Long-term comparative effectiveness of first-line infliximab vs corticosteroids in ulcerative colitis: Remission and colectomy risk

Abstract

BACKGROUND

Intravenous corticosteroids (IVCS) are the conventional first-line therapy for moderate-to-severe ulcerative colitis (UC), but are often limited by nonresponse, steroid dependence, and relapse. Infliximab (IFX) is an alternative first-line option; however, long-term comparative data between IFX and IVCS remain limited.

AIM

To compare the long-term durability of disease remission and surgical outcomes associated with first-line IFX vs IVCS in patients with moderate-to-severe UC.

METHODS

In this multicenter prospective cohort study, adult patients (≥ 18 years) with moderate-to-severe UC were enrolled between September 2019 and September 2025 from three tertiary care hospitals in Jeddah, Saudi Arabia. Disease severity was assessed using the Mayo score and Geboes histologic grading. Patients received either first-line IFX (5 mg/kg at weeks 0, 2, and 6, followed by 8-weekly maintenance) or first-line IVCS (methylprednisolone 40-60 mg/day or hydrocortisone 300-400 mg/day). Participants were followed for at least 24 months, with clinical, biochemical, endoscopic, and histologic assessments performed at predefined intervals up to 144 weeks. Colectomy was considered for patients with nonresponse or loss of remission despite medical therapy.

RESULTS

Of 152 enrolled patients, 139 completed follow-ups (68 IFX, 71 IVCS). At weeks 12, clinical response rates were significantly higher with IFX than IVCS (77.9% vs 56.3%, P = 0.007), and this benefit was sustained throughout follow-up. By weeks 144, IFX-treated patients demonstrated significantly higher rates of clinical remission (36.8% vs 15.5%, P = 0.006), endoscopic remission (25.0% vs 9.9%, P = 0.006), and histologic remission (20.6% vs 7.0%, P = 0.026). Adverse events were comparable overall. Cumulative incidence of colectomy was significantly lower in the IFX group than in the IVCS group (5.9% vs 18.3%, P = 0.037). In adjusted analyses, IFX treatment was independently associated with a lower hazard of colectomy compared with IVCS (hazard ratio = 0.53, 95%CI: 0.20-0.88; P = 0.043).

CONCLUSION

In this real-world tertiary-care cohort, IFX was associated with superior and sustained remission and with a significantly reduced long-term colectomy risk compared with IVCS in moderate-to-severe UC, supporting consideration of early biologic therapy in selected high-risk patients.

Key Words: Ulcerative colitis; Infliximab; Corticosteroids; Biologic therapy; Long-term outcomes; Colectomy

Core Tip: This prospective multicenter study compared first-line infliximab with intravenous corticosteroids in patients with moderate-to-severe ulcerative colitis. Early use of infliximab was associated with higher rates of sustained clinical, endoscopic, and histologic remission, along with a significantly reduced risk of colectomy over long-term follow-up. In contrast, intravenous corticosteroids was linked to lower remission durability and higher rates of severe adverse events. These findings support a shift toward early biologic therapy in selected high-risk patients to improve long-term outcomes and reduce steroid-related complications.

INTRODUCTION

Ulcerative colitis (UC) is a chronic inflammatory disease of the colon characterized by recurrent periods of remission and relapse that often require long-term medical management. Although the exact cause of UC remains unclear, current evidence suggests that disease development results from complex interactions among immune dysregulation, genetic predisposition, impaired epithelial barrier function, and environmental influences. The disease usually begins in the rectum and progresses proximally in a continuous pattern, with considerable variation in disease extent at initial presentation. While many patients present with inflammation confined to the rectosigmoid region, others develop left-sided or extensive colitis, which is commonly associated with greater disease severity and less favorable long-term outcomes[1-3].

Patients with UC commonly present with bloody diarrhea, rectal bleeding, fecal urgency, tenesmus, abdominal discomfort, and weight loss. The severity of clinical manifestations generally reflects both the extent and intensity of colonic inflammation. Beyond intestinal disease, UC may also be associated with extraintestinal manifestations involving the musculoskeletal system, skin, eyes, and hepatobiliary tract, all of which may substantially affect patient morbidity and quality of life[4,5].

Diagnosis is based on colonoscopic assessment together with histopathological confirmation. Endoscopic findings in active UC typically include diffuse erythema, mucosal edema, friability, vascular pattern loss, and superficial ulceration. Histopathological examination commonly demonstrates mucosal inflammation with lamina propria edema, vascular congestion, neutrophilic crypt infiltration, cryptitis, and crypt abscess formation. In a longstanding disease, architectural distortion and fibrosis may become evident. Disease severity is frequently evaluated using established clinical and endoscopic scoring systems, including the Mayo Disease Activity Index, in combination with standardized histological assessment tools[6-8].

Tumor necrosis factor-alpha (TNF-α) has a central role in UC pathogenesis through its contribution to inflammatory cell recruitment, endothelial activation, and impairment of epithelial barrier integrity. Increased mucosal TNF-α expression in active disease has therefore provided a strong therapeutic rationale for anti-TNF-targeted treatment strategies in moderate-to-severe UC[9,10].

Infliximab (IFX), a chimeric monoclonal antibody targeting TNF-α, has shown substantial efficacy in inducing and maintaining remission, achieving mucosal healing, and decreasing rates of hospitalization and colectomy, particularly among patients with severe or treatment-refractory disease[11-13].

