Comparative Effectiveness of Ozanimod and Vedolizumab as First-Line Advanced Therapies in Ulcerative Colitis: A Propensity-Matched Cohort Analysis

Abstract

Introduction

There is limited real-world evidence comparing the effectiveness of ozanimod to vedolizumab as first-line advanced therapies in patients with ulcerative colitis (UC).

Methods

We conducted a retrospective cohort study using TriNetX, a multi-institutional US database in adults with UC who were initiated on ozanimod compared to vedolizumab between January 1, 2021 and 22 June, 2024. The primary outcome was to compare the risk of a composite outcome of corticosteroid use, colectomy, or change to another advanced therapy between the 2 cohorts within 12 months. 1:1 propensity score matching (PSM) was performed for demographics, comorbid conditions, disease extent, laboratory parameters, and previous corticosteroid use. The risk was expressed as an adjusted odds ratio (aOR) with 95% CIs.

Results

We identified 222 patients in the ozanimod cohort (mean age 41.2 ± 15.7, 46.3% male sex, 68% White, and 22.5% ulcerative proctitis), and 4145 patients in the vedolizumab cohort (mean age 47.4 ± 18.3, 45.2% male sex, 69.7% White, and 17.2% ulcerative proctitis). After PSM, there was no significant difference in the risk of the composite outcome (aOR 0.92, 95% CI, 0.63-1.36) and corticosteroid use (aOR 0.80, 95% CI, 0.53-1.18) between the 2 cohorts within 12 months. There was a higher risk of change in therapy in the ozanimod cohort (aOR 1.95, 95% CI, 1.09-3.49) compared to the vedolizumab cohort. Colectomy rates were low in both cohorts (<0.04%).

Conclusions

Our real-world study showed that ozanimod use is associated with similar corticosteroid use but higher odds of a change in therapy compared to vedolizumab when used as first-line therapy in patients with UC. Further prospective studies are needed to understand long-term outcomes.

Introduction

Ulcerative colitis (UC) is a chronic inflammatory bowel disease (IBD) characterized by continuous mucosal inflammation affecting the colon. The disease typically manifests as recurrent episodes of abdominal pain, bloody diarrhea, and weight loss, which can profoundly impact patients’ quality of life.¹ The exact etiology of UC remains unclear, but it is believed to involve a complex interplay of genetic, immunological, and environmental factors.² Due to the chronic relapsing nature of UC and its potential to develop into colorectal cancer over time, effective long-term management strategies are critical.

Over the years, the therapeutic landscape of UC has evolved considerably. The development of biological therapies, particularly antitumor necrosis factor (anti-TNF) agents, vedolizumab marked a turning point. However, many patients are either refractory to anti-TNF agents and/or vedolizumab, or experience adverse effects, prompting the need for alternative treatments. More recently, a new generation of biologics and small-molecule drugs have shown promise in achieving and sustaining remission.³

Ozanimod is an oral sphingosine-1-phosphate (S1P) receptor modulator that selectively targets lymphocyte trafficking and potentially reduces inflammation by preventing lymphocyte egress from lymph nodes and reaching the gut.⁴ In clinical trials, ozanimod has demonstrated effectiveness in inducing and maintaining clinical remission in patients with moderate-to-severe UC. In May 2021, ozanimod (Zeposia, Bristol Myers Squibb) was the first sphingosine-1 phosphate (S1P) receptor modulator approved by the US Food and Drug Administration (FDA) for the treatment of moderately-to-severely active UC.⁵, Vedolizumab is a monoclonal antibody that targets explicitly α4β7 integrin, a molecule integral to lymphocyte migration into the gastrointestinal tract. By blocking this pathway, vedolizumab provides gut-selective immunosuppression, which reduces inflammation while minimizing systemic immunosuppression.⁶ Vedolizumab has been well-documented in inducing and maintaining remission in patients with moderate-to-severe UC.⁷ As a result, on May 20, 2014, the FDA approved intravenous vedolizumab for moderately-to-severely active UC (https://www.fda.gov/drugs/drug-approvals-and-databases/drug-trials-snapshot-entyvio-vedolizumab-treat-ulcerative-colitis). In 2013, subcutaneous vedolizumab was approved for maintenance treatment of adults with moderately-to-severely active UC. A recent systematic review and meta-analysis by Attauabi et al found vedolizumab effective and safe as a first-line biological therapy for bionaïve patients with UC. In patients with UC, 40% achieved clinical remission by week 14 and 63.9% by week 52. Bionaïve patients with UC had a higher remission probability at week 52.⁸ The VARSITY study compared vedolizumab with adalimumab, revealing higher clinical remission and endoscopic improvement rates for vedolizumab at week 52.⁹

Both ozanimod and vedolizumab are effective treatments for UC, but head-to-head data are lacking. Our study evaluates the real-world outcomes for the treatment of advanced therapy naïve UC. We conducted a retrospective propensity-matched cohort study comparing the effectiveness of ozanimod and vedolizumab as first-line advanced therapies. Our primary aim was to assess the risk of corticosteroid use, colectomy, and change to another advanced therapy within 1 year.

