Abstract
Inflammatory bowel disease (IBD) is a chronic inflammatory disorder of the gastrointestinal tract, typically classified as either ulcerative colitis (UC) or Crohn’s disease (CD). The treatment approaches for both UC and CD have changed substantially over the last 10 years, going from step-up to top-down approaches and culminating in treat-to-target algorithms. In addition to the evolution of treatment strategies for IBD, disease endpoints have also changed from symptom-based outcomes to more objective measures, including endoscopy, with histology and transmural healing currently under investigation. In this narrative review, we describe recent advances in the treatment of patients with IBD, with a focus on vedolizumab, the first and only anti-lymphocyte trafficking agent approved as a gut-selective biologic for induction and maintenance therapy in IBD. Furthermore, with the approval of more biologic agents, there remains a need to determine the sequence in which biologics should be used, both individually and in combination. Substantial progress has been made in the last decade, and even though biomarkers and algorithmic trials have been largely disappointing, other avenues, such as clinical decision support tools, prognostic tools, and new approaches incorporating computational biology, have the potential to more effectively curate a tailored approach to IBD therapy in the coming years.
Plain language summary
Inflammatory bowel disease, also known as IBD, is a life-long illness caused by inflammation of the gut. This means that the gut becomes swollen and irritated. For reasons we don’t understand, the immune system makes gut inflammation worse over time. In 2014, a medication called vedolizumab was approved for IBD. Vedolizumab works by preventing certain immune cells from reaching the gut, which can help reduce inflammation. This review describes improvements in IBD treatment over the last 10 years, with a focus on vedolizumab. Studies have shown that a person’s symptoms don’t always tell doctors if their IBD is getting better or worse. Instead, experts say that treatments should aim to heal the gut, not just reduce symptoms of IBD. Many studies using vedolizumab have looked at gut healing in people with IBD. Also, many new medications for IBD have become available in the last decade. However, doctors need better guidance on when to use each medication, in what order, and whether they can be used together. Research has looked at how well these medications, including vedolizumab, work and how safe they are. This information can help doctors create better step-by-step treatment plans for people with IBD. Looking to the future, doctors hope to choose medications based on what will work best for each person. This is called precision medicine. To support this, researchers have created a tool that can help doctors predict how well someone will respond to vedolizumab by using details about their medical history. Overall, our understanding of IBD and how to treat it has changed a lot in the last 10 years. Vedolizumab will continue to play an important role in IBD research and patient care.
Background
Inflammatory bowel disease (IBD) involves chronic inflammation of the gastrointestinal (GI) tract, typically classified as ulcerative colitis (UC) or Crohn’s disease (CD). 1 UC and CD share similar clinical and pathological features but differ markedly in clinical presentation, underlying genetic and environmental factors, and response to treatment. 2 Traditional management of IBD involves stepwise use of 5-aminosalicylates (UC only) and corticosteroids followed by immunosuppressives (e.g., azathioprine (AZA) or methotrexate) in individuals unable to taper corticosteroids, or whose disease recurred rapidly after corticosteroid withdrawal. 3 IBD treatment approaches have since been supplemented by more targeted therapies. In 1998, infliximab (IFX) became the first monoclonal antibody approved to target tumor necrosis factor (TNF), a key pro-inflammatory cytokine in gut inflammation.4,5 The approval of IFX and subsequent advanced therapies expanded physicians’ understanding of IBD pathophysiology and enabled treatment optimization beyond symptomatic relief. However, limitations of IFX, together with issues in clinical trial design and treatment strategies, necessitated an evolution in the therapeutic approach to IBD.
Aim
The aim of this expert-driven narrative review was to describe treatment advances in IBD over the last 10 years and the role of vedolizumab (VDZ; Figure 1), a gut-selective biologic for the treatment of IBD, along with other approved therapies that have become available.
Timeline of key vedolizumab studies for the treatment of patients with IBD.
Search strategy
A PubMed search identified clinical trials and real-world studies in patients with UC or CD, published over the last 10 years, including those involving VDZ. Articles describing advances in clinical trial design or therapeutic strategies were primarily included. As many VDZ studies were informed by the limitations of TNF antagonists, select studies involving TNF antagonists were included for historical context. Key studies involving other advanced therapies were included to contextualize the role of VDZ in the current treatment landscape.
Drug development in IBD
Development of vedolizumab
VDZ is a humanized monoclonal antibody that limits the recruitment of pro-inflammatory lymphocytes to the gut. 6 VDZ development was designed upon the identification of two novel factors contributing to the inflammatory milieu in IBD: mucosal addressin cell adhesion molecule 1 (MAdCAM-1) and α4β7 integrin. The binding of lymphocyte-expressed α4β7 integrin to MAdCAM-1 on the lumen of mucosal blood vessels contributes to lymphocyte trafficking to the gut in response to pathogens. 7 MAdCAM-1 expression is upregulated in IBD, promoting lymphocyte migration and chronic GI inflammation. 7
VDZ targets and blocks the α4β7 integrin, preventing its interaction with MAdCAM-1. Furthermore, VDZ is gut-selective and does not affect systemic lymphocyte trafficking; consequently, VDZ does not confer a predisposition to progressive multifocal leukoencephalopathy, a serious demyelinating disease associated with previous anti-integrin therapies.6,8–10 Furthermore, VDZ exhibits low immunogenicity and does not require concomitant immunosuppressant therapy to prevent development of anti-drug antibodies. 11 The GEMINI 1, GEMINI 2, and GEMINI 3 studies in patients with moderate-to-severely active UC or CD demonstrated the benefit of VDZ on clinical outcomes,12–14 and the GEMINI long-term safety (LTS) trial reported up to 9 years of follow-up. 15 Notably, a meta-analysis including four placebo-controlled studies reported that adverse events (AEs) occurred more frequently in placebo-treated than VDZ-treated patients with IBD (419.4 vs 247.8 per 100 patient years; n = 504 and 2830 patients, respectively). 10
Immunogenicity
