Clinical effectiveness and safety of vedolizumab versus infliximab in biologic-naïve patients with ulcerative colitis: A comparative real-world multicentric observational study

Abstract

Objective

Vedolizumab (VDZ) and infliximab (IFX) are first-line therapies for moderate-to-severe ulcerative colitis (UC). Despite their widespread use, there are no direct comparative studies, and real-world data, particularly in Latin America, are limited. This study compared the effectiveness and safety of VDZ and IFX in biologic-naïve UC patients.

Methods

This retrospective cohort study included patients with moderate-to-severe UC (Mayo score 6–12, endoscopic sub-score ≥2) treated with VDZ or IFX. Primary endpoints were clinical remission (partial Mayo score ≤2), endoscopic remission (Mayo sub-score = 0), and steroid-free clinical remission at week 52. Secondary endpoints included clinical response, endoscopic response, biological therapy optimization, adverse events (AEs), hospitalizations, and biochemical remission at week 52. Propensity score adjustment (1/PS) was used to adjust for potential confounders.

Results

A total of 297 UC patients (156 IFX, 141 VDZ) were analyzed. Clinical remission at week 52 was 82.3% for VDZ and 77.6% for IFX (p = 0.11), while endoscopic remission was higher in VDZ patients (47.4% vs. 33.1%, p = 0.03). Steroid-free clinical remission rates were similar between groups (p = 0.98). Endoscopic response at week 52 favored VDZ (78.4% vs. 62.7%, p < 0.001), and VDZ had higher treatment persistence (80.8% vs. 61.8%, p < 0.001). AEs and hospitalizations were more frequent in IFX patients (p < 0.001).

Conclusions

Both VDZ and IFX are effective in biologic-naïve UC patients, however VDZ demonstrated superior endoscopic outcomes, higher treatment persistence, and a better safety profile, supporting its use as a first-line therapy.

Resumen

Objetivo

Vedolizumab (VDZ) e infliximab (IFX) son terapias de primera línea para la colitis ulcerosa (CU) moderada a grave. Este estudio comparó la efectividad y seguridad de VDZ e IFX en pacientes con CU bionaífs.

Métodos

Estudio de cohorte retrospectivo que incluyó pacientes con CU moderada a grave (Mayo score 6-12, endoscopic sub-score ≥ 2) tratados con VDZ o IFX. Los objetivos primarios fueron la remisión clínica (partial Mayo score ≤ 2), la remisión endoscópica (Mayo sub-score = 0) y la remisión clínica sin esteroides en la semana 52. Los objetivos secundarios incluyeron las respuestas clínica y endoscópica, la optimización de la terapia biológica, los eventos adversos (EA) y las hospitalizaciones en la semana 52. Se utilizó el ajuste de la puntuación de propensión (1/PS) para ajustar los posibles factores de confusión.

Resultados

Se analizó un total de 297 pacientes (156 IFX, 141 VDZ). La remisión clínica en la semana 52 fue 82,3% para VDZ y 77,6% para IFX (p = 0,11), mientras que la remisión endoscópica fue mayor en los pacientes VDZ (47,4% vs. 33,1%, p = 0,03). Las tasas de remisión clínica sin esteroides fueron similares entre los grupos (p = 0,98). La respuesta endoscópica en la semana 52 favoreció a VDZ (78,4% vs. 62,7%, p < 0,001), y VDZ tuvo una mayor persistencia del tratamiento (80,8% vs. 61,8%, p < 0,001). Los eventos adversos y las hospitalizaciones fueron más frecuentes en los pacientes que recibieron IFX (p < 0,001).

Conclusiones

Tanto VDZ como IFX son eficaces en pacientes con CU bionaífs; sin embargo, VDZ demostró resultados endoscópicos superiores, mayor persistencia del tratamiento y un mejor perfil de seguridad.

Introduction

Ulcerative colitis (UC) is a chronic, progressive, immune-mediated inflammatory disease that affects the colon's mucosa. This disease is clinically characterized by periods of remission and relapses.^1,2 Globally the incidence of UC has been increasing, especially in newly industrialized countries leading to higher healthcare costs, reduced quality of life, and increased hospitalizations, particularly in moderate-to-severe cases. Presently, the medical therapy for UC aims to reduce mucosal inflammation by inducing and maintaining both clinical and endoscopic remission.^3,4

Over the past two decades, UC management has evolved with early diagnosis, objective therapeutic targets, improved disease monitoring, and the expanded use of biologic agents, following the regulatory approval of the first biological drugs, such as infliximab (IFX).^5,6 Anti-TNF agents have dramatically altered the management of moderate-to-severe UC – not only promoting mucosal healing in a substantial number of patients but also significantly decreasing rates of colectomy and hospitalization.⁷ In recent years, newly approved molecules and mechanisms of action have emerged, including anti-integrins, interleukin inhibitors, and Janus kinase (JAK) inhibitors.^4,5 Current guidelines recommend early biologic therapy for patients with severe disease or failure of conventional treatments.⁵