Methods

Database

A retrospective cohort study was conducted using the US Collaborative Network in TriNetX (Cambridge, MA, USA), a multi-institutional database. TriNetX is a global federated research network that provides real-time access to de-identified electronic health records of more than 110 million patients within 64 health care organizations in the United States. Most health care organizations are large academic medical institutions that contain inpatient and outpatient facilities. The data represent the entire patient population of the organization. The de-identification process is determined and done at a network level and attested to through a formal determination by a qualified expert as defined in the HIPAA Privacy Rule. TriNetX obfuscates patient counts <10 to ensure patient anonymity. Clinical variables are derived directly from electronic health records of included health care organizations as well as retrieved through a built-in natural language processing system that extracts variables from clinical documents. Robust quality assurance is achieved at the time of extraction from electronic health records before inclusion in the database, in a systematic and standardized format. The process also includes data cleaning which rejects patient records that don’t meet the TriNetX quality standards. The database does not include claims data or data collected from randomized clinical trials. The database contains inpatient and outpatient claims along with prescription drug claims. The interface only provides aggregate counts and statistical summaries to protect patient health information and ensures that the data remain de-identified at all levels of data retrieval and dissemination.

Study Participants and Cohorts

We conducted a real-time search and analysis of the US Collaborative Network in the TriNetX platform between June 1, 2021 and April 18, 2024. The study start date was chosen as ozanimod was approved on May 27, 2021. Patients who were diagnosed with UC included adults age ≥18 years old who had an International Classification of Disease, Tenth Revision, Clinical Modification (ICD-10-CM) codes in their EHR for Ulcerative colitis (K51*). Patients in the ozanimod cohort and vedolizumab cohort were identified using Rxnorm codes for ozanimod (2 288 236) and vedolizumab (1 538 097), respectively. Patients who had Rxnorm codes for other advanced therapies approved for UC prior to ozanimod or vedolizumab were excluded from the respective cohorts. This was done to ensure ozanimod and vedolizumab were first-line advanced therapies as our study focused exclusively on bionaïve patients who were initiating these therapies. Additionally, patients who had Rxnorm codes for advanced therapies within 1 month of ozanimod or vedolizumab were also excluded. This was done to account for potential insurance denials, delay in initiation of therapy from prescription, and inadequate trial of the therapy of interest. Patients who underwent colectomy within 30 days of ozanimod or vedolizumab were excluded from the study.

Study Outcomes

The primary aim of the study was to assess the risk of oral and/or intravenous steroid use, change in therapy, or colectomy between 1 and 12 months in the ozanimod and vedolizumab cohorts. The index event of the study was the date of prescription of ozanimod and vedolizumab. Outcome assessment was after 30 days from the index event to avoid inclusion of patients who were initiated on corticosteroids at the time or soon after the initial prescription of ozanimod and vedolizumab. Patients who required corticosteroids were identified using Rxnorm code for Prednisone (8640) or Budesonide (19 831) and/or intravenous methylprednisolone (6902) or hydrocortisone (5492). Patients who required total colectomy were identified using Current Procedural Terminology (CPT) or ICD-10 Procedure Coding System codes which can be found in Supplementary Table 1. Change in therapy included patients who were switched to an alternative advanced therapy.

The secondary aim of the study was to assess the odds of fecal calprotectin <200 µg/g and health care resource utilization by evaluating the risk of any-cause hospitalization or emergency department (ED) visit. Additional secondary aims were to assess the incidence of adverse events in the ozanimod cohort. Adverse events included bradycardia or atrioventricular blocks, macular edema, herpes zoster, newly elevated aspartate aminotransferase and/or alanine aminotransferase (≥100 U/L), grade 3 or 4 lymphopenia (absolute lymphocyte count <500 lymphocytes per microliter), prolonged QTc interval, hypertensive crisis, ischemic heart disease and/or cerebrovascular accident, sepsis, acute upper respiratory infections, and urinary tract infections within 12 months in the ozanimod cohort.

