 =49%, random-effect model) (Fig. 3A, Table 3).
3) Steroid-Free Remission
Information on steroid-free remission outcome was available from 10 studies including a combined total of 1,349 patients. The aggregated analysis indicated that, among biologic-naive patients with UC, rates of corticosteroid-free remission at 12 months were not significantly different between those receiving vedolizumab and those receiving infliximab (OR, 1.34; 95% CI, 0.95–1.90; P=0.090, I 2 =41%, random-effect model) (Fig. 3B, Table 3).
4) Endoscopic Remission
Seven studies, encompassing a total of 929 biologic-naive patients with UC, provided data on endoscopic remission outcomes. Pooled analysis of these reports demonstrated that treatment with vedolizumab resulted in a greater likelihood of achieving endoscopic remission when compared with infliximab (OR, 1.48; 95% CI, 1.09–1.99; P=0.010, I 2 =0%, random-effect model) (Fig. 3C, Table 3).
5) Hospitalization
Four publications reported hospitalization rates at 12 months of follow-up, while 2 studies each reported rates at 24 and 36 months of follow-up. Meta-analysis of these studies demonstrated lower hospitalization rates at 12 (OR, 0.49; 95% CI, 0.29–0.83; P=0.008, I 2 =40%, random-effect model) and 36 months (OR, 0.40; 95% CI, 0.19–0.85; P=0.020, I 2 =0%, randomeffect model) among biologic-naive UC patients treated with vedolizumab compared to those treated with infliximab. In contrast, hospitalization rates did not differ significantly between the 2 groups at 24 months of follow-up (OR, 0.64; 95% CI, 0.30–1.35; P=0.240, I 2 =0%, random-effect model) (Table 3). Sensitivity analyses using fixed-effect models at 24 months (OR, 0.63; 95% CI, 0.30–1.32; P =0.220, I 2 =0%, fixed-effect model) and 36 months (OR, 0.40; 95% CI, 0.19–0.85; P=0.020, I 2 =0%, fixed-effect model) yielded consistent results, indicating that the findings were robust.
6) IBD-Related Surgery
Data on IBD-related surgical outcomes were available from 4 studies involving a total of 665 participants. Combined analysis of these reports indicated that the likelihood of requiring IBD-related surgery was comparable between biologic-naive UC patients treated with vedolizumab and those treated with infliximab (OR, 0.74; 95% CI, 0.27–2.04; P=0.570, I 2 =32%, random-effect model) (Table 3).
7) Dose Optimization
Five studies provided data on dose optimization at 12 months of follow-up. Meta-analysis of these publications demonstrated that biologic-naive UC patients receiving vedolizumab were less likely to require dose optimization at 12 months (OR, 0.45; 95% CI, 0.24–0.83; P=0.010, I 2 =47%, random-effect model) (Table 3). Meanwhile, only 1 study each reported this outcome at 24, 36, and 48 months; therefore, a meta-analysis could not be performed for these time points.
8) Total AEs
Unlike efficacy outcomes, total AEs were not consistently reported at fixed assessment time points across studies; therefore, the pooled total AEs analyses reflect cumulative events during the respective study periods as can be seen in Table 1. An analysis combining data from 8 studies, involving a total of 1,271 participants, indicated that biologic-naive UC patients treated with vedolizumab experienced a significantly lower overall rate of AEs compared with those treated with infliximab (OR, 0.71; 95% CI, 0.52–0.97; P=0.030, I 2 =5%, random-effect model) (Table 3).
9) Serious AEs
As most included studies did not report serious AEs at uniform prespecified time points, these outcomes were extracted as cumulative events over each study’s reported follow-up period, as summarized in Table 1. Combined data from 6 studies, comprising a total of 1,490 patients, showed that biologic-naive individuals with UC who received vedolizumab had a significantly lower incidence of serious AEs than those treated with infliximab (OR, 0.19; 95% CI, 0.12–0.30; P<0.001, I 2 =0%, random-effect model) (Table 3).
4. Meta-Regression
The results of the meta-regression are summarized in Supplementary Table 2. This analysis indicated that treatment persistence at 12 months was not significantly influenced by study-level variables, such as disease extent (P=0.362), MES (P=0.882), patient age (P=0.856), sex (P=0.488), duration of UC (P=0.495), smoking status (P=0.675), or use of concomitant corticosteroids (P=0.115) (Supplementary Fig. 1).
The meta-regression further demonstrated that, among the evaluated study-level variables, concomitant steroid use was the sole factor significantly associated with treatment persistence at 24 months of follow-up (beta coefficient, 0.018; 95% CI, 0.002–0.034; P=0.031). In contrast, variables including extent of disease (P=0.822), patient age (P=0.733), sex (P=0.733), and duration of UC (P=0.509) showed no meaningful impact on this outcome (Supplementary Fig. 2).
Furthermore, the meta-regression analyses for treatment persistence at 36 and 48 months did not identify any significant associations with study-level variables, including extent of disease, patient age, sex, or duration of UC (Supplementary Table 2, Supplementary Figs. 3 and 4).
Regarding clinical remission at 12 months, the meta-regression demonstrated that MES (beta coefficient, –0.025; 95% CI: –0.043 to –0.006; P =0.008), patient age (beta coefficient, –0.246; 95% CI, –0.482 to –0.010; P=0.041), and sex (beta coefficient, 0.035; 95% CI, 0.001 to 0.070; P=0.046) were significantly associated with this outcome. In contrast, variables such as extent of disease (P=0.745), duration of UC (P=0.214), smoking status (P=0.792), and concomitant steroid use (P=0.447) did not show a meaningful effect (Supplementary Fig. 5).
The meta-regression further indicated that the duration of UC was significantly associated with steroid-free remission at 12 months outcome (beta coefficient, 0.026; 95% CI, 0.008–0.044; P=0.005). In contrast, other evaluated study-level factors—including extent of disease (P=0.302), MES (P=0.167), age (P=0.804), sex (P=0.201), smoking status (P=0.967), and use of concomitant steroids (P=0.834)—did not demonstrate a significant impact on this outcome (Supplementary Fig. 6).
