Short- and Long-Term Effectiveness and Safety of Vedolizumab in Crohn’s Disease: A Multicenter Retrospective Study from the Hokkaido Phoenix Cohort

Abstract

Introduction

Real-world data on the long-term effectiveness of vedolizumab (VDZ) in East Asian patients with Crohn’s disease (CD) remain limited. This study aimed to assess treatment persistence, clinical and endoscopic effectiveness, safety, and predictors of VDZ discontinuation in a multicenter Japanese cohort.

Methods

A retrospective study was conducted across seven hospitals between June 2019 and October 2025, including patients with CD who initiated VDZ. Outcomes included VDZ continuation, factors associated with discontinuation, longitudinal changes in clinical and endoscopic indices, and adverse events.

Results

Seventy-three patients were included. VDZ continuation was 62.3% and 37.8% at 1 and 3 years, respectively. In a multivariable Cox analysis, higher baseline C-reactive protein (CRP), higher body mass index (BMI), and female sex were independently associated with increased risk of VDZ discontinuation, with a trend toward better persistence among patients with ileal disease. In the 1-year analysis, female sex and prior biologic exposure predicted early discontinuation. CD activity index and its subscores improved during early treatment, and albumin levels increased gradually over time. Simple endoscopic score for CD improved at 1 and 2 years. Among patients with active disease, >40% achieved remission from week 6 onward. Adverse events occurred in 16.4%, with two (2.7%) patients discontinuing VDZ due to infusion reactions or psoriasiform rashes.

Conclusion

VDZ demonstrated acceptable long-term durability, clinically meaningful and endoscopic improvement, and a favorable safety profile in Japanese real-world practice. Baseline CRP, BMI, and sex were important predictors of treatment persistence.

Introduction

Crohn’s disease (CD) is a chronic, relapsing inflammatory bowel disease characterized by transmural inflammation and a wide spectrum of complications, including stricturing and penetrating phenotypes [1]. Although therapeutic advances have expanded available options, CD remains a challenging condition that requires long-term immunomodulatory strategies [2, 3]. Among contemporary biologics, vedolizumab (VDZ), a humanized monoclonal antibody targeting the α4β7 integrin, provides gut-selective immunomodulation by inhibiting lymphocyte trafficking to the intestinal mucosa, thereby reducing inflammatory activity while maintaining a favorable systemic safety profile [4].

The GEMINI 2 trial, a global phase 3 randomized controlled trial, established the efficacy of VDZ for induction and maintenance therapy in patients with moderate-to-severe CD, supporting its broad adoption in clinical practice [5, 6]. In Japan, a corresponding phase 3 trial demonstrated a numerically greater induction efficacy with VDZ compared with placebo as induction therapy; however, the difference did not achieve statistical significance, underscoring potential regional or population-specific variability in treatment response [7].

Despite evidence from randomized controlled trials, real-world data on the long-term effectiveness and persistence of VDZ in CD remain limited, particularly in East Asian populations. Although several observational studies have confirmed the clinical effectiveness and safety of VDZ across diverse clinical settings [8–10], most originate from Western cohorts. Notably, the VIOLET study from Taiwan remains the only published real-world evaluation of VDZ in East Asian patients with CD [11]. To date, no real-world outcomes have been reported from Japan.

Predictive factors for VDZ treatment persistence and therapeutic response also remain incompletely defined. A clinical decision tool derived from a post hoc analysis of the GEMINI 2 identified baseline characteristics associated with favorable outcomes, namely absence of prior bowel surgery, no exposure to anti-tumor necrosis factor (TNF), absence of fistulizing disease, higher serum albumin levels, and lower C-reactive protein (CRP) [12]. However, the applicability of these predictors to routine clinical practice remains uncertain, and evidence in Asian populations is particularly limited.

Furthermore, the clinical effectiveness of VDZ in patients with intestinal structural complications, such as strictures or fistulas, remains insufficiently defined. The therapeutic effect of VDZ in penetrating or fibrostenotic disease phenotypes may be attenuated because VDZ regulates gut-homing lymphocyte migration rather than exerting broad systemic immunosuppression. Nevertheless, few studies have systematically assessed how these anatomical features affect the durability of VDZ treatment in routine practice.

To address these knowledge gaps, we conducted a multicenter, retrospective observational study across seven institutions in Hokkaido, Japan. Our objectives were to characterize real-world treatment persistence, evaluate the clinical effectiveness and safety of VDZ in patients with CD, and identify baseline factors associated with long-term treatment continuation.

Methods

Study Design and Patients

This multicenter, retrospective observational study included patients with CD who initiated VDZ between June 2019 and May 2025 at seven institutions participating in the study. Clinical follow-up continued until treatment discontinuation, loss to follow-up, or end of the observation period (October 2025).

