Efficacy and Safety of Natalizumab and Vedolizumab in the Treatment of Crohn’s Disease: A Systematic Review and Meta-Analysis

Introduction

Crohn’s disease (CD) is a chronic inflammatory disease of the gastrointestinal tract that affects all layers of the intestinal wall, primarily the small and large intestines [1, 2]. It belongs to the category of inflammatory bowel diseases (IBD) and is characterized mainly by severe wasting, weight loss, prolonged diarrhea, and abdominal pain [3 ,4]. Compared with other regions of the world, the prevalence of CD among Caucasians in North America and Europe is higher than in Asia. In Asian countries, the incidence ranges from 0.07 to 3.12 per 100,000 [5].

Although the exact cause of CD is unknown, it is closely related to a complex interaction of environmental factors, genetic factors, and an abnormal immune response to intestinal microbiota [6, 7]. Mutation in the nucleotide-binding oligomerization domain 2 (NOD2) gene is one of the risk factors for the development of CD [8]. Environmental factors such as smoking and high-fat diet are also known to aggravate the condition. Treatment options for CD primarily include corticosteroids, immunomodulators, and biologics such as tumor necrosis factor (TNF) inhibitors [9]. However, not all patients respond adequately to these treatments or achieve remission, creating a need for additional therapeutic approaches to improve efficacy and minimize side effects.

Natalizumab and vedolizumab are two drugs that act by inhibiting the adhesion of certain integrin molecules, thereby preventing the migration of immune cells into inflamed intestinal tissues [10]. According to a study [11], natalizumab administered three times during weeks 4, 8, and 12 successfully resulted in remission and clinical response in patients with CD. With a single administration of natalizumab (300 mg) at Week 4, the rate of remission failure was lower in the natalizumab group (76%) than in the placebo group (83%) [11]. Two infusions at Week 8 resulted in a lower rate of remission failure in the natalizumab group (66%) than in the placebo group (77%) [11]. Finally, three injections at Week 12 showed that the natalizumab group had a lower rate of remission failure (61%) than in the placebo group (73%), indicating that natalizumab is significantly more effective than placebo in inducing remission in CD [11].

In addition, vedolizumab is effective in inducing remission and maintaining response to therapy in patients with CD [12]. By Week 6 of the GEMINI II clinical trial, nearly 31% of patients receiving vedolizumab had a higher clinical response rate than 26% of patients in the placebo group [12]. Given that only 15% of patients receiving vedolizumab achieved remission compared to 7% in the placebo group, vedolizumab is useful in helping patients achieve early remission in the induction phase [12].

Both natalizumab and vedolizumab are effective in the treatment of CD, but have different safety profiles and mechanisms of action. Although natalizumab inhibits lymphocyte migration into intestinal and central nervous system tissues by targeting α4 integrin, it also increases the risk of progressive multifocal leukoencephalopathy (PML); consequently, it is only used in patients who did not respond to other treatments. In contrast, vedolizumab specifically targets α4 integrin, which only affects the intestine, thereby reducing the risk of PML and making it safer for long-term use.

However, several previous studies have shown mixed results to date, and there have been no recent meta-analyses of the efficacy and safety of natalizumab and vedolizumab in the treatment of CD. The previous meta-analysis was conducted in 2015 and included a small number of studies. This meta-analysis presents more recent and updated studies, as well as various outcome measures including clinical response, clinical remission, and the incidence of adverse events including nasopharyngitis, arthralgia, nausea, abdominal pain, fatigue, headache, pyrexia, upper respiratory tract infection, vomiting, and serious adverse events compared with placebo. Therefore, we aimed to compare these two drugs to determine the most appropriate treatment option based on the safety profile, efficacy, and needs of patients with CD.

Material and Methods

This meta-analysis follows the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines [13]. Ethical approval was not required because the study relies solely on data from previously published research. We did not use artificial intelligence to assist in writing the manuscript. In the process of compiling the meta-analysis, we used search engines to search for literature, data analysis applications such as Review Managers, and Grammarly to check the correctness of English grammar. This study has been registered with PROSPERO (CRD42024624052).

