Abstract
Background:
Approximately one-third of inflammatory bowel disease (IBD) patients fail to respond to anti-tumor necrosis factor (anti-TNF) therapy, presenting a significant challenge. The two management strategies are changing to a second anti-TNF agent or a drug with a different mechanism of action.
Methods:
A retrospective study was conducted on patients with IBD who initiated biologic therapy, between January 2016 and July 2024. Data were extracted from the records of King Fahad Medical City (KFMC). Categorical data were presented as frequencies and percentages. The χ² test was used to evaluate the association between the choice of first, second, and third biologics with C.D. and U.C. Multivariable hazard ratios (HRs) were calculated to identify predictors of drug change. Kaplan–Meier curves were used to compare time to biological change, stratified by disease type.
Results:
The study included 607 Crohn’s disease (CD) and 392 ulcerative colitis (UC) patients. Most CD patients were male (56.3%), aged 31–45 (45.5% years), with ileocolonic disease (64.9%), and with a penetrating phenotype (38.6%). UC predominantly affected females (55.6%) aged 31–45 (44.4% years), with extensive colitis (63.3%). Infliximab was the most used biologic, with median persistence times of 69 months (CD) and 47 months (UC). In CD, Infliximab is commonly switched to Adalimumab and Ustekinumab. In UC, Infliximab was primarily switched to Ustekinumab, Adalimumab, and Vedolizumab. Persistence times varied; in CD patients treated with Infliximab, persistence was longer before switching compared with those treated with Adalimumab (P = 0.007). In UC, Infliximab persistence (47 months) exceeded that of Adalimumab (23 months; P = 0.016). Regression analysis identified key predictors of switching. In CD, a family history of IBD significantly increased the risk (HR = 1.46, 95% CI: 1.01–2.10; P = 0.042). In UC, younger age (18–30 years) increased risk (HR = 5.94, 95% CI: 1.09–32.27; P = 0.039), while being single reduced it (HR = 0.50, 95% CI: 0.28–0.89; P = 0.019).
Conclusion:
Anti-TNF therapies remain the first- and second-line biologics. Non-anti-TNF agents (Vedolizumab and Ustekinumab) are mainly employed as third-line options for IBD patients.
Keywords: Crohn’s disease, inflammatory bowel disease, primary failure, treatment patterns, ulcerative colitis
INTRODUCTION
Inflammatory bowel diseases (IBDs), encompassing ulcerative colitis (UC) and Crohn’s disease (CD), are chronic, idiopathic inflammatory disorders of the gastrointestinal tract. The introduction of anti-tumor necrosis factor (anti-TNF) agents has markedly revolutionized the management of IBD. Their administration obviates the need for corticosteroid treatment, promotes mucosal healing, and diminishes hospitalization rates and the need for surgical interventions, thereby significantly improving the quality of life for patients with IBD.[1]
Conversely, approximately one-third of patients exhibit a lack of response to anti-TNF induction therapy, classified as primary nonresponders.[2,3,4] Additionally, a subset of patients who initially demonstrate a positive response to anti-TNF agents may subsequently discontinue therapy due to loss of response (secondary nonresponders) or the onset of intolerable adverse events.[5] Moreover, in certain countries, the restricted duration of therapy can influence the clinical trajectory of the disease, often necessitating the reinitiation of biological therapy.[6]
For individuals who experience a loss of response to an anti-TNF agent, switching to an alternative anti-TNF agent to attempt remission is a well-established approach.[7,8] However, limited research has assessed the efficacy of a second anti-TNF agent in IBD patients who failed to achieve remission (i.e. primary nonresponders) with their initial anti-TNF therapy.[9]
Previous studies have shown that prior anti-TNF therapy increases the likelihood of primary treatment failure with subsequent anti-TNF agents.[2]
Patients who exhibit primary nonresponse may have altered pharmacokinetics (such as faster drug clearance leading to lower trough levels) or pharmacodynamics (such as mechanistic failure due to non-TNF-mediated inflammatory pathways), which decreases the likelihood of responding to a second anti-TNF agent.[10] However, recent advances have led to the approval of new therapies targeting various inflammatory pathways, including two biologics (vedolizumab and ustekinumab) and a small molecule (tofacitinib).[11,12,13] As treatment options for IBD continue to expand, understanding the factors that influence response will be crucial for optimizing and personalizing therapies.[14] While transitioning to a second anti-TNF agent or replacing it with a drug targeting a different mechanism of action may effectively induce and sustain remission in IBD patients unresponsive to initial anti-TNF therapy, the optimal timing and order of these strategies after primary anti-TNF failure remain unclear. As more therapies become available for IBD, choosing the correct sequence of treatments becomes increasingly essential. Both switching and swapping strategies are considered feasible options for other immune-mediated conditions.[15] Despite widespread biologic use, real-world data on the utilization and sequencing of biologics in IBD patients remain limited in the Middle East, specifically in Saudi Arabia. Existing studies primarily focus on treatment effectiveness in biologic-naive patients, without addressing switching patterns or predictors of treatment.[16,17] A notable gap exists in the investigation of non-TNF-α inhibitor biologics, their role in treatment transitions at third- and fourth-line management, and the associated patterns of treatment modifications across multiple therapy lines. Along with widely recognized international guidelines, regional studies and consensus statements from Saudi Arabia and the Middle East have also provided valuable insights into biologic therapy outcomes, emphasizing the importance of collecting local practice data.[16,17,18] This study aimed to explore patterns of biological drug changes and identify predictors of such changes in patients with IBD.