Although TNF antagonists were revolutionary at the time, they were not without limitations. In 2004, the ACCENT I study showed that IFX, a chimeric mouse-human antibody, was immunogenic. 16 Episodic IFX administration was associated with the development of anti-IFX antibodies, which led to infusion and delayed hypersensitivity reactions. 17 Concomitant use of immunosuppressives with IFX significantly reduced the formation of anti-IFX antibodies and the incidence of infusion reactions. 17 The SONIC study helped physicians understand how to avoid immunogenicity and improve IFX trough levels. In SONIC, anti-IFX antibodies were detected at week 30 in 0.9% of patients receiving IFX plus AZA and 14.4% of patients receiving IFX alone, suggesting AZA may reduce anti-drug antibodies. The benefits of combination therapy were related to elevated serum IFX concentrations, 18 with higher corticosteroid-free remission rates in patients with increased trough concentrations. 19 Alternatively, adalimumab (ADA) was formulated as a fully human anti-TNF antibody but has also been shown to be immunogenic. 20
Immunogenicity remains a significant factor in patient discontinuation of TNF antagonists. 21 However, the formulation of VDZ combined with its safety profile allowed its use without immunosuppressants and resulted in long-term treatment persistence.22–24 Very low rates of immunogenicity were reported in the GEMINI 1 and GEMINI 2 trials. Among 620 VDZ-treated patients with UC in GEMINI 1, 23 (3.7%) exhibited anti-VDZ antibodies at any time and 6 (1.0%) had samples that were persistently positive (⩾2 consecutive positive samples) through week 52. 12 Similarly, among 814 VDZ-treated patients with CD in GEMINI 2, 33 (4.1%) had at least one sample with a positive test and 3 (0.4%) patients were persistently positive, although these findings had no impact on clinical effectiveness. 13
Among other available biologics, ustekinumab (UST), a monoclonal antibody targeting the p40 subunit of interleukin (IL)-12 and IL-23, was also reported to have low immunogenicity in the IM-UNITI trial (2.3% (27/1154) of patients developed anti-drug antibodies). 25 Evaluation of monoclonal antibodies targeting the p19 subunit of IL-23, including risankizumab, guselkumab (GUS), and mirikizumab, has also been performed. In the M15-989 study (NCT02031276), 12.3% (8/65) of patients with CD developed anti-risankizumab antibodies. 26 Similarly, the GALAXI-2 and GALAXI-3 trials revealed that 5.1% (30/592) of patients with CD developed anti-GUS antibodies by week 48, and this had no effect on clinical outcomes in these patients. 27 A greater proportion of patients with UC in the LUCENT-1 and LUCENT-2 trials developed anti-mirikizumab antibodies (23.3%; 88/378), although only 5/88 patients with anti-mirikizumab antibodies failed to achieve clinical response by week 52. 28
The Janus kinase inhibitors tofacitinib, upadacitinib, and filgotinib are small-molecule agents and therefore have a minimal risk of immunogenicity; however, undesirable drug–drug interactions among small-molecule agents can occur, primarily due to competitive metabolic and clearance mechanisms. 29 Furthermore, these agents are systemic immunosuppressives that significantly increase rates of varicella-zoster infection. 30
Safety
Infections
Data from the TREAT registry confirmed that patients with CD had a higher risk of serious infections, including tuberculosis. 31 A further increased risk of serious infection and/or death was reported with IFX, while other independent risk factors included disease activity, prednisone use, and exposure to narcotic pain relievers. Alternatively, two different meta-analyses of 2932 and 4268 patients reported that rates of serious infections after VDZ administration were similar to those of placebo.10,32 Consequently, VDZ may be a better option than immunosuppressants, corticosteroids, or TNF antagonists for patients predisposed to serious infections, including the elderly, patients with HIV, or those with a history of upper-airway infections or chronic lung disease.
Elderly patients
Managing elderly patients with IBD is a growing challenge due to the aging population and a high frequency of comorbidities. One recent study of patients with IBD from the prospective IBDREAM registry evaluated the safety and efficacy of first anti-TNF use in elderly patients (aged ⩾60 years; n = 81) compared to younger (aged <40 years; n = 546) or middle-aged (40–59 years; n = 268) patients. 33 During a median follow-up of 46 months, 450 patients (50.3%) discontinued therapy, with increasing age being associated with a higher discontinuation rate. In the elderly patient cohort, the incidence of serious adverse events (SAEs; incidence rate ratio (95% confidence interval, CI); vs patients <40 years, 2.1 (1.4–3.0); p < 0.001) and rate of serious infections (vs patients <40 years, 61.2 vs 12.4 per 1000 patient years; p < 0.001) were significantly elevated. 33
Alternatively, the IG-IBD LIVE study in patients with moderate-to-severe IBD compared cohorts of elderly (⩾65 years, n = 198) and matched (sex, disease, site, and duration of disease) non-elderly (18–64 years, n = 396) patients who initiated VDZ. 34 Treatment persistence was higher in non-elderly patients compared with the elderly cohort for UC but not CD (UC, 67.6% vs 51.4%, p = 0.02; CD, 59.4% vs 52.4%, p = 0.32). However, 130 AEs reported by 107 patients were comparable between groups (elderly vs non-elderly, 0.13 vs 0.10 per patient-year). 34 Multivariate analysis showed that CD versus UC and Charlson Comorbidity Index >2 were the only factors associated with an increased risk of AEs, regardless of patients’ age. 34
Cancer and lymphoma
IBD is associated with colorectal,35,36 small intestine, 37 and extraintestinal cancers, including cancers of the skin, hepatobiliary system, oral cavity, breast, and lung.38,39 Of these, colorectal cancer is most common, with a prevalence of 7.7/1000 patients with CD and 37/1000 patients with UC. 39 The risk of lymphoma is also increased in patients with IBD. A registry in Sweden and Denmark (1969–2019), including 164,716 patients with IBD matched with 1,639,027 individuals without IBD, reported that the incidence of lymphoma was significantly higher in patients with CD (35 per 100,000 person years) and UC (34 per 100,000 person years) compared with matched reference individuals (hazard ratio (HR) for CD: 1.32, 95% CI: 1.16–1.50; HR for UC, 1.09, 95% CI: 1.00–1.20). 40
The data evaluating cancer risk related to IBD treatment show mixed results. The TREAT registry reported a non-significant increase in malignancy risk with immunosuppressants alone or in combination with IFX. 41 Similarly, a study of patients with IBD and prior cancer diagnosis found that treatment with VDZ or UST was not associated with subsequent cancer manifestations over a 7-year follow-up. 42 Alternatively, a French nationwide cohort study showed that among adults with IBD, thiopurine and TNF antagonist monotherapies were associated with small but statistically significant increases in lymphoma risk compared with naïve patients; the risk of lymphoma was higher with combination than with monotherapy. 43 However, this study did not evaluate the potential risk of disease severity and may have been confounded by indication. Longer-term data are needed to clarify the link between malignancies and advanced therapies for IBD.