Prior studies and guidelines have shown that in cases of UC patients who are refractory or intolerant to conventional medical therapy, or steroid-dependent, IFX and vedolizumab (VDZ) are two biological therapy options. These are generally reserved for moderate-to-severe active disease in UC biological-naïve patients.^8,9 Besides current recommendations, previous network meta-analyses suggest that both IFX and VDZ rank highest among biological therapeutic options for inducing clinical remission in biological therapy-naïve UC patients.¹⁰

Despite the recommendation to use IFX or VDZ as first-line biologic agents for biologic-naïve patients with moderate-to-severe UC, there are no prospective, head-to-head controlled studies comparing the efficacy and safety of IFX and VDZ in treating moderate-to-severe UC. Furthermore, only a handful of real-world retrospective studies with conflicting results have been published so far.^11–14 However, there might be genetically determined physiological and pharmacodynamic differences between Latin American and North American or European populations. These differences could limit the generalization of the findings from the previous studies, as the data primarily originate from North American or European investigations. So far, no studies from Latin America that compare the efficacy and safety of VDZ or IFX in patients with moderately to severely active UC have been published. Thus, the goal of this multicenter retrospective real-world study is to compare the efficacy and safety of IFX and VDZ in biologic-naïve patients with moderate-to-severe UC.

Material and methods

Study design and population

This multicenter retrospective real-world study included adult outpatients (≥18 years) with moderate-to-severe UC, defined by a total Mayo score of 6–12 and an endoscopic sub-score ≥2, assessed within 3 months prior to starting biological therapy. Patients were unexposed to advanced therapies (biologics or small molecules) and had received at least the induction doses of IFX or VDZ.

Participants were enrolled from 13 tertiary IBD centers across Brazil. IFX enrollment occurred from August 2008 to September 2023, and VDZ enrollment from June 2017 to December 2023. Patients received standard intravenous induction protocols: IFX (5 mg/kg) or VDZ (300 mg) at weeks 0, 2, and 6, followed by maintenance every 8 weeks.

Exclusion criteria included Crohn's disease, undetermined colitis, remission or mild disease at baseline (Mayo score 0–5 or endoscopic sub-score 0–1), acute severe colitis, prior exposure to advanced therapies, colectomy, colorectal cancer, or incomplete data.

Data collection

Demographic, clinical, laboratory, and endoscopic data were collected through a review of electronic medical records from all participating IBD centers. Patients were selected during routine clinical follow-up, utilizing data from medical records of all participating centers. The choice of biological agent was determined by the physician, taking into account the efficacy and safety of the available agents. Baseline characteristics collected included sex, race, age at inclusion and diagnosis, disease duration, location (Montreal classification¹⁵), Mayo scores, smoking status, steroid dependency, biomarkers (CRP, fecal calprotectin), hemoglobin, serum albumin, extra-intestinal manifestations (EIM), and prior/concomitant therapies (e.g., azathioprine, steroids, 5-ASA derivatives).

For patients using steroids at baseline, the decision to discontinue this therapy at each participating center was left to the physician's discretion.

Adverse events (AEs), serious AEs (SAEs), UC-related hospitalizations, relapses, colectomies, primary non-response (PNR), secondary loss of response, reasons for drug discontinuation, dose optimization, and treatment persistence were assessed. SAE were those that necessitated hospitalization or resulted in drug discontinuation. Dose optimization was defined as dose escalation to 10 mg/kg or interval reduction to 4 weeks for IFX, or interval reduction to 4 weeks for VDZ, as determined by the treating physician.

Outcomes and definitions

Primary endpoints included clinical remission at weeks 12, 26, and 52, defined by a PMS ≤2, and steroid-free clinical remission, achieved without systemic steroids during the observation period. Endoscopic remission (EMS = 0) at week 52 in patients who had endoscopic reevaluation at week 52 (±4 weeks) and the final follow-up, defined as the last visit or therapy discontinuation, were also primary endpoints.

Secondary endpoints included clinical response (PMS reduction ≥2 points from baseline) at weeks 12, 26, 52, and final follow-up, endoscopic response (EMS = 0 or 1) at week 52, deep remission (clinical remission plus mucosal healing) at week 52, therapy persistence, biochemical remission (CRP ≤5 mg/dL and/or FC ≤150 μg/g) at week 52, and the occurrence of any AE. Therapy optimization was defined as dose escalation or interval reduction. PNR was determined by unchanged or worsened PMS after 16 weeks, leading to therapy discontinuation, while secondary loss of response was active disease recurrence after initial improvement, with inflammation confirmed by endoscopy, CRP >0.5 mg/dL, or FC >250 μg/g.

