Vedolizumab clinical decision support tool predicts clinically relevant targets in patients with Crohn’s disease

Abstract

Background:

The vedolizumab-specific clinical decision support tool (VDZ-CDST) was developed to predict the probability of a response to vedolizumab treatment in patients with Crohn’s disease (CD).

Objective:

This post hoc analysis aimed to characterize the VDZ-CDST’s ability to predict clinical outcomes and quality of life (QoL) among vedolizumab-treated patients.

Design:

Post hoc analysis of pooled clinical trial data.

Methods:

Pooled data from GEMINI 2, VERSIFY, and VISIBLE 2 clinical trials were utilized, and patients with CD were classified using the VDZ-CDST as having a low, intermediate, or high probability of response to vedolizumab. Mean patient-reported outcome 2 score, symptomatic response, symptomatic remission, and QoL (Inflammatory Bowel Disease Questionnaire) were assessed. Data up to 1 year were analyzed.

Results:

Overall, 1189 patients with CD were included. The VDZ-CDST classified 240, 548, and 401 patients as having a low, intermediate, and high probability of response to vedolizumab, respectively. Mean patient-reported outcome 2 scores decreased from baseline to week 52 in all groups, with fastest and greatest reductions in the high probability group. Symptomatic response at week 6 was achieved by 40.4%, 23.9%, and 9.6% in the high, intermediate, and low probability groups, respectively, and rates in the high probability group were greatest through week 52. Rates of symptomatic remission were highest in the high probability group, followed by intermediate and low at week 6 through week 52. The greatest improvements in QoL were observed in the high probability group.

Conclusion:

These findings provide confirmatory evidence of the utility of the VDZ-CDST in identifying patients with CD who are most likely to achieve optimal clinical and QoL outcomes with vedolizumab treatment.

Plain language summary

Vedolizumab clinical decision support tool to predict improvements in symptoms and quality of life in people living with Crohn’s disease

Crohn’s disease is a long-term condition that causes inflammation in the digestive tract. This can lead to symptoms such as abdominal pain and diarrhea. People with Crohn’s disease often report that they have a lower quality of life than those without Crohn’s disease. Vedolizumab is a medicine used to treat adults with moderate to severe Crohn’s disease. It works by reducing inflammation by targeting the immune system in the gut. The aim of this study was to find out if a decision support tool could help predict which people were likely to have improvements in symptoms and quality of life if they were treated with vedolizumab. The Vedolizumab Clinical Decision Support Tool (VDZ-CDST) uses information on baseline characteristics such as history of bowel surgery, previous treatments, disease history and levels of certain biomarkers. The VDZ-CDST classifies people as having a low, intermediate or high likelihood of vedolizumab keeping Crohn’s disease under control. This study used information on 1189 people taking part in three clinical trials to see how well the VDZ-CDST predicted long-term improvements in symptoms and quality of life. People in the high likelihood group had the quickest and best improvements in symptoms after one year of vedolizumab treatment. People in the intermediate likelihood group had better improvements than those in the low likelihood group. This group had lower improvements in comparison with the high likelihood group, but greater improvements in comparison with the low likelihood group. The greatest improvements in quality of life were in people in the high likelihood group. These results suggest that clinicians could use the VDZ-CDST to identify which people with Crohn’s disease are most likely to respond to the therapy and have long-term improvements in symptoms and quality of life when treated with vedolizumab. This is one of the first examples of grouping patients with IBD based on how they are likely to respond to future treatment.

Keywords

clinical decision support tool Crohn’s disease quality of life symptomatic remission symptomatic response vedolizumab

Introduction

Crohn’s disease (CD) is a chronic inflammatory bowel disease characterized by abdominal pain, diarrhea, fatigue, and weight loss, with an increasing prevalence worldwide.^1
–3 Additionally, patients with CD may report lower quality of life (QoL) than individuals without CD.⁴ Vedolizumab is a gut-selective antilymphocyte trafficking anti-α₄β₇ integrin monoclonal antibody approved in the United States, Europe, and other regions to treat moderately to severely active CD.^5
–8 The efficacy and safety of vedolizumab therapy in patients with CD has been well established in clinical studies and real-world settings.^9
–14

Beyond symptomatic response and remission, improvements in QoL are long-term treatment targets for patients with CD, and are recommended by the Selecting Therapeutic Targets in Inflammatory Bowel Disease Initiative (STRIDE-II) guidelines.¹⁵ To date, however, there is no single prognostic indicator that is able to predict a response to any advanced therapy in patients with CD, and no currently available treatment that has broken the efficacy ceiling, highlighting the need for such tools.^16,17

A vedolizumab-specific clinical decision support tool (VDZ-CDST) has previously been developed and validated to help predict the likelihood of a response to vedolizumab treatment in patients with CD.^18,19 The VDZ-CDST categorizes patients as having a low, intermediate, or high probability of a response to vedolizumab based on: prior bowel surgery; prior exposure to antitumor necrosis factor (TNF) therapy; prior fistulizing disease; and baseline serum concentrations of albumin and C-reactive protein (CRP).^18,19

The aim of this analysis was to better characterize the relevance of the VDZ-CDST by utilizing pooled data from three vedolizumab-controlled clinical trials in patients with CD: GEMINI 2, VISIBLE 2, and VERSIFY. This represents the largest sample used for validation to date. The previous validation studies used the VDZ-CDST to predict the likelihood of achieving clinical response and remission at week 26 and the current analysis extends these findings to predict the likelihood at week 52. This analysis also extends prior analyses by using the VDZ-CDST to predict symptomatic response and remission based on patient reported outcomes rather than Crohn’s Disease Activity Index. Additionally, it further evaluates QoL outcomes of relevance to STRIDE-II using the Inflammatory Bowel Disease Questionnaire for the first time, as well as biological remission and more detailed pharmacokinetics measures.