Intravenous corticosteroids (IVCS) have long been considered standard first-line therapy for hospitalized patients with moderate-to-severe UC. Nevertheless, a considerable proportion of patients either fail to respond adequately to corticosteroids or become steroid-dependent, ultimately requiring biologic therapy escalation or surgical management. Consequently, long-term clinical outcomes with corticosteroid-based treatment remain unsatisfactory for many patients[14,15].

Emerging evidence suggests that early initiation of biologic therapy, including IFX, may provide more durable disease control and improved long-term outcomes compared with conventional step-up strategies relying initially on corticosteroids. Nevertheless, data directly comparing the long-term effectiveness of IFX vs IVCS when used as first-line monotherapy in moderate-to-severe UC remain limited[16,17].

In this context, the present prospective study aims to compare the long-term durability of disease remission and surgical outcomes associated with first-line IFX vs IVCS in a real-world Saudi cohort with moderate-to-severe UC. The primary objectives are to assess sustained remission control and colectomy risk over extended follow-up, thereby providing evidence to guide optimal early therapeutic decision-making and improve long-term patient outcomes.

MATERIALS AND METHODS

Study design and setting

This multicenter cohort study was conducted among adult patients with moderate-to-severe UC admitted to the Departments of Gastroenterology at Ibn Sina National College Hospitals, Al Jedaani Al Safa Hospital, and the New Jedaani Hospital-Ghulail, tertiary referral centers in Jeddah, Saudi Arabia, between September 2019 and September 2025. Patients were allocated into two treatment groups according to first-line therapy: (1) Group 1 received IFX; and (2) Group 2 received IVCS.

Eligibility criteria

Inclusion criteria: Adult patients (≥ 18 years) with moderate-to-severe UC, diagnosed according to the Mayo Score (Supplementary Table 1) and confirmed by histological evaluation using the Geboes grading system, who had not previously received any systemic or biologic UC-targeted therapies[18]; (2) Patients with at least a follow-up period of 24 months; (3) Written informed consent provided; and (4) Patients were assigned to the first-line IFX group if they were identified as high-risk for poor outcomes or steroid refractoriness. The specific criteria for high-risk categorization included: (1) Clinical/biochemical markers: Severe clinical symptoms (Partial Mayo Score > 6) combined with a C-reactive protein (CRP) > 30 mg/L; (2) Endoscopic severity: Presence of deep, large mucosal ulcerations (Mayo Endoscopic Subscore of 3); and (3) Disease extent: Extensive pancolitis. The final treatment decision was made through a shared decision-making process between the clinician and the patient, prioritizing IFX in those who met the aforementioned high-risk criteria to achieve rapid mucosal healing and avoid the complications of prolonged steroid exposure.

Exclusion criteria: (1) Incomplete or missing clinical data; (2) Pregnant or lactating women and pediatric patients; (3) History of previous colonic surgery; (4) Severe comorbidities limiting long-term follow-up, including malignancy, ischemic colitis, or infectious colitis; (5) Absolute contraindications to IVCS or IFX therapy; and (6) Non-compliant or uncooperative patients.

Sampling and recruitment

A consecutive sampling strategy was employed. All adult patients admitted with a new diagnosis of moderate-to-severe UC during the study period were systematically screened for eligibility at the participating centers and enrolled prospectively. Treatment allocation was guided by predefined institutional treatment algorithms incorporating disease severity, endoscopic findings, inflammatory burden (CRP and fecal calprotectin), serum albumin levels, risk of steroid refractoriness, and contraindications to corticosteroid therapy. Patients considered to be at increased risk of corticosteroid failure or steroid-related complications were preferentially considered for early initiation of IFX. However, treatment allocation was not determined solely by these criteria. Final therapeutic decisions were made through a structured shared decision-making process between the treating physician and the patient after standardized counseling regarding the expected benefits and potential risks of each treatment strategy. Consequently, treatment assignment reflected routine real-world clinical practice and remained independent of study participation. Because treatment assignment was not randomized, baseline demographic and clinical characteristics were carefully evaluated to assess comparability between study groups. Potential confounding factors were subsequently addressed using multivariable regression modeling. All included patients completed a minimum follow-up duration of 24 months.

Sample size

An a priori sample size calculation was performed using G*Power software (version 3.1.9.7) based on the χ² test of independence. A small-to-moderate effect size (Cohen’s w = 0.25) was selected to represent clinically meaningful differences in the primary outcomes between treatment groups. Assuming a two-sided significance level (α) of 0.05 and a statistical power of 85%, the estimated minimum required sample size was 144 patients (72 patients per group). To compensate for potential attrition, incomplete records, or loss to follow-up, the target sample size was increased by 10%, resulting in a planned enrollment of approximately 158 participants.

During the study period, 152 eligible patients were enrolled, with 76 patients assigned to each treatment group. Over the course of follow-up, some participants were lost to follow-up, withdrew consent, or had incomplete longitudinal data. Accordingly, the final analysis included 139 patients, comprising 68 patients in group 1 and 71 patients in group 2 (Supplementary Figure 1). Attrition occurred progressively during follow-up, leading to smaller sample sizes at later assessment time points.

Methods

Eligible patients underwent a standardized and comprehensive evaluation protocol that included the following components.

