Real-world Effectiveness of Vedolizumab Compared to Anti-TNF Agents in Biologic-naïve Patients With Ulcerative Colitis

A Two-year Propensity-score-adjusted Analysis From the Prospective, Observational VEDOibd-Study

Bernd Bokemeyer; Sandra Plachta-Danielzik; Romina di Giuseppe; Philipp Efken; Wolfgang Mohl; Thomas Krause; Martin Hoffstadt; Robert Ehehalt; Leo Trentmann; Axel Schweitzer; Petra Jessen; Petra Hartmann; Stefan Schreiber

Disclosures

Aliment Pharmacol Ther. 2023;58(4):429-442.

Abstract and Introduction

Abstract

Background: This observational real-world evidence (RWE) study is based on prospectively collected data from the VEDO_IBD registry study.

Aim: To compare the effectiveness of vedolizumab and anti-TNF agents in biologic-naïve patients with ulcerative colitis (UC) at the end of induction and during maintenance treatment.

Methods: Between 2017 and 2020, we enrolled 512 patients with UC starting therapy with vedolizumab or an anti-TNF agent in 45 IBD centres across Germany. We excluded biologic-experienced patients and those with missing partial Mayo (pMayo) outcomes; this resulted in a final sample of 314 (182 on vedolizumab and 132 on an anti-TNF agent). The primary outcome was clinical remission measured using pMayo score; any switch to a different biologic agent was considered an outcome failure (modified ITT analysis). We used propensity score adjustment with inverse probability of treatment weighting to correct for confounding.

Results: During induction therapy, clinical remission was relatively low and similar in vedolizumab- and anti-TNF-treated patients (23% vs. 30.4%, p = 0.204). However, clinical remission rates after two years were significantly higher for vedolizumab-treated patients than those treated with an anti-TNF agent (43.2% vs. 25.8%, p < 0.011). Among patients treated with vedolzumab, 29% switched to other biologics, versus 54% who had received an anti-TNF agent.

Conclusion: After two years of treatment, vedolizumab resulted in higher remission rates than anti-TNF agents.

Introduction

Ulcerative colitis (UC) and Crohn's disease (CD) are chronic inflammatory bowel diseases (IBDs) characterised by alternating phases of spontaneous flares, drug-induced remission and various degrees of chronic activity.^[1] Persistent disease activity promotes the development of post-inflammatory intestinal complications, which may, in turn, cause gut complications such as neoplasia, extraintestinal manifestations and psychosocial impairments.^[2,3] Altogether, these can lead to an increasing number of unmet medical needs, particularly after failure of therapy with conventional drugs or biologics, which occurs in a substantial number of patients.^[4,5] This calls for optimising the use of existing biologics in real-world settings.^[6]

The anti-α4β7-integrin monoclonal antibody (mAb) vedolizumab (VEDO) was approved in Europe in 2014 for use in moderately to severely active UC and CD in adults failing conventional therapy. Since its approval, VEDO has shown a more favourable safety profile than anti-TNF^[7–11] and demonstrated efficacy superior to that of adalimumab (ADA) in UC through the VARSITY trial (RCT).^[10] Higher effectiveness in comparison with anti-TNFs has been reported in RWE studies.^[11,12] Among the anti-TNFs, infliximab (IFX) is a chimeric human-murine and ADA and golimumab (GOL) are fully humanised immunoglobulin G1 (IgG1) monoclonal antibodies. The therapeutic effect of IFX, ADA and GOL is mediated via tumour necrosis factor-alpha (TNFα) blockade. IFX neutralises proinflammatory cytokine activity and is indicated for and used to treat CD, UC, rheumatoid arthritis, ankylosing spondylitis, psoriatic arthritis and plaque psoriasis.^[13]

Due to their narrow inclusion and exclusion criteria, RCTs such as the pivotal approval studies for these drugs yield results that cannot always be generalised to real life and therefore do not necessarily reflect real-world clinical practice. Moreover, special patient groups, such as the elderly, patients with complications or extraintestinal manifestations, and patients with surgical resections, are often excluded from RCTs. Thus, in addition to RCTs,^[14,15] more real-world comparative studies^[16,17] focusing, for example, on VEDO versus anti-TNF therapies are needed.^[18]

