Key Points
Question
Among biologic-naive patients with ulcerative colitis initiating advanced therapy, how do long-term outcomes compare between tumor necrosis factor (TNF) antagonists and vedolizumab as first-line treatment?
Findings
In this cohort study of 3263 patients followed up for as long as 5 years, there were no significant differences in burden of hospitalization or surgery among those initiating vedolizumab compared with those initiating infliximab or adalimumab; however, initiation of vedolizumab was associated with greater time spent in corticosteroid-free disease stability compared with infliximab or adalimumab. Results were consistent in sensitivity analyses using propensity score matching.
Meaning
The findings of this study suggest that a vedolizumab-first strategy may be associated with improved long-term clinical outcomes compared with a TNF antagonist–first strategy in biologic-naive patients with ulcerative colitis.
This cohort study compares the long-term clinical outcomes among patients in Sweden with ulcerative colitis who initiate a tumor necrosis factor (TNF) antagonist (ie, infliximab or adalimumab) vs those who initiate vedolizumab as their first biologic treatment.
Abstract
Importance
Tumor necrosis factor (TNF) antagonists and vedolizumab are the most commonly used advanced therapies for management of moderate-to-severe ulcerative colitis (UC). There are limited data evaluating long-term outcomes with a TNF antagonist–first vs vedolizumab-first treatment strategy.
Objective
To compare long-term clinical outcomes among biologic-naive patients with UC initiating infliximab or adalimumab vs vedolizumab as first-line advanced therapy.
Design, Setting, and Participants
This Swedish nationwide population-based cohort study used linked health registers. Biologic-naive patients with UC who initiated infliximab, adalimumab, or vedolizumab between January 1, 2015, and June 30, 2021, were included and followed up until June 30, 2023, for a median of 4.3 to 5.4 years. Data analysis was conducted from April to September 2025.
Exposures
Initiation of a TNF antagonist (ie, infliximab or adalimumab) or vedolizumab as first-line advanced therapy.
Main Outcomes and Measures
The primary outcome was the incidence rate of all-cause hospitalizations (including recurrent events). Secondary outcomes included incidence rates of colectomy, corticosteroid dispensations, advanced therapy switching, and proportion of quarters spent in corticosteroid-free disease stability (defined as no oral corticosteroid dispensation, no initiation of a new advanced therapy or immunomodulator, and no hospitalization or colectomy within a 3-month period). Groups were balanced using inverse probability of treatment weighting (IPTW) accounting for disease characteristics, health care utilization, comorbidities, and medication use.
Results
Overall, 3263 patients (1393 infliximab, 1583 adalimumab, and 287 vedolizumab initiators) were included. After IPTW, there were with 1385 infliximab initiators (mean [SD] age, 38 [18] years; 43% female) compared with 320 vedolizumab initiators (mean [SD] age, 36 [19] years; 35% female) and 1575 adalimumab initiators (mean [SD] age, 41 [16] years; 48% female) compared with 318 vedolizumab initiators (mean [SD] age, 38 [19] years; 43% female). All-cause hospitalization did not differ significantly between infliximab and vedolizumab initiators (incidence rate ratio [IRR], 1.14 [95% CI, 0.91-1.43]); however, infliximab initiators spent a significantly lower proportion of time in corticosteroid-free disease stability over 5 years than vedolizumab initiators (60% vs 67% of quarters; risk ratio [RR], 0.88 [95% CI, 0.82-0.94]). Similarly, all-cause hospitalization did not differ significantly between adalimumab and vedolizumab initiators (IRR, 1.08 [95% CI, 0.86-1.35]), although adalimumab initiators had greater therapy switching (IRR, 1.21 [95% CI, 1.02-1.43]) and less time in corticosteroid-free stability (67% vs 70% of quarters; RR, 0.92 [95% CI, 0.87-0.98]) than vedolizumab initiators. Results were broadly consistent in 1:1 propensity score–matched sensitivity analyses.
Conclusions and Relevance
In this cohort study of biologic-naive patients with UC initiating advanced therapy, the burden of hospitalization did not differ significantly between patients receiving vedolizumab and those receiving a TNF agonist; however, a vedolizumab-first strategy was associated with a greater proportion of time spent in corticosteroid-free disease stability over 5 years than a TNF antagonist–first strategy. These findings may inform first-line treatment selection in UC.
Introduction
As treatment options for the management of moderate-to-severe ulcerative colitis (UC) have expanded significantly, positioning and sequencing of different therapies have garnered interest.1 Tumor necrosis factor (TNF) antagonists and vedolizumab are the most commonly used advanced therapies in patients with UC.2 Several observational studies have suggested that infliximab and vedolizumab may have comparable effectiveness in treating outpatients starting advanced therapies, measured either as time to treatment failure, treatment persistence, or achievement of clinical and/or endoscopic remission within 6 to 12 months.3,4,5 While the VARSITY trial demonstrated higher rates of achieving clinical and endoscopic remission with vedolizumab vs adalimumab in a subset of biologic-naive patients with moderate-to-severe UC at 1 year, rates of corticosteroid-free clinical remission were not different.6 However, most of these studies stop patient follow-up once a patient achieves any outcome on index therapy and do not adequately inform long-term outcomes with subsequent treatment decisions and treatment sequencing.
Beyond comparing effectiveness of specific medications, comparing treatment strategies in an intention-to-treat approach (whether to use infliximab or adalimumab first vs vedolizumab first in the patient’s treatment course) may be more appropriate to inform whether one approach is more likely to achieve lower long-term burden of hospitalization, surgery, and corticosteroid use (ie, not only longer time to first event). This approach intrinsically accounts for subsequent decisions regarding treatment optimization, switching to alternative therapies in case of failure of, or intolerance to, first-line therapy. It is biologically possible that exposure to one medication, rather than another, may prime the body to respond favorably or unfavorably to subsequent medications, making a comparison of treatment strategies additionally interesting.7 Hence, we conducted a nationwide register-based cohort study comparing long-term outcomes in biologic-naive patients with UC who initiated either infliximab or adalimumab vs vedolizumab as first-line advanced therapy.
Methods
Study Design
This nationwide cohort study was based on population-based registers with prospectively collected data from routine medical practice. The study was approved by the Regional Ethics Committee in Stockholm (DNR 2014/1287-31/4, 2021-06209-01, 2022-04384-02, and 2023-04868-02). Under the Swedish Act concerning the Ethical Review of Research Involving Humans (SFS 2003:460), individual informed consent is not required for strictly register-based research, and a waiver of consent was granted by the Swedish Ethical Review Authority. Reporting followed the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) reporting guideline.