Statistical Analysis

All statistical analyses were conducted using the TriNetX software using the browser-based real-time analytics feature, TriNetx Live (TriNetX LLC, Cambridge, MA). Baseline characteristics of cohorts were described using means, SDs, and proportions. Covariates based on demographics, comorbid diseases, laboratory parameters, and historical IBD medication use were identified. One-to-one (1:1) propensity score matching (PSM) was performed to balance the following covariates between groups: age, gender, race, diabetes mellitus, nicotine dependence, obesity, primary sclerosing cholangitis, past or recent history of Clostridium difficile infection, oral, and intravenous steroid use prior to initiation of ozanimod and vedolizumab. Propensity score matching was also performed for mean hemoglobin, C-reactive protein, and calprotectin. TriNetX platform utilizes input matrices of the user-identified covariates to conduct logistic regression analysis to obtain propensity scores for all individual subjects. The propensity scores generated are used to match patients using greedy nearest-neighbor algorithms with a caliper width of 0.1 pooled standard deviations. TriNetX randomizes the order of rows to eliminate bias resulting from nearest-neighbor algorithms. Standardized mean difference after PSM indicates the success of matching a covariate between the 2 cohorts. A standardized mean difference <0.1 indicates that the difference between the cohorts for the covariate is small. After PSM, the risk of each outcome was calculated and expressed as adjusted odds ratios (aOR) with 95% CIs. Kaplan-Meier curves for steroid-free survival were generated for the ozanimod and vedolizumab cohort after 1:1 PSM.

Results

Characteristics of Study Cohorts

We identified 222 patients in the ozanimod cohort and 4145 patients in the vedolizumab cohort. The mean disease duration before initiating ozanimod and vedolizumab was 2.87 and 2.43 years, respectively. In our cohorts, 66.6% and 53.9% of patients were on mesalamine prior to initiation of ozanimod and vedolizumab, respectively. Twenty-three percent and 13.9% of patients had concomitant prednisone use at the time or within 30 days of the prescription of ozanimod and vedolizumab, respectively. There were 0.08% and <0.04% of patients who had concomitant or within 30-day prescriptions of immunomodulators in the ozanimod and vedolizumab cohorts, respectively. In our cohorts, 50.4% had ulcerative pancolitis, and 22.5% had ulcerative proctitis in the ozanimod cohort. 45.6% had ulcerative pancolitis, and 17.2% had ulcerative proctitis in the vedolizumab cohort. There was no difference in the mean disease duration prior to initiation of therapy in the ozanimod and vedolizumab cohorts (1049.1 vs 890 days, P = .08). A complete list of demographic parameters, comorbid conditions, and laboratory parameters before and after PSM can be found in Table 1.

Efficacy of Ozanimod Versus Vedolizumab

There were 81 (36.4%) patients in the ozanimod cohort and 85 (38.2%) patients in the vedolizumab cohort that met the study’s primary outcome. The mean follow-up in the ozanimod cohort was 303.05 days, and in the vedolizumab cohort was 337.5 days. After PSM, the 2 cohorts had no difference in the risk of composite outcome, which included oral and/or intravenous steroid use, change to another advanced therapy, or colectomy (aOR 0.92, 95% CI, 0.63-1.36). There was no difference in the risk of oral steroid use (aOR 1.07, 95% CI, 0.70-1.63) and intravenous steroid use (aHR 0.59, 95% CI, 0.34-1.02) between the 2 cohorts (Table 2). Steroid-free survival was similar between the 2 cohorts (aHR 0.88, 95% CI, 0.64-1.23; log-rank P value = .47; Figure 1). Fewer than 10 patients required colectomy in both cohorts. There was also no difference in the odds of fecal calprotectin <200 µg/g in the ozanimod cohort compared to the vedolizumab cohort (aOR 0.81, 95% CI, 0.34-1.93). The 2 cohorts had no difference in the risk of any-cause hospitalization or ED visit (Table 2).

Thirty-six (16.2%) patients in the ozanimod cohort and 20 (9.0%) patients in the vedolizumab cohort required a change to another advanced therapy. There was a higher chance of a change in treatment in the ozanimod cohort than in the vedolizumab cohort. Drug persistence up to 12 months was lower in the ozanimod cohort compared to the vedolizumab cohort (aHR 2.15, 95% CI, 1.24-3.72; log-rank P value .005; Figure 2).