The results of the VARSITY trial—the first head-to-head RCT comparing the efficacy of VEDO and ADA in patients with UC—were published in 2019.^[10] These showed that the clinical remission rate for VEDO was superior to that for ADA after one year (31.3% vs. 22.5%, p = 0.006). However, head-to-head RCTs are time-consuming, costly and logistically challenging, which is why real-world head-to-head comparisons between different biologics are needed. For these studies to deliver meaningful results, however, it is crucial for them to have appropriate structural frameworks in place.^[16] They should be multi-centric, and data should be collected prospectively and consecutively. Remote monitoring and at least some on-site spot checks are essential to ensure high-data quality. Moreover, because the patients in non-interventional studies are not randomly assigned to the different treatment groups, advanced causal analysis techniques such as propensity score (PS) adjustment should be applied to reduce the impact of confounding factors.

The aim of this comparative RWE study was to draw upon prospectively collected data from the VEDO_IBD registry study to investigate the effectiveness of VEDO versus anti-TNF in biologic-naïve patients with UC at the end of induction therapy and during maintenance treatment at one and two years.

1 2 3 4 5

Next Section

Table 1. Baseline characteristics of biologic-naïve patients with ulcerative colitis treated with vedolizumab vs. anti-TNF: effectiveness at the end of the induction phase (week 14) before and after inverse probability of treatment weighting.
Characteristics	anti-TNF	Original cohort (pre-IPTW)	p-value	anti-TNF	Pseudo cohort (post-IPTW)	p-value	Post-IPTW
Characteristics	anti-TNF	Vedolizumab	p-value	anti-TNF	Vedolizumab	p-value	Standardised difference^a
n	132	182		132	182
Age, years	35.3 (26.9–48.0)	41.4 (29.9–55.1)	0.002	41.5 (28.7–50.1)	39.6 (28.3–53.2)	0.986	0.002
Male, %	50.0	51.7	0.773	49.0	50.4	0.803	−0.029
Smokers, %	7.6	6.0	0.586	6.3	7.4	0.925	0.045
Caucasian, %	93.9	96.7	0.242	95.6	95.5	0.961	0.005
Disease duration, years	3.1 (0.9–10.8)	7.0 (1.3–14.5)	0.005	4.4 (1.1–14.7)	4.8 (1.0–11.8)	0.738	−0.041
Age at diagnosis, years	28.1 (21.8–38.5)	29.1 (22.6–42.9)	0.186	28.9 (22.7–40.1)	28.9 (21.7–42.2)	0.820	0.026
Partial Mayo score	5.5 (4.0–7.0)	4.0 (3.0–6.0)	<0.0001	5.0 (3.0–6.0)	5.0 (3.0–6.0)	0.907	0.014
C-reactive protein, mg/dl	0.65 (0.10–2.27)	0.49 (0.13–1.66)	0.349	0.50 (0.10–2.26)	0.59 (0.13–1.79)	0.713	0.041
Body mass index, kg/m²	24.3 (21.5–28.3)	23.9 (21.5–27.5)	0.392	24.4 (21.6–28.3)	24.0 (21.3–27.5)	0.695	−0.044
Montreal classification
Extent, %			0.373			0.808
E1: Ulcerative proctitis	9.1	7.1		8.0	7.6		−0.016
E2: Left-sided UC (distal UC)	32.6	40.1		39.8	36.6		0.068
E3: Extensive UC (pancolitis)	58.3	52.8		52.2	55.9		−0.075
Severity, %			0.340			0.596
S0: Clinical remission	1.5	2.2		2.6	1.9		0.046
S1: Mild UC	21.2	28.6		26.3	25.0		0.038
S2: Moderate UC	61.4	58.2		60.6	61.1		−0.028
S3: Severe UC	15.9	11.0		10.3	11.9		0.040
Perianal fistula/anal abscess, %	1.4	2.1	0.663	1.3	1.7	1.000	−0.034
Proctocolectomy, %	0.7	0.5	0.819	0.6	0.6	1.000	−0.001
Extra intestinal manifestation, %	18.4	11.2	0.070	13.7	13.3	0.909	0.013
Family history of ulcerative colitis, %	9.2	12.8	0.365	12.7	12.5	0.951	0.007
Comorbidities, %	39.0	40.6	0.689	42.7	39.5	0.575	0.065
Previous medication use, %
Corticosteroids	93.6	94.1	0.995	94.3	94.0	0.893	0.015
Immunosuppressants	45.4	43.9	0.792	44.7	41.3	0.546	0.069