Long-term safety
The longest open-label study in IBD, GEMINI LTS, followed 2243 VDZ-treated patients (CD, n = 1349; UC, n = 894) for up to 9 years, including 1822 patients from prior VDZ clinical trials and 421 new patients. 15 During that time, 46.2% and 39.7% of patients with CD or UC experienced an AE attributed to treatment (inclusive of previous trials). SAEs attributed to VDZ were reported in 5.9% and 4.1% of patients with CD or UC, respectively. In addition, GEMINI LTS reported no new trends for infections, malignancies, infusion-related reactions, or hepatic events. 15
LTS of VDZ was further reported in the XAP extension study, where 311 patients (CD, n = 169; UC, n = 142) from GEMINI LTS rolled over with most (88.4%) reducing VDZ-dosing frequency from every 4 weeks (Q4W) to every 8 weeks. 44 At 2 years, treatment persistence was 78.7% or 83.8% for patients with CD or UC, respectively. During that time, 4.1% of patients with CD and 2.1% with UC experienced an AE attributed to VDZ. SAEs attributed to VDZ occurred in 1 (0.6%) patient with CD and 1 (0.7%) with UC. Data from the XAP 2-year extension study further demonstrated the LTS of VDZ use for patients with IBD. 44
The prospective, multicenter SVEAH observational extension study in Sweden also reported LTS of VDZ. Patients with CD (n = 68) or UC (n = 46) treated with VDZ were followed for up to 3 years. 45 At week 156, treatment persistence was 83.8% and 78.3% for patients with CD or UC, respectively, and two SAEs were reported in one patient. 45
Complications
Fistulae
Fistulas are a frequent and serious complication of CD, with perianal fistulas being the most common. The ACCENT II trial demonstrated that maintenance therapy with IFX every 8 weeks significantly reduced hospitalizations, surgeries, and procedures in patients with fistulizing CD compared with placebo. 46 Abscess development in these patients was independent of cumulative IFX exposure. 47 External fistulas (especially perianal fistulas) exhibited a higher closure rate in response to IFX than other fistula types. 48
VDZ was investigated in an observational study and in the phase IV ENTERPRISE study. The observational study enrolled 151 patients with active or inactive perianal CD and a history of fistulizing or draining perianal lesions. 49 VDZ clinical success (no medical or surgical intervention for perianal disease) was achieved in 23/102 (22.5%) patients with active perianal CD at 6 months, with responses in 20/53 (37.7%) patients who underwent pelvic magnetic resonance imaging. 49 In the ENTERPRISE study (NCT02630966), 28 patients with moderate-to-severely active CD and 1–3 active perianal fistulas received VDZ (300 mg IV) at weeks 0, 2, 6, 14, and 22, with some patients receiving an additional VDZ dose at week 10. 50 Sustained improvements in fistulizing CD were observed in both treatment cohorts, and the additional VDZ dose did not affect treatment outcomes over 30 weeks. 50 At week 30, 15/28 (53.6%) patients achieved the primary endpoint of ⩾50% decrease in the number of draining fistulas. 50 Eighteen patients (64.3%) achieved complete perianal fistula closure at any point during the study, primarily during the first few weeks. 50 The study demonstrated that clinically meaningful reductions in draining fistulas were observed as early as week 2 that persisted through week 30. 50
The results of the ENTERPRISE study showed that patients with fistulizing CD can achieve clinical success with VDZ, whereas a single-center retrospective study of 68 patients (VDZ, n = 20; UST, n = 48) suggested that VDZ may be less effective than UST for patients with fistulizing CD. In that study, patients were more likely to discontinue VDZ compared with UST, generally due to inadequate clinical response. 51 In addition, a large systematic literature review of randomized clinical trials reported that TNF antagonists, but not VDZ or UST, were superior to placebo for inducing clinical remission in patients with fistulizing CD. 52 Currently, the American Gastroenterology Association recommends IFX as first-line therapy for patients with CD and active perianal fistula, with ADA, UST, and VDZ suggested as reasonable alternatives. 53
Post-ileocolic CD recurrence
Most patients experience CD recurrence after ileocolonic resection, with endoscopic lesions usually preceding symptoms within a few months. 54 The PREVENT study was the first large, multicenter, placebo-controlled trial in patients with postoperative CD treated with a biologic. 55 Patients were included if they had undergone ileocolonic resection with ileocolonic anastomosis and had no evidence of active CD elsewhere in the GI tract. 55 However, the study was terminated at week 104 (of 208) because the primary outcome (clinical recurrence at or before week 76) was not met (IFX, 12.9% vs placebo, 20.0%). 55 Notably, the key secondary endpoint of endoscopic recurrence at week 76 was significantly lower with IFX than with placebo (30.6% vs 60.0%; p < 0.001). 55
The prospective, placebo-controlled, REPREVIO study was the first to investigate the prophylactic efficacy of VDZ on CD recurrence after ileocolonic resection. 54 Eighty-four patients were assigned within 4 weeks of surgery to receive VDZ or placebo over 24 weeks. Ileocolonoscopy was performed at week 26 to assess recurrent lesions. 54 The primary endpoint—the probability of a lower modified Rutgeerts score at week 26—was 77.8% higher with VDZ than with placebo (p < 0.0001). Severe endoscopic recurrence occurred in 10/43 (23.3%) of VDZ-treated patients and 23/37 (62.2%) of placebo-treated patients (difference −38.9% (95% CI: −56.0 to −17.3); p = 0.0004). Thus, VDZ treatment within 4 weeks of ileocolonic resection was more likely than placebo to prevent endoscopic recurrence, demonstrating that VDZ is effective for postoperative management in patients at risk for disease recurrence. 54
Treatment of chronic active pouchitis
Restorative proctocolectomy with ileal pouch-anal anastomosis (IPAA) is routinely performed in patients with UC who require colectomy. The most common long-term complication is chronic pouchitis, where lymphocytes infiltrate and exacerbate inflammation of the pouch. Based on its relevant mechanism of action, VDZ was evaluated and became the only biologic approved for the treatment of pouchitis. This approval was supported by the phase IV, randomized, double-blind, placebo-controlled EARNEST study that evaluated the safety and efficacy of VDZ over 34 weeks in patients with chronic pouchitis following IPAA for UC. 56 Active chronic pouchitis was defined as a modified Pouchitis Disease Activity Index (mPDAI) total score of ⩾5 with an endoscopic subscore of ⩾2 (on the basis of findings outside the staple or suture line). 56 Patients also had either ⩾3 episodes of pouchitis within 1 year prior to screening that were treated with ⩾2 weeks of antibiotics or other prescription medicines, or otherwise required continuous antibiotic treatment for at least 4 weeks prior to baseline endoscopy. 56 The primary endpoint, mPDAI remission at week 14 (mPDAI score ⩽4 and reduction from baseline of ⩾2 points in mPDAI total score), was significantly higher with VDZ than with placebo (16/51; 31.4% vs 5/51; 9.8%; p = 0.01). 56 Furthermore, VDZ-treated patients had fewer ulcers in the pouch than placebo-treated patients at weeks 14 (mean: 5.0 vs 13.4) and 34 (mean: 2.7 vs 9.7). 57 Finally, a higher proportion of VDZ-treated than placebo-treated patients achieved mucosal healing at week 14 (16.7% (7/42) vs 2.5% (1/40)). 57 All seven patients who achieved mucosal healing with VDZ were in PDAI remission at weeks 14 and 34. 57 These results demonstrated the efficacy of VDZ in achieving clinical remission, endoscopic improvement, and mucosal healing of the ileal pouch in patients with UC and chronic pouchitis.
Evolution of clinical trial design
Although the conventional drug development model is logical, it is also inefficient, and high placebo response rates are common in IBD clinical trials. 58 Efficient drug development models in IBD are increasingly based on strategies integrating objective measures of inflammation and patient-based dosing centered on pharmacokinetic/pharmacodynamic evidence. 59
SPIRIT consensus
The SPIRIT consensus (2021), organized by the International Organization for the Study of Inflammatory Bowel Diseases (IOIBD), evaluated the lack of objective measures to define IBD disease progression. Consequently, the SPIRIT consensus proposed three objective endpoints for future IBD clinical trials. The ultimate therapeutic goal in IBD was to prevent disease impact on patient’s lives (health-related quality of life, disability, fecal incontinence), followed by midterm complications (encompassing bowel damage in CD, IBD-related surgery and hospitalizations, disease extension in UC, extraintestinal manifestations, permanent stoma, short bowel syndrome), and long-term complications (GI and extraintestinal dysplasia or cancer, mortality). 60 SPIRIT laid the groundwork for designing new disease modification trials.
Establishing objective criteria for IBD disease modification
Disease assessments have changed over the last decade from symptom-based clinical outcomes to more objective measures, including endoscopy, histology, cross-sectional imaging, biomarkers, and transmural healing, evaluated by intestinal ultrasound (IUS; Table 1). 61 This shift in therapeutic approach from managing patient symptoms to alleviating the underlying inflammation could alter the course of the disease and result in disease modification.
Milestone clinical trials incorporating evolving efficacy outcomes over time.
Disease clearance is the combination of CR, ER, and absence of active histological disease.