Selection of publications

Literature searches were conducted in PubMed, ScienceDirect, Google Scholar, Cochrane Library, SpringerLink, and EB

SCO databases to identify relevant studies up to July 2024, using the following keywords: “Crohn’s disease” OR “Inflammatory Bowel Disease” OR “IBD” AND “Natalizumab” AND “Vedolizumab” AND “Efficacy” AND “Safety”.

Inclusion and exclusion criteria

The inclusion criteria for this meta-analysis were as follows: (1) selected studies were randomized controlled trials (RCT) with or without blinding, published in English, either domestically or internationally, while observational studies (prospective and retrospective cohorts, case-control, or cross-sectional) were also eligible; (2) selected studies compared natalizumab and vedolizumab; (3) selected studies involved adult patients (18 years or older) diagnosed with CD who met relevant diagnostic criteria; and (4) outcome indicators included clinical response, clinical remission, and incidence of side effects. The exclusion criteria were as follows: (1) duplicate publications; (2) absence of a control group; and (3) conference abstracts and case reports.

Study quality assessment

Three main components of study design were assessed using the modified Newcastle-Ottawa Scale (NOS), which is used to evaluate the quality of RCTs: outcome assessment, comparability of groups, and selection of study groups. This scale has an overall quality score ranging from 0 to 9 pts. However, the Cochrane Risk of Bias Tool (version 2) for RCTs is primarily concerned with assessing the quality of study methodologies. The choice of reported outcomes, handling of missing outcome data, precision of outcome measures, adherence to reported interventions, and randomization method are the five main components of study design that are assessed by this tool.

Data extraction

The following data were extracted from each study: author’s name, publication year, country, study design, sample size, gender distribution, and age. The primary outcome for this meta-analysis was clinical response and clinical remission. Odds ratios (ORs), 95% confidence intervals (CIs), and p-values were also collected from the selected studies. The 95% CI was included as one of the components of this analysis when presenting the data as a survival plot using the Kaplan-Meier curve.

Statistical analyses

RevMan 5.4 software was used for statistical analysis in this study. 95% CI and ORs were determined when calculating the data. To examine the heterogeneity between studies, χ² and I2 tests were employed in this meta-analysis. Fixed effects model analysis was performed when P>0.1 or I2<50% indicated no statistical heterogeneity between studies. This suggests statistical heterogeneity between studies. Further investigation of the causes of heterogeneity was required. After eliminating obvious heterogeneity, random effects model was used for analysis. Publication bias analysis and subgroup analysis based on the type of included studies were performed using funnel plots. The test threshold was: α=0.05.

Results

Selection of publications

PubMed, ScienceDirect, Google Scholar, Cochrane Library, SpringerLink, and EBSCO were among the online databases that yielded 1,029 studies throughout the relevant study search process. After screening for titles and abstracts, 971 studies were found that could be evaluated for eligibility. Ten studies were included after 773 were eliminated due to their incompliance with the inclusion and exclusion criteria. A flowchart that summarizes the entire literature search procedure in accordance with PRISMA Guidelines 2022 is shown in Figure 1.

Figure 1. PRISMA flowchart.

Characterization of included studies

Ten studies (9 RCTs and 1 cohort study) were included in our meta-analysis (Tables 1 and 2). In terms of the design of the included studies evaluating the efficacy and safety of natalizumab in patients with CD, most of them used an RCT design. In terms of the country, most studies were from England. The total sample size was 2,557 patients. The lowest mean age was 34.4 years, and the highest mean age was 38.1 years. Based on the included studies evaluating the efficacy and safety of vedolizumab in patients with CD, most studies used an RCT design. Based on the country, most studies were from the United States, Canada, and Japan. The total sample size was 2,624 patients. The lowest mean age was 33.9 years, and the highest mean age was 38.2 years. The quality assessment of the cohort studies yielded the NOS score of 8 pts implying good quality. The assessment of bias in RCT studies is presented in Figure 2.

Table 1. Characteristics of the natalizumab (NAT) study for the treatment of Crohn's disease (CD)

Table 2. Characteristics of the vedolizumab (VEDO) study for the treatment of Crohn's disease (CD)

Figure 2. Quality assessment of each study.

Efficacy of natalizumab

Clinical response

When comparing patients with CD treated with natalizumab with those treated with placebo, the clinical response was significantly higher in the former (OR=2.14; 95% CI: 1.68-2.71, p<0.00001). A fixed effects model was used to analyze clinical response, as there was no evidence of significant heterogeneity (I2=15%; p=0.32) (Figure 3A, Table 3).