Note: Continuous skewed variables are represented as medians and interquartile ranges; Categorical variables are represented as N (%); Chi-square or Fisher's exact test for categorical variables; Wilcoxon-rank sum test for skewed continuous variables; Maximum likelihood imputation after exclusion of missing outcomes (final sample n = 314); IPTW, inverse probability of treatment weighting.
Abbreviations: TNF, tumour necrosis factor; UC, ulcerative colitis.
^aA standardised difference <0.1 indicates negligible difference in the mean or prevalence of a covariate between treatment groups; Logit propensity score standardised difference = 0.053.

Table 2. Effectiveness of (A) adalimumab and (B) infliximab vs. vedolizumab in biologic-naïve patients with ulcerative colitis at the end of the induction phase (week 14) after inverse probability of treatment weighting.
A
Outcome	Adalimumab	Vedolizumab	Adalimumab	Vedolizumab	p-value
Outcome	Weighted %	Weighted %	OR (95% CI)^a	OR (95% CI)^a	p-value
n	34	182	34	182
Clinical response, %	57.2	51.8	Reference	0.81 (0.36–1.82)	0.604
Clinical remission, %	29.4	23.0	Reference	0.72 (0.27–1.91)	0.507
Clinical steroid-free remission, %	27.2	17.6	Reference	0.57 (0.20–1.60)	0.286
B
Outcome	Infliximab	Vedolizumab	Infliximab	Vedolizumab	p-value
Outcome	Weighted %	Weighted %	OR (95% CI)^a	OR (95% CI)^a	p-value
n	77	182	77	182
Clinical response, %	52.7	51.8	Reference	0.97 (0.53–1.77)	0.909
Clinical remission, %	27.8	23.0	Reference	0.78 (0.40–1.52)	0.462
Clinical steroid-free remission, %	23.1	17.6	Reference	0.71 (0.34–1.49)	0.369

Abbreviations: CI, confidence interval; OR, odds ratio.
^aResults are expressed as weighted OR and 95% CI calculated using a robust sandwich covariance estimate.

Table 3. Longitudinal analysis of EQ–VAS change from baseline (week 0) to year 1 (week 52) in vedolizumab and anti-TNF-treated biologic-naïve patients with ulcerative colitis.
Treatment group	Estimate (SE)^b	p-value	Estimate (SE)^c	p-value
Treatment group	EQ-VAS	p-value	EQ–VAS	p-value
Overall (n = 189)^a
Vedolizumab week-52—week-0 (n = 64)	14.3 (1.6)	<0.0001	14.4 (1.6)	<0.0001
anti-TNF week-52—week-0 (n = 125)	14.8 (2.3)	<0.0001	15.5 (2.2)	<0.0001
Vedolizumab vs. anti-TNF week-52—week-0	−0.6 (2.8)	0.837	−1.0 (2.7)	0.699

Note: Per-protocol analysis: exclusion of switchers.
Abbreviations: SE, standard error; TNF, tumour necrosis factor.
^a n = 12 missing EQ-VAS values.
^bAdjusted for baseline EQ-VAS values.
^cAdjusted for baseline EQ-VAS values and weighted for inverse probability of treatment; Linear mixed model for difference scores.