Target groups: (1) CS-free symptomatic remission; (2) CS-free symptomatic + endoscopic remission; (3) CS-free symptomatic + endoscopic remission + histologic remission.
(1) CS-free IUS outcomes + clinical symptoms + biomarkers; (2) clinical symptoms + biomarkers alone.
Patients receive dual targeted therapy during an induction period followed by VDZ monotherapy during maintenance therapy.
ADA, adalimumab; AZA, azathioprine; CD, Crohn’s disease; CR, clinical remission/response; CRP, C-reactive protein; CS, corticosteroid; DR, deep remission; ER, endoscopic response/remission; FCP, fecal calprotectin; GS, Geboes score; HR, histological response/remission; IFX, infliximab; IUS, intestinal ultrasound; IV, intravenous; MaRIA, Magnetic Resonance Index of Activity; MH, mucosal healing; mPDAI, modified Pouchitis Disease Activity Index; PBO, placebo; RHI, Robarts Histopathology Index; SC, subcutaneous; TNF, tumor necrosis factor; UC, ulcerative colitis; UPA, upadacitinib; UST, ustekinumab; VDZ, vedolizumab.
Endoscopic remission
Endoscopic remission has been defined across clinical trials using different thresholds, including the Crohn’s disease endoscopic index of severity (CDEIS), the simple endoscopic score for CD (SES-CD), or complete absence of ulceration (Table 2).73–75 As endoscopic healing is rarely achieved in patients with CD, endoscopic response (a 50% reduction in SES-CD score from baseline) is often included as an additional endpoint. VERSIFY was the first large-scale trial to evaluate the benefits of VDZ on endoscopic outcomes in patients with CD. 65 The primary endpoint, endoscopic remission (SES-CD ⩽4) at week 26, was achieved by 12/101 (11.9%) patients. 65 Endoscopic remission was consistently higher in patients who were TNF antagonist-naïve than in those who were TNF antagonist-experienced (week 26, 19.6% vs 5.5%; week 52, 25.0% vs 8.3%). 65 The LOVE-CD trial similarly evaluated endoscopic remission in 110 patients with CD treated with VDZ over 1 year, of whom 97 (88.2%) were TNF-antagonist experienced. At week 52, 40/110 (36.4%) patients achieved endoscopic remission. 76
Definitions of scoring systems in CD.
CD, Crohn’s disease; CDEIS, Crohn’s disease endoscopic index of severity; GS, Geboes score; NI, Nancy Index; RHI, Robarts Histopathology Index; SES-CD, simple endoscopic scale for Crohn’s disease; VAS, visual analog scale.
Mucosal healing
The EXTEND study (2012) was the first randomized, placebo-controlled trial to use mucosal healing (defined as the absence of mucosal ulceration at week 12) as a primary endpoint. 62 In patients with moderate-to-severe CD, those treated with ADA tended to be more likely than those given a placebo to achieve mucosal healing (27.4% vs 13.1%; p = 0.056). Mucosal healing was also included as a secondary endpoint in the endoscopy substudy of the UNITI-1, UNITI-2, and IM-UNITI trials evaluating UST for treating CD, 77 defined as the absence of mucosal ulcerations among patients who presented with ulceration in at least one ileocolonic segment at baseline. The proportion of patients with mucosal healing at week 44 was numerically higher in the UST group than in the placebo group (13.0% vs 4.2%, p = 0.409). 77
VICTORY, the first multicenter observational study comparing two classes of drugs in patients with IBD (VDZ vs TNF antagonists) using propensity score matching, reported both mucosal healing and deep remission (clinical remission plus mucosal healing) rates for the first time in patients with moderate-to-severely active CD. 78 The primary endpoint was the proportion of individuals achieving clinical remission (resolution of all CD-related symptoms) or mucosal healing (defined as absence of ulcers and/or erosions assessed by endoscopy). 78 The study demonstrated that clinical remission was achieved with VDZ induction therapy (10.9% of patients at week 6) and after 6, 12, and 18 months of maintenance treatment (cumulative; 18%, 35%, and 54% of patients, respectively). Moreover, after 6 and 12 months of VDZ maintenance treatment, cumulative rates of mucosal healing were 20% and 63%, respectively, while the cumulative rates of deep remission were 14% and 26%. 78 Multivariable Cox proportional hazard regression analysis showed that patients were less likely to achieve: (1) clinical remission if they had severe disease or active perianal disease at baseline, were a previous or current smoker, or had previous TNF antagonist use, and (2) mucosal healing if they had severe baseline disease activity and previous TNF antagonist use. 78
Mucosal healing was also used as a secondary endpoint for the VERSIFY trial, achieved by 9/32 (28.1%) VDZ-treated patients naïve to TNF antagonists and 1/24 (4.2%) TNF antagonist-experienced patients at week 52. 65
Transmural healing in CD
While mucosal healing reflects repair of the superficial GI mucosal layer, an alternative target for patients with CD is transmural healing, which reflects the resolution of inflammation throughout the bowel wall. 79 Cross-sectional imaging of the bowel with magnetic resonance enterography (MRE) is a less invasive alternative to supplement ileocolonoscopy in assessing disease activity beyond the mucosa, extramural disease features, and complications in CD. 80 Magnetic Resonance Index of Activity score (MaRIA) is the most extensively evaluated MRE index, incorporating a composite of mucosal and bowel wall features including bowel edema, wall thickening, presence of ulcers, and relative contrast enhancement.
VERSIFY was the first phase III trial to include a substudy evaluating transmural healing using a standardized MRE protocol to determine how VDZ affects transmural inflammation (bowel edema and wall thickness). 66 In total, 89% of patients at baseline and 83% at follow-up had MRE sequences sufficient to evaluate disease activity. 66 In patients with MRE at baseline and week 26, 75.9% of ileocolonic segments exhibited bowel edema at baseline, decreasing to 36.1% by week 26. 66 This improvement was larger in colorectal segments (30.6%) than in the terminal ileum (52.4%). 66 A similar reduction was seen in the bowel wall thickness (>3 mm), from 100% of patients at baseline to 67.5% at week 26 (colorectal, 61.3%; terminal ileum, 85.7%). 66 This demonstrated the efficacy of VDZ to impact changes in transmural inflammation in patients with CD and established that transmural healing is a feasible clinical outcome in the treatment of CD. 66
Transmural healing is being further evaluated in patients with CD in the continuing SUNRISE study (NCT05192863). A recent interim analysis reported that after 6 months of VDZ treatment, 9/28 (32.1%) and 12/28 (42.9%) of patients achieved transmural remission or response, respectively. 81 In addition, VECTORS (NCT06257706), a phase IV, randomized, controlled, parallel-group, open-label study of VDZ is evaluating the value of treating to transmural healing using IUS in participants with moderate-to-severe CD (ClinicalTrials.gov identifier: NCT06257706).