All the patients had a confirmed diagnosis of CD according to the criteria of the Japanese Ministry of Health, Labor, and Welfare [13]. No exclusion criteria were applied. Demographic characteristics, disease phenotype and history, laboratory data, endoscopic findings, and treatment responses were extracted from electronic medical records. A subset of patients commenced VDZ despite being in clinical remission or having only mild symptoms as defined by the CD activity index (CDAI) [14].

Outcomes

The primary outcome of this study was VDZ treatment continuation rate during the observation period, which was selected as a composite measure reflecting effectiveness, safety, and tolerability in real-world clinical practice. Secondary outcomes included factors associated with VDZ discontinuation, evaluated for the entire observation period and within the first year; stratified continuation rates according to key clinical characteristics; longitudinal changes in clinical indices; endoscopic outcomes assessed by the Simple Endoscopic Score for CD (SES-CD) [15]; induction of clinical remission in patients with baseline CDAI ≥150 with patients lost to follow-up considered non-remitters; and safety outcomes, including adverse events (AEs) and treatment discontinuation due to AEs.

Clinical remission was defined as a CDAI score of <150. Treatment continuation was defined as ongoing VDZ therapy during follow-up. During the study period, switching from intravenous to subcutaneous VDZ administration was considered continuation of treatment.

Treatment discontinuation was defined as the permanent cessation of VDZ due to insufficient effectiveness, AEs, or a switch to an alternative therapy at the discretion of the treating physician. Small bowel disease activity was assessed using ileocolonoscopy (with or without retrograde ileography), balloon-assisted enteroscopy, or capsule endoscopy, depending on routine clinical practice at each participating center.

Statistical Analysis

All statistical analyses were performed using EZR (Saitama Medical Center, Jichi Medical University, Saitama, Japan), a graphical user interface for R (The R Foundation for Statistical Computing, Vienna, Austria). Figures were generated using GraphPad Prism 10 (GraphPad Software Inc., San Diego, CA, USA). Treatment continuation was estimated using the Kaplan-Meier method, and differences between groups were evaluated using the log-rank test. Factors associated with VDZ discontinuation were assessed using Cox proportional hazards models for the entire observation period and logistic regression analyses for discontinuation within the first year. Variables included in multivariable models were selected based on clinical relevance and the results of univariable analyses. The number of variables included in each multivariable model was limited based on the number of discontinuation events. Changes in continuous, non-normally distributed variables, including CDAI and its subscores, serum albumin, and CRP, were evaluated using the Wilcoxon signed-rank test. Stool frequency (SF) and abdominal pain (AP) subscores were calculated as 7-day sums. Endoscopic activity assessed by SES-CD at 1 and 2 years was compared with baseline values using the Wilcoxon signed-rank test in patients with paired evaluations. Baseline characteristics, including CDAI, CRP, and albumin levels, were compared using the Mann-Whitney U test. A two-sided p value <0.05 was considered statistically significant.

Results

Baseline Characteristics of Patients

Seventy-three patients with CD (54 males) who initiated VDZ therapy during the study period were included in the analysis (Table 1). Median age at treatment initiation was 38 years, and median disease duration was 10 years. With respect to disease location, 51 (69.9%) patients had ileocolonic disease (L3), 17 (23.3%) had ileal disease (L1), and 5 (6.8%) had isolated colonic disease (L2). Perianal fistulas or abscesses were observed in 11 (15.1%) patients, and intestinal structural complications (strictures and/or fistulas) were identified in 23 (31.5%). Prior exposure to biologics was frequent: 54 (74.0%) patients had received at least one prior biologic, and 43 had been treated with ≥2 biologics. Among patients with prior biologic exposure at the time of VDZ initiation, infliximab (n = 18), adalimumab (n = 10), and ustekinumab (n = 23) were the most commonly used therapies. Median CDAI at baseline was 119 (range: 11.7–402). Median serum albumin and CRP levels were 3.7 g/dL (range: 2.3–4.8) and 0.27 mg/dL (range: 0.01–15.9), respectively. Median baseline SES-CD was 9.5 (range: 0–32).

Table 1.