Clinical evaluation: Detailed medical histories were obtained, including age at disease onset, sex, smoking status, presenting manifestations, previous opportunistic infections (e.g., tuberculosis, human immunodeficiency virus, hepatitis B virus), previous medications, and family history of inflammatory bowel disease.

Clinical examination: A comprehensive systemic and gastrointestinal examination was performed to assess disease activity and detect extraintestinal manifestations.

Laboratory investigations: Baseline laboratory assessment included complete blood count, erythrocyte sedimentation rate, CRP, and serum albumin levels. Fecal calprotectin was measured as a non-invasive biomarker of intestinal inflammation. Stool testing for Clostridioides difficile (C. difficile) was performed in clinically suspected cases. Therapeutic drug monitoring and assessment of anti-drug antibodies were performed in selected patients exhibiting suboptimal response to IFX and was not routinely used to guide standardized dose intensification across all patients[19].

Endoscopic and histological assessment: A baseline colonoscopy with targeted biopsies was conducted to determine disease extent according to the Montreal classification and to assess disease activity and associated complications (Figures 1 and 2). Endoscopic disease severity was evaluated using the Mayo UC Activity Index (Supplementary Table 1), while histopathological activity was assessed using Geboes scoring system (histologic remission defined as a Geboes score < 2.0 with absent neutrophils in the epithelium). Scheduled follow-up colonoscopies were performed at 3 months and 12 months and annually thereafter to evaluate mucosal healing, histologic remission, and to identify disease-related complications, including cytomegalovirus (CMV) colitis when clinically indicated[18,20]. Histopathological assessments were performed by senior gastrointestinal pathologists at each center. To minimize interobserver variability, a standardized interpretation manual based on the Geboes histologic grading system was utilized across all sites. Pathologists at each participating center were blinded to treatment allocation and clinical outcomes.

Figure 1 Endoscopic images of severe ulcerative colitis. A-C: They show severe diffuse inflammation, friable mucosa, spontaneous bleeding and ulcerations, loss of the normal vascular pattern with granular mucosal pattern, pseudopolyps; D-F: Deep ulceration with mucosal bridge and necrosis.

Figure 2 Endoscopic images of complicated ulcerative colitis. A: Colorectal adenocarcinoma; B and C: High-grade dysplasia.

Imaging: Transabdominal ultrasonography and cross-sectional imaging of the abdomen and pelvis, including magnetic resonance imaging or computed tomography enterography (Figure 3), were utilized as part of the diagnostic evaluation and longitudinal monitoring to assess disease activity, progression, and to detect disease-related complications[21].

Figure 3 Contrast-enhanced computed tomography abdomen and pelvis in case of active ulcerative colitis. A: Axial computed tomography (CT) image showed sigmoid colonic segmental wall thickening with mucosal enhancement and submucosal edema; B: Axial CT image showed descending colonic segmental wall thickening with mucosal enhancement and submucosal edema; C: Coronal CT image showed descending colonic segmental wall thickening with hyperemia, prominence of the pericolonic mesenteric vessels (arrow) and pericolic fat stranding is present.

Data collection and monitoring: Demographic, clinical, endoscopic, and histopathological data were systematically collected. Patients with severe UC were closely monitored throughout the follow-up period for disease-related complications, including toxic megacolon and infectious events. For patients who reached the endpoints of mortality or colectomy, laboratory parameters (CRP, fecal calprotectin, albumin, and hemoglobin) were analyzed based on the last available measurements recorded within the 14 days prior to the event, to ensure the clinical status at the time of the outcome was accurately captured.

Treatment strategies

Patients with moderate-to-severe UC were managed according to institutional treatment protocols reflecting routine clinical practice. Group 1 received IFX as first-line therapy during the index hospitalization at a dose of 5 mg/kg intravenously over 2 hours at weeks 0, 2, and 6 for induction, followed by maintenance every 8 weeks[22]. In this study, ‘first-line IFX’ was defined as the initiation of biologic therapy as the initial induction treatment during the index hospitalization, without prior exposure to systemic corticosteroids or immunomodulators for the current flare. Accordingly, all patients in the IFX group were corticosteroid-naïve at the time of treatment induction. This strategy was specifically reserved for patients considered at increased risk of steroid refractoriness or poor clinical outcomes based on predefined clinical criteria, including extensive colitis (pancolitis), high inflammatory burden (CRP > 30 mg/L), hypoalbuminemia, and severe endoscopic activity (Mayo endoscopic subscore of 3 with deep ulcerations), where rapid disease control was prioritized to reduce the risk of early treatment failure. Group 2 received IVCS (methylprednisolone 40-60 mg/day or hydrocortisone 300-400 mg/day). To ensure consistency across the multicenter study, corticosteroid administration followed a standardized institutional protocol. Patients in the IVCS group received either methylprednisolone (0.8-1 mg/kg/day) or hydrocortisone (300-400 mg/day in divided doses). The choice of agent was predetermined by the participating center’s standard operating procedure to minimize inter-physician variability and to achieve equivalent anti-inflammatory potency. The tapering schedule was uniform across all sites, starting after a clinical response was achieved, based on clinical response and institutional protocols typically within 3-5 days of initiation.