Histological healing
Histological healing in patients with IBD reflects a lack of active inflammation assessed by mucosal biopsies. 82 A Cochrane review in 2017 concluded that the development of a validated histological scoring system for CD was a clinical and research priority. 83 Currently, three different scoring systems are commonly used: the Geboes Score (GS), the Nancy Index (NI), and the Robarts Histopathology Index (RHI; Table 2). 84
Initially, histological outcomes were only included as exploratory endpoints in clinical trials. However, in VARSITY, histological remission was the biggest difference between VDZ and ADA: at week 52, it occurred in 10.4% of the VDZ group and 3.1% of the ADA group using GS <2.0; and in 37.6% and 19.9% of patients, respectively, using the RHI score <3. 63 LOVE-CD took a further step by including histological outcomes at week 52 in the primary objective of VDZ treatment.67,76 Histological outcomes were assessed using paired biopsies at weeks 0 and 26 and analyzed using the GS and the RHI. 76 Of 67 biopsies with active inflammation at baseline (GS score ⩾3.1), 43 (64.2%) showed histological remission (GS <3.1) at week 26. 76 Of the 56 biopsies with active inflammation at baseline according to the RHI (score >7), 37 (66.1%) showed histological remission (RHI ⩽6) at week 26. 76 Corticosteroid-free histological remission is also included as a treatment target in one of the groups of the ongoing VERDICT study. 72
Composite endpoints in UC
Endpoints that combine endoscopic and histologic disease activity in clinical trials include the Histologic Endoscopic Mucosal Improvement (HEMI) and the Histologic Endoscopic Mucosal Remission (HEMR). 85 Early achievement of HEMI after induction therapy has been shown to be a better predictor of clinical and symptomatic outcomes at the end of maintenance therapy than either histologic improvement or endoscopic remission alone. Early achievement of HEMR was associated with a numerically higher likelihood of improved clinical outcomes at the end of maintenance therapy than achieving HEMI without HEMR. 86
Disease clearance (DC) is another novel composite endpoint combining clinical and endoscopic remission with minimal histologic disease activity (Figure 2). The IOIBD released a systematic review and expert consensus (2023) defining DC as simultaneous clinical, endoscopic, and histological remission of disease (clinical remission, partial Mayo score = 0; endoscopic remission, MES = 0; histologic remission, NI = 0). 87 It was believed that achieving DC could block long-term disease progression and prevent disease-related complications. 87
Disease clearance (DC).
A VARSITY subanalysis (2021) investigated the value of DC as an endpoint. 64 DC was defined as clinical remission (partial Mayo score ⩽2 and no individual subscore >1 (excluding sigmoidoscopy subscore)), endoscopic improvement (MES ⩽1), and absence of active histological disease (RHI score of <5). 64 29.2% (112/383) of VDZ-treated patients achieved DC at week 52 versus 16.3% (63/386) of ADA-treated patients. 64 Compared with patients whose treatment failed, those who achieved DC had lower C-reactive protein (CRP) concentrations at baseline. 64
Can early intervention with VDZ improve clinical outcomes?
In addition to establishing the need for objective measures to evaluate IBD, the SPIRIT consensus also introduced the concept of a “window of opportunity” where early disease control can prevent disease progression. Multiple studies have since evaluated the efficacy of VDZ in patients with early versus late-stage IBD.
The LOVE-CD study examined the efficacy of VDZ treatment in early (<2 years since diagnosis and naïve to advanced therapy, n = 86) versus late (>2 years and prior exposure to TNF antagonists, n = 74) CD. The stringent primary endpoint (clinical and endoscopic remission at weeks 26 and 52) was reached in 31.4% versus 8.6% (p = 0.001) of patients with early versus late CD, respectively.67,88 Similarly, a retrospective review of patients from the VICTORY Consortium registry evaluated VDZ treatment outcomes in those with early (⩽2 years, n = 62) versus late (>2 years, n = 588) years CD. 89 At 6 months, significantly more VDZ-treated patients with early versus late CD met the primary endpoints of clinical remission (38% vs 23%), corticosteroid-free clinical remission (43% vs 14%), and endoscopic remission (29% vs 13%; all p < 0.05). 89 By contrast, the LOVE-UC trial reported no significant differences in clinical or endoscopic and histologic outcomes between VDZ-treated patients with early (<4 years) or late (>4 years) disease. 71 In agreement with these findings, retrospective review of VDZ-treated patients from the VICTORY Consortium registry with early (⩽2 years, n = 109) versus late (>2 years, n = 328) UC found that disease duration was not a significant predictor of clinical outcomes. 89
These results indicate that early intervention with VDZ in patients with CD but not UC provides a better opportunity to prevent disease prevention. However, there remains an unmet need to define how early “early” is with regard to treatment to determine whether physicians can halt the progression of IBD.
Comparison of biologic therapies
Another important unmet need in IBD is determining the optimal sequence of biologics, both as monotherapy and in combination. Initial RCTs used a placebo as a comparator, with head-to-head comparisons limited to real-world studies. VICTORY (2016) was the first registry to compare two classes of treatment in patients with IBD (VDZ vs TNF antagonist) using propensity score matching. 78 It showed that mucosal healing and deep remission rates were possible after VDZ treatment of patients with moderate-to-severe CD. 78 VARSITY (2019), the first head-to-head study of biologics in patients with IBD, demonstrated that not all biologics were equally effective. 63 Intravenous (IV) VDZ was found to be superior to subcutaneous (SC) ADA at week 52 in adults with moderate-to-severe UC who were TNF antagonist-naïve and in whom conventional therapy had failed, both for clinical remission (primary endpoint; VDZ, 120/383; 31.3% vs ADA, 87/386; 22.5%; p = 0.006) and endoscopic improvement (VDZ, 152/383; 39.7% vs ADA, 107/386; 27.7%; p < 0.001), but not corticosteroid-free clinical remission (VDZ, 14/111; 12.6% vs ADA, 26/119; 21.8%; p > 0.05). 63
The EVOLVE study compared the real-world effectiveness and safety of VDZ and IFX in biologic-naïve patients with moderate-to-severe CD. 90 The results in the inverse probability weighted dataset (VDZ, n = 99; IFX, n = 63) showed that there was no significant difference in adjusted 12-month clinical remission rates between treatments (73.1% vs 55.2%; p = 0.31). 90 However, AEs were significantly less likely with VDZ treatment than with IFX.
The VEDOIBD registry investigated the real-world effectiveness of VDZ and TNF antagonists in 314 biologic naïve patients with UC at the end of induction (week 14) and during 1–2 years of maintenance treatment. 91 The primary outcome, clinical remission (partial Mayo ⩽1 + bleeding subscore = 0), was relatively similar in VDZ- and TNF antagonist-treated patients during induction therapy (23.0% vs 30.4%; p = 0.2). 91 However, clinical remission rates were significantly higher after 2 years in VDZ- than TNF antagonist-treated patients (43.2% vs 25.8%; p < 0.011). 91 A similar analysis from the VEDOIBD registry was conducted in 327 biologic-naïve patients with CD at 14 weeks, and at 1 and 2 years. 92 While the rate of clinical remission was lower with VDZ than TNF antagonists at week 14 (56.3% vs 73.9%; p < 0.05), after 2 years clinical remission was significantly higher with VDZ than with TNF antagonists (74.2% vs 44.7%; p < 0.05). 92
Other studies which compared biologics in patients with CD include the SEAVUE, SEQUENCE, GALAXI, and VIVID trials.93,94 The SEAVUE trial found that in biologic-naïve patients with CD, UST and ADA showed similar efficacy in helping patients achieve clinical remission at week 52 (primary endpoint; UST, 124/191; 64.9% vs ADA, 119/195; 61.0%). 94 Endoscopic remission was also similar between treatments (UST, 51/179; 28.5% vs ADA, 55/179; 30.7%). 94 In the SEQUENCE study where patients (n = 255) had prior experience of TNF antagonist failure, risankizumab was superior to UST for the co-primary endpoint of endoscopic remission at week 48 (31.8% vs 16.2%; p < 0.001). 93 Similarly, in the GALAXI-2 and 3 trials, which included comparisons between GUS and UST, GUS was reported to be superior to UST at 48 weeks for endoscopic response (mean difference (CI); 100 mg GUS vs UST, 10.6% (2.7,18.5); 200 mg GUS vs UST, 15.6% (7.9, 23.4)), endoscopic remission (8.5% (1.1, 15.9); 12.3% (4.9, 19.7)), and deep remission (7.4% (0.3, 14.6); 11.3% (4.2, 18.5)). 27 Alternatively, the VIVID-1 study included comparisons between mirikizumab and UST in patients with CD; mirikizumab (n = 579) was non-inferior to UST (n = 287) at week 52 for clinical remission (54.1% vs 48.4%), and also led to similar outcomes for endoscopic response (48.4% vs 46.3%). 95
Evolution of treatment strategies
STRIDE II guidelines
The STRIDE II guidelines (2021) reaffirmed the importance of a treat-to-target approach and emphasized the role of intermediate targets in IBD, identifying clinical and endoscopic remission as the main therapeutic goals in disease management (Table 3). 75 Transmural healing in CD and histological healing in UC were recognized as important measures representing a deeper level of healing, but were not recommended as formal treatment targets.