Methods

Patients

This analysis utilized pooled data from adult patients with CD who were treated with vedolizumab in in three clinical trials (GEMINI 2 (NCT00783692), VERSIFY (NCT02425111), and VISIBLE 2 (NCT02611817). The primary aim of the pivotal randomized, double-blind, placebo-controlled, phase III GEMINI 2 study was to evaluate the effect of vedolizumab induction treatment on clinical response and remission at 6 weeks and the effect of maintenance treatment on clinical remission at 52 weeks among patients with moderately to severely active CD.⁹ VISIBLE 2 was a phase III, randomized, double-blind, placebo-controlled study that evaluated the efficacy and safety of subcutaneous vedolizumab maintenance therapy in patients with moderately to severely active CD who had achieved a clinical response to vedolizumab intravenous induction therapy.¹¹ VERSIFY was an open-label phase IIIb study that assessed endoscopic remission and mucosal healing in patients with moderately to severely active CD treated with intravenous vedolizumab that consisted of two study parts: part A up to week 26 and part B post week 26 up to week 52. Consent for part B of the study was optional.¹⁴

Vedolizumab-specific clinical decision support tool

The methodology for the development and validation of the VDZ-CDST has been described previously.¹⁹ In brief, the VDZ-CDST assigns a score at baseline to each patient by summing the following items: no prior bowel surgery (+2 points); no prior exposure to anti-TNF therapy (+3 points); no prior fistulizing disease (+2 points); baseline albumin (+0.4 points per g/L); and baseline CRP (−0.5 points if 3.0–10.0 mg/L; −3.0 points if >10 mg/L).

Based on the derived CDST score, patients are classified into the following groups: low probability (total score of ⩽13 points); intermediate probability (total score of >13 to 19 points); or high probability (total score >19 points) of a response to vedolizumab treatment.

The VDZ-CDST originally predicted the probability of a response to vedolizumab treatment based on steroid-free clinical remission at week 26. This analysis extends the assessment of the VDZ-CDST utility to consider further endpoints up to week 52.

Outcomes

Clinical and QoL outcomes up to 12 months were analyzed using VDZ-CDST response probability groups for time points common to all three studies, expanding on previous work assessing VDZ-CDST predictions for the probability of a response at week 26.

Patient-reported outcome 2 (PRO2) scores were assessed among VDZ-CDST groups at baseline (week 0) and weeks 6, 14, 22/26, 38, 46, and 52. PRO2 is a patient-reported outcome measure commonly used in adult patients with CD, and PRO2 scores are derived as the sum of the weighted daily stool frequency and abdominal pain items contained in the Crohn’s Disease Activity Index.²⁰

Symptomatic response was defined, according to STRIDE-II, as a decrease in PRO2 scores of ⩾50% from baseline, and was assessed at weeks 6, 14, 22/26, 38, and 52.¹⁵ Symptomatic remission was defined as having a patient diary-reported average daily stool frequency of ⩽3 and an average abdominal pain score of ⩽1, and was assessed at weeks 0, 6, 14, 22/26, 38, and 52.

QoL was assessed at weeks 0, 6, 26/30, and 52 using the Inflammatory Bowel Disease Questionnaire (IBDQ),²¹ with higher scores indicating better QoL. IBDQ remission was defined as a total score of ⩾170.

Biological remission was defined by CRP levels of <5 mg/L and fecal calprotectin levels of ⩽250 µg/g determined using PhiCal™ quantitative enzyme-linked immunosorbent assay at Covance Laboratories among patients enrolled in VISIBLE 2 and VERSIFY who were not in biological remission at baseline.

To determine whether the VDZ-CDST probability groups reflected vedolizumab serum concentrations in different dosing regimens, trough and peak vedolizumab concentrations were assessed using an enzyme-linked immunosorbent assay.

Statistical analysis

All outcomes were compared between the VDZ-CDST probability groups; for some outcomes, further subgroup analyses were performed, as described hereafter.

Mean (standard deviation (SD)) PRO2 scores at each time point were derived by VDZ-CDST probability groups and also evaluated by vedolizumab dosing regimen (every 4 weeks (Q4W) and every 8 weeks (Q8W)) among patients enrolled in GEMINI 2.

The proportions of patients achieving symptomatic response and symptomatic remission were summarized by VDZ-CDST probability groups. In addition, these proportions were further evaluated by prior anti-TNF treatment experience (experienced and naive), and by CD disease localization (ileum only, colon only, ileocolonic, and unknown/other). Symptomatic response was further evaluated by Q4W and Q8W vedolizumab dosing regimen among patients enrolled in GEMINI 2.

Mean (SD) IBDQ scores and the proportions of patients achieving IBDQ remission at each time point were summarized by VDZ-CDST probability group.