As this is a real-life experience, for the calculations of the various primary and secondary endpoints, data was permitted to be captured at various centers up to ±4 weeks apart from the established key dates (weeks 12, 26, and 52).

Safety outcomes

Safety outcomes included serious and non-serious AEs and infections. SAEs were defined as events requiring hospitalization, therapy discontinuation, significant disability, or resulting in death. Serious infections necessitated intravenous antibiotics, therapy discontinuation, or hospitalization. AE data were collected during the follow-up period while the patients were receiving treatment with IFX or VDZ.

Statistical analysis

Fisher's exact test and Student's t-test were used to compare categorical and continuous variables, respectively. Non-responder imputation (NRI) classified patients who discontinued early as non-responders for clinical, biochemical, and endoscopic endpoints.

To address selection bias, propensity score (PS) analysis simulated randomization. Covariates displaying baseline imbalances were adjusted using inverse probability weighting (IPTW). Logistic models calculated weights (1/PS for VDZ and [1/(1 − PS)] for IFX), ensuring balanced covariates (SMD <0.1). Weighted comparisons estimated relative risks using Poisson regression with robust variance, presented as odds ratios (OR) with 95% confidence intervals (CI).

Treatment persistence was analyzed via Kaplan–Meier curves and log-rank tests. Statistical significance was set at p <0.05. Analyses were conducted using SAS 9.4, with graphs generated in RStudio.

Result

Baseline population characteristics

Out of 304 initially identified patients (160 IFX, 144 VDZ), seven were excluded due to a diagnosis change (n = 1), prior colectomy (n = 3), or incomplete data (n = 4), resulting in 297 patients (156 IFX, 141 VDZ). Table 1 summarizes baseline characteristics.

Table 1.

Baseline clinical and demographic characteristics of biologic-naïve patients with ulcerative colitis treated with infliximab and vedolizumab before and after 1/IPTW* adjustment.

Characteristics	Original cohort (pre-1/IPTW)			Post-1/IPTW
	Infliximab	Vedolizumab	p	Infliximab	Vedolizumab	p
Gender, female, n (%)	96 (61.5)	81 (57.4)	0.01	91 (58.7)	77 (54.9)	0.51
Mean BMI, kg/m^2^†(SD)**	25.23 (4.74)	25.65 (5.28)	0.47	25.23 (4.74)	25.65 (5.28)	0.47
Mean age, years (IQR)	42.8 (18–78)	44.3 (18–80)	0.39	42.8 (18–78)	44.3 (18–80)	0.39
Mean disease duration, months (IQR)	111.7 (12–552)	90.4 (2–392)	0.02	103.8 (12–552)	106.5 (2–392)	0.81
Current smoker, n (%)	5 (3.4)	3 (2.1)	0.56	5 (3.4)	3 (2.1)	0.56
Extra-intestinal manifestations, n (%)	38 (24.3)	18 (12.8)	0.01	31 (19.6)	30 (21.5)	0.68
Montreal classification, n (%)
E1	8 (5.2)	3 (3.6)	0.56	8 (5.2)	3 (3.6)	0.56
E2	52 (33.5)	52 (37.7)	0.52	52 (33.5)	52 (37.7)	0.52
E3	95 (61.3)	81 (58.7)	0.54	95 (61.3)	81 (58.7)	0.54
Increased biomarkers, n (%)	119 (76.3.0)	110 (78.0)	0.72	119 (76.3.0)	110 (78.0)	0.72
Mean C-reactive protein, mg/dL (IQR)	2.6 (0.04–25)	2.5 (0.01–58)	0.86	2.6 (0.04–25)	2.5 (0.01–58)	0.86
Mean fecal calprotectin, μg/g (IQR)	1,368.0 (40–6,000)	1,609.1 (83–6,200)	0.16	1,368.0 (40–6,000)	1,609.1 (83–6,200)	0.16
Mean albumin serum levels, g/dL (IQR)	3.9 (1.7–5.6)	4.2 (2.9–5.3)	0.06	3.9 (1.7–5.6)	4.2 (2.9–5.3)	0.06
Mean hemoglobin levels, g/dL (IQR)	12.4 (7.1–16.1)	12.8 (8.1–16.2)	0.07	12.4 (7.1–16.1)	12.8 (8.1–16.2)	0.07
Total Mayo score, mean (IQR)	8.9 (3–12)	8.7 (2–12)	0.65	8.9 (3–12)	8.7 (2–12)	0.65
Partial Mayo score, mean (IQR)	6.3 (1–9)	6.1 (1–9)	0.48	6.3 (1–9)	6.1 (1–9)	0.48
Endoscopic sub-score 2 (moderate disease), n (%)	53 (35.3.0)	62 (43.9)	0.07	53 (35.3.0)	62 (43.9)	0.07
Endoscopic sub-score 3 (severe disease), n (%)	92 (61.3)	76 (53.9)	0.37	92 (61.3)	76 (53.9)	0.37
Concomitant use of corticosteroids, n (%)	122 (78.2)	107 (75.8)	0.63	122 (78.2)	107 (75.8)	0.63
Previous use of azathioprine, n (%)	63 (40.7)	65 (46.1)	0.32	63 (40.7)	65 (46.1)	0.32
Concomitant use of azathioprine, n (%)	65 (41.9)	14 (9.9)	<0.001	46 (26.9)	40 (28.2)	0.81
Previous use of oral mesalazine, n (%)	86 (55.5)	91 (64.5)	0.09	86 (55.5)	91 (64.5)	0.09
Concomitant use of oral mesalazine, n (%)	50 (32.3)	46 (32.6)	0.9	50 (32.3)	46 (32.6)	0.91
Concomitant use of topical mesalazine	24 (15.5)	20 (14.2)	0.77	24 (15.5)	20 (14.2)	0.77
Previous use of topical mesalazine	61 (39.4)	54 (38.3)	0.88	61 (39.4)	54 (38.3)	0.92