In addition, the meta-regression demonstrated that extent of disease was the only study-level variable significantly associated with endoscopic remission at 12 months outcome (beta coefficient, 0.045; 95% CI, 0.002–0.087; P=0.040). Conversely, factors such as MES (P=0.637), patient age (P=0.128), sex (P=0.514), duration of UC (P=0.420), smoking status (P=0.607), and concomitant steroid use (P=0.802) did not exhibit a significant influence on this outcome (Supplementary Fig. 7).
With regard to hospitalization outcomes, meta-regression analysis could be conducted only using patient age as an available variable, and this factor was not found to have a significant effect (P=0.742) (Supplementary Fig. 8).
Meta-regression analysis showed that the dose optimization outcome at 12 months was influenced only by disease extent (beta coefficient, 0.084; 95% CI, 0.011–0.158; P=0.025). Other variables, including MES (P=0.720), age (P=0.356), sex (P=0.918), UC duration (P=0.853), and concomitant steroid use (P=0.920) (Supplementary Fig. 9), showed no significant influence on this outcome. Although disease extent covariate was statistically associated with dose optimization outcome at 12 months, this meta-regression result was based on a small number of studies and should therefore be interpreted cautiously.
For the IBD-related surgery outcome, our meta-regression analysis did not identify any significant influence from disease extent (P=0.371), age (P=0.905), sex (P=0.108), or UC duration (P=0.060) (Supplementary Fig. 10).
Our meta-regression analysis also revealed that the sole factor that exerted a significant influence on the total AEs outcome was the use of concomitant steroids (beta coefficient, 0.014; 95% CI, 0.000–0.027; P=0.049). Meanwhile, other factors, including disease extent (P=0.998), MES (P=0.538), age (P=0.205), sex (P=0.567), UC duration (P=0.912), and smoking status (P=0.912), did not show any significant influence on this outcome (Supplementary Fig. 11). While concomitant steroid use showed a statistically significant association with total AEs outcome, this result was derived from a small number of studies and should be interpreted cautiously.
Finally, our meta-regression analysis indicated that the serious AEs outcome was not significantly influenced by study-level variables, such as disease extent (P=0.758), MES (P=0.216), age (P=0.178), sex (P=0.418), UC duration (P=0.103), smoking status (P=0.744), and concomitant steroids (P=0.230) (Supplementary Fig. 12).
5. Publication Bias
Assessment of publication bias using funnel plots demonstrated generally symmetrical distributions for the outcomes of treatment persistence at 12 and 24 months, as well as for steroid-free remission at 12 months, suggesting no evident publication bias (Supplementary Fig. 13). Evaluation of potential publication bias for the other outcomes was not performed because the number of available studies was insufficient (fewer than 10), and robust assessment of bias typically requires a greater number of studies [34,35].
DISCUSSION
This meta-analysis synthesizes available comparative data on vedolizumab versus infliximab in biologic-naive UC and provides a coherent picture across effectiveness and safety domains. The principal findings are that vedolizumab is associated with higher treatment persistence (12–48 months), greater odds of clinical remission at 12 months, and higher rates of endoscopic remission, while steroid-free remission at 12 months did not differ significantly between agents. Because steroid-free remission is a more stringent and clinically relevant treat-to-target outcome in UC, the observed advantage in clinical remission should be interpreted cautiously and should not be assumed to reflect a clear steroid-sparing benefit. Vedolizumab treatment was also linked to lower hospitalization at selected time points (12 and 36 months), reduced need for dose optimization at 12 months, and a lower incidence of SAEs. By contrast, the need for IBD-related surgery was similar between groups. Together, these results suggest that in biologic-naive UC populations vedolizumab may offer advantages in long-term treatment retention, mucosal healing, and serious safety events compared with infliximab.
Several biologic and clinical explanations could account for the superiority of vedolizumab on persistence, remission, and endoscopic outcomes. Vedolizumab’s mechanism–selective blockade of α4β7 integrin and consequent inhibition of guttropic lymphocyte trafficking–produces a gut-focused immunomodulation that directly targets intestinal inflammation while sparing systemic immune compartments; this property plausibly promotes mucosal healing with fewer systemic immune effects [36-38]. Empirical and mechanistic studies have shown that α4β7-expressing lymphocytes are preferentially gut-homing and that their inhibition limits intestinal inflammation [36-38].
By contrast, infliximab exerts broad systemic anti-TNFα activity [39,40]. While anti-TNF agents often bring rapid induction of clinical response, their systemic immunosuppression can increase risks of infections, infusion reactions, and other systemic AEs–factors that may lead to treatment interruption, switching, or dose changes in routine practice [39,40]. Infusion reactions and systemic infectious complications are well documented with infliximab and are recognized contributors to drug discontinuation [39,40].
Taken together, vedolizumab’s gut-selectivity may explain both (1) higher long-term persistence and better endoscopic outcomes, through durable mucosal targeting, and (2) a more favorable SAE profile because of reduced systemic immune perturbation. Real-world cohort data have similarly reported longer persistence with vedolizumab compared with infliximab in biologic-naive UC, supporting the plausibility of these mechanistic links in clinical practice [26,30].
If confirmed in higher-quality prospective studies, the observed pattern (greater persistence and mucosal healing with fewer serious AEs) supports considering vedolizumab as a strong first-line biologic option for selected biologic-naive UC patients–particularly for those in whom long-term safety and mucosal healing are priorities. Greater treatment persistence may translate into fewer switches and lower cumulative exposure to multiple agents, with potential downstream benefits for disease control and healthcare utilization. However, these implications must be tempered by the nature of the underlying evidence. Much of the data are derived from observational cohorts and non-randomized comparators; such designs are susceptible to channeling bias (physicians selecting 1 drug over another based on unmeasured patient characteristics), indication bias, and residual confounding despite statistical adjustment. Therefore, while the data are hypothesis-generating and clinically informative, they do not establish causality, and prescribing decisions should integrate individual patient factors, induction-phase urgency, comorbidity profile, infection risk, and patient preference.