Biologic remission was compared between the high probability group and the combined low/intermediate probability groups at weeks 26/30 and 52. Vedolizumab serum concentrations were assessed by probability groups among patients enrolled in VISIBLE 2 who received vedolizumab every 2 weeks (Q2W), and those enrolled in GEMINI 2 who received Q4W and Q8W dosing.

PRO2 scores and IBDQ scores were summarized based on observed data. Missing data for the binary outcomes symptomatic response, symptomatic remission, and IBDQ remission were imputed as nonresponse. For patients enrolled in the VERSIFY study, missing data after week 26 were only imputed for patients who had consented for study part B.

Exploratory left-sided Cochran-Armitage trend tests were performed to test the association between symptomatic response and symptomatic remission rates at weeks 22/26 and 52 and the VDZ-CDST probability groups.

Results

Patient characteristics

A total of 1189 patients with CD were included in this analysis (813 from GEMINI 2, 275 from VISIBLE 2, and 101 from VERSIFY). Of these, the VDZ-CDST classified 240 (20.2%) patients as having a low probability, 548 (46.1%) an intermediate probability, and 401 (33.7%) a high probability of a response to vedolizumab treatment. Patient demographic and disease characteristics are displayed in Table 1. Disease duration, mean Crohn’s Disease Activity Index score, and prior anti-TNF treatment exposure were lowest in the high probability group. Patient disposition overall and by qualifying study is shown in Table 2.

Table 1.

Patient demographic and disease characteristics.

Characteristic	Low probability group (n = 240)	Intermediate probability group (n = 548)	High probability group (n = 401)
Age, mean (SD), years	33.6 (11.2)	37.0 (12.3)	37.0 (13.6)
Sex, n (%)
Male	107 (44.6)	259 (47.3)	221 (55.1)
Female	133 (55.4)	289 (52.7)	180 (44.9)
Disease duration, mean (SD), years	10.8 (7.9)	10.1 (7.9)	7.5 (7.6)
CDAI score, mean (SD)	342.9 (70.0)	325.7 (65.2)	305.8 (60.1)
PRO2 score, mean (SD)	22.4 (7.1)	22.3 (6.8)	21.0 (6.3)
Prior anti-TNF exposure, n (%)	225 (93.8)	400 (73.0)	129 (32.2)
Prior bowel surgery, n (%)	147 (61.3)	244 (44.5)	51 (12.7)
Baseline CS and/or IMM use, n (%)
CS only	82 (34.2)	173 (31.6)	89 (22.2)
IMM only	33 (13.8)	91 (16.6)	87 (21.7)
CS + IMM	37 (15.4)	77 (14.1)	54 (13.5)
None	88 (36.7)	207 (37.8)	171 (42.6)
Disease localization, n (%)
Ileum only	19 (7.9)	103 (18.8)	67 (16.7)
Colon only	48 (20.0)	108 (19.7)	81 (20.2)
Ileocolonic	99 (41.3)	222 (40.5)	145 (36.2)
Other	74 (30.8)	79 (14.4)	42 (10.5)
Unknown	0 (0.0)	36 (6.6)	66 (16.5)

PRO2 score is the sum of the CDAI stool frequency and CDAI abdominal pain scores.

CDAI, Crohn’s Disease Activity Index; CS, corticosteroid; IMM, immunosuppressant; PRO2, patient-reported outcome 2; SD, standard deviation; TNF, tumor necrosis factor.

Table 2.

Patient disposition overall and by qualifying study.

Patients, n (%)	Low probability group	Intermediate probability group	High probability group
Pooled (total)	n = 240	n = 548	n = 401
Discontinued	165 (68.8)	302 (55.1)	139 (34.7)
Completed W52	75 (31.3)	246 (44.9)	262 (65.3)
GEMINI 2	n = 226	n = 414	n = 173
Discontinued	156 (69.0)	240 (58.0)	67 (38.7)
Completed W52	70 (31.0)	174 (42.0)	106 (61.3)
VERSIFY	n = 0	n = 36	n = 65
Discontinued^a	0 (0.0)	25 (69.4)	29 (44.6)
Completed W26	0 (0.0)	30 (83.3)	54 (83.1)
Completed W52	0 (0.0)	11 (30.6)	36 (55.4)
VISIBLE 2	n = 14	n = 98	n = 163
Discontinued	9 (64.3)	37 (37.8)	43 (26.4)
Completed W52	5 (35.7)	61 (62.2)	120 (73.6)

Includes patients in VERSIFY who did not consent for study part B.

W, week.

Change in PRO2 scores

Mean (SD) PRO2 scores decreased from baseline to week 52 in all three VDZ-CDST probability groups (Figure 1). The high probability group had the fastest onset and greatest reduction in baseline mean PRO2 scores after vedolizumab treatment compared with the intermediate and low probability groups. In turn, the intermediate probability group had a greater decrease in mean PRO2 scores compared with the low probability group.

Figure 1.

Mean (SD) PRO2 score by VDZ-CDST probability group (observed data). Error bars represent SDs; n represent baseline patient numbers per VDZ-CDST group.

Change in mean (SD) PRO2 scores in each probability group by vedolizumab dosing regimen are shown in Supplemental Figure 1. Decreases in PRO2 scores were similar for patients in the high probability group receiving vedolizumab Q4W or Q8W. However, patients in the intermediate probability group receiving Q8W dosing had a numerically greater reduction in mean PRO2 scores than those treated Q4W, whereas those receiving Q8W dosing in the low probability group had higher mean PRO2 scores at weeks 14–38 than those treated Q4W, but lower mean scores at week 52.