None of the patients in either treatment group (IFX or IVCS) received concomitant 5-aminosalicylates during the study period, as the severity of the disease at inclusion (moderate-to-severe UC) warranted the use of more intensive therapeutic interventions. Treatment was initiated shortly after diagnosis, with a median interval of approximately 1-2 weeks across both groups, reflecting real-world clinical workflows rather than immediate same-day initiation. All acute severe UC patients received broad-spectrum antibiotics. Surgical consultation was reserved for nonresponders or patients with complications, including perforation, toxic megacolon, uncontrolled bleeding, medical therapy failure, high-grade dysplasia, colorectal malignancy, or structural complications[23].

Outcomes measures

Clinical outcomes included clinical response (≥ 3-point reduction in total Mayo score), clinical remission (full Mayo ≤ 2 with no subscore > 1), endoscopic remission (mucosal healing, endoscopic subscore 0-1), and histologic remission (absence of microscopic inflammation). Sustained remission was defined as maintenance ≥ 12 months. Relapse was defined as a partial Mayo score ≥ 5, a ≥ 3-point increase from baseline, or need for therapy escalation. Safety outcomes were the proportions of patients experiencing adverse events (AEs), serious AEs, serious infections, and AEs leading to treatment discontinuation. Primary endpoints were clinical response, clinical remission, and safety. Secondary endpoints included sustained clinical, endoscopic, and histologic remission, colectomy, and mortality[24,25].

Assessments of short-term and long-term outcomes

Disease activity was longitudinally assessed via partial Mayo scores and clinical evaluation at weeks 0, 2, 6, 8, 12, 24, 48, and as clinically indicated. Full Mayo scores, inflammatory markers, and histologic/endoscopic assessments were performed in weeks 0, 12, and 48. Long-term outcomes were evaluated at 14 weeks, 48 weeks, 96 weeks, and 144 weeks. Predictive factors for remission, complications, and mortality were analyzed relative to demographics and biomarkers, including serum albumin, CRP, erythrocyte sedimentation rate, and fecal calprotectin. This approach provided a comprehensive assessment of the comparative efficacy and long-term outcomes of IFX vs IVCS therapy[26,27].

Ethical considerations

This study was approved by the Institutional Review Boards of Ibn Sina National College Hospitals and the Al-Jedaani Group of Hospitals (No. IRRB-EC/ER-02-01082019). All study procedures were conducted in accordance with the Declaration of Helsinki and relevant institutional and national ethical guidelines for research involving human participants. Participant safety, confidentiality, and privacy were strictly maintained throughout the study. All participants received comprehensive information regarding study objectives, procedures, and potential risks and benefits. Written informed consent was obtained from all participants prior to enrollment. Participants were informed that their participation was entirely voluntary and that they could withdraw from the study at any time without penalty or justification. All data were handled confidentially and used exclusively for academic and research purposes.

Statistical analysis

Data were analyzed using IBM SPSS Statistics version 26.0 (IBM Corp., Armonk, NY, United States). Normality was assessed with the Shapiro-Wilk test. Continuous variables were expressed as medians with interquartile ranges and compared using the Mann-Whitney U test. Categorical variables were expressed as frequencies and percentages and compared using the χ² or Fisher’s exact tests, as appropriate. Repeated within-group comparisons were analyzed using the Friedman test. Time-to-event outcomes were evaluated using multivariable Cox proportional hazards regression. Variables included in the multivariable model were selected based on a combination of clinical relevance and univariable analysis. Clinically important covariates related to disease severity, disease extent, inflammatory burden, infectious complications, and treatment group were included a priori, while additional variables with P < 0.10 in univariable analysis were considered for inclusion. Results were reported as hazard ratios (HRs) and 95%CIs. Analyses were conducted using a complete-case approach. No patients were lost to follow-up during the study period. Longitudinal remission outcomes were calculated using the original study cohorts as denominators. Patients who underwent colectomy or died before a scheduled assessment were classified as treatment failures and were not considered to be in remission at subsequent follow-up time points. For time-to-event analyses, participants were followed until the occurrence of the event of interest or completion of follow-up, and censoring was applied accordingly. All tests were two-sided, and P values < 0.05 were considered statistically significant.

RESULTS

Baseline demographic, clinical, disease-related, and laboratory characteristics were well balanced between the two groups. No statistically significant differences were observed in age, sex distribution, disease extent, disease activity (Mayo and Geboes scores), interval from diagnosis to treatment initiation, or baseline inflammatory and laboratory parameters (P > 0.05 for all; Table 1).

Table 1 Demographic and baseline clinical and laboratory characteristics of both groups, median (interquartile range)/n (%).