Setting and Data Sources
In Sweden, every interaction with the health care system is documented in nationwide medical registers, which are linked using a unique personal identity number.8 These registers include data on vital status, migration, morbidity, and mortality and provide virtually complete follow-up. Sweden’s health care system is tax funded and ensures equal access for all residents, with no or minimal copayment for prescription drugs, outpatient services, and inpatient care.9 Baseline and follow-up information—including demographic characteristics, disease characteristics, treatments, and outcomes—was obtained from multiple national databases: the Swedish Inflammatory Bowel Disease Register (SWIBREG),10 the Swedish National Patient Register,11,12 the Prescribed Drug Register,13,14 the Swedish Cancer Register,15 the Swedish nationwide histopathology cohort (ESPRESSO),16 and the Total Population Register.17
Study Population
We identified patients with UC as those having 2 or more records with a relevant International Statistical Classification of Diseases and Related Health Problems, Tenth Revision (ICD-10) code or 1 such record plus a colorectal biopsy report with a morphology code suggestive of inflammatory bowel disease (IBD).18,19,20 From these patients, we selected those who initiated infliximab, adalimumab, or vedolizumab as their first-line advanced therapy between January 1, 2015, and June 30, 2021, and had at least 2 years follow-up after initiation of advanced therapy (to ensure we capture long-term outcomes with different treatment strategies). We excluded patients with colectomy prior to initiation of advanced therapy; patients who initiated index advanced therapy during a hospitalization (since this would typically represent acute severe colitis and infliximab initiation); patients who initiated an alternative advanced therapy as first-line therapy; patients with HIV, chronic liver or kidney disease, solid organ or bone marrow transplantation, or cancer (excluding nonmelanoma skin cancers); and those who resided outside Sweden in the preceding 5 years.
Exposure and Comparator
Our primary exposures of interest were initiation of infliximab (exposure 1, first-line infliximab strategy) or adalimumab (exposure 2, first-line adalimumab strategy) as first-line advanced therapy for patients with UC, and our primary comparator was initiation of vedolizumab (first-line vedolizumab strategy). Since our objective was to compare long-term outcomes with different treatment strategies, rather than comparing specific therapies, patients were analyzed in the group they were originally assigned (based on initiation of first-line advanced therapy), regardless of whether they continued to take the index advanced therapy, switched to an alternative therapy, or experienced clinical events such as colectomy or hospitalization. Date of initiation of first advanced therapy was defined as the index date, and patients were followed up until end of the study period (June 30, 2023), emigration, or death.
Outcomes
To evaluate disease course after treatment initiation over a long horizon, our primary outcome was the incidence rate (IR) of any (ie, not only first) all-cause hospitalization. Our secondary outcomes were IRs of IBD-related hospitalization, colectomy, systemic oral corticosteroid dispensation, and switching to alternative advanced therapy, after initiation of infliximab or adalimumab vs vedolizumab, over the course of follow-up.21 IBD-related hospitalization was defined as any hospitalization where the primary discharge diagnosis was UC. We also compared the proportion of eligible quarters of follow-up in which the patient was deemed to be in corticosteroid-free disease stability (no dispensed systemic oral corticosteroid prescription, no start of new advanced therapies or immunomodulator, and no hospitalization or colectomy during a quarter was considered corticosteroid-free disease stability in that quarter) between the 2 treatment strategies. We did not have access to patient- or physician-reported outcomes, biomarkers of inflammation, or endoscopic reports. Serious infection was included as a secondary safety outcome and was defined as hospitalization with infection as the primary diagnosis, ascertained from the National Patient Register.21 Switching of advanced therapy was defined as dispensation of a different advanced-therapy molecule; dose escalation of the index agent, biosimilar-to-originator (or originator-to-biosimilar) substitution of the same molecule, and gaps in or restarts of the same therapy were not counted as switches. Steroid excess use (>1 course) was defined using an assumed course duration of 83 days: a new dispensed oral corticosteroid prescription within 83 days of the preceding fill was counted as excess use, whereas a new fill more than 83 days after the preceding fill was counted as a separate course. All-cause and IBD-related death were examined using the Cause of Death Register but were too infrequent to analyze (IBD-related deaths: infliximab, 2; adalimumab, 0; vedolizumab, 3); patients were censored at death and emigration.
Covariates
We extracted baseline covariates (at time of initiation of advanced therapy). First, demographic characteristics (age at diagnosis of UC, age at time of advanced therapy initiation, sex, education level, and country of birth) were collected. Disease and treatment characteristics at initiation of infliximab, adalimumab, or vedolizumab were also extracted. These included time since diagnosis, disease extent at treatment initiation, all-cause and IBD-related hospitalization, IBD-related office visits in 2 years prior to index date, prescription medication use in preceding 2 years, any prior IBD-related medication use, systemic corticosteroid use in preceding 12 months examined as prednisolone equivalents in milligrams, and concomitant treatment with immunomodulators (azathioprine, mercaptopurine, or methotrexate prescription within 0-6 months before biologic initiation). Finally, health care utilization, including comorbidity burden measured by modified Charlson Comorbidity index (CCI) and frailty status (based on Hospital Frailty Risk score) based on codes, within the preceding 5 years was extracted.22
Statistical Analysis
Baseline characteristics are presented with number and proportion for categorical variables and as mean and SD or median and IQR for continuous variables. The incidence of outcomes of interest during follow-up according to treatment strategy exposure is presented as IR (number of events [both first and repeated events] per 100 person-years), and treatment strategies are compared as incidence rate ratios (IRRs) with 95% CIs. IRRs for recurrent-event count outcomes were estimated using a negative binomial regression model with a log link and an offset for the logarithm of person-time, to account for overdispersion in the count data; because models clustering repeated events on the individual did not converge (owing to high variability in follow-up time and event counts), events and total person-time were aggregated per patient. IR differences are not reported for these count outcomes because there is no straightforward approach to account for overdispersion when directly estimating a rate difference. Corticosteroid-free disease stability, a proportion, was estimated using a repeated-measures binomial generalized estimating equation model with an exchangeable working correlation structure and robust SEs (clustering on the individual), yielding marginal risk differences (RDs) and risk ratios (RRs). Propensity scores were estimated separately for each pairwise comparison (infliximab vs vedolizumab and adalimumab vs vedolizumab). Because weighting reweights rather than adds patients, the weighted vedolizumab pseudo-population differed slightly between comparisons (320 and 318 vs an unweighted 287); unweighted numbers are reported in the text and weighted numbers in the table and figure legends. All tests were 2-sided, with statistical significance defined as P < .05.