PRO2 symptomatic response

Symptomatic response at week 6 was achieved by 162 (40.4%), 131 (23.9%), and 23 (9.6%) patients in the high, intermediate, and low VDZ-CDST probability groups, respectively (Figure 2). The rates of symptomatic response in the high probability group continued to exceed those of the intermediate and low probability groups through to week 52 (high 47.2%, intermediate 28.3%, low 11.7%). Similarly, rates of symptomatic response in the intermediate probability group were higher than those in the low probability group at each time point. The differences in rates of symptomatic response between VDZ-CDST probability groups at weeks 22/26 and week 52 were statistically significant (p < 0.0001).

Figure 2.

Rates of PRO2 symptomatic response by vedolizumab-specific clinical decision support tool probability group (missing data imputed as nonresponse). Symptomatic response was defined as a decrease in PRO2 scores of ⩾50% from baseline. For patients enrolled in VERSIFY, missing data after week 26 were only imputed for patients who had provided consent for study part B (up to week 52).

PRO2 symptomatic response by prior anti-TNF exposure

Owing to the criteria used in the VDZ-CDST to assign patients to probability groups, anti-TNF-experienced patients were more highly represented in the low and intermediate groups than the high probability group: 625 and 129 patients, respectively. Patients who were anti-TNF naive at baseline had higher rates of symptomatic response compared with patients who had prior anti-TNF exposure at time points through week 52, with the exception of the low probability group at week 6 and the intermediate probability group at weeks 6 and 14 (Supplemental Figure 2). Interestingly, prior anti-TNF exposure had a lower impact on the high probability group, as both anti-TNF-naive and anti-TNF-experienced patients had higher rates of symptomatic response compared with those in the low or intermediate probability groups.

PRO2 symptomatic response by disease localization

Rates of symptomatic response were highest in the high probability group, followed by the intermediate and then low probability groups, regardless of disease localization (Figure 3). The proportions of patients in each VDZ-CDST group were similar by ileum only, colon only, ileocolonic, and unknown/other disease location.

Figure 3.

Proportion of patients achieving patient-reported outcome 2 symptomatic response by disease localization (missing data imputed as nonresponse). For patients enrolled in VERSIFY, missing data after week 26 were only imputed for patients who had provided consent for study part B (up to week 52).

PRO2 symptomatic response by dosing regimen

Among patients enrolled in GEMINI 2, those in the high probability group who received vedolizumab Q8W had higher rates of symptomatic response than those receiving vedolizumab Q4W (Supplemental Figure 3). However, patients who received Q4W dosing had higher rates of symptomatic response than patients who received Q8W dosing in both the low and intermediate probability groups, except at week 6 in the intermediate probability group and week 52 in the low probability group.

PRO2 symptomatic remission

The rates of symptomatic remission were greater in the high probability group at week 6 (high 34.7%, intermediate 17.2%, low 7.5%), and continued to exceed those of the low and intermediate probability groups at each time point through week 52 (high 40.3%, intermediate 22.2%, low 9.2%; Figure 4). Rates of symptomatic remission were also higher in the intermediate probability group compared with the low probability group at weeks 6 through 52. These differences in the rates of symptomatic remission between VDZ-CDST groups at weeks 22/26 and week 52 were statistically significant (p < 0.0001).

Figure 4.

Rates of patient-reported outcome 2 symptomatic remission by vedolizumab-specific clinical decision support tool probability group (missing data imputed as nonresponse). Symptomatic remission was defined as an average daily stool frequency of ⩽3 and an average abdominal pain score of ⩽1. For patients enrolled in VERSIFY, missing data after week 26 were only imputed for patients who had provided consent for study part B (up to week 52).

PRO2 symptomatic remission by prior anti-TNF exposure

Compared with patients who had prior anti-TNF treatment exposure, patients who were anti-TNF naive had higher rates of symptomatic remission in all three probability groups at weeks 6 to 52 (Figure 5). Among patients who were both anti-TNF naive and anti-TNF experienced, higher rates of symptomatic remission were observed in the high probability group versus the low or intermediate probability groups.

Figure 5.

Proportion of patients achieving patient-reported outcome 2 symptomatic remission by prior antitumor necrosis factor treatment exposure (missing data imputed as nonresponse). Symptomatic remission was defined as an average daily stool frequency of ⩽3 and an average abdominal pain score of ⩽1. For patients enrolled in VERSIFY, missing data after week 26 were only imputed for patients who had provided consent for study part B (up to week 52).

PRO2 symptomatic remission by disease localization

Regardless of disease localization, rates of symptomatic remission were highest in the high probability group followed by the intermediate then low probability groups at all-time points (Supplemental Figure 4).

Quality of life

The mean (SD) IBDQ total score increased from baseline in all three VDZ-CDST probability groups, with the fastest and greatest increase observed in the high probability group compared with the low and intermediate probability groups (Figure 6(a)).

Figure 6.