Continuous variables are represented as median and interquartile range (IQR); categorical variables are represented as n (%); χ² or Fisher's exact test for categorical variables.

*

1/IPTW inverse of the weighting by the propensity score.

**

Standard deviation.

†

BMI: body mass index.

The cohort included 59.6% females, with an average age of 43.5 years and a mean disease duration of 101.6 months. Extensive colitis was present in 59.3% of cases, 18.9% had EIM, 77.1% were on steroids, and 26.6% were using immunomodulators. The average Mayo score was 8.4, with 35% having an EMS of 2 and 59.3% an EMS of 3.

After PS adjustment, baseline differences (e.g., gender, EIM, disease duration, AZA use) were balanced. IFX and VDZ groups showed similar characteristics: 61.5% vs. 57.4% female, median age 42.8 vs. 44.3 years, and extensive colitis in 61.3% vs. 58.7%. Mean CRP levels were 2.6 vs. 2.5 mg/dL, and elevated FC levels were seen in 42.3% vs. 59.6%.

Primary endpoints

Clinical and endoscopic outcomes: The clinical remission rates at weeks 12 [53.03% for VDZ versus 50.7% for IFX (p = 0.89; OR: 0.98; 95% CI, 0.75–1.28)], 26 [75.6% for VDZ versus 72.8% for IFX (p = 0.52; OR: 0.95; 95% CI, 0.80–1.12)], and 52 [82.3% for VDZ versus 77.6% for IFX (p = 0.11; OR: 0.90; 95% CI, 0.79–1.03)] were comparable between the two groups. Clinical remission by the end of the follow-up was achieved by 82.2% and 78.1% of the patients receiving VDZ and IFX respectively (p = 0.16; OR: 0.92; 95% CI, 0.82–1.03).

Steroid-free clinical remission during the same durations was observed in 58.1% vs. 63.0% (p = 0.99; OR: 1.0; 95% CI, 0.65–1.54), 62.5% vs. 63.6% (p = 0.99; OR: 0.99; 95% CI, 0.83–1.10), and 64.5% vs. 64.7% (p = 0.98; OR: 0.98; 95% CI, 0.87–1.10) of patients using VDZ and IFX at weeks 12, 26, and 52, respectively.

Conversely, patients who received VDZ exhibited a significantly higher endoscopic remission rate at week 52 compared to those treated with IFX (47.4% vs. 33.1%, respectively; p = 0.03; OR: 0.69; 95% CI, 0.49–0.97) (Fig. 1).

Figure 1.

Clinical, steroid-free and endoscopic remission rates at 52 weeks comparing vedolizumab and infliximab.

Secondary endpoints

Clinical, biochemical, and endoscopic outcomes: Among the secondary outcomes, clinical response at weeks 12 [83.4% for IFX vs. 75.4% for VDZ (p = 0.50; OR: 1.05; 95% CI, 0.92–1.19)], 26 [91.7% for IFX vs. 90.0% for VDZ (p = 0.67; OR: 0.98; 95% CI, 0.91–1.06)] and 52 [90.7% for IFX vs. 88.3% for VDZ (p = 0.80; OR: 1.01; 95% CI, 0.93–1.10)] were comparable between the two groups.