Due to the nonrandomized nature of this study, inverse probability of treatment weighting (IPTW) with stabilized weights was used to estimate the pairwise average treatment effect of infliximab-first vs vedolizumab-first and adalimumab-first vs vedolizumab-first treatment strategies. Briefly, we used a logistic regression model to estimate the propensity score defined as the probability of initiating infliximab vs vedolizumab, or adalimumab vs vedolizumab, conditional on 13 variables including: age at treatment initiation, sex, years of education (≤9 10-12, >12, or missing), time since diagnosis (0 to <2 years or ≥2 years), disease extent (extensive UC vs left-sided UC or proctitis), type 1 and 2 diabetes and/or hypertension in the past 5 years (yes or no), hospital frailty risk score (HFRS; no frailty with HFRS 0 or frailty with HFRS >0), hospitalization in past 2 years (yes or no), number of IBD-related office visits in past 2 years, number of unique prescription drugs used in the past 2 years, oral corticosteroid use in the past 2 years (prednisolone equivalents of 0, >0 to <1500, 1500 to <3000, or ≥3000 mg), and concomitant use of immunomodulators at time of advanced therapy initiation. To assess the covariate balance between the 2 groups, covariates with an absolute standardized mean difference of less than 0.1 were considered balanced.23 Matching variables with residual imbalance between the groups were included in the generalized linear regression model for estimating the IRR and for corticosteroid-free disease stability, RD and RR. As a sensitivity analysis, we performed 1:1 nearest-neighbor propensity score matching without replacement, using a caliper of 0.2 SDs of the logit of the propensity score. Balance was assessed using standardized mean differences before and after matching. Statistical analyses were conducted using SAS version 9.4 (SAS Institute).
Results
Between 2015 and 2021, we identified 2222, 1955, and 426 patients who initiated infliximab, adalimumab, and vedolizumab, respectively, as first-line advanced therapy for UC. After applying exclusion criteria, we included 1393, 1583, and 287 patients who initiated infliximab, adalimumab, and vedolizumab, respectively, in our cohort (eFigure 1 in Supplement 1).
Infliximab-First vs Vedolizumab-First Treatment Strategy
After IPTW, we compared 1385 infliximab initiators (mean [SD] age, 38 [18] years; 43% female; mean [SD] disease duration, 5.6 [6.7] years; 55% with extensive colitis) vs 320 vedolizumab initiators (mean [SD] age, 36 [19] years; 35% female; mean [SD] disease duration, 5.0 [6.7] years; 49% with extensive colitis). Both groups were largely comparable at baseline, although patients initiated on first-line infliximab were slightly older, more likely to be female, more likely to have extensive colitis, and less likely to be taking concomitant immunomodulators (Table 1; eFigure 2 in Supplement 1). Among patients who initiated infliximab as first-line advanced therapy, 35% did not have frailty, 44% were hospitalized in the preceding 2 years, 32% had used 3000 mg prednisolone-equivalent or more in the preceding 12 months, and 53% were taking concomitant immunomodulators. Among patients who initiated vedolizumab as first-line advanced therapy, 36% did not have frailty, 43% were hospitalized in the preceding 2 years, 39% had used 3000 mg prednisolone-equivalent or more in the preceding 12 months, and 60% were taking concomitant immunomodulators. The distribution of the propensity score before and after weighting is shown in eFigure 3 in Supplement 1.
Table 1. Baseline Characteristics of Biologic-Naive Patients With Ulcerative Colitis Starting Infliximab, Adalimumab, or Vedolizumab as First-Line Treatment Strategy.
| Characteristic | Patients initiating infliximab or vedolizumab | Patients initiating adalimumab vs vedolizumab | ||||
|---|---|---|---|---|---|---|
| Infliximab, % | Vedolizumab, % | Standardized mean difference, % | Adalimumab, % | Vedolizumab, % | Standardized mean difference, % | |
| No. unweighted | 1393 | 287 | NA | 1583 | 287 | NA |
| No. weighted | 1385 | 320 | NA | 1575 | 318 | NA |
| Sex | ||||||
| Female | 43.3 | 35.2 | 16.5 | 47.8 | 43.2 | 9.4 |
| Male | 56.7 | 64.8 | 16.5 | 52.2 | 56.8 | 9.4 |
| Age at treatment start, y | ||||||
| Mean (SD) | 38.2 (17.6) | 36.3 (18.7) | 10.8 | 41.2 (15.9) | 37.8 (18.5) | 19.3 |
| Median (IQR) | 35.1 (24.0-51.6) | 30.7 (23.6-44.0) | NA | 40.1 (28.2-52.4) | 32.1 (24.7-48.5) | NA |
| Range | 2.5-86.8 | 14.3-87.2 | NA | 3.3-86.3 | 14.3-87.2 | NA |
| Categories | ||||||
| <18 | 11.0 | 5.5 | 20.1 | 4.6 | 3.7 | 4.8 |
| 18 to <40 | 47.6 | 62.6 | 30.6 | 45.1 | 60.4 | 31.0 |
| 40 to <60 | 26.6 | 16.7 | 24.2 | 36.0 | 20.0 | 36.2 |
| ≥60 y | 14.8 | 15.2 | 0.9 | 14.3 | 15.9 | 4.7 |
| Disease duration, y | ||||||
| Mean (SD) | 5.6 (6.7) | 5.0 (6.7) | 8.9 | 7.6 (8.3) | 6.1 (7.1) | 19.1 |
| Median (IQR) | 2.7 (0.7-8.3) | 2.6 (0.7-7.7) | NA | 4.7 (1.1-12.0) | 4.4 (0.9-8.6) | NA |
| Range | 0.0-45.8 | 0.0-38.0 | NA | 0.0-47.5 | 0.0-38.0 | NA |