(a) Mean (SD) IBDQ total score (observed data) and (b) proportion of patients achieving IBDQ remission by vedolizumab-specific clinical decision support tool probability group (missing data imputed as nonresponse). IBDQ remission was defined as a total score of ⩾170. Dotted line in (a) shows threshold for IBDQ remission. For patients enrolled in VERSIFY, missing data after week 26 were only imputed for patients who had provided consent for study part B (up to week 52).

In the high probability group, 26 (6.5%), 140 (41.7%), 185 (46.1%), and 167 (44.3%) patients were in IBDQ remission at baseline, week 6, week 26/30, and week 52, respectively (Figure 6(b)). Twenty-seven (4.9%), 142 (27.7%), 147 (26.8%), and 128 (24.3%) patients in the intermediate probability group, and 7 (2.9%), 41 (17.1%), 45 (18.8%), and 35 (14.6%) in the low probability group, were in IBDQ remission at baseline, week 6, week 26/30, and week 52, respectively.

Biologic remission

Patients in the high probability group had significantly higher odds of biological remission at weeks 26/30 and 52 than patients in the combined low and intermediate probability group (Table 3).

Table 3.

Biologic remission rates by vedolizumab-specific clinical decision support tool probability group.

Week	Data used	Low and intermediate combined (%)	High (%)	OR—high versus low/intermediate (95% CI)
26/30	Observed data	1.7	8.8	5.46 (1.21–24.49)
	NRI	1.4	7.4	5.81 (1.30–25.97)
52	Observed data	2.7	12.3	4.98 (1.11–22.32)
	NRI	1.4	8.4	6.71 (1.52–29.68)

Biological remission was defined by C-reactive protein levels of <5 mg/L and fecal calprotectin levels of ⩽250 µg/g.

CI, confidence interval; NRI, nonresponse imputation; OR; odds ratio.

Vedolizumab pharmacokinetics

Among patients treated with vedolizumab subcutaneous Q2W enrolled in VISIBLE 2, median vedolizumab serum concentrations were highest among patients in the high probability group followed by the intermediate then low probability groups at each assessment (Supplemental Figure 5(a)). Similarly, after week 2, vedolizumab concentrations were highest in the high probability group followed by the intermediate and low probability groups in patients treated with vedolizumab intravenous Q4W enrolled in GEMINI 2 (Supplemental Figure 5(b)). However, this relationship between vedolizumab concentrations and VDZ-CDST probability groups was not consistently observed among patients treated Q8W in GEMINI 2 (Supplemental Figure 5(c)). Additionally, concentrations in the intermediate probability group at Q8W dosing were more comparable with the low probability group, whereas they were closer to the high probability group at Q4W dosing, supporting the notion that dose intensification might be beneficial for patients in the intermediate probability group.

Discussion

In this pooled analysis of data from the GEMINI 2, VERSIFY, and VISIBLE 2 studies, patients classified using the VDZ-CDST as having a high probability of a response to vedolizumab treatment had faster onset and higher rates of symptomatic response and remission, and improvements in QoL than those classified as having an intermediate or low probability of a response. Similarly, patients classified as having an intermediate probability of a response had better outcomes than those classified as having a low response probability.

The categorization of patients using the VDZ-CDST provided consistent predictions for outcomes, supporting previous studies that found the VDZ-CDST was able to accurately predict the likelihood of achieving symptomatic response and remission in patients with CD.^18,19,22 The utility of the VDZ-CDST to predict clinical outcomes in real-world settings has also been demonstrated.^23
–28 A recent real-world study by Alsoud et al. confirmed the utility of the VDZ-CDST in routine clinical practice to discriminate the probability of patients achieving a clinical and endoscopic response and remission, and those most likely to persist with vedolizumab treatment without requiring dose escalation or surgery.²⁹ Considering that improved QoL is an important patient-centric long-term goal,¹⁵ the current study expands on earlier data showing that the VDZ-CDST was able to distinguish between patients with CD with low, intermediate, and high probability of improvements in QoL and IBDQ remission.

In addition, the VDZ-CDST was also able to differentiate response probability groups in both anti-TNF-naive and anti-TNF-experienced patients; anti-TNF-naive patients had higher rates of symptomatic response and symptomatic remission over time. As expected, given the criteria for assigning patients to VDZ-CDST probability groups, most patients in the low and intermediate probability groups were anti-TNF experienced, whereas more patients in the high probability groups were anti-TNF naive, in line with previous studies reporting that prior anti-TNF exposure is associated with lower rates of response and remission.^10,30,31 Although prior anti-TNF exposure is one of the criteria for assigning patients to VDZ-CDST probability groups, approximately one-third of patients in the high probability group were anti-TNF experienced. This suggests that a personalized approach to treatment would be of benefit, regardless of prior anti-TNF exposure. Disease location did not play a role in the categorization using the VDZ-CDST, echoing the recent analysis of LOVE-CD, which showed location is not a reliable predictor of endoscopic healing with vedolizumab.³² Differences in observed symptomatic response among patients receiving vedolizumab Q4W and Q8W within each probability group raise questions about the use of the VDZ-CDST in dose optimization during the management of patients with IBD. Although real-world evidence has shown that dose optimization can result in clinical remission in approximately half of patients who have lost remission,³³ there is currently limited evidence regarding dose optimization in intermediate patients as classified by the VDZ-CDST. Among these patients, it would be logical to recommend close follow-up, with early optimization of partial responders as described in the European label for vedolizumab. Although exposure-efficacy relationships for VDZ IV and SC have been shown to be comparable,³⁴ the lack of currently available prospective data means that dose-related findings in this study should be interpreted as hypothesis generating rather than prescriptive.