Patient characteristics at the time of VDZ initiation

Characteristic	​
n (%)	73 (100.0)
Age, years	38 (17–81)
Male, n (%)	54 (74.0)
BMI	20.2 (13.8–29.4)
Current smoker, n (%)	10 (13.7)
Duration of disease, years	10 (0–42)
Disease location, n (%)
L1: ileal disease	17 (23.3)
L2: colonic disease	5 (6.8)
L3: ileocolonic disease	51 (69.9)
Anal fistula or abscess, n (%)	11 (15.1)
Intestinal structural complications, n (%)	23 (31.5)
Stricture, n (%)	22 (30.1)
Fistula, n (%)	1 (1.4)
Skin manifestations, n (%)	4 (5.5)
Joint symptoms, n (%)	3 (4.1)
History of bowel resection, n (%)	31 (42.5)
≥1 prior biologics, n (%)	54 (74.0)
≥2 prior biologics, n (%)	43 (58.9)
History of malignancy, n (%)	5 (6.8)
Disease activity
CDAI	119 (11.7–402)
SF/week	14 (0–70)
Abdominal pain score/week	0 (0–14)
Albumin, g/dL	3.7 (2.3–4.8)
CRP, mg/dL	0.27 (0.01–15.9)
SES-CD	9.5 (0–32)
Concomitant medications at induction, n (%)
5-ASA	46 (63.0)
Elemental diet (≥900 kcal/day)	25 (34.2)
Steroids (prednisolone or budesonide)	31 (42.5)
Immunomodulators	28 (38.3)

Values are presented as median (range) or number (%).

CDAI, Crohn’s disease activity index; CRP, C-reactive protein; SES-CD, Simple Endoscopic Score for Crohn’s Disease; 5-ASA, 5-aminosalicylic acid.

Concomitant medications at VDZ initiation included 5-aminosalicylic acid in 46 (63.0%) patients, elemental nutrition (≥900 kcal/day) in 25 (34.2%), oral steroids in 31 (42.5%), and immunomodulators in 28 (38.3%). The median follow-up duration was 15.4 months (range: 0.47–75.0). VDZ was initiated for various reasons, most commonly due to secondary loss of response to prior biologics (n = 34), followed by primary non-response (n = 8), steroid dependence (n = 9), intolerance to prior therapies (n = 7), and other reasons (n = 5).

Treatment Persistence and Predictors of VDZ Discontinuation

The cumulative probabilities of continued VDZ therapy at 1, 3, and 5 years were 62.3%, 37.8%, and 20.6%, respectively (Fig. 1). The reasons for VDZ discontinuation included secondary loss of response (n = 28), primary non-response (n = 19), surgery (n = 3), and AEs (n = 2).

Fig. 1.

Kaplan-Meier curve of VDZ treatment continuation.

In a univariable Cox proportional hazards analysis, several baseline factors were associated with treatment discontinuation (Table 2). Female sex was linked to a higher risk of treatment discontinuation (hazard ratio [HR], 2.09; 95% confidence interval [CI]: 1.17–3.72). Additionally, higher SF was associated with increased discontinuation risk (HR, 1.31; 95% CI: 1.15–1.49). Among laboratory markers, elevated CRP predicted poorer treatment persistence (HR, 1.07; 95% CI: 1.01–1.13), whereas higher serum albumin was protective (HR, 0.58; 95% CI: 0.36–0.93). Concomitant corticosteroid or budesonide use at induction was associated with a higher hazard of discontinuation (HR, 1.78; 95% CI: 1.06–3.09). Prior exposure to two or more biologics was also associated with an increased risk of treatment discontinuation (HR, 1.79; 95% CI: 1.01–3.17).

Table 2.

Baseline factors associated with VDZ discontinuation (time-to-event analysis)

Variables	Univariate analysis			Multivariate analysis
	HR	95% CI	p value	HR	95% CI	p value
Age	0.99	0.98–1.02	0.95	​	​	​
Female	2.09	1.17–3.72	0.013	2.11	1.13–3.94	0.018
BMI	1.08	0.99–1.19	0.086	1.12	1.01–1.23	0.019
Current smoking	1.06	0.48–2.37	0.88	​	​	​
Duration of disease	1.02	0.99–1.04	0.15	​	​	​
L1: ileal disease	0.54	0.27–1.09	0.090	0.49	0.23–1.04	0.061
Anal fistula or abscess	1.41	0.66–3.02	0.37	​	​	​
Intestinal structural complications (strictures or fistulas)	1.68	0.94–2.98	0.075	1.63	0.92–2.89	0.096
History of bowel resection	0.74	0.43–1.31	0.31	​	​	​
Stoma	0.77	0.34–1.71	0.51	​	​	​
≥1 prior biologics	1.45	0.77–2.72	0.24	​	​	​
≥2 prior biologics	1.79	1.01–3.17	0.047	​	​	​
Prior anti-TNF-α antibodies	1.45	0.73–2.91	0.29	​	​	​
Disease activity
CDAI	1.00	1.00–1.01	0.23	​	​	​
SF	1.31	1.15–1.49	<0.01	​	​	​
Abdominal pain score	0.99	0.97–1.01	0.44	​	​	​
CRP	1.07	1.01–1.13	0.033	1.30	1.13–1.50	<0.01
Albumin	0.58	0.36–0.93	0.025	​	​	​
SES-CD	1.03	0.99–1.07	0.11	​	​	​
Concomitant medication
5-ASA	1.31	0.69–2.47	0.41	​	​	​
Elemental diet ≥900 kcal/day	0.79	0.44–1.41	0.43	​	​	​
Corticosteroids or budesonide	1.78	1.03–3.09	0.039	​	​	​
Immunomodulators	0.91	0.51–1.62	0.74	​	​	​