Biochemical remission at 1 year was noted in 41.7% of patients on VDZ and 34.4% of patients on IFX (p = 0.93; OR: 1.02; 95% CI, 0.61–1.72). Conversely, an endoscopic response at week 52 was attained by 78.4% of VDZ patients and 62.7% of IFX patients (p < 0.001; OR: 0.77; 95% CI, 0.66–0.90) (Fig. 2). Similarly, patients treated with VDZ exhibited a significantly higher deep remission rate at week 52 compared to who received IFX (48.6% vs. 32.8%, respectively; p = 0.04; OR: 0.68; 95% CI, 0.48–0.98).

Figure 2.

Clinical response, biochemical remission and endoscopic response rates at 1 year comparing vedolizumab and infliximab.

Treatment persistence, biological therapy optimization, UC relapse, and UC-related hospitalization

At 52 weeks, the requirement for optimization was less for VDZ than for IFX. Twenty-seven VDZ patients (19.2%) needed a dose escalation to 300 mg every 4 weeks, while 50 (32.1%) IFX patients required a dose escalation of IFX therapy to 5 mg/kg every 4 weeks or 10 mg/kg every 8 weeks (p = 0.03; OR: 1.71; 95% CI, 1.06–2.78). The proportion of UC patients who regained a clinical response after increasing the VDZ dose to 300 mg every 4 weeks (Q4W) was 51.5% (n = 14). Among the patients who began with VDZ, twelve (8.6%) were identified as PNR. In contrast, out of those who started IFX, twenty-nine (18.8%) were categorized as PNR (p = 0.06; OR: 2.19; 95% CI, 0.96–4.98). The rate of secondary loss of response was similar in the two groups (15.0% for VDZ and 20.3% for IFX, p = 0.24; OR: 1.39; 95% CI, 0.80–2.43). Meanwhile, twenty-seven (19.2%) VDZ patients and fifty-five (38.1%) IFX patients discontinued their respective medications, primarily due to ineffectiveness.

The Kaplan–Meier curve analysis depicted in Fig. 3 shows the cumulative probability of continued VDZ or IFX treatment. The probability of maintaining the treatment at week 52 was greater with VDZ than with IFX (VDZ 80.8% vs. IFX 61.8%) [p < 0.001; OR: 0.74; 95% CI, 0.62–0.87].

Figure 3.

Kaplan–Meier curve for treatment persistence of infliximab and vedolizumab in biologic-naïve patients with Ulcerative colitis at 52 weeks.

UC relapses, characterized by bloody diarrhea without criteria for hospitalization, but necessitating the use of oral corticosteroids, occurred more frequently in patients using IFX than VDZ during 1 year of follow-up (9.1% vs. 0.71%, p < 0.001; OR: 17.16; 95% CI: 2.25–130.8). Concerning UC-related hospitalizations that required the administration of intravenous corticosteroids, patients on IFX experienced higher rates than those on VDZ (26.0% vs. 8.51%, p < 0.001; OR: 2.56; 95% CI: 1.25–5.52). During the 1-year follow-up period, colectomy was recommended for 9 (5.84%) patients in the IFX group and one (0.71%) patient in the VDZ group.

Safety events

AEs were observed in 72 patients (46.1%) treated with IFX, and 48 patients (34.0%) treated with VDZ, of which 29 patients (18.6%) on IFX and nine patients (6.4%) on VDZ were considered to have SAEs. For those treated with IFX, the most common AEs were lower urinary tract infections (9.6%), respiratory infections (6.4%), pseudomembranous colitis (5.2%), and infusion reactions (5.2%). For VDZ-treated patients, the most common AEs were arthralgia (12.1%), lower urinary tract infections (5.7%), and pseudomembranous colitis (2.8%). Overall, there were no statistically significant differences in the risk of infections (p = 0.11; OR: 1.51; 95% CI, 0.91–2.59), or other AEs (p = 0.17; OR: 1.28; 95% CI, 0.90–1.83), between VDZ and IFX-treated patients. VDZ-treated patients, however, were associated with a significantly lower risk of serious infection (p = 0.0004; OR, 3.43; 95% CI, 1.48–7.95) and a non-significantly lower risk for other SAEs (p = 0.65; OR: 1.50; 95% CI, 0.25–8.98) than IFX-treated patients. The most common SAEs were pseudomembranous colitis (5.2%), tuberculosis (1.9%), CMV infection (1.9%), and pyelonephritis (1.9%) for IFX-treated patients, and pseudomembranous colitis (2.8%), tuberculosis (0.7%) and bacterial cholangitis (0.7%) for those treated with VDZ. There were no fatalities reported in this study. These results can be found summarized in Table 2.

Table 2.