A prior, influential meta-analysis that compared infliximab and vedolizumab across IBD populations (including both CD and UC) reported that infliximab had superior efficacy during the induction phase, with comparable maintenance-phase efficacy and similar overall safety profiles between agents [41]. That analysis aggregated RCTs and indirect comparisons across indications and phases of therapy and therefore emphasized short-term induction outcomes and trial-based safety signals [41]. Key differences between that earlier work and our analysis help explain divergent conclusions. First, our meta-analysis focused specifically on biologic-naive UC patients and included more real-world cohort data, with persistence assessed at longer predefined follow-up intervals (12–60 months), whereas the prior meta-analysis combined RCT data across disease types and largely emphasized induction/maintenance endpoints defined in trial protocols. Second, our inclusion of observational studies and real-world registries allowed assessment of long-term persistence, hospitalization, dose optimization, and safety outcomes as observed in routine care over extended follow-up periods. However, although such designs may better reflect real-world practice and longer-term treatment exposure, serious AEs reporting in observational studies may be less standardized and potentially less complete than in RCTs, where AEs are prospectively monitored according to predefined protocols. Finally, the earlier meta-analysis relied on an indirect comparison that combined data from separate randomized trials–those evaluating infliximab versus placebo and those evaluating vedolizumab versus placebo–without applying formal statistical methods designed for adjusted indirect treatment comparisons [41]. The clinical trials of infliximab and vedolizumab differed in several important respects, such as inclusion criteria, baseline disease characteristics, use of concomitant therapies, definitions of outcomes, and timing of outcome assessments. Directly comparing pooled results from 2 independent groups of trials may therefore be methodologically limited, as this approach does not adequately account for variations in study design, patient populations, or analytical rigor. These methodological and population differences likely account for the contrasting emphasis on long-term persistence and safety in our results versus induction-phase effectiveness in the prior meta-analysis.
Several limitations should be acknowledged. First, because most included studies were retrospective cohorts, our findings are inherently susceptible to channeling bias, indication bias, and residual confounding. Thus, the observed differences between vedolizumab and infliximab may partly reflect nonrandom treatment allocation rather than the true treatment effect alone. In clinical practice, biologic selection may be influenced by disease severity, physician preference, patient characteristics, and local prescribing patterns—factors that are not uniformly measured or adjusted for across studies. Therefore, although statistical adjustment methods (including meta-regression analysis) may reduce measured confounding, they cannot fully account for unmeasured bias. Accordingly, the comparative estimates reported in this meta-analysis should be interpreted cautiously and viewed as hypothesis-generating, supporting the need for more robust prospective comparative studies or randomized head-to-head trials in biologic-naive UC. Second, some outcomes were informed by a limited number of studies, preventing robust assessment of publication bias and reducing confidence in pooled estimates (funnel plot assessment was feasible only for outcomes with ≥10 studies). Third, because AE outcomes were generally not reported at standardized time points and may have been collected using different methods across studies, the pooled safety estimates should be interpreted cautiously. Fourth, meta-regression analyses were limited by the small number of studies available for several models and by incomplete reporting of study-level covariates. Consequently, even statistically significant associations may be unstable and should be interpreted with caution. Finally, the effect sizes for some outcomes (e.g., the non-significant steroid-free remission at 12 months) were imprecise and could change with additional data.
Definitive assessment of comparative effectiveness and safety would benefit from prospective head-to-head randomized trials in biologic-naive UC populations that include both short-term induction and long-term maintenance endpoints (clinical, steroid-free, and endoscopic remission), standardized definitions of outcomes, and collection of safety events with sufficient follow-up to capture serious infections and other infrequent harms. Where RCTs are not feasible, large, well-designed prospective registries that capture baseline disease activity, comorbidity, biomarkers, and standardized outcome measures–coupled with advanced causal inference methods (e.g., propensity scores, instrumental variables, target trial emulation)–can strengthen causal inference from observational data. Research to identify biomarkers predictive of differential response (to enable personalized selection between gut-selective and systemic agents) and cost-effectiveness analyses that incorporate long-term persistence and safety profiles will also be valuable.
In biologic-naive UC patients, available comparative realworld and observational data suggest that vedolizumab is associated with greater long-term treatment persistence, higher rates of clinical and endoscopic remission at 1 year, and fewer SAEs compared with infliximab. However, no significant difference was observed in steroid-free remission, a more stringent and clinically important treat-to-target endpoint, indicating that the clinical relevance of the observed advantage in clinical remission should be interpreted cautiously and should not be assumed to reflect a clear steroid-sparing benefit. These findings are biologically plausible given vedolizumab’s gut-selective mechanism and the systemic immunomodulatory profile of anti-TNF therapy, but residual confounding in non-randomized studies requires that conclusions remain cautious. Prospective head-to-head trials and robust longitudinal registry studies are needed to confirm these observations and to guide evidence-based selection of first-line biologic therapy in UC.
NOTES
-
Funding Source
The authors received no financial support for the research, authorship, and/or publication of this article.
-
Conflict of Interest
No potential conflict of interest relevant to this article was reported.
-
Data Availability Statement
The authors confirm that the data supporting the findings of this study are available within the article and/or its supplementary materials.
-
Author Contributions
Conceptualization: Vatvani AD, Hariyanto TI, Mosquera DG, Stephanie S. Methodology: Vatvani AD, Hariyanto TI. Formal analysis: Vatvani AD, Hariyanto TI, Mosquera DG, Stephanie S. Investigation: Vatvani AD, Hariyanto TI, Mosquera DG, Stephanie S. Resources: Vatvani AD, Hariyanto TI. Supervision: Mosquera DG, Stephanie S. Writing–original draft: Vatvani AD, Hariyanto TI. Writing–review & editing: Mosquera DG, Stephanie S. Approval of final manuscript: all authors.
Supplementary Material
Supplementary materials are available at the Intestinal Research website (https://www.irjournal.org).
Supplementary Fig. 1.
Bubble–plot for meta-regression. Meta-regression analysis showed that the treatment persistence at 12 months outcome was not significantly affected by disease extent (A), Mayo endoscopic subscore (B), age (C), sex (D), ulcerative colitis (UC) duration (E), smokers (F), and concomitant steroids use (G).
ir-2026-00100-Supplementary-Fig-1.pdf
Supplementary Fig. 2.