| Categories | ||||||
| 0 to <2 | 43.3 | 45.5 | 4.5 | 34.8 | 33.2 | 3.3 |
| ≥2 | 56.7 | 54.5 | 4.5 | 65.2 | 66.8 | 3.3 |
| Education level, y | ||||||
| <9 | 19.7 | 18.1 | 4.0 | 15.5 | 13.7 | 4.9 |
| 10-12 | 45.3 | 50.0 | 9.5 | 44.0 | 45.3 | 2.7 |
| >12 | 34.9 | 31.9 | 6.4 | 40.5 | 40.9 | 0.9 |
| Missing | 0.1 | 0.0 | 5.0 | 0.1 | 0.0 | 3.7 |
| Comorbidity in past 5 y | ||||||
| Diabetesa | 4.7 | 2.8 | 10.0 | 4.1 | 2.4 | 9.8 |
| Hypertensionb | 14.6 | 12.7 | 5.4 | 16.4 | 14.1 | 6.4 |
| Hospital frailty score | ||||||
| Mean (SD) | 1.7 (2.5) | 1.7 (3.0) | 1.0 | 1.5 (2.2) | 1.6 (2.9) | 4.9 |
| Median (IQR) | 0.8 (0.0-2.4) | 0.7 (0.0-2.3) | NA | 0.5 (0.0-2.3) | 0.6 (0.0-2.3) | NA |
| Range | 0.0-25.4 | 0.0-21.3 | NA | 0.0-21.5 | 0.0-21.3 | NA |
| Categories | ||||||
| No frailty (score of 0) | 34.6 | 35.5 | 1.9 | 37.4 | 36.0 | 2.9 |
| Frailty (score >0) | 65.4 | 64.5 | 1.9 | 62.6 | 64.0 | 2.9 |
| Inpatient care in past 2 y, d | ||||||
| Mean (SD) | 3.2 (6.1) | 4.1 (8.3) | 12.2 | 2.1 (5.5) | 3.0 (7.3) | 14.8 |
| Median (IQR) | 0 (0-4) | 0 (0-6) | NA | 0 (0-2) | 0 (0-4) | NA |
| Range | 0-58 | 0-74 | NA | 0-85 | 0-74 | NA |
| Categories | ||||||
| Yes | 43.8 | 42.5 | 2.5 | 34.4 | 32.3 | 4.6 |
| No | 56.2 | 57.5 | 2.5 | 65.6 | 67.7 | 4.6 |
| IBD-related outpatient care visits in past 2 y | ||||||
| Mean (SD) | 6.1 (5.0) | 5.9 (5.0) | 2.8 | 5.5 (4.9) | 5.7 (4.7) | 3.5 |
| Median (IQR) | 5 (3-8) | 5 (2-8) | NA | 4 (2-7) | 5 (2-8) | NA |
| Range | 0-39 | 0-30 | NA | 0-47 | 0-30 | NA |
| Categories | ||||||
| 0 d | 4.6 | 4.3 | 1.1 | 7.3 | 3.8 | 15.6 |
| 1-2 d | 16.9 | 22.2 | 13.5 | 20.8 | 23.3 | 6.0 |
| ≥3 d | 78.6 | 73.4 | 12.0 | 71.8 | 72.9 | 2.4 |
| Prescription drugs in past 2 y, No. | ||||||
| Mean (SD) | 15.7 (8.7) | 14.9 (8.8) | 8.6 | 16.8 (9.4) | 16.2 (9.3) | 6.8 |
| Median (IQR) | 14 (9-20) | 13 (9-19) | NA | 15 (10-21) | 14 (10-20) | NA |
| Range | 0-56 | 3-64 | NA | 0-80 | 3-64 | NA |
| Categories | ||||||
| 0 | 0.4 | 0.0 | 8.8 | 0.1 | 0.0 | 3.3 |
| 1-4 | 2.2 | 5.7 | 18.1 | 1.5 | 4.7 | 18.4 |
| 5-9 | 23.4 | 22.7 | 1.8 | 21.0 | 19.4 | 3.9 |
| ≥10 | 74.0 | 71.6 | 5.4 | 77.4 | 75.9 | 3.7 |
| Oral corticosteroids treatment within 12 mo, mg prednisolone equivalents | ||||||
| Mean (SD) | 2145 (1717) | 2309 (1861) | 9.1 | 1808 (1682) | 1857 (1774) | 2.8 |
| Median (IQR) | 2000 (1000-3000) | 2000 (1000-3500) | NA | 1500 (0-3000) | 1500 (0-3050) | NA |
| Range | 0-10 000 | 0-11 000 | NA | 0-10 000 | 0-11 000 | NA |
| Categories | ||||||
| 0 | 18.3 | 16.3 | 5.4 | 26.6 | 25.1 | 3.5 |
| >0 to <1500 mg | 15.9 | 16.5 | 1.5 | 16.6 | 18.8 | 5.6 |
| 1500 to <3000 mg | 33.4 | 28.4 | 10.9 | 30.5 | 27.6 | 6.4 |
| ≥3000 mg | 32.3 | 38.9 | 13.7 | 26.3 | 28.6 | 5.2 |
| Concomitant use of immunomodulators within 180 d before start of follow-up | ||||||
| Yes | 52.9 | 59.8 | 14.0 | 44.1 | 51.3 | 14.3 |
| No | 47.1 | 40.2 | 14.0 | 55.9 | 48.7 | 14.3 |
| Montreal stage at treatment start | ||||||
| E1 (ulcerative proctitis)/E2 (left-sided UC) | 31.7 | 37.8 | 12.9 | 31.6 | 38.2 | 13.9 |
| E3 (extensive UC) | 54.9 | 49.3 | 11.3 | 53.3 | 49.5 | 7.6 |
| EX (extent not defined) | 12.7 | 12.2 | 1.6 | 12.9 | 11.5 | 4.4 |
| Missing | 0.7 | 0.7 | 0.1 | 2.1 | 0.7 | 11.6 |
Abbreviations: IBD, inflammatory bowel disease; NA, not applicable; UC, ulcerative colitis.
Defined as 2 or more main diagnoses of diabetes in the National Patient Register or 2 or more dispensations of antidiabetic medication in the Prescribed Drug Register.
Defined as 2 or more dispensings of antihypertensive medication in the Prescribed Drug Register.
With a mean (SD) follow-up of 5.3 (1.8) years for infliximab initiators and 4.4 (1.6) years in vedolizumab initiators, the IR of all-cause hospitalization was 30.5 (95% CI, 29.2-31.7) per 100 person-years vs 25.2 (95% CI, 22.6-27.8) per 100 person-years, corresponding to an IRR of 1.14 (95% CI, 0.91-1.43), which was not statistically significant (Table 2). Neither IBD-related hospitalization (IRR, 1.28 [95% CI, 0.92-1.77]) nor colectomy (3.6 vs 2.5 per 100 person-years; IRR, 1.06 [95% CI, 0.61-1.83]) differed significantly between strategies. The mean (SD) time on index therapy was comparable (infliximab vs vedolizumab, 2.3 [2.2] years vs 2.4 [1.9] years), and IR of switching advanced therapies was similar (IRR, 1.09 [95% CI, 0.91-1.30]). Adalimumab and vedolizumab were the most common second-line advanced therapies in infliximab-initiators, whereas infliximab, adalimumab, and ustekinumab were the most common second-line advanced therapies in vedolizumab initiators (Figure 1). Corticosteroid dispensations did not differ significantly (70.9 vs 64.9 per 100 person-years; IRR, 1.06 [95% CI, 0.89-1.28]); however, patients who initiated infliximab spent a significantly lower proportion of quarters in corticosteroid-free disease stability over the course of follow-up (60.0% vs 66.6%; RR, 0.88 [95% CI, 0.82-0.94]) (Table 3 and Figure 2). There was no difference in the incidence of serious infections between the 2 strategies (5.0 vs 3.0 per 100 person-years; IRR, 1.16 [95% CI, 0.77-1.74]) (Table 2).