Figure 3. Forest plot of the meta-analyses of the efficacy natalizumab vs. placebo for clinical response (A) and clinical remission (B).

Table 3. Forest plot of the meta-analyses of the efficacy natalizumab vs. placebo for clinical response

Clinical remission

Patients with CD treated with natalizumab had a significantly higher rate of clinical remission (OR=2.47; 95% CI: 1.12–5.42, p=0.02) than patients treated with placebo. Due to high heterogeneity (I2=82%; p<0.0001), clinical response was analyzed using a random effects model (Figure 3B, Table 4).

Table 4. Forest plot of the meta-analyses of the efficacy natalizumab vs. placebo for clinical remission

Efficacy of vedolizumab

Clinical response

Vedolizumab demonstrated a better clinical response than placebo (OR=0.60; 95% CI: 1.28-1.99, p<0.0001). Due to the lack of evidence of statistically significant heterogeneity (I2=20%; p=0.28), clinical response was analyzed using a fixed effects model (Figure 4A, Table 5).

Figure 4. Forest plot of the meta-analyses of the efficacy vedolizumab vs. placebo for clinical response (A) and clinical remission (B).

Table 5. Forest plot of the meta-analyses of the efficacy vedolizumab vs. placebo for clinical response

Heterogeneity: Chi²=6.28, df=5 (P=0.28); l²=20%. Test for overall effect: Z=4.11 (P<0.0001).

Clinical remission

Vedolizumab was superior to placebo in clinical remission (OR=1.86; 95% CI: 1.42-2.43, p<0.00001). Since no evidence of significant heterogeneity was observed, clinical response was analyzed using a fixed effects model (I2=0%; p=0.94) (Figure 4B, Table 6).

Table 6. Forest plot of the meta-analyses of the efficacy vedolizumab vs. placebo for clinical remission

Heterogeneity: Chi²=1.25, df=5 (P=0.94); l²=0%. Test for overall effect: Z=4.54 (P<0.00001).

Safety of natalizumab

Natalizumab was safer than placebo regarding the incidence of nasopharyngitis (OR=1.30; 95% CI: 0.62-2.71, p=0.49), arthralgia (OR=1.99; 95% CI: 0.79-5.00, p=0.14), nausea (OR=0.08; 95% CI: 0.56-1.16, p=0.24), abdominal pain (OR=0.88; 95% CI: 0.62-1.26, p=0.50), fatigue (OR=1.08; 95% CI: 0.71-1.64, p=0.74), headache (OR=1.21; 95% CI: 0.90-1.63, p=0.22) and serious adverse events (OR=0.65; 95% CI: 0.41-1.05, p=0.08). Other adverse events, such as infections (OR=0.79; 95% CI: 0.18-3.55, p=0.76) and pharyngitis (OR=1.17; 95% CI: 0.73-1.86, p=0.51), were similar between natalizumab and placebo (Figure 5, Table 7-15).

Figure 5. Forest plot of the meta-analyses of the safety of natalizumab vs. placebo for nasopharyngitis (A), arthralgia (B), nausea (C), abdominal pain (D), fatigue (E), headache (F), serious adverse events (G), infections (H) and pharyngitis (I).

Table 7. Forest plot of the meta-analyses of the safety of natalizumab vs. placebo for nasopharyngitis.

Table 8. Forest plot of the meta-analyses of the safety of natalizumab vs. placebo for arthralgia

Table 10. Forest plot of the meta-analyses of the safety of natalizumab vs. placebo for abdominal pain

Table 11. Forest plot of the meta-analyses of the safety of natalizumab vs. placebo for fatigue

Table 12. Forest plot of the meta-analyses of the safety of natalizumab vs. placebo for headache

Table 13. Forest plot of the meta-analyses of the safety of natalizumab vs. placebo for serious adverse events

Table 14. Forest plot of the meta-analyses of the safety of natalizumab vs. placebo for infections

Table 15. Forest plot of the meta-analyses of the safety of natalizumab vs. placebo for pharyngitis