Characteristics	Group 1 (IFX), n = 68	Group 2 (IVCS), n = 71	P value
Age (years)	49 (45, 55)	55 (49, 62)	0.101
Gender			0.487
Male	41 (60.3)	47 (66.2)
Female	27 (39.7)	24 (33.8)
Body weight (kg)	68.5 (65, 82)	70 (66, 90)	0.422
Smoking habits			0.950
Heavy smoker	10 (14.7)	9 (12.7)
Moderate smoker	9 (13.2)	9 (12.7)
Ex smoker	13 (19.1)	12 (16.9)
Nonsmoker	36 (52.9)	41 (57.7)
Family history of IBD			0.832
Crohn’s disease	10 (14.7)	8 (11.3)
UC	12 (17.6)	16 (22.5)
UC extent (Montreal)			0.655
E1 (proctitis)	21 (30.9)	17 (23.9)
E2 (left-sided colitis)	29 (42.6)	33 (46.5)
E3 (extensive colitis)	18 (26.5)	21 (29.6)
Mayo score	10 (9, 12)	10.5 (9, 12)	0.542
Geboes score	4 (3.8, 5)	4 (4, 4)	0.920
Interval from diagnosis to treatment initiation (weeks)	1.4 (1.0, 2.2)	1.2 (0.7, 2.0)	0.117
Disease severity			1.000
Moderate UC	36 (52.9)	38 (53.5)
Severe UC	32 (47.1)	33 (46.5)
Extraintestinal manifestations			0.850
Aphthous ulcers	3 (4.4)	2 (2.8)
Erythema nodosum	2 (2.9)	1 (1.4)
Peripheral arthritis	2 (2.9)	3 (4.2)
Laboratory findings
CRP (mg/dL)	28 (23.5, 35)	30 (25, 35)	0.058
Albumen (g/dL)	3 (2.5, 3.1)	3 (3, 3.3)	0.098
Fecal calprotectin (μg/g)	903 (800, 1100)	895 (700, 989)	0.137
Hb (g/dL)	11 (10, 12)	11 (10, 12)	0.787
Leucocytes (109/L)	10 (7, 11)	8 (7, 11)	0.829
Platelets (cm2)	250 (160, 450)	245 (170, 433)	0.847

Values present as n (%) were analyzed by χ² or Fisher’s exact tests. Values present as median (interquartile range) were analyzed by Mann-Whitney test. IFX: Infliximab; IVCS: Intravenous corticosteroids; IBD: Inflammatory bowel disease; UC: Ulcerative colitis; CRP: C-reactive protein; Hb: Hemoglobin.

At weeks 12, patients in group 1 (IFX) demonstrated significantly superior short-term efficacy compared with group 2 (IVCS), highlighting an early advantage of IFX in achieving clinical, endoscopic, and histologic disease control. A good clinical response was achieved in 77.9% of patients in group 1 (IFX) vs 56.3% in group 2 (IVCS) (P = 0.007). Clinical remission occurred in 60.3% vs 31% (P = 0.001). Similarly, endoscopic remission was observed significantly more often in group 1 (IFX) than in group 2 (IVCS) (51.5% vs 28.2%, P = 0.006). Histologic remission rates were also higher in group 1 (IFX) (41.2% vs 23.9%, P = 0.045). Additionally, colectomy rates and mortality at weeks 12 were comparable, with both outcomes occurring in two patients in each group (P = 1.000 for both; Figure 4).

Figure 4 Short-term outcomes in weeks 12 among both groups. IFX: Infliximab; IVCS: Intravenous corticosteroids.

Follow-up data were available for all enrolled patients until completion of follow-up, death, or colectomy, and no patients were lost to follow-up during the study period. In the long-term follow-up analysis, the cumulative reduction in colectomy risk and sustained superiority in disease control were observed in group 1 (IFX) compared to group 2 (IVCS) over long-term follow-up. At weeks 48, clinical remission was achieved in 57.4% of patients in group 1 (IFX) compared with 29.6% in group 2 (IVCS) (P = 0.001). Similarly, endoscopic remission was significantly more frequent in group 1 (IFX) (41.2% vs 23.9%, P = 0.045), whereas histologic remission did not differ significantly between groups at this time point (P = 0.179).

By weeks 96, group 1 (IFX) continued to demonstrate significantly higher sustained remission rates compared with group 2 (IVCS), regarding clinical remission (44.1% vs 26.8%, P = 0.035), endoscopic remission (38.2% vs 19.7%, P = 0.024), and histologic remission (30.9% vs 14.1%, P = 0.024).

At the end of follow-up (weeks 144), these differences persisted. Sustained remission remained significantly higher in group 1 (IFX) than group 2 (IVCS) (36.8% vs 15.5%, P = 0.006), as did endoscopic remission (25% vs 9.9%, P = 0.006) and histologic remission (20.6% vs 7%, P = 0.026).

Importantly, colectomy rates diverged over time. While colectomy rates were similar in weeks 48 and weeks 96, by weeks 144, colectomy occurred significantly less frequently in group 1 (IFX) than group 2 (IVCS) (5.9% vs 18.3%, P = 0.037). Mortality rates remained low and did not differ significantly between groups throughout the follow-up period (Table 2 and Figure 5).

Figure 5 Long-term outcomes at the end of the follow-up period (weeks 144) among both groups. IFX: Infliximab; IVCS: Intravenous corticosteroids.

Table 2 Long-term outcomes at different follow-up periods among both groups, n (%).

Outcomes	Weeks 48			Weeks 96			Weeks 144
	Group 1 (IFX), n = 68	Group 2 (IVCS), n = 71	P value	Group 1 (IFX), n = 68	Group 2 (IVCS), n = 71	P value	Group 1 (IFX), n = 68	Group 2 (IVCS), n = 71	P value
Clinical remission	39 (57.4)	21 (29.6)	0.001b	30 (44.1)	19 (26.8)	0.035a	25 (36.8)	11 (15.5)	0.006b
Endoscopic remission	28 (41.2)	17 (23.9)	0.045a	26 (38.2)	14 (19.7)	0.024a	17 (25.0)	7 (9.9)	0.024a
Histologic remission	22 (32.4)	15 (21.1)	0.179	21 (30.9)	10 (14.1)	0.024a	14 (20.6)	5 (7.0)	0.026a
Colectomy1	4 (5.9)	4 (5.6)	1.000	4 (5.9)	8 (11.3)	0.367	4 (5.9)	13 (18.3)	0.037a
Mortality2	2 (2.9)	2 (2.8)	1.000	2 (2.9)	4 (5.6)	0.681	2 (2.9)	6 (8.5)	0.275

^aP < 0.05.