Table 2. Adverse Events During Follow-Up Using Inverse-Probability of Treatment Weights in Biologic-Naive Patients With Ulcerative Colitis Starting Infliximab, Adalimumab, or Vedolizumab as First-Line Treatment Strategy.
| Outcome | Events, No. | Incidence rate per 100 PY (95% CI) | IRR (95% CI)a | ||
|---|---|---|---|---|---|
| Infliximab | Vedolizumab | Infliximab | Vedolizumab | ||
| Infliximab vs vedolizumab | |||||
| All-cause hospital admission | 2239 | 370 | 30.5 (29.2-31.7) | 25.2 (22.6-27.8) | 1.14 (0.91-1.43) |
| All-cause hospital admission ≥2 d | 1611 | 272 | 21.9 (20.8-23.0) | 18.6 (16.4-20.8) | 1.11 (0.86-1.44) |
| IBD-related hospital admission | 853 | 140 | 11.6 (10.8-12.4) | 9.5 (7.9-11.1) | 1.28 (0.92-1.77) |
| IBD-related hospital admission ≥2 d | 707 | 124 | 9.6 (8.9-10.3) | 8.4 (7.0-9.9) | 1.18 (0.83-1.67) |
| IBD-related surgery | 244 | 34 | 3.6 (3.1-4.0) | 2.5 (1.7-3.4) | 1.06 (0.61-1.83) |
| Serious infection | 367 | 44 | 5.0 (4.5-5.5) | 3.0 (2.1-3.9) | 1.16 (0.77-1.74) |
| Medication switches | 1599 | 285 | 21.8 (20.7-22.8) | 19.4 (17.2-21.7) | 1.09 (0.91-1.30) |
| Steroid dispensations | 5209 | 951 | 70.9 (68.9-72.8) | 64.9 (60.7-69.0) | 1.06 (0.89-1.28) |
| Steroid excess use (>1 course) | 964 | 208 | 13.1 (12.3-13.9) | 14.2 (12.2-16.1) | 1.07 (0.88-1.30) |
| Adalimumab vs vedolizumab | |||||
| All-cause hospital admission | 2 023 | 340 | 28.8 (27.6-30.1) | 23.9 (21.4-26.4) | 1.08 (0.86-1.35) |
| All-cause hospital admission ≥2 d | 1 376 | 238 | 19.6 (18.6-20.6) | 16.7 (14.6-18.9) | 1.01 (0.78-1.32) |
| IBD-related hospital admission | 568 | 114 | 8.1 (7.4-8.8) | 8.0 (6.6-9.5) | 0.89 (0.63-1.25) |
| IBD-related hospital admission ≥2 d | 464 | 99 | 6.6 (6.0-7.2) | 7.0 (5.6-8.3) | 0.83 (0.58-1.19) |
| IBD-related surgery | 214 | 31 | 3.2 (2.8-3.7) | 2.4 (1.5-3.2) | 0.79 (0.46-1.38) |
| Serious infection | 284 | 44 | 4.0 (3.6-4.5) | 3.1 (2.2-4.0) | 0.97 (0.62-1.52) |
| Medication switches | 1 661 | 260 | 23.7 (22.5-24.8) | 18.3 (16.1-20.5) | 1.21 (1.02-1.43) |
| Steroid dispensations | 5 329 | 886 | 75.9 (73.9-77.9) | 62.3 (58.2-66.4) | 1.14 (0.94-1.39) |
| Steroid excess use (>1 course) | 966 | 180 | 13.8 (12.9-14.6) | 12.7 (10.8-14.6) | 1.11 (0.91-1.36) |
Abbreviations: IBD, inflammatory bowel disease; IRR, incidence rate ratio; PY, person-years.
Adjusted for sex, age, diabetes status, oral corticosteroid treatment, concomitant use of immunomodulators, and extensive ulcerative colitis at treatment start.
Figure 1. Sankey Diagram of Treatment Switch Patterns in Patients With Ulcerative Colitis Starting Infliximab, Adalimumab, or Vedolizumab as First-Line Treatment Strategies.
Numbers indicate the number of patients by drug and treatment line. AT indicates advanced therapies.
Table 3. Corticosteroid-Free Disease Stability Among Biologic-Naive Patients With Ulcerative Colitis Starting Infliximab, Adalimumab, or Vedolizumab as First-Line Treatment Strategy.
| Comparison | Sum of eligible quarters | Quarters in CS-free disease stability (95% CI), %b | RD (95% CI), percentage points | RR (95% CI)a | ||
|---|---|---|---|---|---|---|
| TNF antagonist | Vedolizumab | TNF antagonist | Vedolizumab | |||
| Infliximab vs vedolizumab | 29 131 | 5798 | 60.0 (59.3 to 60.8) | 66.6 (65.1 to 68.1) | −8.3 (−13.2 to −3.3) | 0.88 (0.82 to 0.94) |
| Adalimumab vs vedolizumab | 27 719 | 5614 | 66.6 (65.9 to 67.3) | 70.0 (68.6 to 71.4) | −5.5 (−9.6 to −1.4) | 0.92 (0.87 to 0.98) |
Abbreviations: RD, risk difference; RR, risk ratio.
Adjusted for sex, age, diabetes status, oral corticosteroid treatment, concomitant use of immunomodulators, and extensive ulcerative colitis at treatment start.
Figure 2. Line Graphs of Proportion of Patients With Corticosteroid-Free Disease Stability and Adjusted Proportion Difference in Propensity Score–Matched Biologic-Naive Patients With UC Starting Infliximab or Vedolizumab as First-Line Treatment Strategy.
Overall findings were similar in 1:1 propensity score-matched groups of infliximab and vedolizumab initiators (eTable 1 in Supplement 1), although the IR of IBD-related hospitalization was significantly higher in patients who initiated infliximab as first-line advanced therapy (eTable 2 in Supplement 1). Findings were also essentially unchanged when stabilized weights were truncated at the 1st and 99th percentiles (eTable 3 in Supplement 1).