Adverse Events in Ozanimod Versus Vedolizumab Cohort

The incidence of bradycardia or AV block, macular edema, sepsis, and herpes zoster was low, occurring in fewer than 10 patients in both cohorts. There was no difference in acute upper respiratory tract infection incidence between the 2 cohorts (ozanimod 4.9% vs vedolizumab 6.7%, P = .41). Thirty-six patients (18.1%) developed grade 3 or higher lymphopenia in the ozanimod cohort compared to fewer than 10 patients in the vedolizumab cohort (Table 3).

Discussion

Our study utilized a prospectively maintained database to assess 1-year outcomes between patients receiving ozanimod compared to vedolizumab as a first-line advanced treatment for UC. As the treatment of UC becomes more personalized and the early initiation of advanced technology increases, comparative outcomes between medications will be important. Our study did not find a difference in the risk of corticosteroid use between the ozanimod and vedolizumab cohorts. Overall, rates of colectomy were low across both groups. However, we observed higher odds of change in therapy in patients on ozanimod compared to vedolizumab. 18.1% of patients developed grade 3 or higher lymphopenia which is expected given ozanimod’s mechanism of action. It is unclear if this potentially impacted the higher odds of change in advanced therapy observed in the ozanimod cohort. Additionally, given the option of increased dosing frequency with vedolizumab, this could have led to increased drug persistence in this cohort. With respect to adverse events, the rates of bradycardia/atrioventricular block, macular edema, and infections including herpes zoster were low in both cohorts.

There is scarce head-to-head data comparing ozanimod to vedolizumab in UC. A study by Dubinsky et al published in abstract format compared the efficacy of patients on ozanimod in the TRUE NORTH trial to patients on vedolizumab in GEMINI 1 for moderately-to-severely active UC. After utilizing matching-adjusted indirect comparisons (MAICs), there was no difference in the rates of clinical remission, clinical response, and endoscopic improvement between TNFi-naïve patients on ozanimod and vedolizumab in the induction phase. However, patients on vedolizumab had higher clinical responses than those on ozanimod (OR 0.40, 95% CI, 0.21-0.70), which was felt due to differences in placebo response.¹⁰ Similarly, Dubinsky et al. also performed a MAIC study of patients on ozanimod in the TRUE NORTH trial in patients on ustekinumab in the UNIFI study. The authors found comparable efficacy in bionaïve patients on ozanimod and ustekinumab during the induction and maintenance phase.¹¹ Another recent systematic review and meta-analysis testing on the rapidity of onset of efficacy within the induction period of all biological therapies and small molecules approved for patients with moderate-to-severe UC found that on indirect comparison, no significant difference was found between ozanimod and vedolizumab (1.09 [0.76, 1.56]) at the end of week 2 of induction therapy.¹² However, it is to be noted that in this network meta-analysis, filgotinib, ustekinumab, and ozanimod ranked lowest in all the analyses, particularly depicting delay in induction as compared to other agents.¹²

With respect to adverse events in the ozanimod cohort, the incidence of primary cardiovascular events was low. However, it is to be noted that ozanimod use has been restricted to patients with no known prior cardiovascular risk factors, including recent myocardial infarction, unstable angina or other clinically significant cardiovascular disease, or active or chronic infection according to the phase 3 trial results.¹³ Hence, we can only conclude that cases of incident cardiovascular events in people with no known prior CV events are low and not in the general population as a whole.

Ozanimod and vedolizumab each offer unique advantages due to their distinct mechanisms of action and administration routes. Ozanimod provides a novel approach with benefits such as a rapid onset of action,¹³ absent immunogenicity compared to biologics,¹⁴ and patient preference for oral administration.¹⁵ Its favorable safety profile, including lower risks of serious infections¹⁶ and cardiovascular events compared to other advanced therapies, further justifies its early use in the treatment pathway. These features make ozanimod suitable for patients after failure of aminosalicylates and steroids, potentially before progressing to biologics.^17,18 Vedolizumab is highly effective in maintaining remission over the long term, particularly after anti-TNF failure.^7,19 Hence, the choice between these therapies can be influenced by whether the patient is biologic-naive or has prior biologic exposure. For biologic-naive patients, ozanimod’s rapid onset of action and oral administration may be particularly appealing, making it a strong candidate for early treatment after failure of conventional therapies such as corticosteroids or 5-aminosalicylic acid (5-ASA). Moreover, the convenience of oral administration can enhance adherence and patient satisfaction. On the other hand, for patients with prior exposure to biologics, vedolizumab has shown consistent efficacy and safety. Dubinsky et al found that ozanimod had significantly improved clinical response in biologic-naive patients during induction compared to adalimumab (ADA), but for patients with prior anti-TNF exposure, ozanimod’s efficacy was not significantly different from vedolizumab during the initial 10-week induction period.¹⁰ This suggests that while ozanimod can be an effective treatment option for biologic-exposed patients, its relative benefit may be greater in biologic-naive individuals.¹⁰ The choice between ozanimod and vedolizumab should be individualized based on patient preference for administration route, previous treatment history, and specific clinical considerations.