The VDZ-CDST intermediate probability group could also be considered as the ‘indeterminate’ group as the VDZ-CDST was unable to categorize these patients as having a low or high probability of responding to VDZ. In this scenario, clinicians might consider active (tight) monitoring of these patients, as evidence indicates that the benefits of VDZ are established in the first few months of treatment.³⁵ This suggests that dose optimization and other management strategies should be taken early in the treatment course rather than delayed to maximize therapeutic response. Therefore, the VDZ-CDST can be a useful tool to help determine the type of clinical management needed. This is of particular importance given that VDZ is not as effective in patients with prior exposure to anti-TNF medications than in anti-TNF naive patients.³⁶ In the context of shared decision-making, deciding to start first-line advance therapy with VDZ may offer a long-term clinical benefit in patients categorized by the VDZ-CDST as intermediate.³⁷

Despite the simplicity of the VDZ-CDST, it has been shown to be specific to patients initiating vedolizumab treatment,^22,38,39 and has limited use among patients receiving alternative therapies such as ustekinumab.²² Additionally, while tools for other treatments have been developed, the VDZ-CDST has been demonstrated to have greater predictive utility than the ustekinumab CDST in the relevant patient populations.²⁹ Given that high CRP/albumin ratio has been associated with disease severity in inflammatory bowel disease, the fact that albumin plays such an important role in the VDZ-CDST might be a reason for its specificity to vedolizumab.²³ As with most monoclonal antibodies, albumin appears in the pharmacokinetic/pharmacodynamic model of vedolizumab and is an important factor in the clearance of the drug.⁴⁰ However, the mechanism of action of vedolizumab requires its presence in the vascular compartment (as endocytes are lining the vessels), which could account for the importance of albumin as an indicator of the likelihood of satisfying vedolizumab plasmatic levels.

Results presented in this analysis are of relevance for payers when considering the cost-effectiveness of therapies for CD. In addition to the ability to predict clinical outcomes and QoL, the VDZ-CDST has previously been shown to predict healthcare resource utilization, with per-patient costs 2.5-times lower in patients with a high probability of a response to vedolizumab than those with a low probability.³⁸

This analysis has some limitations that should be considered when interpreting these results. In clinical trials, patient eligibility and treatment regimens are constrained by the study protocol, and consequently the data may lack generalizability to patients encountered in real-world clinical practice. However, the VDZ-CDST has previously demonstrated utility in predicting clinical outcomes in real-world settings.^23
–28 While the pooling of clinical trial data with differing inclusion criteria introduces heterogeneity into our sample, this suggests that the utility of the VDZ-CDST may apply across different study settings and patient populations. Additionally, only data up to 1 year were analyzed; thus, the ability of the VDZ-CDST to predict a longer-term response to vedolizumab could not be assessed.

In conclusion, in this analysis, the VDZ-CDST was able to predict the probability of symptomatic response and symptomatic remission, and improvements in QoL, in patients with CD treated with vedolizumab. These findings add further confirmatory support to the clinical utility of the VDZ-CDST and its ability to identify patients with CD who are most likely to achieve the best clinical and QoL outcomes with vedolizumab treatment.

Supplemental Material

sj-docx-1-tag-10.1177_17562848261436120 – Supplemental material for Vedolizumab clinical decision support tool predicts clinically relevant targets in patients with Crohn’s disease

Supplemental material, sj-docx-1-tag-10.1177_17562848261436120 for Vedolizumab clinical decision support tool predicts clinically relevant targets in patients with Crohn’s disease by Parambir S. Dulai, Giorgos Bamias, Vipul Jairath, Anthony Buisson, Marlies Meyer, Dirk Lindner, Christian Agboton and Laurent Peyrin-Biroulet in Therapeutic Advances in Gastroenterology

Supplemental Material

sj-eps-2-tag-10.1177_17562848261436120 – Supplemental material for Vedolizumab clinical decision support tool predicts clinically relevant targets in patients with Crohn’s disease

Supplemental material, sj-eps-2-tag-10.1177_17562848261436120 for Vedolizumab clinical decision support tool predicts clinically relevant targets in patients with Crohn’s disease by Parambir S. Dulai, Giorgos Bamias, Vipul Jairath, Anthony Buisson, Marlies Meyer, Dirk Lindner, Christian Agboton and Laurent Peyrin-Biroulet in Therapeutic Advances in Gastroenterology

Supplemental Material

sj-eps-3-tag-10.1177_17562848261436120 – Supplemental material for Vedolizumab clinical decision support tool predicts clinically relevant targets in patients with Crohn’s disease

Supplemental material, sj-eps-3-tag-10.1177_17562848261436120 for Vedolizumab clinical decision support tool predicts clinically relevant targets in patients with Crohn’s disease by Parambir S. Dulai, Giorgos Bamias, Vipul Jairath, Anthony Buisson, Marlies Meyer, Dirk Lindner, Christian Agboton and Laurent Peyrin-Biroulet in Therapeutic Advances in Gastroenterology

Supplemental Material

sj-eps-4-tag-10.1177_17562848261436120 – Supplemental material for Vedolizumab clinical decision support tool predicts clinically relevant targets in patients with Crohn’s disease