Adalimumab-First vs Vedolizumab-First Treatment Strategy
After IPTW, we compared 1575 adalimumab initiators (mean [SD] age, 41 [16] years; 48% female; mean [SD] disease duration, 7.6 [8.3] years; 53% with extensive colitis) vs 318 vedolizumab initiators (mean [SD] age, 38 [19] years; 43% female; mean [SD] disease duration, 6.1 [7.1] years; 50% with extensive colitis). Both groups were largely comparable at baseline, although patients initiated on first-line adalimumab were older and less likely to be taking concomitant immunomodulators (Table 1). Among patients who were initiated on adalimumab as first-line advanced therapy, 37% did not have frailty, 34% were hospitalized in the preceding 2 years, 26% had used 3000 mg prednisolone-equivalent or more in the preceding 12 months and 44% were taking concomitant immunomodulators. Among matched patients who were initiated on vedolizumab as first-line advanced therapy, 36% did not have frailty, 32% were hospitalized in the preceding 2 years, 29% had used 3000 mg prednisolone-equivalent or more in the preceding 12 months, and 51% were taking concomitant immunomodulators. The distribution of the propensity score before and after weighting is shown in eFigure 4 in Supplement 1.
After a mean (SD) follow-up of 4.5 (1.9) years for adalimumab initiators and 4.4 (1.6) years for vedolizumab initiators, the IR of all-cause hospitalization was 28.8 (95% CI, 27.6-30.1) per 100 person-years vs 23.9 (95% CI, 21.4-26.4) per 100 person-years, corresponding to an IRR of 1.08 (95% CI, 0.86-1.35), which was not statistically significant (Table 2). There was no difference in the IR of IBD-related hospitalization. The IR of colectomy was comparable (3.2 vs 2.4 per 100 person-years; IRR, 0.79 [95% CI, 0.46-1.38]). The mean (SD) time on index therapy was shorter in patients who started adalimumab vs vedolizumab (1.8 [1.7] years vs 2.4 [1.9] years), with a higher IR of switching advanced therapies among adalimumab initiators (23.7 vs 18.3 per 100 person-years; IRR, 1.21 [95% CI, 1.02-1.43]). Vedolizumab was the most common second-line advanced therapy in adalimumab initiators, whereas infliximab, adalimumab, and ustekinumab were the most common second-line advanced therapies in vedolizumab initiators (Figure 1). Patients who initiated adalimumab had higher IR of corticosteroid dispensations (75.9 vs 62.3 per 100 person-years; IRR, 1.14 [95% CI, 0.94-1.39]) and spent a lower proportion of quarters in corticosteroid-free disease stability over the course of follow-up (66.6% vs 70.0%; RR, 0.92 [95% CI, 0.87-0.98]) (Table 3; eFigure 5 in Supplement 1). There was no difference in the incidence of serious infections between the two strategies (4.0 vs 3.1 per 100 person-years; IRR, 0.97 [95% CI, 0.62-1.52]) (Table 2).
Overall findings were similar in 1:1 propensity score–matched groups of adalimumab and vedolizumab initiators (eTable 4 in Supplement 1), with no significant difference in hospitalization between strategies (eTable 5 in Supplement 1). Findings were also essentially unchanged when stabilized weights were truncated at the 1st and 99th percentiles (eTable 6 in Supplement 1).
Discussion
As highly effective treatment options for the management of UC become available, focus has shifted to comparing the outcomes associated with different treatment strategies and therapeutic sequencing. In this population-based study, we compared long-term outcomes associated with infliximab or adalimumab–first vs vedolizumab-first treatment strategy incorporating clinical treatment patterns. Over approximately 5 years of follow-up, in biologic-naive patients with UC starting an advanced therapy, vedolizumab initiators spent significantly more time with stable disease not taking corticosteroids than infliximab or adalimumab initiators, whereas the burden of hospitalization did not differ significantly between strategies in the primary analysis. Findings were consistent in propensity score–matched sensitivity analyses.
Most prior studies comparing TNF antagonists with vedolizumab have emphasized short-term outcomes, such as treatment failure, clinical or endoscopic remission, or persistence on therapy.3,4,24 In a multicenter cohort, Lukin et al4 reported that patients with UC treated with vedolizumab were more likely to achieve remission and had fewer adverse events than those receiving TNF antagonists during a median follow-up of 11 months, while allowing for dose escalation. In contrast, relatively few investigations have evaluated long-term outcomes associated with a TNF antagonist–first or vedolizumab-first treatment strategy, accounting for treatment optimization and switching. In a small single-center cohort, Haynesworth and colleagues5 observed similar outcomes with an infliximab-first and vedolizumab-first treatment strategy in patients with UC with 3-year follow-up. The UK IBD BioResource prospective cohort, which included 3967 patients with UC initiating a first-line biologic, found that 5-year treatment survival without discontinuation or failure was 1.9-fold higher with vedolizumab compared with infliximab and 3.1-fold higher compared with adalimumab.25 After 5 years, discontinuation rates were approximately 30% for vedolizumab, approximately 55% for infliximab, and approximately 70% for adalimumab. While the investigators adjusted for key confounding variables including disease extent and duration, there was limited balancing for surrogates for baseline disease severity, such as IBD-related hospitalization and corticosteroid use. Switching patterns were broadly similar across studies. In our study, patients starting vedolizumab commonly transitioned to a TNF antagonist if needed, while those initiated with a TNF antagonist often switched to another TNF antagonist or vedolizumab. Similarly, in the BioResource cohort, vedolizumab users most often switched to infliximab, and infliximab users switched to vedolizumab or adalimumab. Importantly, no loss of effectiveness (treatment persistence free of treatment discontinuation or treatment failure defined as major abdominal surgery and clinician’s coding for treatment failure, without any explicit mention of corticosteroid use or IBD-related hospitalization) was observed across successive lines of therapy. Supporting these findings, an Australian population-based study of 420 patients with UC26 also demonstrated greater treatment persistence with vedolizumab compared to infliximab.
Strengths and Limitations
Our study has several strengths and limitations linked to its design and the use of data from nationwide registers. Our approach offers a novel perspective by comparing treatment strategies and sequencing. With our population-based cohort, we have high positive predictive value for IBD diagnosis and virtually complete, long-term follow-up, and very low risk of misclassifying first-line advanced therapy exposures.
There are also several limitations. First, as with all observational studies, prescribing decisions for infliximab, adalimumab, or vedolizumab as first-line therapy were subject to clinical factors that may not have been fully captured, such as clinical disease activity, desire for rapidity of response, or burden of extraintestinal manifestations. We attempted to balance covariates related to these factors using IPTW, but residual confounding likely remained. For example, infliximab is often preferred in patients with more severe disease or high inflammatory burden, where a rapid response is critical to prevent hospitalization. Residual confounding may have attenuated observed differences between treatment strategies, potentially biasing estimates toward the null. Second, the analysis relied on administrative data, which lack granular information on clinical, endoscopic, and biomarker activity. Third, variability in therapeutic positioning limited our ability to systematically assess predefined treatment sequences. Swedish guidelines recommend TNF antagonists as the first-line advanced treatment for UC, although vedolizumab can be considered if there are specific reasons to choose an anti-integrin over a TNF antagonist.27,28 Reasons for choosing vedolizumab over TNF antagonists as first-line agent in specific patients were not fully captured in our database, although we adjusted for comorbidity index and frailty. Our data do not capture patient preferences for drug and route of administration. Fourth, despite the population-based design, we were unable to perform subgroup analyses to explore heterogeneity of treatment effects, which may have revealed patient subsets more responsive to one approach over another. Furthermore, our findings may be applicable mainly to high-income countries with tax-funded health care and with patients mainly of European heritage.