^bP < 0.01.

¹In group 1 (infliximab) (n = 4), indications include refractoriness to medical therapy (n = 2), colonic perforation (n = 1), and acute fulminant ulcerative colitis with refractory hemorrhage (n = 1); whereas in group 2 (Intravenous corticosteroids) (n = 13), indications include refractoriness to medical therapy (n = 9), high grade dysplasia (n = 2), colorectal cancer (n = 1), and toxic megacolon (n = 1).

²In group 1 (infliximab) (n = 2), causes include colonic perforation with septicemia (n = 1) and acute fulminant ulcerative colitis with refractory hemorrhage and post operative septicemia (n = 1); whereas in group 2 (intravenous corticosteroids) (n = 6), causes include colonic perforation with septicemia (n = 4), colorectal cancer (n = 1), and toxic megacolon (n = 1).

Values were analyzed by Fisher’s exact test. IVCS: Intravenous corticosteroids; IFX: Infliximab.

Overall, AEs were comparable between groups, but their profiles differed. Dermatologic effects occurred in 11.7% of group 1 (IFX) (rash/pruritus) vs 11.3%-15.7% in group 2 (IVCS) (cushingoid features, striae, bruising, thinning skin, or acne). Gastrointestinal symptoms were slightly more frequent in group 2 (IVCS) (19.7% vs 17.6%), while elevated liver enzymes (15.4%) and general symptoms (14.7%) occurred only in group 1 (IFX). Edema (12.6%), hypertension (15.5%), and diabetes mellitus (9.9%) were reported exclusively in group 2 (IVCS). Musculoskeletal complaints were slightly more frequent in group 2 (IVCS) (12.7% vs 9.8%). Infectious complications affected both groups, but serious infections, including C. difficile and CMV infections, were more common in group 2 (IVCS) (8.5%-15.5%), upper respiratory tract infections were slightly more frequent in group 1 (IFX) (7.4%), and pneumonia was observed only in 2 patients (2.8%) of group 2 (IVCS). Grade ≥ 3 AEs occurred in both groups. In the IFX group, these occurred in 7 patients (10.3%) and were primarily serious infections, including CMV and C. difficile infection. In the IVCS group, grade ≥ 3 events occurred in 26 patients (36.6%) and included opportunistic infections as well as corticosteroid-related metabolic complications, such as new-onset hypertension and diabetes mellitus. No deaths directly attributable to the study medications were reported (Supplementary Table 2).

During follow-up, group 1 (IFX) showed sustained improvement in inflammatory markers, with reduced CRP and fecal calprotectin and increased albumin (P < 0.001), which correlated with disease activity. Patients requiring colectomy or experiencing mortality had persistently unfavorable biomarker profiles. In contrast, group 2 (IVCS) demonstrated early biochemical improvement followed by partial rebound over long-term follow-up (P < 0.001), reflecting a more heterogeneous disease course.

Higher CRP and fecal calprotectin, along with lower albumin and hemoglobin, were significantly associated with colectomy and mortality at early time points (weeks 12) (P ≤ 0.024). However, these associations were less consistent and not statistically significant at later follow-up (weeks 144) (Supplementary Tables 3 and 4).

Figure 6 demonstrates longitudinal changes in Mayo scores. Baseline scores were comparable between group 1 (IFX) and group 2 (IVCS) [median (interquartile range): 10 (9, 12) vs 10.5 (9, 12); P = 0.542]. From weeks 12 onward, group 1 (IFX) consistently exhibited significantly lower Mayo scores than group 2 (IVCS), with sustained separation throughout follow-up at weeks 48, 96, and 144 (P ≤ 0.001), indicating durable improvement in disease activity.

Figure 6 Comparison of Mayo scores between group 1 (infliximab) and group 2 (intravenous corticosteroids) at baseline and during follow-up (weeks 12, 48, 96, and 144). Horizontal lines indicate the median, and boxes represent the interquartile range. IFX: Infliximab; IVCS: Intravenous corticosteroids.

In the adjusted multivariable Cox proportional hazards regression analysis, higher CRP (HR = 1.05, 95%CI: 1.01-2.04, P = 0.040), fecal calprotectin (HR = 1.45, 95%CI: 1.11-4.01, P = 0.003), severe disease activity (HR = 2.04, 95%CI: 1.24-4.46, P = 0.042), CMV infection (HR = 5.86, 95%CI: 1.09-12.53, P < 0.001), C. difficile infection (HR = 1.68, 95%CI: 1.22-5.48, P = 0.018), and lower serum albumin (HR = 0.78, 95%CI: 0.13-0.93, P = 0.035) were independently associated with an increased risk of colectomy. Disease extent was not significantly associated with colectomy. IFX treatment was independently associated with a reduced hazard of colectomy compared with IVCS (HR = 0.53, 95%CI: 0.20-0.88; P = 0.043; Table 3).

Table 3 Multivariable Cox regression analysis of independent predictors of colectomy in patients with moderate-to-severe ulcerative colitis.