Supplemental material, sj-eps-4-tag-10.1177_17562848261436120 for Vedolizumab clinical decision support tool predicts clinically relevant targets in patients with Crohn’s disease by Parambir S. Dulai, Giorgos Bamias, Vipul Jairath, Anthony Buisson, Marlies Meyer, Dirk Lindner, Christian Agboton and Laurent Peyrin-Biroulet in Therapeutic Advances in Gastroenterology

Supplemental Material

sj-eps-5-tag-10.1177_17562848261436120 – Supplemental material for Vedolizumab clinical decision support tool predicts clinically relevant targets in patients with Crohn’s disease

Supplemental material, sj-eps-5-tag-10.1177_17562848261436120 for Vedolizumab clinical decision support tool predicts clinically relevant targets in patients with Crohn’s disease by Parambir S. Dulai, Giorgos Bamias, Vipul Jairath, Anthony Buisson, Marlies Meyer, Dirk Lindner, Christian Agboton and Laurent Peyrin-Biroulet in Therapeutic Advances in Gastroenterology

Footnotes

Acknowledgements

Medical writing support was provided by Jon Waldron, PhD, of Envision Catalyst, an Envision Medical Communications agency, a part of Envision Pharma Group, and funded by Takeda.

Declarations

ORCID iDs

Parambir S. Dulai

Christian Agboton

Laurent Peyrin-Biroulet

Supplemental material

Supplemental material for this article is available online.

References

Baumgart

Sandborn

WJ.

Crohn’s disease. Lancet 2012; 380: 1590–1605.

Torres

Mehandru

Colombel

, et al. Crohn’s disease. Lancet 2017; 389: 1741–1755.

Shi

Hamidi

, et al. Worldwide incidence and prevalence of inflammatory bowel disease in the 21st century: a systematic review of population-based studies. Lancet 2017; 390: 2769–2778.

van der Have

van der Aalst

Kaptein

, et al. Determinants of health-related quality of life in Crohn’s disease: a systematic review and meta-analysis. J Crohns Colitis 2014; 8: 93–106.

Rath

Billmeier

Ferrazzi

, et al. Effects of anti-integrin treatment with vedolizumab on immune pathways and cytokines in inflammatory bowel diseases. Front Immunol 2018; 9: 1700.

Soler

Chapman

Yang

, et al. The binding specificity and selective antagonism of vedolizumab, an anti-alpha4beta7 integrin therapeutic antibody in development for inflammatory bowel diseases. J Pharmacol Exp Ther 2009; 330: 864–875.

Takeda Pharmaceuticals USA, Inc. Entyvio (vedolizumab) prescribing information, https://www.accessdata.fda.gov/drugsatfda_docs/label/2021/125476s038s039lbl.pdf (2021, accessed 8 April 2025).

Takeda Pharma A/S. Entyvio INN-vedolizumab summary of product characteristics, https://www.ema.europa.eu/en/documents/product-information/entyvio-epar-product-information_en.pdf (2021, accessed 8 April 2025).

Sandborn

Feagan

Rutgeerts

, et al. Vedolizumab as induction and maintenance therapy for Crohn’s disease. N Engl J Med 2013; 369: 711–721.

10.

Sands

Feagan

Rutgeerts

, et al. Effects of vedolizumab induction therapy for patients with Crohn’s disease in whom tumor necrosis factor antagonist treatment failed. Gastroenterology 2014; 147: 618–627.e3.

11.

Vermeire

D’Haens

Baert

, et al. Efficacy and safety of subcutaneous vedolizumab in patients with moderately to severely active Crohn’s disease: results from the VISIBLE 2 randomised trial. J Crohns Colitis 2022; 16: 27–38.

12.

Vermeire

Loftus

Jr Colombel

, et al. Long-term efficacy of vedolizumab for Crohn’s disease. J Crohns Colitis 2017; 11: 412–424.

13.

Engel

Ungar

Yung

, et al. Vedolizumab in IBD-lessons from real-world experience; a systematic review and pooled analysis. J Crohns Colitis 2018; 12: 245–257.

14.

Danese

Sandborn

Colombel

, et al. Endoscopic, radiologic, and histologic healing with vedolizumab in patients with active Crohn’s disease. Gastroenterology 2019; 157: 1007–1018.e7.

15.

Turner

Ricciuto

Lewis

, et al. STRIDE-II: an update on the Selecting Therapeutic Targets in Inflammatory Bowel Disease (STRIDE) initiative of the International Organization for the Study of IBD (IOIBD): determining therapeutic goals for treat-to-target strategies in IBD. Gastroenterology 2021; 160: 1570–1583.

16.

Gisbert

Chaparro

Predictors of primary response to biologic treatment [anti-TNF, vedolizumab, and ustekinumab] in patients with inflammatory bowel disease: from basic science to clinical practice. J Crohns Colitis 2020; 14: 694–709.

17.

Raine

Danese

Breaking through the therapeutic ceiling: what will it take?

Gastroenterology 2022; 162: 1507–1511.

18.

Dulai

Amiot

Peyrin-Biroulet

, et al. A clinical decision support tool may help to optimise vedolizumab therapy in Crohn’s disease. Aliment Pharmacol Ther 2020; 51: 553–564.

19.