Safety events with infliximab and vedolizumab treatment during follow-up.

Adverse events	Infliximab	Vedolizumab
Any adverse events, n (%)	72 (46.1)	48 (34.0)
Adverse events, n (%)	49 (31.4)	41 (29.1)
Infections, n (%)	32 (20.5)	19 (15.8)
Lower urinary tract infection	15 (9.6)	8 (5.7)
Nasopharyngitis	5 (3.2)	3 (2.1)
Community-acquired pneumonia	5 (1.9)	2 (1.4)
Cellulitis/erysipelas	3 (1.9)	–
Mild COVID-19	2 (1.2)	2 (1.4)
Hansen's disease	1 (0.6)	–
Dengue	–	2 (1.4)
Chikungunya	1 (0.6)	–
Vaginal candidiasis	–	1 (0.7)
Recurrent sinusitis	–	1 (0.7)
Other AEs, n (%)	17 (10.9)	22 (15.6)
Arthralgia	6 (3.8)	17 (12.1)
Infusion reaction	8 (5.1)	–
Paradoxical psoriasis	3 (1.9)	–
Alopecia	–	3 (2.1)
Headache	–	2 (1.4)
Serious adverse events, n (%)	28 (18.5)	9 (6.4)
Clostridioides difficile infection	8 (5.1)	4 (2.8)
Tuberculosis	3 (1.9)	1 (0.7)
Pyelonephritis	3 (1.9)	–
Cytomegalovirus infection	3 (1.9)	–
Bacterial cholangitis	2 (1.2)	1 (0.7)
Shingles	1 (0.6)	–
Nosocomial pneumonia	1 (0.6)	–
Meningoencephalitis	–	1 (0.7)
Other SAEs, n (%)	8 (4.9)	2 (1.4)
Infusion reaction*	4 (2.5)	–
Paradoxical psoriasis**	2 (1.2)	–
Arthralgia^&	–	2 (1.4)
Cervical neoplasia^#	1 (0.6)	–
Prostate neoplasm^#	1 (0.6)	–

*

Severe infusion reaction with dyspnea, skin rash in 3 patients and severe anaphylactic reaction in 1 patient.

**

Psoriasis relapse after reintroduction of IFX, requiring change of biological therapy mechanism to ustekinumab in 2 patients.

&

Severe arthralgia, requiring suspension of VDZ treatment in 2 patients.

#

Patients using infliximab in combination with azathioprine.

Discussion

In this retrospective, multicentric Brazilian cohort study, we compared, for the first time, the effectiveness and safety of IFX versus VDZ in biologically naïve patients with moderate-severe UC in a Latin American population. We found that both IFX and VDZ demonstrated robust and comparable effectiveness in inducing clinical remission, clinical response, steroid-free remission, and biochemical remission at the 52-week mark. However, VDZ was superior in inducing endoscopic remission and response and exhibited greater treatment persistence. Additionally, VDZ presented a better safety profile and a reduced necessity for dose escalation.

Direct comparisons are crucial, given the numerous treatment options currently available for IBD. So far, no prospective head-to-head trials of anti-TNF vs. anti-integrin agents in UC have been published, except for the VARSITY study. This study compared VDZ and adalimumab (ADA) in a randomized, double-blind trial involving subjects with moderate-to-severe UC. In this study, VDZ was found superior to ADA in achieving clinical remission (31.3% vs. 22.5%) and endoscopic response (39.7% vs. 27.7%) at week 52. Additionally, rates of AEs were lower with VDZ than ADA.¹⁶ However, it is worth noting a limitation of the VARSITY study – it compared ADA with VDZ, and the results for ADA probably cannot be generalized to IFX. Also, there was no dose adjustment in the trial for either drug (ADA or VDZ), which does not reflect how clinicians use these drugs in actual clinical practice.¹⁷ Furthermore, without head-to-head comparisons, retrospectively evaluating real-world data may serve as a feasible alternative for comparing a drug's effectiveness and safety.

To date, few studies have compared two intravenous biologics with different mechanisms of action (anti-TNF and anti-integrins) in biologic-naïve patients with UC.^12–14 In our study, the rate of clinical remission after 1 year of treatment for VDZ was slightly higher than for IFX (82.3% vs. 77.4%), but this difference was not significant. A clinical response was observed in 88.3% of VDZ patients and 90.7% of IFX patients. These findings are consistent with previously published studies. Indeed, a real-world comparison of the effectiveness of VDZ vs. IFX induction therapy among patients with moderately to severely active UC at a tertiary IBD center in the US revealed a generally higher response rate with VDZ (78% vs. 67%).¹¹ A recent Italian real-world study evaluating biologic-naïve patients with UC found that by week 52, CR was achieved in a higher percentage of VDZ patients than IFX patients (75% for VDZ vs. 32% for IFX).¹² Bokemeyer et al. showed, using 1-year data, higher clinical remission rates for VDZ than anti-TNF (38.2% vs. 26.0%). But this difference was not significant.¹⁸ A multicenter, retrospective, observational cohort study compared the clinical remission rate in 454 patients treated with VDZ and 268 with IFX, ADA, or golimumab (GOL) in biologic-naïve and biologic-exposed UC patients. After PS adjustment in the biologic-naïve sub-cohort of 143 VDZ and 114 IFX patients, rates of clinical remission by week 52 were higher and statistically significant in patients treated with VDZ.¹⁴ This is likely due to the higher rate of drug persistency when comparing VDZ to IFX.