Bubble–plot for meta-regression. Meta-regression analysis showed that the treatment persistence at 24 months outcome was significantly influenced only by concomitant steroid use (A), but not by disease extent (B), age (C), sex (D), and ulcerative colitis (UC) duration (E).
ir-2026-00100-Supplementary-Fig-2.pdf
Supplementary Fig. 3.
Bubble–plot for meta-regression. Meta-regression analysis showed that the treatment persistence at 36 months outcome was not significantly affected by disease extent (A), age (B), sex (C), and ulcerative colitis (UC) duration (D).
ir-2026-00100-Supplementary-Fig-3.pdf
Supplementary Fig. 4.
Bubble–plot for meta-regression. Meta-regression analysis showed that the treatment persistence at 48 months outcome was not significantly affected by disease extent (A), age (B), sex (C), and ulcerative colitis (UC) duration (D).
ir-2026-00100-Supplementary-Fig-4.pdf
Supplementary Fig. 5.
Bubble–plot for meta-regression. Meta-regression analysis showed that the clinical remission at 12 months outcome was significantly affected by Mayo endoscopic subscore (A), age (B), and sex (C), but not by disease extent (D), ulcerative colitis (UC) duration (E), smokers (F), nor concomitant steroids use (G).
ir-2026-00100-Supplementary-Fig-5.pdf
Supplementary Fig. 6.
Bubble–plot for meta-regression. Meta-regression analysis showed that the steroid-free remission at 12 months outcome was significantly affected only by ulcerative colitis (UC) duration (A), but not by disease extent (B), Mayo endoscopic subscore (C), age (D), sex (E), smokers (F), nor concomitant steroids use (G).
ir-2026-00100-Supplementary-Fig-6.pdf
Supplementary Fig. 7.
Bubble–plot for meta-regression. Meta-regression analysis showed that the endoscopic remission at 12 months outcome was significantly affected only by disease extent (A), but not by Mayo endoscopic subscore (B), age (C), sex (D), ulcerative colitis (UC) duration (E), smokers (F), nor concomitant steroids use (G).
ir-2026-00100-Supplementary-Fig-7.pdf
Supplementary Fig. 8.
Bubble–plot for meta-regression. Meta-regression analysis showed that the hospitalization outcome was not significantly affected by age.
ir-2026-00100-Supplementary-Fig-8.pdf
Supplementary Fig. 9.
Bubble–plot for meta-regression. Meta-regression analysis showed that the dose optimization at 12 months outcome was significantly affected by disease extent (A), but not by Mayo endoscopic subscore (B), age (C), sex (D), ulcerative colitis (UC) duration (E), nor concomitant steroids use (F).
ir-2026-00100-Supplementary-Fig-9.pdf
Supplementary Fig. 10.
Bubble–plot for meta-regression. Meta-regression analysis showed that the inflammatory bowel disease-related surgery outcome was not significantly affected by disease extent (A), age (B), sex (C), and ulcerative colitis (UC) duration (D).
ir-2026-00100-Supplementary-Fig-10.pdf
Supplementary Fig. 11.
Bubble plot for meta-regression. Meta-regression analysis showed that the total adverse events outcome was significantly affected only by concomitant steroid use (A), but not by disease extent (B), Mayo endoscopic subscore (C), age (D), sex (E), ulcerative colitis (UC) duration (F), or smoking status (G).
ir-2026-00100-Supplementary-Fig-11.pdf
Supplementary Fig. 12.
Bubble–plot for meta-regression. Meta-regression analysis showed that the serious adverse events outcome was not significantly affected by disease extent (A), Mayo endoscopic subscore (B), age (C), sex (D), ulcerative colitis (UC) duration (E), nor smokers (F), and concomitant steroids (G).
ir-2026-00100-Supplementary-Fig-12.pdf
Supplementary Fig. 13.
(A) Funnel plot analysis for the treatment persistence at 12 months outcome. (B) Funnel plot analysis for the treatment persistence at 24 months outcome. (C) Funnel plot analysis for the steroid-free remission at 12 months outcome.
ir-2026-00100-Supplementary-Fig-13.pdf
Fig. 1.PRISMA diagram of the detailed process of selection of studies for inclusion in the systematic review and meta-analysis.
Fig. 2.Forest plot of long-term treatment persistence at 12, 24, 36, 48, and 60 months comparing vedolizumab versus infliximab in biologic- naive ulcerative colitis using a random-effects model. M-H, Mantel-Haenszel; CI, confidence interval.
Fig. 3.Forest plots comparing (A) clinical remission, (B) steroid-free remission, and (C) endoscopic remission at 12 months between vedolizumab and infliximab in biologic-naive ulcerative colitis using a random-effects model. M-H, Mantel-Haenszel; CI, confidence interval.