HR and 95% CI were calculated using the Cox proportional hazards model.

TNF, tumor necrosis factor; CDAI, Crohn’s disease activity index; CRP, C-reactive protein; SES-CD, Simple Endoscopic Score for Crohn’s Disease; 5-ASA, 5-aminosalicylic acid; HR, hazard ratio; CI, confidence interval.

In the multivariable Cox model, higher body mass index (BMI) (HR, 1.12; 95% CI: 1.01–1.23) and elevated CRP (HR, 1.30; 95% CI: 1.13–1.50) remained independent predictors associated with VDZ discontinuation. Female sex continued to show a significant association with poorer treatment persistence (HR, 2.11; 95% CI: 1.13–3.94). An ileal disease phenotype demonstrated a trend toward longer continuation (HR, 0.49; 95% CI: 0.23–1.04), while intestinal structural complications (strictures and/or fistulas) tended toward increased discontinuation risk (HR: 1.63; 95% CI: 0.92–2.89), although neither reached statistical significance in adjusted analyses.

During the overall observation period, 52 patients discontinued VDZ. Multivariable analysis was performed including five variables.

Subgroup Analyses of VDZ Continuation

Kaplan-Meier analyses stratified according to baseline clinical characteristics are shown in Figure 2. When stratified according to prior biologic exposure (biologic-naïve, 1 prior biologic, and ≥2 prior biologics), VDZ continuation did not differ significantly among the three groups (p = 0.078, Fig. 2a). Patients with one prior biologic tended to show more favorable persistence compared with both biologic-naïve patients and those with ≥2 prior biologics. When stratified according to sex, male patients exhibited significantly better treatment persistence than females (p = 0.011, Fig. 2b). Patients without intestinal structural complications tended to have higher continuation rates than those with strictures or fistulas, although this difference did not reach statistical significance (p = 0.061, Fig. 2c). Similarly, patients with an ileal disease phenotype showed a trend toward better treatment persistence compared with those with non-ileal disease (p = 0.085, Fig. 2d). At baseline, patients with ileal disease had significantly higher albumin levels compared with non-ileal disease (p = 0.022). CRP tended to be lower in patients with ileal disease (p = 0.070). CDAI did not differ significantly between the groups (p = 0.19).

Fig. 2.

Subgroup analyses of VDZ continuation. a Stratified according to prior biologic exposure (biologic-naïve, 1 prior biologic, and ≥2 prior biologics; log-rank p = 0.078). b Male vs. female (log-rank p = 0.011). c With vs. without intestinal structural complications (SC) (log-rank p = 0.061). d Ileal disease vs. non-ileal disease (log-rank p = 0.085).

Factors Associated with VDZ Discontinuation within 1 Year

Univariable logistic regression analysis showed that female patients were significantly more likely to discontinue VDZ than males (odds ratio [OR], 3.67; 95% CI: 1.21–11.10) (Table 3). Prior exposure to ≥1 biologic was also associated with an increased risk of 1-year discontinuation (OR, 3.47; 95% CI: 1.02–11.90).

Table 3.