STRIDE II guidelines for CD and UC.
If target is not reached, consider changing treatment.
Clinical remission: CD: PRO2 (abdominal pain ⩽1 and stool frequency ⩽3) or HBI <5; UC: PRO2 (rectal bleeding = 0 and stool frequency = 0) or partial Mayo (<3 and no score >1).
Clinical response: decrease of ⩾50% in PRO2 (CD: abdominal pain and stool frequency; UC: rectal bleeding and stool frequency).
CRP to values under the upper limit of normal and FCP to 100–250 mg/g.
Endoscopic healing: CD: SES-CD <3 points or absence of ulcerations (e.g., SES-CD ulceration subscores = 0); UC: Mayo endoscopic subscore = 0 points or UCEIS ⩽1 point.
Could be used as an adjunct to endoscopic remission to represent a deeper level of healing.
Transmural healing assessed by CTE, MRE, or bowel ultrasound.
CD, Crohn’s disease; CRP, C-reactive protein; CTE, computerized tomography enterography; ESR, erythrocyte sedimentation rate; FCP, fecal calprotectin; HBI, Harvey–Bradshaw Index; HRQoL, health-related quality of life; MRE, magnetic resonance enterography; PRO2, patient-reported outcomes—version 2; SES-CD, simple endoscopic scale for Crohn’s disease; UC, ulcerative colitis; UCEIS, Ulcerative Colitis Endoscopic Index of Severity.
Treat-to-target approach
Treat-to-target approaches aim to optimize treatment outcomes and maximize the effectiveness of biologic therapies, ensuring patients are not maintained on ineffective treatments. 96 This replaced progressive, stepwise treatment intensification that could put patients at risk of bowel damage by delaying highly effective therapies. Moreover, the recognition of early disease states allowed for regular disease assessment using objective outcome measures to guide treatment adjustments, looking beyond symptoms to treat endoscopic lesions and alter the natural course of the disease.97,98 The treat-to-target approach is informed by prior and ongoing studies evaluating optimal algorithms for treating patients with IBD.
Algorithmic trials
To elevate the therapeutic ceiling of current IBD therapies, algorithmic trials have been developed to define optimal treatment, dosing protocols, and treatment targets for patients with IBD. An important early algorithmic trial in patients with IBD was the REACT study, which evaluated treatment algorithms for early combined immunosuppression (n = 22 practices) or conventional management (n = 19 practices) in adult patients with CD. Although the primary endpoint of symptomatic remission was similar between practices, lower rates of major adverse outcomes were observed for patients with combined immunosuppression. 99 These results suggested that a modified treatment algorithm incorporating early induction with an effective therapy can alter the course of disease in patients with CD.
This was further demonstrated by the results of the PROFILE study. PROFILE was an open-label, biomarker-stratified RCT in adults with newly diagnosed active CD in which eligible patients were randomized in a 1:1 ratio to a top-down (IFX + immunomodulator) or accelerated step-up treatment approach. 100 Randomization was stratified by biomarker subgroup, disease location (colon vs other), and mucosal inflammation (mild, moderate, or severe). 100 The primary endpoint, the incidence of continued surgery-free and steroid-free remission through week 48 after an initial (⩽8-week) steroid induction course, was more frequent in the top-down group than in the accelerated step-up group (78.8% vs 15.3%; p < 0.0001). 100 The results of the PROFILE study—similar to LOVE-CD—highlighted the importance of early treatment in CD.
Similarly, a tight control algorithm utilizing treatment escalation in response to worsening clinical symptoms and inflammatory biomarkers is also key to successful therapeutic decisions. This was demonstrated in the CALM study, which showed that tight monitoring was feasible in biologic-naïve patients with CD. 101 Responsive escalation of ADA based on clinical symptoms together with biomarkers (CRP and fecal calprotectin (FCP)) and prednisone use in patients with early CD (median disease duration of 1 year) resulted in better clinical and endoscopic outcomes than symptom driven (Crohn’s disease activity index and prednisone use) decisions alone. 101 Significantly more patients in the tight control group than the clinical management group achieved mucosal healing (primary endpoint; CDEIS <4 and no deep ulcers) at week 48 (45.9% vs 30.3%). 101
A tight control algorithm was also evaluated in REACT-2, where early combination therapy with treatment intensification targeting the absence of ulcers (>5 mm in size) in 29 practices was compared with step-up care with treatment intensification targeting clinical remission. 102 There was no difference between groups in the primary endpoint of time to first occurrence of CD-related complications at 2 years (step-up, 40.9% vs early combined, 43.1%). 102 However, there was a 25% reduction in risk of CD-related complications in patients with active inflammation at baseline (CRP >5 mg/L) and a 29% reduction in the risk of CD-related complications in patients with CRP >5 mg/L and centrally confirmed baseline ulcers (early combined group only). 102
The continuing VERDICT study (NCT04259138; 2020) is the first algorithmic trial evaluating the best target for disease modification in UC. 72 Patients (n = 660) with active UC are being randomly assigned in a 2:3:5 ratio to one of three treatment target groups: (1) corticosteroid-free symptomatic remission (Mayo rectal bleeding subscore = 0), (2) corticosteroid-free endoscopic remission (MES ⩽1) + symptomatic remission, and (3) corticosteroid-free histologic remission (GS <2B.0) + endoscopic + symptomatic remission. 72 Induction treatment algorithms varied based on patients’ prior treatment regimen at screening. 72 Early treatment with IV VDZ and dose escalation up to 300 mg Q4W until target achievement is a central part of each algorithm. 72 The primary endpoint is the time from treatment target achievement to a disease-related complication among patients who achieved their pre-defined target remission. 72 As of September 5, 2024, 90.3% (84/93), 77.3% (102/132), and 67.2% (123/183) of patients with observed data achieved their treatment targets by week 48 in Groups 1–3, respectively, with a higher percentage in biologic-naïve versus biologic-experienced patients for Group 3 (70.1% vs 50.0%). 103 These interim results highlight that DC is an achievable target for VDZ-treated patients with UC.