Predictors	B	P value	Exp(B)	95%CI for Exp(B)
CRP	0.05	0.040a	1.05	1.01-1.14
Albumin	-0.25	0.035a	0.78	0.13-0.93
Fecal calprotectin	0.37	0.003b	1.45	1.11-4.01
Severity (severe UC)	0.71	0.042a	2.04	1.24-4.46
Extensive colitis	0.17	0.134	1.18	0.17-3.16
CMV infection	1.77	< 0.001b	5.86	1.09-12.53
C. difficile infection	0.52	0.018a	1.68	1.22-5.48
Treatment group (IFX vs IVCS)	-0.64	0.043a	0.53	0.20-0.88

^aP < 0.05.

^bP < 0.01.

B: Regression coefficients, Exp(B): Hazard ratio. Hazard ratios < 1 indicate a reduced risk of colectomy. C. difficile: Clostridioides difficile; CRP: C-reactive protein; UC: Ulcerative colitis; CMV: Cytomegalovirus; IVCS: Intravenous corticosteroids; IFX: Infliximab.

DISCUSSION

This multi-center prospective cohort study compared first-line IFX with IVCS in patients with moderate-to-severe UC. We found that IFX provided superior short and long-term clinical, endoscopic, and histologic disease control, with a significantly lower cumulative colectomy risk over 144 weeks of follow-up. These findings support and extend the conclusions of recent meta-analyses or randomized controlled trials, highlighting the sustained advantage of IFX in long-term remission[17,28,29].

Early efficacy and induction of remission

At weeks 12, patients treated with IFX demonstrated significantly better therapeutic outcomes than those who received IVCS, including higher rates of clinical response (77.9% vs 56.3%), clinical remission (60.3% vs 31.0%), endoscopic remission (51.5% vs 28.2%), and histological remission (41.2% vs 23.9%). These findings are consistent with previous studies reporting that anti-TNF therapy provides more rapid symptom control and promotes earlier mucosal healing compared with corticosteroid-based therapy in patients with moderate-to-severe UC. The superior early efficacy of IFX may be explained by its targeted inhibition of TNF-α, a central mediator of intestinal inflammation and tissue injury in UC[30,31].

Sustained long-term disease control

The clinical benefits observed with IFX during induction therapy persisted throughout the follow-up period. By weeks 48, sustained clinical remission was achieved in 57.4% of patients in the IFX group compared with 29.6% in the IVCS group, with comparable improvements noted for endoscopic and histological remission outcomes. At weeks 144, IFX continued to demonstrate superior rates of sustained remission across all evaluated clinical endpoints. These observations are supported by accumulating real-world evidence indicating that biologic therapy provides durable disease control and sustained mucosal healing in UC[32].

An important long-term finding of the present study was the significantly lower cumulative colectomy rate observed among patients treated with IFX (5.9% vs 18.3%), accompanied by a significant reduction in colectomy risk over time. This observation is consistent with recent systematic reviews demonstrating that early biologic therapy may reduce the likelihood of colectomy compared with conventional corticosteroid-based approaches, particularly during prolonged follow-up periods. These findings further emphasize the importance of evaluating long-term time-to-event outcomes rather than relying exclusively on short-term response measures[11,29]. In contrast, Oh et al[32] reported higher mortality rates among patients receiving IFX compared with those treated with IVCS. Differences in study design, patient characteristics, disease severity, and duration of follow-up may partly explain this discrepancy.

In the present study, mortality remained relatively uncommon and did not differ significantly between treatment groups throughout follow-up. Therefore, although numerically fewer deaths were observed in the IFX group, the small number of mortality events limits definitive interpretation, and no firm conclusions regarding the effect of treatment strategy on long-term survival can be drawn. Furthermore, because the present study was observational in nature and included a relatively limited number of colectomy events, the findings should be interpreted cautiously and require validation in large prospective randomized studies.

Biomarkers and predictors of outcomes

Multivariable analysis identified several independent predictors of colectomy, including elevated CRP levels, increased fecal calprotectin, severe baseline disease activity, hypoalbuminemia, and concomitant CMV or C. difficile infection. These factors have previously been associated with severe disease behavior and unfavorable clinical outcomes in UC, highlighting the importance of close biomarker monitoring for early identification of patients at increased risk of disease progression and surgical intervention[33,34]. The observed partial rebound in biochemical markers (CRP and fecal calprotectin) in the IVCS group during long-term follow-up correlates with the period of corticosteroid tapering and eventual discontinuation. Unlike IFX, which demonstrated sustained suppressive effects on inflammatory pathways, corticosteroids primarily target acute systemic inflammation. Their withdrawal likely unmasked residual underlying mucosal activity in patients who did not achieve complete endoscopic remission, further explaining the higher long-term relapse and colectomy rates in this group.

In our study, we observed that while fecal calprotectin is a strong indicator of mucosal inflammation, its levels in the mortality group at certain time points were lower than expected. This discrepancy can be explained by the fact that mortality was often related to systemic complications or post-surgical issues rather than active colonic inflammation alone at the terminal stage.