Baseline factors associated with 1-year VDZ discontinuation

Variables	Univariate analysis			Multivariate analysis
	OR	95% CI	p value	OR	95% CI	p value
Age	0.99	0.96–1.02	0.53	​	​	​
Female	3.67	1.21–11.10	0.022	4.92	1.41–17.2	0.013
BMI	1.13	0.96–1.33	0.14	​	​	​
Current smoking	0.96	0.25–3.76	0.95	​	​	​
Duration of disease	1.02	0.98–1.06	0.36	​	​	​
L1: ileal disease	0.36	0.10–1.24	0.10	0.25	0.06–1.03	0.056
Anal fistula or abscess	1.54	0.40–5.90	0.53	​	​	​
Intestinal structural complications (strictures or fistulas)	1.93	0.70–5.29	0.20	​	​	​
History of bowel resection	0.75	0.42–1.30	0.31	​	​	​
Stoma	0.96	0.25–3.76	0.95	​	​	​
≥1 prior biologics	3.47	1.02–11.90	0.047	3.93	1.04–14.8	0.043
≥2 prior biologics	1.90	0.72–5.04	0.20	​	​	​
Prior anti-TNF-α antibodies	2.22	0.63–7.82	0.22	​	​	​
Disease activity
CDAI	1.00	0.99–1.01	0.77	​	​	​
SF	0.97	0.94–1.01	0.11	​	​	​
Abdominal pain score	1.04	0.93–1.17	0.50	​	​	​
CRP	1.39	0.91–2.13	0.13	​	​	​
Albumin	0.58	0.26–1.29	0.18	​	​	​
SES-CD	1.02	0.95–1.10	0.54	​	​	​
Concomitant medication
5-ASA	0.66	0.24–1.87	0.44	​	​	​
Elemental diet ≥900 kcal/day	1.05	0.39–2.86	0.92	​	​	​
Corticosteroids or budesonide	1.74	0.67–4.54	0.26	​	​	​
Immunomodulators	0.92	0.34–2.45	0.86	​	​	​

OR and 95% CI were calculated using logistic regression analysis.

TNF, tumor necrosis factor; CDAI, Crohn’s disease activity index; CRP, C-reactive protein; SES-CD, Simple Endoscopic Score for Crohn’s Disease; 5-ASA, 5-aminosalicylic acid; OR, odds ratio; CI, confidence interval.

In the multivariable analysis, female sex remained an independent predictor of early discontinuation (OR, 4.92; 95% CI: 1.41–17.2), and prior exposure to biologics continued to be significantly associated with 1-year discontinuation (OR, 3.93; 95% CI: 1.04–14.8). Ileal disease showed a borderline association with improved treatment persistence (OR, 0.25; 95% CI: 0.06–1.03). Within 1 year, 29 patients discontinued VDZ, and three variables were included in the multivariable model.

Time-Course Changes in CDAI, Symptom Subscores, and Laboratory Markers

Longitudinal changes in clinical activity and inflammatory markers following VDZ initiation are summarized in Figure 3. CDAI showed sustained improvement, with significant reductions from baseline evident by week 2 and maintained through weeks 6, 14, 30, 52, 104, and 156 (Fig. 3a). SF subscores declined significantly at weeks 6, 14, 52, 104, and 156 (Fig. 3b), while AP subscores improved at weeks 6, 14, and 104 (Fig. 3c). Albumin increased gradually, with significant gains observed at weeks 6, 30, 52, and 156 (Fig. 3d). CRP did not significantly decrease during the first 104 weeks, but a significant reduction emerged at week 156 (Fig. 3e). Steroid-free remission rates were 37.1% (26/70) at baseline, 63.3% (38/60) at 3 months, and 68.3% (28/41) at 12 months.

Fig. 3.

Longitudinal changes in clinical and laboratory parameters during VDZ treatment. Crohn’s disease activity index (CDAI) (a), stool frequency (SF)/week (b), abdominal pain (AP) subscore/week (c), albumin (d), and C-reactive protein (CRP) (e) were evaluated at baseline and weeks 2, 6, 14, 30, 52, 104, and 156. Data are presented as mean ± 95% confidence interval. Asterisks indicate statistical significance compared with baseline using the Wilcoxon signed-rank test (*p < 0.05, **p < 0.01).

Endoscopic Improvement and Clinical Remission

Endoscopic evaluation was available in 60 patients at baseline, 35 patients at 1 year, and 18 patients at 2 years. Paired endoscopic assessments were available in 30 patients at baseline and 1 year and in 15 patients at baseline and 2 years.

Among patients who underwent follow-up endoscopy, SES-CD scores at 1 year were significantly reduced compared with those at baseline (Fig. 4). A similar reduction was observed at 2 years, indicating sustained endoscopic improvement during longer-term follow-up. A supplementary analysis focusing on ileal SES-CD subscores showed no significant differences.

Fig. 4.

Endoscopic outcomes after VDZ initiation. Paired data were available for 30 patients (baseline–1 year) and 15 patients (baseline–2 years). Simple Endoscopic Score for Crohn’s Disease (SES-CD) significantly decreased from baseline at 1- and 2-year evaluations (Wilcoxon signed-rank test, *p < 0.05).

Among patients with baseline CDAI ≥150, the clinical remission rate increased over time: 3.7% (2/27) at week 2, 44.4% (12/27) at week 6, and 48.1% (13/27) at weeks 14 and 30 (Fig. 5). At week 52, 44.4% (12/27) of the patients remained in remission.