Table 4. Effectiveness of (A) adalimumab and (B) infliximab vs. vedolizumab in biologic-naïve patients with ulcerative colitis during maintenance therapy two years after baseline assessment and after adjustment with inverse probability of treatment weighting.
A
Outcome	Adalimumab	Vedolizumab	Adalimumab	Vedolizumab	p-value
Outcome	Weighted %	Weighted %	OR (95% CI)^a	OR (95% CI)^a	p-value
n	26	124	26	124
Clinical remission, %	37.9	45.2	Reference	1.89 (0.68–5.26)	0.225
Clinical steroid-free remission, %	37.9	44.5	Reference	1.86 (0.66–5.23)	0.242
B
Outcome	Infliximab	Vedolizumab	Infliximab	Vedolizumab	p-value
Outcome	Weighted %	Weighted %	OR (95% CI)^a	OR (95% CI)^a	p-value
n	68	124	68	124
Clinical remission, %	16.5	43.1	Reference	3.82 (1.84–7.93)	0.0003
Clinical steroid-free remission, %	15.5	42.4	Reference	4.01 (1.90–8.44)	0.0003

Note: Categorical variables are reported as weighted percentages.
Abbreviations: CI, confidence interval; OR, odds ratio.
^aResults are expressed as weighted OR and 95% CI calculated using a robust sandwich covariance estimate.

Table 5. Safety profile (adverse events/serious adverse events) for vedolizumab vs. anti-TNF in biologic-naïve patients with ulcerative colitis during maintenance therapy two years after baseline assessment.
	Vedolizumab n = 204	Anti-TNF n = 115	p-value
Adverse events	46 (22.5%)	24 (20.9%)	0.779
Serious adverse events	27 (13.2%)	16 (13.9%)	0.866
Death	0 (0%)	0 (0%)	>0.999
Life threatening	3 (11.1%)	1 (6.3%)	>0.999
Hospitalisation	23 (85.2%)	12 (75.0%)	0.443

Authors and Disclosures

Bernd Bokemeyer^1–3, Sandra Plachta-Danielzik², Romina di Giuseppe², Philipp Efken⁴, Wolfgang Mohl⁵, Thomas Krause⁶, Martin Hoffstadt⁷, Robert Ehehalt⁸, Leo Trentmann⁹, Axel Schweitzer¹⁰, Petra Jessen¹¹, Petra Hartmann⁴ and Stefan Schreiber^2,3,*

¹Interdisciplinary Crohn Colitis Centre Minden, Minden, Germany
²Competence Network IBD, Kiel, Germany
³Clinic of General Internal Medicine I, University Hospital Schleswig-Holstein, Campus Kiel, Kiel, Germany
⁴Gastroenterology Practice, Minden, Germany
⁵Center for Gastroenterology Saar MVZ, Saarbruecken, Germany
⁶Gastroenterology Practice, Kassel, Germany
⁷Gastroenterology Practice, Iserlohn, Germany
⁸Gastroenterology Practice, Heidelberg, Germany
⁹Gastroenterology Practice, Bremen, Germany
¹⁰Gastroenterology Practice, Münster, Germany
¹¹Group Practice at Medicum, Altenholz, Germany

Correspondence
Bernd Bokemeyer, Interdisciplinary Crohn Colitis Centre Minden, Märchenweg 17, 32439 Minden and Department of Medicine I, UKSH, Kiel University, Kiel, Germany. Email: bernd.bokemeyer@t-online.De

Bernd Bokemeyer and Sandra Plachta-Danielzik contributed equally.

*For the German Inflammatory Bowel Diseases Study Group [GISG].