Table 1.Baseline Characteristics of the Included Studies
|
Author (year) |
Country |
Design |
Follow-up (mo) |
Disease extent |
MES |
Study arms |
Sample size |
Age (yr) |
Male (%) |
UC duration (yr) |
Smokers (%) |
Steroids (%) |
|
Alghamdi et al. (2026) [11] |
Saudi Arabia |
RC |
12 |
E1 (7.4%), E2 (29.3%), E3 (63.3%) |
NR |
VDZ |
29 |
40.8 ± 14.2 |
44.4 |
12.3 ± 10.6 |
7.7 |
NR |
|
IFX |
93 |
41.5 ± 12.9 |
|
|
|
|
|
Atia et al. (2025) [12] |
Israel |
RC |
60 |
NR |
NR |
VDZ |
553 |
34.4 ± 17.7 |
49 |
2.5 ± 0.8 |
NR |
NR |
|
IFX |
1,186 |
28.3 ± 16.0 |
48 |
1.8 ± 0.5 |
|
|
|
Bokemeyer et al. (2023) [13] |
Germany |
PC |
24 |
E1 (7.6%), E2 (37.8%), E3 (54.6%) |
NR |
VDZ |
182 |
40.4 ± 18.6 |
50.4 |
5.9 ± 8.0 |
7.4 |
94.0 |
|
IFX |
77 |
40.0 ± 16.1 |
49 |
6.8 ± 10.2 |
6.3 |
94.3 |
|
Bokemeyer et al. (2025) [14] |
Germany |
PC |
12 |
E3 (51.2%) |
NR |
VDZ |
138 |
40.9 ± 14.6 |
48.9 |
10.2 ± 9.7 |
8.8 |
93.4 |
|
IFX |
83 |
40.8 ± 13.9 |
47.9 |
10.0 ± 9.4 |
9.0 |
93.1 |
|
Cassinotti et al. (2022) [15] |
Italy |
RC |
12 |
E1 (11.7%), E2 (34.8%), E3 (53.5%) |
1 (2.8%), 2 (44.9%), 3 (52.3%) |
VDZ |
34 |
47.0 ± 16.9 |
66.9 |
9.3 ± 7.4 |
4.2 |
42.3 |
|
IFX |
73 |
39.1 ± 14.4 |
56.2 |
5.8 ± 5.2 |
3.8 |
58.1 |
|
Krugliak Cleveland et al. (2024) [16] |
USA |
RC |
12 |
E1 (41.4%), E2 (9.4%), E3 (49.2%) |
NR |
VDZ |
213 |
47.2 ± 18.5 |
53.8 |
NR |
12.2 |
NR |
|
IFX |
661 |
43.6 ± 16.8 |
46.8 |
NR |
18.0 |
NR |
|
da Costa Ferreira et al. (2025) [7] |
Brazil |
RC |
12 |
E1 (3.7%), E2 (35.0%), E3 (61.3%) |
2 (38.7%), 3 (61.3%) |
VDZ |
141 |
44.7 ± 11.8 |
45.1 |
9.5 ± 6.2 |
2.1 |
75.8 |
|
IFX |
156 |
43.2 ± 11.3 |
41.3 |
9.9 ± 8.5 |
3.4 |
78.2 |
|
Dalal et al. (2022) [17] |
USA |
RC |
48 |
E1 (12.4%), E2-E3 (87.6%) |
0 (4%), 1 (14.6%), 2 (44.2%), 3 (37.2%) |
VDZ |
195 |
47.6 ± 24.6 |
44.6 |
9.1 ± 11.9 |
1.5 |
71.5 |
|
IFX |
332 |
37.1 ± 18.6 |
44.3 |
4.7 ± 6.7 |
4.8 |
88.0 |
|
Dalal et al. (2025) [18] |
USA |
RC |
12 |
E1 (100%) |
0 (7%), 1 (29%), 2 (43%), 3 (21%) |
VDZ |
34 |
41.0 ± 16.6 |
29 |
NR |
1.0 |
46.0 |
|
IFX |
8 |
|
|
|
|
|
|
Dignass et al. (2025) [19] |
Germany |
RC |
60 |
NR |
NR |
VDZ |
248 |
44.9 |
NR |
NR |
NR |
NR |
|
IFX |
419 |
41.0 |
NR |
NR |
NR |
NR |
|
Haynesworth et al. (2025) [20] |
USA |
RC |
36 |
E1 (9.7%), E2 (43.1%), E3 (47.2%) |
0-1 (4.8%), 2 (54.2%), 3 (40.9%) |
VDZ |
72 |
41.8 ± 15.7 |
50 |
8.0 ± 1.9 |
34.7 |
40.3 |
|
IFX |
72 |
40.3 ± 16.5 |
43.1 |
6.6 ± 0.7 |
34.7 |
44.4 |
|
Huang et al. (2024) [21] |
China |
RC |
12 |
E1 (6.2%), E2 (25.6%), E3 (68.2%) |
1 (10.5%), 2 (57.7%), 3 (31.8%) |
VDZ |
117 |
40.7 ± 15.3 |
56.3 |
4.5 ± 4.9 |
6.1 |
40.6 |
|
IFX |
82 |
41.1 ± 14.6 |
62.6 |
5.0 ± 4.4 |
5.0 |
40.7 |
|
Jensen et al. (2023) [22] |
Denmark |
RC |
24 |
NR |
NR |
VDZ |
98 |
60.3 ± 22.5 |
58 |
5.6 ± 6.7 |
NR |
15.0 |
|
IFX |
2,677 |
39.7 ± 19.3 |
51 |
3.0 ± 5.2 |
NR |
21.0 |
|
Kapizioni et al. (2024) [23] |
UK |
RC |
60 |
E2 (59.6%), E3 (40.4%) |
NR |
VDZ |
621 |
48.3 ± 23.0 |
59 |
11.3 ± 8.2 |
13.0 |
3.2 |
|
IFX |
2,339 |
38.0 ± 17.0 |
53 |
10.0 ± 6.7 |
15.0 |
17.0 |
|
Koo et al. (2023) [24] |
South Korea |
RC |
24 |
NR |
NR |
VDZ |
164 |
44.1 ± 13.7 |
67.7 |
NR |
NR |
39.6 |
|
IFX |
3,019 |
42.8 ± 11.8 |
62.8 |
NR |
NR |
65.6 |
|
Ko et al. (2021) [25] |
Australia |
RC |
60 |
NR |
NR |
VDZ |
167 |
45.7 ± 11.2 |
51.5 |
NR |
NR |
NR |
|
IFX |
399 |
40.5 ± 9.3 |
53.1 |
NR |
NR |
NR |
|
Lukin et al. (2022) [26] |
USA |
RC |
12 |
E1 (5.6%), E2 (37.5%), E3 (56.9%) |
1 (9.9%), 2 (48.9%), 3 (41.2%) |
VDZ |
143 |
42.8 ± 18.8 |
39.9 |
6.0 ± 11.5 |
2.8 |
51.7 |
|
IFX |
114 |
39.2 ± 15.9 |
51.2 |
4 ± 10 |
5.0 |
52.5 |
|
Matsuoka et al. (2025) [27] |
Japan |
RC |
24 |
NR |
NR |
VDZ |
335 |
40.4 ± 13.7 |
60.9 |
2.4 ± 2.2 |
NR |
67.1 |
|
IFX |
639 |
39.0 ± 13.5 |
65.8 |
1.9 ± 1.9 |
NR |
74.1 |
|
Meng et al. (2024) [28] |
China |
RC |
12 |
E2 (31.9%), E3 (68.1%) |
2 (81.9%), 3 (18.1%) |
VDZ |
57 |
45.4 ± 16.7 |
66.7 |
4.2 ± 5.7 |
31.6 |
10.5 |
|
IFX |
65 |
39.5 ± 15.1 |
60 |
2.7 ± 3.0 |
18.5 |
7.7 |
|
Patel et al. (2019) [29] |
USA |
RC |
12 |
E1 (15.8%), E2 (27.1%), E3 (57.1%) |
NR |
VDZ |
247 |
51.3 ± 16.8 |
55.9 |
4.2 ± 4.5 |
NR |
NR |
|
IFX |
469 |
51.3 ± 16.8 |
55.9 |
4.2 ± 4.5 |
NR |
NR |
|
Pudipeddi et al. (2022) [30] |
Australia |
RC |
48 |
NR |
NR |
VDZ |
169 |
44.3 ± 17.2 |
49.7 |
NR |
NR |
NR |
|
IFX |
251 |
39.8 ± 19.4 |
54.2 |
NR |
NR |
NR |
|
Sablich et al. (2023) [6] |
Italy |
RC |
12 |