Fig. 5.

Induction of clinical remission among patients with baseline CDAI ≥150. Clinical remission (CDAI <150) was achieved in 3.7%, 44.4%, 48.1%, 48.1%, and 44.4% at weeks 2, 6, 14, 30, and 52, respectively. CDAI, Crohn’s disease activity index.

Extraintestinal manifestations occurred infrequently and had heterogeneous clinical courses. Perianal disease improved in 3 of 11 patients, remained stable in 7, and worsened in 1. Skin and joint symptoms exhibited similarly variable trajectories. Given the small number of affected patients, no firm conclusions regarding extraintestinal manifestations of VDZ can be drawn.

Safety Profile

AEs observed during VDZ treatment are summarized in Table 4. Overall, VDZ was well tolerated, with AEs occurring in 12 of 73 patients (16.4%) and treatment discontinuation due to AEs in 2 (2.7%). The most common AE category was infection (n = 3), which included 2 patients with COVID-19. Gastrointestinal complications included bowel obstruction (n = 2), worsening anal fistula (n = 1), and gastrointestinal bleeding (n = 1). Infusion-related AEs were infrequent, with 1 patient with an infusion reaction and one with pruritus. Other events included a psoriasiform rash, ischemic stroke, and elevated liver enzymes. Among these, infusion reaction (n = 1) and psoriasiform rash (n = 1) led to permanent discontinuation of VDZ. No deaths or opportunistic infections were observed.

Table 4.

Adverse events

Adverse event	Events, n (%)	Events leading to discontinuation, n (%)
Any adverse event	12 (16.4)	2 (2.7)
Bowel obstruction	2 (2.7)	0
Gastrointestinal bleeding	1 (1.4)	0
Worsening of anal fistula	1 (1.4)	0
Infusion reaction	1 (1.4)	1 (1.4)
Infections	3 (4.1)	0
COVID-19	2 (2.7)	0
Psoriasiform rash	1 (1.4)	1 (1.4)
Pruritus	1 (1.4)	0
Ischemic stroke	1 (1.4)	0
Elevated liver enzymes	1 (1.4)	0

Discussion

This multicenter retrospective study provides a real-world evaluation of the short- and long-term effectiveness, durability, and safety of VDZ in Japanese patients with CD. Treatment persistence was acceptable, with continuation rates of 62.3% at 1 year, 37.8% at 3 years, and 20.6% at 5 years. Over the entire observation period, higher baseline CRP, higher BMI, and female sex were independently associated with VDZ discontinuation. During 1-year analyses, female sex and prior exposure to a biologic remained significant predictors of early discontinuation. Kaplan-Meier analyses further demonstrated significantly lower persistence in female. There were also trends toward better persistence in patients with ileal disease and in those without stricturing or fistulizing disease.

With respect to effectiveness, VDZ produced sustained reductions in CDAI and its subscores, along with gradual improvements in albumin. Endoscopic activity decreased significantly at 1- and 2-year follow-up. Among patients with active disease at baseline, clinical remission was achieved in more than 40% beginning at week 6 and maintained through week 52. Safety outcomes were favorable: AEs occurred in 16.4% of patients, and only two individuals discontinued therapy because of treatment-related events. Collectively, these findings provide important insight into VDZ performance in routine practice in East Asia, where real-world evidence remains limited.

The relatively low baseline CDAI observed in this cohort (median 119) likely reflects real-world prescribing patterns of VDZ in Japan. In clinical practice, VDZ is often introduced not only for patients with moderate-to-severe active disease but also for those with mild disease activity, intolerance to prior biologics, or a preference for a safer long-term treatment option. Therefore, our cohort included a broader spectrum of disease severity than that typically seen in randomized controlled trials, which may partly explain the lower baseline CDAI values and may limit the generalizability of our findings to the broader Japanese CD population.

Treatment discontinuation should be interpreted with caution, as it may reflect multiple factors beyond treatment response, including AEs, physician decision-making, and patient preference. Therefore, treatment persistence represents a composite outcome rather than a direct surrogate for therapeutic efficacy.

The persistence rates observed in our cohort were consistent with prior real-world data. A recent meta-analysis reported pooled 1- and 2-year continuation rates of 65.3% and 54.8%, respectively [16], which are broadly consistent with the early treatment durability observed here. In contrast, a nationwide claims analysis from Japan reported lower continuation rates [17], suggesting that VDZ durability in our multicenter clinical cohort may be somewhat more favorable than that in administrative datasets.