Conclusions
In this cohort study of biologic-naive patients with UC, over 5 years of follow-up, first-line advanced therapy with vedolizumab was associated with a greater proportion of time in corticosteroid-free disease stability than infliximab- or adalimumab-first strategies, without a significant difference in the burden of hospitalization or surgery. These findings are associational and hypothesis generating. With the treatment landscape continuing to evolve, future empirical investigations and clinical trials are needed to evaluate alternative therapeutic strategies and sequencing over extended follow-up periods.
eTable 1. Baseline Characteristics in Biologic-Naive Patients With UC Starting Infliximab or Vedolizumab as First-Line Treatment Strategy
eTable 2. Adverse Events During Follow-Up in PS-Matched Biologic-Naive Patients With UC Starting Infliximab or Vedolizumab as First-Line Treatment Strategy
eTable 3. Adverse Events During Follow-Up Using Inverse Probability of Treatment Weights Truncated at the 1st and 99th Percentile in Biologic-Naive Patients With Ulcerative Colitis Starting Infliximab or Vedolizumab as First-Line Treatment Strategy
eTable 4. Baseline Characteristics in Biologic-Naive Patients With UC Starting Adalimumab or Vedolizumab as First-Line Treatment Strategy
eTable 5. Adverse Events During of Follow-Up in PS-Matched Biologic-Naive Patients With UC Starting Adalimumab or Vedolizumab as First-Line Treatment Strategy
eTable 6. Adverse Events During Follow-Up Using Inverse-Probability of Treatment Weights Truncated at the 1st and 99th Percentile in Biologic-Naive Patients With Ulcerative Colitis Starting Adalimumab or Vedolizumab as First-Line Treatment Strategy
eFigure 1. Flow Chart of Identified Biologic-Naive Patients With Ulcerative Colitis Starting Infliximab, Adalimumab, or Vedolizumab as First-Line Treatment Strategy
eFigure 2. Standardized Mean Difference (%) for Variables Included in the Propensity Score Models Before and After Weighting
eFigure 3. Distribution of the Propensity Score (PS) Before and After PS-Matching in Biologic-Naive Patients With Ulcerative Colitis Starting Infliximab or Vedolizumab as First-Line Treatment Strategy
eFigure 4. Distribution of the Propensity Score (PS) Before and After PS-Matching in Biologic-Naive Patients With Ulcerative Colitis Starting Adalimumab or Vedolizumab as First-Line Treatment Strategy
eFigure 5. Proportion of Patients With Corticosteroid-Free Disease Stability by Quarter Since Start of Treatment and Adjusted Proportion Difference in PS-Matched Biologic-Naive Patients With UC Starting Adalimumab or Vedolizumab as First-Line Treatment Strategy
Nonauthor Collaborators
Data Sharing Statement
References
- 1.Fudman DI, McConnell RA, Ha C, Singh S. Modern advanced therapies for inflammatory bowel diseases: practical considerations and positioning. Clin Gastroenterol Hepatol. 2025;23(3):454-468. doi: 10.1016/j.cgh.2024.06.050 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Singh S, Desai A, Goodwin S, Jairath V, Kochhar G. Trends in utilization of advanced therapies in patients with inflammatory bowel diseases in the USA, 2021-2025. J Crohns Colitis. 2025;19(10):jjaf186. doi: 10.1093/ecco-jcc/jjaf186 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Bressler B, Yarur A, Silverberg MS, et al. Vedolizumab and anti-tumour necrosis factor α real-world outcomes in biologic-naïve inflammatory bowel disease patients: results from the EVOLVE study. J Crohns Colitis. 2021;15(10):1694-1706. doi: 10.1093/ecco-jcc/jjab058 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.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. doi: 10.1016/j.cgh.2020.10.003 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Haynesworth A, Yeh KH, Lee HH, et al. Long-term outcomes of an infliximab-first versus vedolizumab-first treatment strategy in biologic-naïve patients with ulcerative colitis. Aliment Pharmacol Ther. 2025;61(4):693-701. doi: 10.1111/apt.18441 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Sands BE, Peyrin-Biroulet L, Loftus EV Jr, et al. ; VARSITY Study Group . Vedolizumab versus adalimumab for moderate-to-severe ulcerative colitis. N Engl J Med. 2019;381(13):1215-1226. doi: 10.1056/NEJMoa1905725 [DOI] [PubMed] [Google Scholar]
- 7.Lee HH, Solitano V, Singh S, et al. Differential efficacy of advanced therapies in inducing remission in ulcerative colitis based on prior exposure to TNF antagonists. Clin Gastroenterol Hepatol. 2025;23(12):2102-2114.e5. doi: 10.1016/j.cgh.2024.12.007 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Ludvigsson JF, Otterblad-Olausson P, Pettersson BU, Ekbom A. The Swedish personal identity number: possibilities and pitfalls in healthcare and medical research. Eur J Epidemiol. 2009;24(11):659-667. doi: 10.1007/s10654-009-9350-y [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Ludvigsson JF, Bergman D, Lundgren CI, et al. The healthcare system in Sweden. Eur J Epidemiol. 2025;40(5):563-579. doi: 10.1007/s10654-025-01226-9 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Ludvigsson JF, Andersson M, Bengtsson J, et al. Swedish Inflammatory Bowel Disease Register (SWIBREG)— a nationwide quality register. Scand J Gastroenterol. 2019;54(9):1089-1101. doi: 10.1080/00365521.2019.1660799 [DOI] [PubMed] [Google Scholar]
- 11.Everhov AH, Frisell T, Osooli M, et al. Diagnostic accuracy in the Swedish national patient register: a review including diagnoses in the outpatient register. Eur J Epidemiol. 2025;40(3):359-369. doi: 10.1007/s10654-025-01221-0 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Ludvigsson JF, Andersson E, Ekbom A, et al. External review and validation of the Swedish national inpatient register. BMC Public Health. 2011;11:450. doi: 10.1186/1471-2458-11-450 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Bröms G, Söderling J, Sachs MC, et al. ; SWIBREG study group . Capturing biologic treatment for IBD in the Swedish Prescribed Drug Register and the Swedish National Patient Register—a validation study. Scand J Gastroenterol. 2021;56(4):410-421. doi: 10.1080/00365521.2021.1884894 [DOI] [PubMed] [Google Scholar]