E1 (3.6%), E2 (46.3%), E3 (50.1%) |
1 (30.5%), 2 (59.7%), 3 (9.8%) |
VDZ |
32 |
47.2 ± 17.8 |
53 |
8.0 ± 8.5 |
NR |
56.0 |
|
IFX |
50 |
35.4 ± 13.8 |
66 |
9.5 ± 9.3 |
NR |
42.0 |
|
Shehab et al. (2024) [31] |
Kuwait |
RC |
12 |
E1 (19%), E2 (34%), E3 (49%) |
NR |
VDZ |
21 |
34.5 ± 11.4 |
54.3 |
1.4 ± 2.7 |
NR |
NR |
|
IFX |
72 |
|
|
|
|
|
|
Supovec et al. (2025) [32] |
Slovenia |
RC |
48 |
E1 (3.7%), E2 (37.4%), E3 (58.9%) |
NR |
VDZ |
113 |
47.6 ± 20.3 |
49.6 |
8.2 ± 6.9 |
NR |
NR |
|
IFX |
152 |
37.6 ± 14.9 |
47.3 |
5.3 ± 6.0 |
NR |
NR |
|
Wagner et al. (2024) [33] |
Germany |
RC |
48 |
E1 (16.5%), E2 (33.9%), E3 (49.5%) |
NR |
VDZ |
45 |
39.5 |
56.1 |
9.5 |
NR |
NR |
|
IFX |
125 |
|
|
|
|
|
Table 2.Newcastle–Ottawa Quality Assessment of Observational Studies
|
Author (year) |
Study design |
Selectiona
|
Comparabilityb
|
Outcomec
|
Total score |
Result |
|
Alghamdi et al. (2026) [11] |
Cohort |
*** |
** |
** |
7 |
Good |
|
Atia et al. (2025) [12] |
Cohort |
*** |
** |
*** |
8 |
Good |
|
Bokemeyer et al. (2023) [13] |
Cohort |
*** |
** |
*** |
8 |
Good |
|
Bokemeyer et al. (2025) [14] |
Cohort |
*** |
** |
** |
7 |
Good |
|
Cassinotti et al. (2022) [15] |
Cohort |
*** |
** |
** |
7 |
Good |
|
Krugliak Cleveland et al. (2024) [16] |
Cohort |
*** |
** |
** |
7 |
Good |
|
da Costa Ferreira et al. (2025) [7] |
Cohort |
*** |
** |
** |
7 |
Good |
|
Dalal et al. (2022) [17] |
Cohort |
*** |
** |
*** |
8 |
Good |
|
Dalal et al. (2025) [18] |
Cohort |
*** |
** |
** |
7 |
Good |
|
Dignass et al. (2025) [19] |
Cohort |
*** |
** |
** |
7 |
Good |
|
Haynesworth et al. (2025) [20] |
Cohort |
*** |
** |
*** |
8 |
Good |
|
Huang et al. (2024) [21] |
Cohort |
*** |
** |
** |
7 |
Good |
|
Jensen et al. (2023) [22] |
Cohort |
*** |
** |
*** |
8 |
Good |
|
Kapizioni et al. (2024) [23] |
Cohort |
*** |
** |
*** |
8 |
Good |
|
Koo et al. (2023) [24] |
Cohort |
*** |
** |
*** |
8 |
Good |
|
Ko et al. (2021) [25] |
Cohort |
*** |
** |
*** |
8 |
Good |
|
Lukin et al. (2022) [26] |
Cohort |
*** |
** |
** |
7 |
Good |
|
Matsuoka et al. (2025) [27] |
Cohort |
*** |
** |
*** |
8 |
Good |
|
Meng et al. (2024) [28] |
Cohort |
*** |
** |
** |
7 |
Good |
|
Patel et al. (2019) [29] |
Cohort |
*** |
** |
** |
7 |
Good |
|
Pudipeddi et al. (2022) [30] |
Cohort |
*** |
** |
** |
7 |
Good |
|
Sablich et al. (2023) [6] |
Cohort |
*** |
** |
** |
7 |
Good |
|
Shehab et al. (2024) [31] |
Cohort |
*** |
** |
** |
7 |
Good |
|
Supovec et al. (2025) [32] |
Cohort |
*** |
** |
*** |
8 |
Good |
|
Wagner et al. (2024) [33] |
Cohort |
*** |
** |
*** |
8 |
Good |
Table 3.Summary of Meta-Analysis Results for Each Outcome of Interest
|
Outcome |
Studies |
Patients analyzed
|
OR (95% CI) |
P-value |
I2 (%) |
|
VDZ |
IFX |
|
Treatment persistence |
|
|
|
|
|
|
|
12 Months |
20 |
2,095 |
7,942 |
1.73 (1.50–1.99) |
< 0.001 |
8 |
|
24 Months |
10 |
872 |
4,380 |
1.78 (1.35–2.34) |
< 0.001 |
46 |
|
36 Months |
7 |
303 |
815 |
1.67 (1.23–2.26) |
0.001 |
0 |
|
48 Months |
6 |
300 |
786 |
1.84 (1.37–2.47) |
< 0.001 |
0 |
|
60 Months |
3 |
20 |
214 |
2.19 (0.82–5.84) |
0.120 |
0 |
|
Clinical remission |
|
|
|
|
|
|
|
12 Months |
8 |
655 |
720 |
1.96 (1.37–2.81) |
< 0.001 |
49 |
|
Steroid-free remission |
|
|
|
|
|
|
|
12 Months |
10 |
625 |
724 |
1.34 (0.95–1.90) |
0.090 |
41 |
|
Endoscopic remission |
|
|
|
|
|
|
|
12 Months |
7 |
382 |
547 |
1.48 (1.09–1.99) |
0.010 |
0 |
|
Hospitalization |
|
|
|
|
|
|
|
12 Months |
4 |
386 |
517 |
0.49 (0.29–0.83) |
0.008 |
40 |
|
24 Months |
2 |
80 |
132 |
0.64 (0.30–1.35) |
0.240 |
0 |
|
36 Months |
2 |
102 |
134 |
0.40 (0.19–0.85) |
0.020 |
0 |
|
IBD-related surgery |
|
|
|
|
|
|
|
12 Months |
4 |
280 |
385 |
0.74 (0.27–2.04) |
0.570 |
32 |
|
Dose optimization |
|
|
|
|
|
|
|
12 Months |
5 |
392 |
456 |
0.45 (0.24–0.83) |
0.010 |
47 |
|
Total AEs |
8 |
705 |
566 |
0.71 (0.52–0.97) |
0.030 |
5 |
|
Serious AEs |
6 |
675 |
815 |
0.19 (0.12–0.30) |
< 0.001 |
0 |
REFERENCES
- 1. Bhattacharya A, Osterman MT. Biologic therapy for ulcerative colitis. Gastroenterol Clin North Am 2020;49:717–729.ArticlePubMed
- 2. Rutgeerts P, Sandborn WJ, Feagan BG, et al. Infliximab for induction and maintenance therapy for ulcerative colitis. N Engl J Med 2005;353:2462–2476.ArticlePubMedPMC
- 3. Feagan BG, Rutgeerts P, Sands BE, et al. Vedolizumab as induction and maintenance therapy for ulcerative colitis. N Engl J Med 2013;369:699–710.ArticlePubMedPMC