Biologic-naïve status was a strong predictor of 1-year VDZ persistence. This finding echoes prior observations indicating greater VDZ effectiveness when introduced before exposure to other biologics [12, 18]. In contrast, in long-term analyses stratified according to prior biologic exposure, persistence did not follow a linear pattern. This discrepancy may reflect heterogeneity in patient selection and treatment positioning in real-world clinical practice.

Female sex consistently predicted a higher likelihood of discontinuation. Although sex-related differences in VDZ response have been described in ulcerative colitis [19], comparable data in CD are limited. Therefore, this finding should be considered exploratory rather than definitive and may also be influenced by the retrospective study design and potential unmeasured confounding. Potential explanations for sex-related differences in treatment persistence may include biological factors, such as the effects of sex hormones and X-linked immune-regulatory genes [20], as well as differences in AE perception, treatment expectations, and decision-making between patients and physicians. Given the small number of women in our cohort, this finding requires validation in larger datasets.

Higher baseline CRP independently predicted VDZ discontinuation, consistent with prior reports linking greater systemic inflammation with poorer VDZ response [12]. Patients with lower inflammatory burden may have a disease better suited to the gut-selective mechanism of VDZ, whereas marked systemic inflammation may attenuate the therapeutic effect.

Higher BMI was also associated with poorer treatment durability. Pharmacokinetic analyses suggest that VDZ trough levels may be lower in individuals with higher BMI [21], raising the possibility that fixed-dose regimens may be less optimal for heavier patients, particularly in Japan, where dose intensification (300 mg every 4 weeks) is not routinely permitted. These findings highlight the need for future studies incorporating pharmacokinetic and body size-adjusted analyses.

Patients with strictures or fistulas exhibited a consistent but nonsignificant trend toward poorer VDZ persistence. This aligns with current models of cumulative bowel damage, in which fibrostenotic or penetrating phenotypes represent a later stage [22, 23], that is anatomically fixed disease less amenable to VDZ’s gut-selective mechanism. Although not statistically significant, the numeric trend supports the notion that structural bowel damage may limit therapeutic potential.

In ileal disease, there was a trend toward better VDZ persistence. MAdCAM-1, the endothelial ligand for α4β7 integrin, is expressed in the colon and small intestine [24], providing a biological rationale for VDZ efficacy in ileal CD. Emerging capsule endoscopy data further suggest that VDZ may improve small bowel inflammation [25]. Although the trend in our cohort did not reach significance, it raises the possibility that VDZ may exert meaningful therapeutic effects in ileal disease. Alternatively, the higher persistence observed in ileal disease may partly reflect differences in symptom presentation, as small bowel lesions are often less symptomatic than colonic disease, potentially leading to continued treatment despite limited objective endoscopic improvement. In our cohort, patients with ileal disease had higher baseline albumin levels and a trend toward lower CRP, suggesting that differences in baseline disease activity may also have contributed to this finding.

Clinical and endoscopic outcomes demonstrated broad therapeutic benefit. CDAI and its components improved steadily, and albumin increased, while SES-CD scores declined significantly at 1 and 2 years, indicating sustained mucosal improvement. More than 40% of patients with active disease achieved clinical remission from week 6 onward, suggesting that VDZ efficacy becomes apparent during the early induction period.

The safety profile of VDZ in this cohort was consistent with established evidence. AEs were infrequent, and treatment discontinuation due to AEs was rare. Bowel obstruction episodes likely reflected underlying stricturing disease rather than treatment toxicity. No recurrence of previous malignancy or new dysplasia was observed. Although one ischemic stroke occurred, it did not prompt discontinuation, and existing data do not support a consistent cerebrovascular signal with VDZ [26].

This study has limitations, including its retrospective design, modest sample size, and limited subgroup numbers. The inclusion of patients with mild or inactive disease may also have influenced estimates of persistence and effectiveness.

In conclusion, this multicenter Japanese cohort provides essential real-world evidence on VDZ for CD in an Asian setting. VDZ demonstrated acceptable long-term durability, meaningful clinical and endoscopic benefits, and a favorable safety profile. Baseline CRP, BMI, sex, and prior exposure to a biologic emerged as key predictors of treatment persistence. Larger prospective studies in Asian populations are warranted to validate these findings and refine patient selection for VDZ therapy.

Acknowledgments

We would like to thank Editage (www.editage.jp) for English language editing.

Statement of Ethics

The study protocol was developed by the Phoenix Study Group and approved by the Ethics Committee of Hokkaido University Hospital (IRB No. 018-0295). The present study was registered before it started (UMIN 000035384, available at http://www.umin.ac.jp/ctr/). The present study was conducted according to the guidelines stipulated in the Declaration of Helsinki and adhered to the STROBE guidelines for retrospective studies. The requirement for obtaining written informed consent was waived by the Ethics Committee of Hokkaido University Hospital owing to the retrospective nature of the study using anonymized data.