Safety and adverse event profiles

Regarding safety, the IFX group demonstrated a more favorable profile compared to the IVCS group, however, these findings should be interpreted with caution given the observational design and potential residual confounding. Grade ≥ 3 AEs were more frequent in the IVCS group (36.6%), driven by new-onset hypertension, diabetes, and serious infections (CMV and pneumonia). Conversely, grade ≥ 3 events in the IFX group were less frequent (10.3%) and primarily limited to infections. No deaths directly attributable to the study medications were reported. These findings suggest that first-line IFX may be associated with fewer corticosteroid-related complications than an IVCS-based treatment strategy. However, because cumulative corticosteroid exposure was not formally quantified, these observations should be interpreted cautiously. These observations are in line with the well-established risk profile of longterm corticosteroid therapy, which is associated with systemic toxicity and opportunistic infections[35,36]. Conversely, biologic agents like IFX have been shown to have an acceptable safety profile in UC with appropriate monitoring, these observations are in line with previous clinical trials[37].

Clinical implications

The findings of the present study support the early introduction of IFX in patients with moderate-to-severe UC, particularly in individuals considered at increased risk for steroid refractoriness or steroid dependence. Early biologic therapy was associated with higher rates of sustained remission and lower rates of surgical intervention, suggesting potential benefits of earlier treatment escalation in appropriately selected patients. Targeted biologic therapy, such as IFX, may facilitate long-term disease control and reduce the likelihood of irreversible disease-related complications in patients with chronic relapsing UC.

Overall, our findings align with the evolving therapeutic approach in moderate-to-severe UC that emphasizes early, effective, and steroid-sparing treatment strategies. In light of accumulating evidence published in recent years, first-line IFX appears capable not only of inducing remission but also of improving long-term outcomes, including durable disease control and reduced colectomy risk.

Study limitations

Several limitations should be acknowledged when interpreting the findings of this study. The non-randomized study design introduces the possibility of residual confounding despite protocol-guided treatment allocation and generally balanced baseline characteristics between groups. Although predefined criteria were used to identify patients at increased risk of steroid refractoriness, treatment assignment also incorporated clinician judgment and patient preferences. Consequently, some factors influencing treatment selection may not have been fully captured by the measured baseline variables, potentially resulting in residual selection bias. Although multivariable regression analyses were performed to adjust for measured confounders, residual confounding from unmeasured clinical factors, physician treatment preferences, and patient-related considerations cannot be entirely excluded.

Furthermore, propensity score-based adjustment methods were not applied, and therefore some degree of treatment-selection bias may have persisted despite statistical adjustment. Accordingly, the observed associations should not be interpreted as establishing a definitive causal treatment effect. The study was conducted exclusively in tertiary-care referral centers within a single geographic region, which may limit the generalizability of the findings. Patients treated in these centers often have more severe disease and may not fully represent the broader UC population. Therefore, the findings should be interpreted cautiously and validated in multicenter studies across diverse healthcare settings and geographic regions.

In addition, treatment-related decisions, including dose optimization and treatment discontinuation, were made according to clinician judgment and routine clinical practice, which may have introduced treatment-selection bias. Variability in corticosteroid regimens and post-response management strategies within the IVCS group may also have influenced clinical outcomes and affected internal consistency. Furthermore, detailed cumulative corticosteroid exposure, duration of therapy, and individual tapering trajectories were not systematically quantified throughout follow-up. As corticosteroid-related AEs such as hypertension, diabetes mellitus, and infectious complications are dose-dependent and duration-dependent, differences in steroid exposure may have contributed to the observed safety outcomes. Therefore, comparisons of adverse event rates between treatment groups should be interpreted cautiously.

Despite follow-up extended to 144 weeks, longer follow-up is needed to better understand late outcomes and unusual AEs. Although we employed standardized scoring systems, formal interobserver agreement testing was not performed. Therefore, the multicenter design may have introduced some degree of interobserver variability in endoscopic and histopathological assessments. However, the involvement of experienced specialists and the use of objective grading criteria were intended to mitigate this risk.

In our cohort, first-line IFX was associated with a significant reduction in the hazard of colectomy. However, these findings should be interpreted with caution due to the relatively small number of surgical events (n = 17 total). While the difference reached statistical significance, the small event size may limit the precision of the HR. Future large-scale, prospective studies or real-world registries are necessary to confirm whether these long-term surgical benefits are sustained across broader and more diverse patient populations. Similarly, mortality analyses were limited by the small number of deaths observed during follow-up (n = 8 overall). Consequently, the study was not adequately powered to detect modest differences in mortality between treatment groups, and mortality-related findings should be interpreted cautiously. Larger studies with greater event numbers are required to determine whether treatment strategy influences long-term survival outcomes in patients with UC. Finally, while several biomarkers were assessed, other genetic and immunologic determinants of treatment response were not examined.

CONCLUSION

In this real-world tertiary-care cohort, first-line IFX therapy was associated with higher rates of sustained clinical, endoscopic, and histological remission than IVCS in patients with moderate-to-severe UC and was linked to a significantly lower long-term risk of colectomy. IFX also demonstrated an acceptable safety profile, although the pattern of AEs differed from that observed with IVCS therapy. These findings support the consideration of early biologic intervention with IFX as a potential alternative first-line treatment strategy in appropriately selected patients with moderate-to-severe UC. Further prospective randomized studies comparing IFX with newer biologic agents, including vedolizumab and ustekinumab, are needed to optimize first-line therapeutic strategies and improve long-term patient outcomes.

ACKNOWLEDGEMENTS

The authors express their appreciation to all participants who took part in this study and to the clinical teams whose commitment to patient care and meticulous data collection made this research possible.