Our study has several notable strengths. This is the first real-world study to provide data regarding the comparative efficacy of ozanimod and vedolizumab in patients with UC. We had a diverse cohort from HCOs across the US, allowing for generalizability when interpreting the results. We utilized a prospectively maintained database, which allowed for PSM to reduce the impact of confounding variables despite the retrospective nature of the study.

However, our study has several limitations that warrant consideration. While useful, the reliance on administrative data might affect the study’s granularity, as detailed information on disease and endoscopic severity was limited, which could influence treatment outcomes. However, key laboratory values, including calprotectin and C-reactive protein, as well as recent oral and intravenous steroid use, were well-matched between the 2 cohorts. In spite of this, there is still a possibility of bias-related disease activity as data on clinical symptoms cannot be extracted from the database. The study design did not include endoscopic assessment of mucosal healing, a critical marker of long-term treatment success, as the database does not have endoscopy data. Additionally, variability in health care access and treatment practices across centers might have influenced the results, as clinicians in different settings may have different strategies for initiating or continuing advanced therapies, particularly given the rapidly evolving evidence around these drugs and no formal guideline yet in terms of placement of these drugs in the treatment ladder of UC. As previously discussed, data regarding dosing and frequency of vedolizumab was not available during the follow-up period to determine the proportion of patients that had dose optimization. This could explain the higher drug persistence observed in the vedolizumab cohort compared to the ozanimod cohort. Although dosing strengths are available for prednisone from the database, it is difficult to ascertain the actual dose taken by patients. There is a possibility of missing data for certain covariates; however, identification of these patients would be difficult to ascertain, and exclusion could significantly affect sample sizes. Additionally, there is a possibility of patients who may have met the study outcomes outside of the health care organizations which are part of the database. Finally, as with all database studies, there is always a concern for misdiagnosis, residual confounding, and under-reporting of some variables.

In conclusion, this real-world study provides valuable insights into the comparative efficacy of ozanimod and vedolizumab as first-line advanced therapies for UC. The overall steroid-sparing effects were similar between the groups and colectomy rates were low, reinforcing that both treatments offer effective options for controlling UC in biologic-naive patients. However, patients on ozanimod had higher odds of change in therapy. Future prospective studies and randomized head-to-head controlled trials are required to validate our findings. For now, either ozanimod or vedolizumab is an effective first-line advanced therapy for patients with moderate-to-severe UC.

Acknowledgments

Dr. Gursimran S. Kochhar is the guarantor of the article. All authors approved the final version of the manuscript.

Author Contributions

G.S.K.: study conception, study design, critical revision and final approval of the manuscript. H.K. and F.T.: manuscript preparation. J.H., M.R., and F.A.F.: critical revision of the manuscript and final approval of the manuscript. A.D.: study design, data collection, data analyses, manuscript preparation, and final approval of the manuscript.

Funding

None declared.

Conflicts of Interest

H.S., F.T., and A.D.: None. J.G.H.: Advisory board for BMS. F.A.F.: consultant for AbbVie, Avalo Therapeutics, Bausch, BMS, Braintree Labs, Fresenius Kabi, GI Reviewers, GSK, IBD Educational Group, Iterative Health, Janssen, Pharmacosmos, Pfizer, Sandoz Immunology, and Viatris. DSMB for Eli Lilly. M.R.: advisory Boards and Consultant (both) for Abbvie, Janssen, UCB, Takeda, Pfizer, BMS, Organon, Amgen, Genentech, Gilead, Salix, Prometheus, Lilly, Celgene, Boehringer Ingelheim Pharmaceuticals Inc. (BIPI), Celltrion, and Roche. G.S.K.: advisory board—CoreVetas Research, Eli Lilly, and GIE Medical; consultant—Boston Scientific Endoscopy, Olympus Endoscopy, Pentax Endoscopy, Takeda Pharmaceuticals, and Pharmacosmos; speaker—Eli Lilly; stock options—DigBi Health.

Data Availability

The data underlying this article will be shared on reasonable request to the corresponding author.

References