Steroid-free clinical remission is an important goal in the treatment of IBD. In our series, both biological therapies, for instance, IFX and VDZ, facilitated achieving this goal in a comparable manner and most patients (84.8% vs. 83.1%). An Italian real-world study assessing biologic-naïve UC patients found that steroid-free clinical remission by week 52 with VDZ was predictably higher than with IFX (94.0% vs. 66.1%).¹² Results akin to our study were shown in a large multicenter study by Lukin et al., where patients treated with VDZ were more likely to achieve steroid-free clinical remission than those treated with anti-TNF. However, statistical significance was not attained in all subgroup analyses.¹⁴ The high rate of steroid-free remission observed in our study could be justified partly by the optimization of biological therapy that was performed for a significant proportion of patients throughout our real-life study.

In our study, we found that the endoscopic remission and response rates at 1 year were 47.4% and 78.4% respectively for the subjects on VDZ, and 33.1% and 64.7% for those on IFX. We used the definitions for endoscopic remission (EMS of zero) and endoscopic response (EMS of zero or one) as recommended by the Selecting Therapeutic Targets in IBD II (STRIDE II).⁶ In line with our results, after adjusting for concomitant steroid use, concomitant immunomodulator (AZA, 6-MP, methotrexate) use, and the number of prior TNF-antagonists used, VDZ-treated patients had statistically significant higher cumulative rates of endoscopic healing at 1 year (50% vs. 42%) than those treated with anti-TNF.¹⁹ The EVOLVE study, a large multicenter retrospective study that compared the effectiveness of IFX and VDZ in 376 and 221 biologic-naïve UC patients respectively, observed a similar percentage of mucosal healing between the two treatments.²⁰

Efficacy and safety are crucial outcomes in clinical practice, particularly concerning the use of biological agents. These outcomes are typically included in the results of clinical trials. Treatment durability serves as a measure of the ongoing effectiveness of biological therapy, given the chronic nature of the diseases being dealt with. Real-world studies that compared the efficacy of VDZ and IFX in biologic-naïve UC cohort patients observed that drug persistency was higher for VDZ than for IFX at 12 and 24 months, respectively.²¹ A recent retrospective Israeli study in the real-world scenario demonstrated significantly longer drug persistency for VDZ than for IFX in UC biological-naïve patients when adjusted for disease severity.²² The EVOLVE study found that the rates of drug persistency were higher for VDZ than for IFX 20. We discovered that the persistence of VDZ was significantly longer than that of IFX at week 52, illustrating greater drug durability as an indirect estimate of safety/tolerance and treatment efficacy. Upon analyzing the causes of dropouts, PNR was higher and showed a tendency to be statistically significant in the IFX group (18.8% vs. 8.6%), while secondary loss of response was comparable between the two groups.

Dose escalation is a part of the treat-to-target approach in IBD, aiming to restore response and remission to different biological therapies. It is permitted for both VDZ (shortening of intervals) and IFX (shortening of intervals, dose escalation, or both). However, this strategy may entail a direct and indirect increase in treatment costs.²³ In our study, VDZ required less optimization than IFX, and the difference was statistically significant (19.2% vs. 36.8%). We also observed a higher frequency of UC relapse and UC-related hospitalizations in patients using IFX than those using VDZ. These results can be attributed to the effectiveness of VDZ in controlling inflammation due to its gut-selective mode of action.²⁴

In the current study, we noted a numerically higher incidence of AEs in patients treated with IFX compared to those treated with VDZ (46.1% vs. 36.0%). However, there was no statistically significant difference. Other studies have similarly demonstrated equivalence when comparing the safety of VDZ with that of anti-TNF agents.^25,26 In contrast, when we assessed the incidence of SAEs, we discovered a higher occurrence with a statistically significant difference in patients treated with IFX compared to those treated with VDZ (18.6% vs. 6.4%). These findings align with other studies comparing the real-world efficacy and safety of VDZ and anti-TNFα in treating IBD. This pattern may be attributable to VDZ's gut-selective mode of action as opposed to the systemic action and immunosuppression of anti-TNFα.^27–29

Inherent to real-life observational studies, our study has a few limitations. First, it is a retrospective study with a non-controlled design, which can result in patient loss, missing data, and potential selection bias. Second, the relatively short follow-up period may have contributed to some degree of variation in the results. Third, the decision to optimize treatment with IFX and VDZ was made at the physician's discretion based on clinical rationale, rather than following a standardized protocol. Moreover, a possible selection bias exists as IFX could be used for more severe patients.