In addition, the continuing VECTORS study will examine two treatment target algorithm groups: corticosteroid-free transmural healing + clinical remission + biomarker remission versus corticosteroid-free remission + biomarker remission (ClinicalTrials.gov identifier: NCT06257706).
Precision medicine
Biomarkers
Serum testing of CRP and stool testing of FCP are the most widely available biomarkers in clinical practice for patients with IBD. 104
CRP expression is upregulated in numerous infectious and inflammatory pathologies, limiting its utility to differentiate IBD from other diagnoses. In the GEMINI 2 study, mean changes in CRP from baseline to week 6 were comparable between the VDZ and placebo groups. 13 In patients with elevated CRP at week 6, a greater proportion of VDZ-treated patients exhibited normal levels by week 52 compared to those treated with placebo. 13 It remains unclear whether CRP is a reliable predictor of treatment response or a marker of disease activity. 105
Alternatively, FCP is associated with neutrophils that infiltrate the intestinal lumen. While informative as a marker for gut inflammation, FCP is less useful for determining the underlying etiology. Nonetheless, FCP is superior to CRP for predicting endoscopic response, predicting relapse in patients in clinical remission, and monitoring response in active disease. 104
The open-label PROFILE (predicting outcomes for CD using a molecular biomarker) trial was stratified by biomarker subgroup (IBDhi vs IBDlo), disease location (colon vs other), and mucosal inflammation (mild vs moderate vs severe). 100 The results showed no significant biomarker-treatment interaction effects for the primary outcome or any secondary outcomes. 100
Moving forward, utilizing biomarkers to evaluate disease activity and to monitor and predict treatment response will require further evaluation in future studies.
Clinical decision support tools
The lack of useful clinical prediction models guiding therapeutic decisions led to the development and validation of a Clinical Decision Support Tool (CDST) to anticipate VDZ response in patients with CD. 106 The scoring system was developed using data from GEMINI 2 and validated using data from the VICTORY study. 106 Investigators evaluated the tool’s ability to identify patients in clinical remission or corticosteroid-free remission with mucosal healing, clinical remission with mucosal healing, or corticosteroid-free remission with mucosal healing after VDZ treatment. 106 The CDST is calculated using the following variables associated with remission: TNF antagonist naïvety (+3 points), no prior bowel surgery (+2 points) or fistulizing disease (+2 points), and baseline albumin (+0.4 points per g/L) and CRP (−0.5 points if 3.0–10.0 mg/L, −3.0 points if >10 mg/L). 106 Patients with ⩽13 points had a low probability of response, whereas patients with >19 points had a high probability of response. 106
The VDZ CDST was further expanded to include clinical remission rates and measure VDZ exposure, rapidity of onset of action, response to dose optimization, and progression to surgery, assessed in three independent datasets (GEMINI, GETAID, and VICTORY), by CDST-defined response groups (low, intermediate, or high). 107 Higher response rates were reported in patients with a high versus a low probability of response for various outcomes, including healthcare resource use and surgery. Highly significant differences in measured VDZ exposure were predicted by the CDST-defined response categories throughout the 52-week study, irrespective of when VDZ exposure was measured. Moreover, the response categories defined by the CDST were strong predictors of differences in the rapidity of onset of action and absolute decrease in Harvey–Bradshaw Index across the 52-week study. These observations allowed identification of patients with low drug exposure and slower onset of action who may benefit from early dose intensification. The VDZ CDST predicted clinical remission and steroid-free remission with VDZ, but not with UST, in patients with CD who were refractory or intolerant to anti-TNF antagonist therapy. 108 As the VDZ CDST is a free, web-based resource based on a simple scoring system, it can easily be incorporated by clinicians without significant training and without increasing their workload. The utility of the VDZ CDST in identifying patients likely to be VDZ responders is crucial to employing a precision medicine approach in daily clinical practice.
Risk identification algorithm
CDPATH™ is a validated tool that was developed to evaluate patients’ risk of complications due to CD by analyzing clinical, genetic, and serologic factors to determine factors influencing the time between diagnosis and the onset of the first complication (bowel stricture, internal penetrating disease, or non-perianal surgery). 109 Patients were divided into low-risk (0%–19.9%), medium-risk (20.0%–59.9%), and high-risk (60.0%–100%) groups. 109 For the calibration cohort, the incidence of CD complications was 0%, 28%, and 58% in patients assigned low-, medium-, and high-risk scores, respectively. 109 The CDPATH tool was developed by healthcare providers as a tool to give physicians and patients a better understanding of a patient’s potential risk of CD-related complications within 3 years. The COMPASS study (NCT04809363) will help elucidate its impact on real-world clinical practice.
Disease trajectory
Most current research is cross-sectional, and longitudinal data focusing on individual responses to drugs (i.e., “trajectories”) are lacking. Clinical endpoints are typically evaluated at discrete time points and sequentially; thus, the responder populations for different endpoints only partially overlap. In cohort trials, patient-reported outcomes are normally averaged across groups and do not include individual changes over time.
Using the individual patient-level data (PRO2 values), machine learning approaches using dynamic learning could more efficiently derive patterns and identify responder subpopulations during early treatment (e.g., at induction). In VARSITY, four types of response patterns (super responders, fast responders, partial responders, and incomplete/non-responders) were identified, culminating in different disease trajectories at week 52. 110 Early responders achieved deeper clinical benefit than other groups. These results support the possibility of precision medicine in patients with IBD, as disease trajectories could become an important tool for personalized medicine and clinical decision-making.
Treatment sequencing
The optimal sequencing of therapies in patients with IBD is crucial for patient outcomes. A real-world comparative effectiveness study evaluated 13,222 patients (UC, n = 4185; CD, n = 9037) registered in the UK-based IBD BioResource. 111 Inverse probability treatment weighting was used to assess treatment outcomes associated with various biologics (IFX, ADA, golimumab, VDZ, UST (in patients with CD only)), including drug sequencing across patient subpopulations. 111 The primary outcome was drug efficacy assessed by treatment persistence. 111 For first-line biologic treatment in patients with UC, VDZ was superior to TNF antagonists for treatment persistence over 5 years. 111 Compared with VDZ, HRs (95% CI) for treatment failure or discontinuation were 1.9 (1.3–2.8) for IFX, 3.1 (2.1–4.5) for ADA, and 3.4 (2.2–5.2) for golimumab (all p < 0.001). 111 For first-line biologic treatment in patients with CD, IFX was superior to ADA over a 10-year follow-up. 111 Alternatively, the low number of patients in the UST and golimumab groups excluded them from the 10-year analysis. 111 Data for VDZ were only available over a 3-year follow-up, where the rate of survival free of treatment discontinuation or failure was higher for VDZ (69.5%) than for IFX (54.5%) or ADA (49.8%). 111 In patients with CD and perianal disease, IFX was superior to ADA but not VDZ, with no significant difference between ADA and VDZ. 111
For second-line treatment after TNF antagonist failure, after a 3-year follow-up in patients with UC previously treated with ADA, VDZ was superior to IFX (p < 0.001). 111 Similarly, VDZ was superior to ADA in patients who received first-line IFX therapy (p < 0.001). 111 In patients with CD treated with non-anti-TNF biologics, UST and VDZ demonstrated superior efficacy compared with second-line TNF antagonist use, although the differences were small. 111 In patients with CD and prior TNF antagonist use, effectiveness outcomes were similar between patients who received either UST or VDZ as second- or third-line treatment. 111 This was the largest real-world study to date that showed a benefit for VDZ in terms of persistence compared with other therapies, suggesting VDZ should be positioned earlier when treating patients with UC. 111 VDZ may also be more effective in patients with CD when used earlier in the treatment sequence. 111
Together, these data suggest that VDZ is optimally used earlier in the treatment sequence before initiating TNF antagonist therapy, particularly when using superior endpoints like endoscopic remission or mucosal healing that better predict long-term disease outcomes compared to clinical remission. 112
Combination therapy
In 2010, the landmark SONIC study showed that the combination of IFX + AZA improved efficacy outcomes in biologic-naïve patients with CD. 19 In SONIC, combination therapy resulted in a significantly higher proportion of patients achieving corticosteroid-free clinical remission at week 26 versus monotherapy (IFX + AZA, 56.8%; IFX, 44.4%; AZA, 30.0%). A post hoc analysis of SONIC showed that deep remission (clinical remission and mucosal healing) was achievable with the combination of IFX and AZA. 113 Patients with objective evidence of inflammation (i.e., elevated baseline CRP) had the best clinical results with IFX.