Safety of vedolizumab

Vedolizumab was safer than placebo in terms of the incidence of nasopharyngitis (OR=1.23; 95% CI: 0.90-1.69, p=0.19), arthralgia (OR=1.02; 95% CI: 0.74-1.42, p=0.89), nausea (OR=1.26; 95% CI: 0.91-1.75, p=0.17), abdominal pain (OR=0.81; 95% CI: 0.60-1.10, p=0.17), exhaustion (OR=0.18; 95% CI: 0.02-2.13, p=0.18), headache (OR=0.97; 95% CI: 0.74-1.29, p=0.85) and serious adverse event (OR=1.06; 95% CI: 0.49-2.31, p=0.88). Additional adverse events comparable with placebo included vomiting (OR=0.81; 95% CI: 0.53-1.24, p=0.33), upper respiratory tract infection (OR=1.34; 95% CI: 0.87-2.07, p=0.18), and pyrexia (OR=1.04; 95% CI: 0.72–1.49, p=0.84 (Figure 6, Tables 16-25).

Figure 6. Forest plot of the meta-analyses of the safety of vedolizumab vs. placebo for nasopharyngitis (A), arthralgia (B), nausea (C), abdominal pain (D), fatigue (E), headache (F), serious adverse events (G), pyrexia (H), upper respiratory tract infections (I), and vomiting (J).

Table 16. Forest plot of the meta-analyses of the safety of vedolizumab vs. placebo for nasopharyngitis

Table 17. Forest plot of the meta-analyses of the safety of vedolizumab vs. placebo for arthralgia

Table 18. Forest plot of the meta-analyses of the safety of vedolizumab vs. placebo for nausea

Table 19. Forest plot of the meta-analyses of the safety of vedolizumab vs. placebo for abdominal pain

Table 20. Forest plot of the meta-analyses of the safety of vedolizumab vs. placebo for fatigue

Table 21. Forest plot of the meta-analyses of the safety of vedolizumab vs. placebo for headache

Table 22. Forest plot of the meta-analyses of the safety of vedolizumab vs. placebo for serious adverse events

Table 23. Forest plot of the meta-analyses of the safety of vedolizumab vs. placebo for pyrexia

Table 24. Forest plot of the meta-analyses of the safety of vedolizumab vs. placebo for upper respiratory tract infections

Table 25. Forest plot of the meta-analyses of the safety of vedolizumab vs. placebo for vomiting

Assessing the risk of bias

This study also assessed a risk of bias in selected publications, which is presented in Figures 7 and 8. The results of the analysis show that the clinical remission (Figure 7B) and arthralgia (Figure 7D) in the natalizumab vs. placebo, and the arthralgia (Figure 8D), fatigue (Figure 8G) and serious adverse events (Figure 8I) in the analysis of vedolizumab vs. placebo, may be subject to publication bias, one of the reasons for which is the study limitations associated with a relatively new research. However, most variables showed a symmetrical distribution in the funnel plots, indicating a low risk of bias.

Figure 7. Funnel plots of natalizumab vs. placebo for clinical response (A), clinical remission (B), nasopharyngitis (C), arthralgia (D), nausea (E), abdominal pain (F), fatigue (G), headache (H), serious adverse events (I), infections (J) and pharyngitis (K).

Figure 8. Funnel plots of vedolizumab vs. placebo for clinical response (A), clinical remission (B), nasopharyngitis (C), arthralgia (D), nausea (E), abdominal pain (F), fatigue (G), headache (H), serious adverse events (I), pyrexia (J), upper respiratory tract infections (K), and vomiting (L).

This meta-analysis also conducted a quality assessment of each included study. The results of the assessment showed that the included studies were of moderate to high quality; hence, they provided accurate data for interpreting the results of this meta-analysis, as shown in Figure 2.

Discussion

Based on the results of this meta-analysis, both natalizumab and vedolizumab are significantly more effective than placebo in the treatment of CD. Among patients receiving natalizumab, clinical response (OR=2.14; 95% CI: 1.68-2.71, p<0.00001) was significantly higher compared with placebo. In addition, natalizumab also demonstrated superiority over placebo in terms of clinical remission (OR=2.47; 95% CI: 1.12-5.42, p=0.02). This was supported by the ENACT-1 study, which demonstrated that natalizumab could improve clinical response (56% vs. 49% in the placebo group) and clinical remission (37% vs. 30% in the placebo group) at Week 10, albeit these results were not statistically significant [14]. However, further analysis of the subpopulation of patients with elevated C-reactive protein (CRP) levels (>2.87 mg/L) showed more significant results in the natalizumab group [14]. At Week 12, the ENCORE study demonstrated that natalizumab was successful as an induction therapy with a clinical response rate of 48% vs. 32% in the placebo group (p<0.001) and a clinical remission rate of 26% vs. 16% in the placebo group (p=0.002) [14].