- 4. Peyrin-Biroulet L, Arkkila P, Armuzzi A, et al. Comparative efficacy and safety of subcutaneous infliximab and vedolizumab in patients with Crohn’s disease and ulcerative colitis included in randomised controlled trials. BMC Gastroenterol 2024;24:121.PubMedPMC
- 5. Laharie D, Filippi J, Roblin X, et al. Impact of mucosal healing on long-term outcomes in ulcerative colitis treated with infliximab: a multicenter experience. Aliment Pharmacol Ther 2013;37:998–1004.ArticlePubMed
- 6. Sablich R, Urbano MT, Scarpa M, Scognamiglio F, Paviotti A, Savarino E. Vedolizumab is superior to infliximab in biologic naive patients with ulcerative colitis. Sci Rep 2023;13:1816.PubMedPMC
- 7. da Costa Ferreira S, Parra RS, Sassaki LY, et al. Clinical effectiveness and safety of vedolizumab versus infliximab in biologic-naive patients with ulcerative colitis: a comparative realworld multicentric observational study. Gastroenterol Hepatol 2025;48:502396.PubMed
- 8. Page MJ, McKenzie JE, Bossuyt PM, et al. The PRISMA 2020 statement: an updated guideline for reporting systematic reviews. BMJ 2021;372–n71.Article
- 9. Stang A. Critical evaluation of the Newcastle-Ottawa scale for the assessment of the quality of nonrandomized studies in meta-analyses. Eur J Epidemiol 2010;25:603–605.ArticlePubMedPMCPDF
- 10. Higgins JP, Thompson SG, Deeks JJ, Altman DG. Measuring inconsistency in meta-analyses. BMJ 2003;327:557–560.ArticlePubMedPMC
- 11. Alghamdi AG, Aldofaian HS, Alahmari MS, et al. Patterns and predictors of biologic therapy switching in inflammatory bowel disease at a tertiary care center: a retrospective observational study. Saudi J Gastroenterol 2026;32:49–59.ArticlePubMedPMC
- 12. Atia O, Friss C, Focht G, et al. Durability of the first biologic in children and adults with ulcerative colitis: a nationwide study from the epi-IIRN. Inflamm Bowel Dis 2025;31:617–624.ArticlePubMedPDF
- 13. Bokemeyer B, Plachta-Danielzik S, di Giuseppe R, et al. Real-world effectiveness of vedolizumab compared to anti-TNF agents in biologic-naive patients with ulcerative colitis: a two-year propensity-score-adjusted analysis from the prospective, observational VEDOIBD-study. Aliment Pharmacol Ther 2023;58:429–442.ArticlePubMed
- 14. Bokemeyer B, Plachta-Danielzik S, Gilman E, et al. Real-world effectiveness of ustekinumab versus anti-TNF or vedolizumab in ulcerative colitis: induction and 12-month maintenance results from the prospective, observational RUN-UC study. J Crohns Colitis 2025;19:jjaf052.ArticlePubMedPDF
- 15. Cassinotti A, Mezzina N, De Silvestri A, et al. Continuous clinical remission with biologics in ulcerative colitis: the ‘AURORA’ comparison study. Eur J Gastroenterol Hepatol 2022;34:1238–1246.ArticlePubMed
- 16. Krugliak Cleveland N, Ghosh S, Chastek B, et al. Real-world persistence of successive biologics in patients with inflammatory bowel disease: findings from ROTARY. Inflamm Bowel Dis 2024;30:1776–1787.ArticlePubMedPMCPDF
- 17. Dalal RS, McClure EL, Marcus J, Allegretti JR. Comparative long-term drug survival of vedolizumab, adalimumab, and infliximab in biologic-naive patients with ulcerative colitis. Dig Dis Sci 2023;68:223–232.ArticlePubMedPDF
- 18. Dalal RS, Clarke LM, Carlin A, Cabral H, Allegretti JR. Real-world comparison of effectiveness, treatment persistence, and safety of first-line advanced therapies at 1 year for ulcerative proctitis. Inflamm Bowel Dis 2025;31:1174–1177.ArticlePubMedPDF
- 19. Dignass A, Teich N, Kaiser S, et al. Treatment sequences, outcomes, healthcare utilization, and costs in patients with inflammatory bowel diseases requiring advanced treatment-real world comparative effectiveness from German claims data. BMC Gastroenterol 2025;25:436.ArticlePubMedPMCPDF
- 20. Haynesworth A, Yeh KH, Lee HH, et al. Long-term outcomes of an infliximab-first versus vedolizumab-first treatment strategy in biologic-naive patients with ulcerative colitis. Aliment Pharmacol Ther 2025;61:693–701.ArticlePubMedPMC
- 21. Huang Z, Tang J, Wu R, et al. Comparison of clinical and endoscopic efficacy between vedolizumab and infliximab in bio-naive patients with ulcerative colitis: a multicenter, real-world study. Ther Adv Gastroenterol 2024;17:17562848241281218.PubMedPMC
- 22. Jensen KJ, Jensen CB, Wennerström C, Burisch J, Petersen J. Drug utilization of biologic therapy in Crohn’s disease and ulcerative colitis: a population-based Danish cohort study 2015-2020. Scand J Gastroenterol 2023;58:726–736.ArticlePubMed