Despite the above-mentioned limitations, our study presents numerous strengths. To our knowledge, it is the first real-world study conducted in Latin America that directly compares two biologics with different mechanisms of action (VDZ and IFX) in biologic-naïve patients with UC over a 52-week observation period. Since patient selection by physicians in other real-world retrospective multicenter studies often results in unbalanced groups, an adjustment using the inverse of the PS (1/PS) was utilized to correct any confounders in the patient assignment, thereby improving group comparability.

Furthermore, in this study, patient data was obtained through a structured survey form, with data capture standardized by the researchers. We also employed objective clinical, biochemical, and endoscopic criteria frequently reported in real-world observational studies to report therapy-induced significant outcomes in UC patients.

In conclusion, comparative effectiveness research is crucial, particularly head-to-head trials comparing IFX versus VDZ in a large population of biologic-naïve patients with moderate-to-severe UC. These trials are needed to confirm our findings.

In conclusion, this retrospective real-life study indicates that IFX and VDZ demonstrate similar efficacy in inducing clinical remission, clinical response, steroid-free clinical remission, and biochemical remission in biologic-naïve patients with UC at week 52. However, VDZ was superior to IFX in inducing endoscopic remission and endoscopic response. Moreover, VDZ appeared to have a better safety profile, with a lower associated risk of treatment-related SAEs and less need for dose escalation. Based on our findings, we suggest the VDZ as the first-line biologic in UC.

Ethical aspects

This study was approved by the Research Ethics Committee of the University Hospital, Ribeirão Preto Medical School, University of São Paulo, Brazil (HCFMRP-USP protocol # CAAE: 76743023.6.0000.5440; Ethics Committee Number: 6.607.596/2024) and informed consent was obtained from all patients included in the study.

Funding

This research was supported by the Fundação de Apoio ao Ensino, Pesquisa e Assistência do Hospital das Clínicas da Faculdade de Medicina de Ribeirão Preto (FAEPA).

Conflicts of interest

Ferreira SC has received fees for serving as a speaker and/or an advisory board member for Janssen, Takeda, and Pfizer.

Parra RS has received fees for serving as a speaker and/or an advisory board member for Takeda, Janssen, Abbvie, and Pfizer.

Sassaki LY has received fees for serving as a speaker and/or an advisory board member for Janssen and Abbvie.

Parente JML has received fees for serving as a speaker and/or an advisory board member for Takeda, Abbvie, and Janssen.

Mello MK has received fees for serving as a speaker and/or an advisory board member for Takeda, Janssen, Abbvie, Sandoz and Ferring.

Chebli LA has received fees for serving as a speaker and/or an advisory board member for Janssen, Takeda and Abbvie.

Luporini RL has received fees for as a speaker and/or an advisory board member for Takeda, Janssen, Abbvie.

Alves Junior AJT has received fees for serving as a speaker and/or an advisory board member for Janssen, Takeda, UCB, and Abbvie.

Nóbrega FJ has received fees for serving as a speaker and/or an advisory board member for Janssen, Takeda and Abbvie.

da Silva BC has received fees for serving as a speaker and/or an advisory board member for Janssen, Takeda and Abbvie.

Miranda EF has received fees for has received fees for serving as a speaker and/or an advisory board member for Janssen, Takeda and Celltrion.

Nicollelli GM EF has received fees for serving as a speaker and/or an advisory board member for Janssen and Takeda.

Gasparini RG has received fees for has received fees for serving as a speaker and/or an advisory board member for Janssen and Takeda.

Magro, DO has received fees for serving as a speaker and/or an advisory board member for Janssen and Takeda.

Imbrizi MR has received fees for serving as a speaker and/or an advisory board member for Janssen and Takeda.

Kotze PG has received fees for serving as a speaker and/or an advisory board member for Janssen, Takeda, Abbvie, and Pfizer.

Chebli JMF has received fees for serving as a speaker and/or an advisory board member for Takeda, Janssen, AbbVie, Abbott, and Sandoz.

Vasconcelos JRO, Dutra RM, da Silva KC, Quaresama AB, Nagasako CK, Féres O and Troncon LEA report no conflicts of interest.