Among patients with CD, vedolizumab was similarly superior to placebo in achieving clinical response (OR=1.60; 95% CI: 1.28-1.99, p<0.0001) and clinical remission (OR=1.86; 95% CI: 1.42-2.43, p<0.00001). Of 4 studies with 1,126 participants, 19.8% of the vedolizumab group achieved remission vs. 11.6% in the placebo group (RR 1.61; 95% CI 1.20-2.17). A number needed to treat (NNT) of 13 indicates high treatment efficacy [15]. Further evidence for this comes from a study that showed that patients who responded to vedolizumab had a higher rate of clinical remission (36–39%) at Week 52 during the maintenance phase than those in the placebo group (22%) [16]. Furthermore, compared with 30% in the placebo group, 44-46% of participants taking vedolizumab had a sustained clinical response [16]. Although it is unclear how this result was maintained from the early stages to Week 52, the GEMINI 1 study showed that vedolizumab successfully achieved corticosteroid-free clinical remission at Week 52 [17].

In terms of safety, natalizumab carries a higher risk of nasopharyngitis, arthralgia, and fatigue. Although serious side effects are generally lower with natalizumab than with placebo, PML remains a serious concern due to its potential to induce PML [14]. One study found that natalizumab use in patients with CD may increase the risk of PML, a rare but serious brain infection. PML is caused by reactivation of the JC polyomavirus (JCV) in the central nervous system of patients with weakened immune systems [18]. Natalizumab works by inhibiting the α4 integrin molecule, which is essential for lymphocyte migration into inflamed tissue. However, inhibition of this integrin also affects the brain, increasing the risk of PML because immune cells are less capable of protecting against JCV reactivation [19]. Therefore, patients taking natalizumab should be closely monitored for neurological symptoms and should be enrolled in the TOUCH registry to monitor the risk of PML and improve early detection [20].

Vedolizumab, on the other hand, has a better safety profile, with adverse events such as nasopharyngitis, arthralgia, and headache not significantly different from placebo. Serious adverse events were also similar to placebo, and to date, no cases of PML have been reported in patients taking vedolizumab [21]. This is further supported by a study that suggests that although vedolizumab has a similar mechanism of action to natalizumab, it selectively targets the α4β7 integrin on lymphocytes, directing them to the gastrointestinal tract [22]. Unlike natalizumab, which targets the central nervous system (CNS)-associated α4 subunit (α4β1 integrin), vedolizumab is considered to be less prone to disruption of immune surveillance in the CNS, thereby reducing the potential risk of developing PML compared to natalizumab [15, 22].

Limitations of this meta-analysis include the variability in the time frame of the included studies, spanning from 2008 to 2022. Such a long period may lead to differences in treatment standards and clinical approaches, which may affect the overall results. Additionally, although all analyzed studies were RCTs, which are considered the gold standard of clinical research, differences in patient population characteristics and variability in treatment duration may lead to heterogeneity in the results.

Conclusion

Both natalizumab and vedolizumab have proven their efficacy in the treatment of Crohn’s disease, demonstrating significant benefits over placebo in terms of clinical response and remission. Natalizumab shows high efficacy as an induction therapy, especially in the subpopulation of patients with elevated CRP levels, but poses a significant risk of PML requiring close monitoring. In contrast, vedolizumab has a better safety profile, with adverse events comparable to placebo and no cases of PML reported, making it a safer option for long-term use. However, differences in study design and duration are limitations of this meta-analysis that should be considered when applying the results to clinical practice.

Acknowledgments

We are grateful to the Division of Gastroenterology and Hepatology, Department of Internal Medicine, School of Medicine, Udayana University, for supporting this article.

Conflict of interest

No conflicts of interest declared by the authors.

Funding

No external funding was involved.