Author Contributions
Bernd Bokemeyer: Conceptualization (equal); data curation (equal); funding acquisition (equal); investigation (equal); methodology (equal); project administration (equal); supervision (equal); validation (equal); writing – original draft (equal); writing – review and editing (equal). Sandra Plachta-Danielzik: Conceptualization (equal); data curation (equal); methodology (equal); project administration (equal); supervision (equal); validation (equal); writing – original draft (equal); writing – review and editing (equal). Romina di Giuseppe: Data curation (supporting); formal analysis (equal); methodology (equal); writing – original draft (supporting); writing – review and editing (supporting). Philipp Efken: Investigation (supporting); project administration (supporting); writing – review and editing (supporting). Wolfgang Mohl: Investigation (equal); project administration (supporting); writing – review and editing (supporting). Thomas Krause: Investigation (supporting); project administration (supporting); writing – review and editing (supporting). Martin Hoffstadt: Investigation (supporting); project administration (supporting); writing – review and editing (supporting). Robert Ehehalt: Investigation (supporting); project administration (supporting); writing – review and editing (supporting). Leo Trentmann: Investigation (supporting); project administration (supporting); writing – review and editing (supporting). Axel Schweitzer: Investigation (supporting); project administration (supporting); writing – review and editing (supporting). Petra Jessen: Investigation (supporting); project administration (supporting); writing – review and editing (supporting). Petra Hartmann: Conceptualization (supporting); investigation (supporting); project administration (supporting); writing – review and editing (supporting). Stefan Schreiber: Conceptualization (equal); data curation (equal); funding acquisition (equal); investigation (equal); project administration (equal); supervision (equal); writing – original draft (supporting); writing – review and editing (equal). All authors approved the final version of the manuscript.

Conflict of Interest Statement
BB has received consulting fees from: AbbVie, MSD, Shire, Ferring, Hospira, Takeda, Movetis, Shield Therapeutics, Pfizer, Biogen, Janssen, Hexal and Celgene, outside the submitted work. Financial support for lectures and teaching from: AbbVie, Ferring, MSD, Merckle, Falk, HLR, Shield Therapeutics, Pfizer, Celltrion, Takeda, Janssen and Mundipharma, outside the submitted work. Grant and financial support for research from AbbVie and Ferring, outside the submitted work. SPD have no relevant financial or non-financial interests to disclose. RdG have no relevant financial or non-financial interests to disclose. PE has received personal fees from: AbbVie, BMS, Galapagos, Janssen and Amgen, outside the submitted work. WM has received consulting fees from: AbbVie, Advanz, Amgen, Biogen, Bristol-Myers Squibb, Ferring, Fresenius Kabi, Galapagos, Hexal, Janssen and Takeda, outside the submitted work. Fees for lectures and further education from: AbbVie, alanta health service, Bristol-Myers Squibb, Galapagos, Janssen, Pfizer, Takeda and Tillotts, outside the submitted work. He is an active member of the following public law bodies: Advisory Commission of the Examination Office for the Saarland, the State Committee of Physicians and Health Insurance Companies and the Extended State Committee, and an employee in a company belonging to alanta health group GmbH, Hamburg, Germany, outside the submitted work. Shares, other fees from: AbbVie, Bayer, Gilead and GlaxoSmithKline, outside the submitted work. TK has received consulting fees from: Takeda, Janssen, Pfizer, Ferring, AbbVie and MSD, outside the submitted work. MH has received fees for studies from: Sandoz, Janssen, and AbbVie, outside the submitted work. RE has received personal fees from: Takeda, Abbvie, MSD, Biogen, Hexal, Amgen, Celltrion and Viatris, outside the submitted work. LT has no relevant financial or non-financial interests to disclose. AS has received personal fees from: AbbVie Deutschland GmbH & Co. KG, Janssen-Cilag GmbH, Tillotts Pharma, outside the submitted work. PJ has received personal fees from: AbbVie, BMS, Janssen, Falk, Medtronic, Pfizer and Ferring. PH has received consulting fees from AbbVie, Amgen, MSD, Ferring, Hospira, Takeda, Pfizer, Mylan, Biogen, Janssen, Hexal, Galapagos, BMS, and Celltrion, outside the submitted work; PH has provided speaking and teaching services for AbbVie, Amgen, Ferring, MSD, Hospira, Pfizer, Mylan, Celltrion, Takeda, Janssen and Mundipharma, outside the submitted work. SS has received personal fees from AbbVie, Arena, BMS, Biogen, Celltrion, Celgene, IMAB, Gilead, HIMA, IMAB-Biopharma, MSD, Mylan, Pfizer, Fresenius, Janssen, Takeda, Theravance, Provention Bio, Protagonist and Falk, outside the submitted work.