Dulai

Boland

Singh

, et al. Development and validation of a scoring system to predict outcomes of vedolizumab treatment in patients with Crohn’s disease. Gastroenterology 2018; 155: 687–695.e10.

20.

Khanna

Zou

D’Haens

, et al. A retrospective analysis: the development of patient reported outcome measures for the assessment of Crohn’s disease activity. Aliment Pharmacol Ther 2015; 41: 77–86.

21.

Guyatt

Mitchell

Irvine

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

22.

Alric

Amiot

Kirchgesner

, et al. Vedolizumab clinical decision support tool predicts efficacy of vedolizumab but not ustekinumab in refractory Crohn’s disease. Inflamm Bowel Dis 2022; 28: 218–225.

23.

Buisson

Nancey

Dodel

, et al. P578 CDST (Clinical Decision Support Tool) is predictive of steroid-free clinical remission in patients with Crohn’s disease treated with vedolizumab in a specific manner which is mainly driven by the serum level of albumin. J Crohns Colitis 2024; 18(Suppl. 1): i1129.

24.

Choon

Yeo

White

, et al. P0836 Real-world application of the Crohn’s disease clinical decision support tool (CDST) in both intravenous and subcutaneous vedolizumab. J Crohns Colitis 2025; 19: i1595–i1596.

25.

Kim

Park

Yoon

, et al. Application of clinical decision support tools for predicting outcomes with vedolizumab therapy in patients with inflammatory bowel disease: a KASID multicentre study. Aliment Pharmacol Ther 2024; 59: 1539–1550.

26.

Konecny

Tichanek

Prokopova

, et al. P1152 Single center study of using clinical decision support tool to predict outcomes of vedolizumab therapy in Crohn’s disease patients. J Crohns Colitis 2025; 19: i2105.

27.

Maček

Molinari

Ocepek

P773 Clinical decision support tool for vedolizumab is useful in predicting endoscopic remission in Crohn’s disease–a retrospective real-life single-centre cohort study. J Crohns Colitis 2024; 18: i1443-i1443.

28.

Ocepek

Molinari

Maček

Clinical decision support tool for vedolizumab can predict treatment persistence in Crohn’s disease but not in ulcerative colitis. Posters/Posterji 2024; 29: 51.

29.

Alsoud

Sabino

Ferrante

, et al. Calibration, clinical utility, and specificity of clinical decision support tools in inflammatory bowel disease. Clin Gastroenterol Hepatol 2024; 23: 1216–1227.e14.

30.

Feagan

Lasch

Lissoos

, et al. Rapid response to vedolizumab therapy in biologic-naive patients with inflammatory bowel diseases. Clin Gastroenterol Hepatol 2019; 17: 130–138.e7.

31.

Stallmach

Langbein

Atreya

, et al. Vedolizumab provides clinical benefit over 1 year in patients with active inflammatory bowel disease – a prospective multicenter observational study. Aliment Pharmacol Ther 2016; 44: 1199–1212.

32.

Hens

Oldenburg

Molnár

, et al. P845 Vedolizumab induces endoscopic healing uniformly irrespective of disease location: post-hoc analysis of the LOVE-CD trial. J Crohns Colitis 2024; 18(Suppl. 1): i1570–i1571.

33.

Peyrin-Biroulet

Danese

Argollo

, et al. Loss of response to vedolizumab and ability of dose intensification to restore response in patients with Crohn’s disease or ulcerative colitis: a systematic review and meta-analysis. Clin Gastroenterol Hepatol 2019; 17: 838–846.e2.

34.

D’Haens

Rosario

Polhamus

, et al. Exposure-efficacy relationship of vedolizumab subcutaneous and intravenous formulations in Crohn’s disease and ulcerative colitis. Expert Rev Clin Pharmacol 2024; 17: 403–412.

35.

D’Haens

Löwenberg

Baert

, et al. Vedolizumab in early and late Crohn's disease (LOVE-CD): a phase 4 open-label cohort study. Lancet Gastroenterol Hepatol 2026; 11: 12–21.

36.

Feagan

Rubin

Danese

, et al. Efficacy of vedolizumab induction and maintenance therapy in patients with ulcerative colitis, regardless of prior exposure to tumor necrosis factor antagonists. Clin Gastroenterol Hepatol 2017; 15: 229–239.e5.

37.

Louis

Litkiewicz

Agboton

, et al. Therapeutic sequencing in inflammatory bowel disease: determining the optimal position of vedolizumab for long-term Crohn’s disease control using real-world evidence. United European Gastroenterol J 2024; 12: 574–584.

38.

Dulai

Wan

Huang

, et al. Probability of response as defined by a clinical decision support tool is associated with lower healthcare resource utilization in vedolizumab-treated patients with Crohn’s disease. Crohns Colitis 360 2022; 4: otac048.

39.

Kim

Won

Kim

, et al. Specificity of clinical decision support tools for vedolizumab and ustekinumab in biologically naïve patients with Crohn’s disease. Scand J Gastroenterol 2022; 57: 446–448.

40.

Rosario

Dirks

Milch

, et al. A review of the clinical pharmacokinetics, pharmacodynamics, and immunogenicity of vedolizumab. Clin Pharmacokinet 2017; 56: 1287–1301.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

1.17 MB

1.14 MB

0.00 MB

1.24 MB

0.95 MB

1.04 MB