- 23. Kapizioni C, Desoki R, Lam D, 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.PubMedPMC
- 24. Koo HM, Jun YK, Choi Y, et al. 10 years of biologic use patterns in patients with inflammatory bowel disease: treatment persistence, switching and dose intensification: a nationwide population-based study. Ther Adv Gastroenterol 2023;16:17562848231201728.ArticlePubMedPMCPDF
- 25. Ko Y, Paramsothy S, Yau Y, Leong RW. Superior treatment persistence with ustekinumab in Crohn’s disease and vedolizumab in ulcerative colitis compared with anti-TNF biological agents: real-world registry data from the Persistence Australian National IBD Cohort (PANIC) study. Aliment Pharmacol Ther 2021;54:292–301.ArticlePubMedPDF
- 26. Lukin D, Faleck D, Xu R, et al. Comparative safety and effectiveness of vedolizumab to tumor necrosis factor antagonist therapy for ulcerative colitis. Clin Gastroenterol Hepatol 2022;20:126–135.ArticlePubMedPMC
- 27. Matsuoka K, Nakajo K, Kawamura S, et al. Persistence of advanced therapies in patients with inflammatory bowel disease: retrospective cohort study using a large healthcare claims database in Japan. Intest Res 2025;23:358–371.ArticlePubMedPMCPDF
- 28. Meng RP, Huang BB, Wei YL, et al. Effectiveness and safety of vedolizumab and infliximab in biologic-naive patients with moderate-to-severe ulcerative colitis: a multicenter, retrospective cohort study. J Dig Dis 2024;25:230–237.PubMed
- 29. Patel H, Latremouille-Viau D, Burne R, Shi S, Adsul S. Comparison of real-world treatment outcomes with vedolizumab versus infliximab in biologic-naive patients with inflammatory bowel disease. Crohns Colitis 360 2019;1:otz022.ArticlePDF
- 30. Pudipeddi A, Ko Y, Paramsothy S, Leong RW. Vedolizumab has longer persistence than infliximab as a first-line biological agent but not as a second-line biological agent in moderate-to-severe ulcerative colitis: real-world registry data from the Persistence Australian National IBD Cohort (PANIC) study. Ther Adv Gastroenterol 2022;15:17562848221080793.ArticlePubMedPMCPDF
- 31. Shehab M, Alfadhli A, Abdullah I, Alostad W, Marei A, Alrashed F. Effectiveness of biologic therapies in achieving treatment targets in inflammatory bowel disease; real-world data from the Middle East (ENROLL study). Front Pharmacol 2024;15:1388043.ArticlePubMedPMC
- 32. Supovec E, Hanžel J, Novak G, Manevski D, Štabuc B, Drobne D. First-line anti-TNF agents, ustekinumab and vedolizumab perform similarly in Crohn’ disease, but not in ulcerative colitis. Eur J Gastroenterol Hepatol 2025;37:557–564.ArticlePubMedPMC
- 33. Wagner K, Müller TM, Vitali F, et al. Treatment trajectories and outcomes in inflammatory bowel disease: a tertiary single-centre experience. Ther Adv Gastroenterol 2024;17:17562848241284051.ArticlePubMedPMCPDF
- 34. Thornton A, Lee P. Publication bias in meta-analysis: its causes and consequences. J Clin Epidemiol 2000;53:207–216.ArticlePubMedPMC
- 35. Terrin N, Schmid CH, Lau J, Olkin I. Adjusting for publication bias in the presence of heterogeneity. Stat Med 2003;22:2113–2126.ArticlePubMed
- 36. Lord JD, Long SA, Shows DM, et al. Circulating integrin alpha4/beta7+ lymphocytes targeted by vedolizumab have a pro-inflammatory phenotype. Clin Immunol 2018;193:24–32.ArticlePubMedPMC
- 37. Arijs I, De Hertogh G, Lemmens B, et al. Effect of vedolizumab (anti-a4b7-integrin) therapy on histological healing and mucosal gene expression in patients with UC. Gut 2018;67:43–52.ArticlePubMed
- 38. Hariyanto TI. From dysbiosis to decolonization: the emerging role of fecal microbiota transplantation in antimicrobial resistance. Infect Med (Beijing) 2026;5:100253.ArticlePubMedPMC
- 39. Lichtenstein L, Ron Y, Kivity S, et al. Infliximab-related infusion reactions: systematic review. J Crohns Colitis 2015;9:806–815.ArticlePubMedPMC
- 40. Rosenblum H, Amital H. Anti-TNF therapy: safety aspects of taking the risk. Autoimmun Rev 2011;10:563–568.ArticlePubMed
- 41. Peyrin-Biroulet L, Arkkila P, Armuzzi A, et al. Comparative efficacy and safety of infliximab and vedolizumab therapy in patients with inflammatory bowel disease: a systematic review and meta-analysis. BMC Gastroenterol 2022;22:291.ArticlePubMedPMCPDF
Citations
Citations to this article as recorded by
KASID
, Timotius Ivan Hariyanto2
PubReader
ePub Link
Cite
XML Download