Currently, clinical trials are evaluating patient outcomes with combined advanced therapies that target non-overlapping disease pathways. The VEGA study showed that for patients with UC (n = 255), golimumab plus GUS improved clinical response versus monotherapy at week 12 (golimumab + GUS 83.1%; GUS 74.6%; golimumab, 61.1%). 114 The phase II DUET-CD (NCT05242471) and DUET-UC (NCT05242484) trials aim to confirm these results. In addition, the EXPLORER study was a small phase IV, single-arm, open-label study evaluating triple therapy with VDZ + ADA + methotrexate. In 55 biologic-naïve patients with newly diagnosed moderate-to-severe CD, 19 (34.5%) were in endoscopic remission and 30 (54.5%) were in clinical remission at week 26. 70 The EXPLORER 2.0 (NCT06045754; 2024) study will assess combined treatment with VDZ plus ADA versus VDZ plus UST, while the VICTRIVA (NCT06227910; 2025) study will evaluate VDZ plus upadacitinib compared to VDZ plus placebo. Combining advanced therapies could potentially elevate the therapeutic ceiling for patients with IBD.
Potential limitations of VDZ therapy
Not all patients respond to VDZ
While numerous studies have reported on the safety and efficacy of VDZ, it remains true that there is no “one-size-fits-all” therapy for patients with IBD. The retrospective EVOLVE study reported rates of primary non-response for anti-TNF naïve patients who were treated with VDZ or TNF antagonists. 115 For patients with CD, 8.8% (19/215) of patients receiving VDZ and 4.9% (13/266) receiving a TNF antagonist reported treatment discontinuation due to primary non-response. 115 Similarly, rates of treatment discontinuation due to primary non-response among patients with UC were 9.8% (37/376) of VDZ-treated patients and 17.2% (38/221) of patients receiving a TNF antagonist. 115 Separately, a real-world study of 290 patients with UC who received VDZ, of whom 63.7% were naïve to prior advanced therapy, reported that 17.6% (51/290) of patients discontinued VDZ due to primary non-response. 116 Looking to the future, tools like the VDZ CDST will be important to help identify patients most likely to respond to VDZ therapy.
Onset of action
It has been suggested that VDZ may have a delayed onset of action compared with other treatment options, possibly due to its mechanism of action preventing leukocyte trafficking rather than directly targeting inflammation. 117 This perception may also be due to the fact that in the GEMINI 2 study of patients with CD, rates of clinical response between VDZ- and placebo-treated patients were not significantly different at Week 6 (31.4% vs 25.7%; p = 0.23). 13 By contrast, a systematic review of real-world studies (20 for CD; 16 for UC) found that 43% of patients with UC and 56% with CD achieved clinical response with VDZ by Week 6. 118 Thus, roughly half of all patients with IBD who were treated with VDZ in real-world settings had symptomatic relief after treatment induction. A separate post hoc analysis of the GEMINI 1, 2, and 3 trials further suggested that some patients with CD or UC can achieve symptomatic improvement as early as Week 2, especially those who are anti-TNF naïve. 119 Nonetheless, small molecules such as Janus kinase inhibitors are considered to be the fastest-acting therapies for patients with IBD, with reports of symptomatic improvement as early as 1–6 days.120,121
VDZ and extraintestinal manifestations of IBD
As many as half of all patients with IBD may experience extraintestinal manifestations, which are disease symptoms that affect organs besides the digestive tract.122,123 To date, it remains unclear whether extraintestinal manifestations are caused by an environmental or genetic predisposition shared with IBD, or else if leukocytes activated in the gut translocate to other sites in the body. 123 It has been suggested that, due to its gut-selective mechanism of action, VDZ may have limited efficacy in preventing extraintestinal manifestations and could potentially be linked to new instances.124–127 However, studies suggesting an association between VDZ and new extraintestinal manifestations were not necessarily causal and could be confounded by the withdrawal of corticosteroids or TNF antagonists in these patients. 123 Alternatively, other studies have suggested that there is either no difference or else a reduction in new and improvement in pre-existing extraintestinal manifestations after treatment with VDZ.126,128–130 Improvements in pre-existing extraintestinal manifestations after taking VDZ were likely related to decreases in IBD-related inflammation. 123
One recent meta-analysis of non-TNF antagonists found that the incidence of new and changes in pre-existing extraintestinal manifestations was not significantly different between VDZ and UST, except for joint manifestations; they reported that UST may be more beneficial for improving pre-existing extraintestinal manifestations of the joints. 131 Current European Crohn’s and Colitis Organization guidelines most widely recommend TNF antagonists for extraintestinal manifestations, especially of the skin, joints, or eyes. 132
The future of IBD management
Despite advances in drug development, clinical trial design, and treatment strategies, a therapeutic ceiling persists for patients with IBD. Greater understanding of disease etiology may help advance drug development, 133 and biologics with new mechanisms of action, including anti-fibrotic drugs targeting anti-TL1A or the TGFβ-RI receptor, are in development for patients with CD; further investigation of their safety and efficacy is needed. 134 Similarly, dietary and microbiome-related factors could also contribute to IBD 135 ; a recent small study suggested that dietary intervention combined with VDZ was superior to VDZ monotherapy for achieving clinical remission, endoscopic response, and mucosal healing after 16 weeks in patients with CD. 136 In addition, there remains an unmet need for treatment of patients with mild disease, as the majority of trials in patients with IBD are in those with moderate-to-severe disease. Finally, despite the approval of several biosimilars, improved patient access to treatment remains an unmet need. 137
Conclusion
Undeniable progress has been achieved over the past decade, driven less by the availability of novel therapeutics than by a deeper understanding of the interaction among disease, patients, and therapy. VDZ has been at the forefront of treatment innovation aimed at improving clinical outcomes, clinical management, and persistence. Due to its gut-selective mechanism of action and unique indication for IBD, VDZ is likely to be included in future approaches using precision medicine. The cultivation of novel tools and strategies, including CDST, biomarkers, predicting disease trajectories, and the treat-to-target approach, provides hope for another decade of progress, defining a tailored approach to IBD treatment.
Footnotes
Acknowledgements
Under the direction of the authors, medical writing support was provided by Chase Hellmer, PhD, and Claire Line, PhD, of Envision Catalyst, an Envision Medical Communications agency, a part of Envision Pharma Group, and funded by Takeda.