Conflict of Interest Statement

Keiya Aoyama, Takehiko Katsurada, Toshifumi Ashida, Yoshihiro Yokoyama, Fumika Orii, Katsuyoshi Ando, and Satoshi Motoya have no conflicts of interest to declare. Shinsuke Otagiri received a research grant from Mochida Pharmaceutical Co. Ltd. Takahiro Ito and Atsuo Maemoto received grants for commissioned/joint study from Janssen Pharmaceutical K.K., Takeda Pharmaceutical Co. Limited, and Pfizer Japan, Inc. Mikihiro Fujiya has received honoraria and had expenses paid to attend or give a presentation or advice at a meeting for EA Pharma Co., Ltd., AYUMI Pharmaceutical Corporation, AbbVie GK, Otsuka Pharmaceutical Factory, Inc., Zeria Pharmaceutical Co., Ltd., JIMRO Co., Ltd., Nippon Kayaku Co., Ltd., Pfizer Japan Inc., Janssen Pharmaceutical K.K., KYORIN Pharmaceutical Co., Ltd., Mochida Pharmaceutical Co., Ltd., Daiichi Sankyo Company, Limited, Mitsubishi Tanabe Pharma Corporation, Takeda Pharmaceutical Co., Ltd., Yakult Honsha Co., Ltd., Olympus Corporation, Celltrion Healthcare, Alfresa Pharma Corporation, Covidien Japan, Inc., FUJIFILM Corporation, Ltd., and JIMRO Co., Ltd. and received research grants from EA Pharma Co., Ltd., AYUMI Pharmaceutical Corporation, AbbVie GK, Otsuka Pharmaceutical Factory, Inc., Zeria Pharmaceutical Co., Ltd., Nippon Kayaku Co., Ltd., Pfizer Inc., Janssen Pharmaceutical K.K., KYORIN Pharmaceutical Co., Ltd., Mochida Pharmaceutical Co., Ltd., Daiichi Sankyo Company, Limited, Mitsubishi Tanabe Pharma Corporation, Takeda Pharmaceutical Co., Ltd., Yakult Honsha Co., Ltd., JIMRO Co., Ltd., and Kamui Pharma, Inc. He is a shareholder in Kamui Pharma, Inc. Masanao Nasuno has received lecture fees from AbbVie GK, Eisai Co., Ltd., Mochida Pharmaceutical Co., Ltd., Medtronic Japan Co., Ltd., and Takeda Pharmaceutical Co., Ltd. Hiroki Tanaka has received lecture fees from AbbVie GK, Mochida Pharmaceutical Co., Ltd., and Takeda Pharmaceutical Co., Ltd. and has received research grants from AbbVie GK, Janssen Pharmaceutical K.K., MSD K.K., and AstraZeneca. Hiroshi Nakase received personal fees from AbbVie Inc., Mitsubishi Tanabe Pharma Corporation, Janssen Pharmaceutical K.K., Takeda Pharmaceutical Co., Ltd., Pfizer Japan Inc., EA Pharma Co., Ltd., Mochida Pharmaceutical Co., Ltd., Daiichi Sankyo Co., Ltd., Gilead Sciences Inc., Viatris Inc., and JIMRO Co., Ltd. and received grants for commissioned/joint research from Hoya Group Pentax Medical, Mitsubishi Tanabe Pharma Corporation, Mochida Pharmaceutical Co., Ltd., and AbbVie Inc. He has been endowed chair by Miyarisan Pharmaceutical Co., Mochida Pharmaceutical Co., Ltd., JIMRO Co., Ltd., and KYORIN Pharmaceutical Co., Ltd. Hiroshi Nakase was a member of the journal’s Editorial Board at the time of submission.

Funding Sources

This study was not supported by any sponsor or funder.

Author Contributions

Shinsuke Otagiri: study concept and design, data acquisition, analysis and interpretation, manuscript drafting, and statistical analysis. Keiya Aoyama and Takehiko Katsurada, Takahiro Ito, Atsuo Maemoto, Fumika Orii, Toshifumi Ashida, Masanao Nasuno, Hiroki Tanaka, Katsuyoshi Ando, Mikihiro Fujiya, Yoshihiro Yokoyama, and Satoshi Motoya: data acquisition and critical revision; Hiroshi Nakase: study supervision, critical revision, and final approval. All the authors have approved the final version of the manuscript.

Funding Statement

This study was not supported by any sponsor or funder.

Data Availability Statement

The data of this study are available upon request due to privacy requirements. Further inquiries can be directed to the corresponding author.