- 14.Wallerstedt SM, Wettermark B, Hoffmann M. The first decade with the Swedish Prescribed Drug Register—a systematic review of the output in the scientific literature. Basic Clin Pharmacol Toxicol. 2016;119(5):464-469. doi: 10.1111/bcpt.12613 [DOI] [PubMed] [Google Scholar]
- 15.Barlow L, Westergren K, Holmberg L, Talbäck M. The completeness of the Swedish Cancer Register: a sample survey for year 1998. Acta Oncol. 2009;48(1):27-33. doi: 10.1080/02841860802247664 [DOI] [PubMed] [Google Scholar]
- 16.Ludvigsson JF, Lashkariani M. Cohort update: ESPRESSO (Epidemiology Strengthened by Histopathology Reports in Sweden). Clin Epidemiol. 2025;17:193-196. doi: 10.2147/CLEP.S499859 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Ludvigsson JF, Almqvist C, Bonamy AK, et al. Registers of the Swedish total population and their use in medical research. Eur J Epidemiol. 2016;31(2):125-136. doi: 10.1007/s10654-016-0117-y [DOI] [PubMed] [Google Scholar]
- 18.Jakobsson GL, Sternegård E, Olén O, et al. Validating inflammatory bowel disease (IBD) in the Swedish National Patient Register and the Swedish Quality Register for IBD (SWIBREG). Scand J Gastroenterol. 2017;52(2):216-221. doi: 10.1080/00365521.2016.1246605 [DOI] [PubMed] [Google Scholar]
- 19.Mouratidou N, Malmborg P, Järås J, et al. ; SWIBREG Study Group . Identification of childhood-onset inflammatory bowel disease in Swedish healthcare registers: a validation study. Clin Epidemiol. 2022;14:591-600. doi: 10.2147/CLEP.S358031 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Nguyen LH, Örtqvist AK, Cao Y, et al. Antibiotic use and the development of inflammatory bowel disease: a national case-control study in Sweden. Lancet Gastroenterol Hepatol. 2020;5(11):986-995. doi: 10.1016/S2468-1253(20)30267-3 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Karlqvist S, Sachs MC, Eriksson C, et al. ; SWIBREG Study Group . Comparative Risk of serious infection with vedolizumab vs anti-tumor necrosis factor in inflammatory bowel disease: results from nationwide Swedish registers. Am J Gastroenterol. 2024;119(12):2480-2492. doi: 10.14309/ajg.0000000000002961 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Gilbert T, Neuburger J, Kraindler J, et al. Development and validation of a Hospital Frailty Risk Score focusing on older people in acute care settings using electronic hospital records: an observational study. Lancet. 2018;391(10132):1775-1782. doi: 10.1016/S0140-6736(18)30668-8 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Austin PC. Balance diagnostics for comparing the distribution of baseline covariates between treatment groups in propensity-score matched samples. Stat Med. 2009;28(25):3083-3107. doi: 10.1002/sim.3697 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Rundquist S, Sachs MC, Eriksson C, Olén O, Montgomery S, Halfvarson J; SWIBREG Study Group . 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(4):471-483. doi: 10.1111/apt.16193 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Kapizioni C, Desoki R, Lam D, et al. ; UK IBD BioResource Investigators . 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(6):790-800. doi: 10.1093/ecco-jcc/jjad203 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.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. doi: 10.1177/17562848221080793 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Svensk Gastroenterologisk Förening. Vuxna med inflammatorisk tarmsjukdom (IBD)—Nationellt vårdprogram, 2023. April 28, 2023. Accessed August 24, 2026. https://svenskgastroenterologi.se/kunskap/nationellt-vardprogram-for-vuxna-med-inflammatorisk-tarmsjukdom-ibd-2023/
- 28.Läkemedelsverket. Läkemedel vid inflammatorisk tarmsjukdom (IBD)—behandlings-rekommendation. Updated June 7, 2022. Accessed August 24, 2026. https://www.lakemedelsverket.se/sv/behandling-och-forskrivning/behandlingsrekommendationer/sok-behandlingsrekommendationer/lakemedel-vid-inflammatorisk-tarmsjukdom-ibd---behandlingsrekommendation
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
eTable 1. Baseline Characteristics in Biologic-Naive Patients With UC Starting Infliximab or Vedolizumab as First-Line Treatment Strategy
eTable 2. Adverse Events During Follow-Up in PS-Matched Biologic-Naive Patients With UC Starting Infliximab or Vedolizumab as First-Line Treatment Strategy
eTable 3. Adverse Events During Follow-Up Using Inverse Probability of Treatment Weights Truncated at the 1st and 99th Percentile in Biologic-Naive Patients With Ulcerative Colitis Starting Infliximab or Vedolizumab as First-Line Treatment Strategy
eTable 4. Baseline Characteristics in Biologic-Naive Patients With UC Starting Adalimumab or Vedolizumab as First-Line Treatment Strategy
eTable 5. Adverse Events During of Follow-Up in PS-Matched Biologic-Naive Patients With UC Starting Adalimumab or Vedolizumab as First-Line Treatment Strategy
eTable 6. Adverse Events During Follow-Up Using Inverse-Probability of Treatment Weights Truncated at the 1st and 99th Percentile in Biologic-Naive Patients With Ulcerative Colitis Starting Adalimumab or Vedolizumab as First-Line Treatment Strategy
eFigure 1. Flow Chart of Identified Biologic-Naive Patients With Ulcerative Colitis Starting Infliximab, Adalimumab, or Vedolizumab as First-Line Treatment Strategy
eFigure 2. Standardized Mean Difference (%) for Variables Included in the Propensity Score Models Before and After Weighting
eFigure 3. Distribution of the Propensity Score (PS) Before and After PS-Matching in Biologic-Naive Patients With Ulcerative Colitis Starting Infliximab or Vedolizumab as First-Line Treatment Strategy
eFigure 4. Distribution of the Propensity Score (PS) Before and After PS-Matching in Biologic-Naive Patients With Ulcerative Colitis Starting Adalimumab or Vedolizumab as First-Line Treatment Strategy
eFigure 5. Proportion of Patients With Corticosteroid-Free Disease Stability by Quarter Since Start of Treatment and Adjusted Proportion Difference in PS-Matched Biologic-Naive Patients With UC Starting Adalimumab or Vedolizumab as First-Line Treatment Strategy
Nonauthor Collaborators
Data Sharing Statement