PATIENTS AND METHODS
Study Design and Setting
A retrospective study reviewed electronic medical records from January 2016 to July 2024 to assess the persistence of biological therapy and small molecules in IBD patients, at King Fahad Medical City (KFMC) in Riyadh, Saudi Arabia. Our study adhered to the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) guidelines.[19]
Study Population and Data Collection
Patients were included in the study if they were diagnosed with CD or UC according to internationally recognized criteria encompassing clinical, endoscopic, radiologic, and histopathological findings in line with the European Crohn’s and Colitis Organisation (ECCO) consensus guidelines and Lennard–Jones criteria; received biological therapy or small molecules at KFMC; were aged 14 years or older; and had available medical records for data collection.
Data were extracted from the medical records of KFMC patients to evaluate the persistence of biological therapy and small molecules in IBD patients. Medical records served as the primary data source.
Exclusion criteria encompassed patients with incomplete or missing medical records, those not receiving biological therapy or small molecules for IBD, patients with primary immunodeficiency disorders, and individuals who had undergone solid organ or bone marrow transplantation.
Sample Size Calculation and Sampling Technique
The sample size was calculated based on the estimated persistence rate of the most adherent biological agent among patients with IBD. A recent study reported a 72% persistence rate after 1 year.[20] Using a 95% confidence level, a 5% margin of error, and a design effect of 2 due to non-random sampling, the minimum required sample size was 620 IBD patients. The calculation was performed using Epi Info software, and convenience sampling was used to recruit participants.
Statistical Analysis
Categorical variables were reported as frequencies and percentages. The χ² test was used to evaluate the association between the selection of first-, second-, and third-line biologics and disease types (CD and UC). The χ² test was also applied to compare the baseline characteristics of the study population across both conditions. Multivariable hazard ratios (HRs) were used to identify predictors of changes in biologic treatment. Baseline variables were initially screened using univariate Cox regression, and those with P < 0.10, along with key clinical factors, were included in the multivariable model. In this study, Kaplan–Meier survival curves, accompanied by log-rank, Breslow, and Tarone–Ware tests, were used to compare the time to biologic treatment change, stratified by disease type. Different statistical methods were employed, each suited to specific event timing: The log-rank test was more sensitive to late events, while the other tests were more effective for early events.[21,22] Because treatment discontinuation and switching occurred at varying times, reporting all three tests allowed for the comprehensive interpretation of both early and late differences in survival curves. A P value of < 0.05 was considered statistically significant. Statistical analysis was performed using SPSS version 26.
Study Definitions
We define switching broadly as any change from one biologic to another. For clarity, we distinguish intra-class switches (anti-TNF → anti-TNF) and inter-class swaps (anti-TNF → non-anti-TNF). Persistence was defined as the duration from the initiation of a biologic therapy until discontinuation or switch to another biologic. A drug was considered discontinued at the date of the last dose plus the expected dosing interval. No additional grace period was applied beyond the approved dosing schedule.
Primary failure was defined as the absence of clinical improvement within 14 weeks of biologic initiation. Secondary failure was defined as loss of response after initial improvement despite adequate therapeutic drug levels.
Ethical Considerations
This study was approved by the Institutional Review Board of King Fahad Medical City, Riyadh (approval number: FWA00018774). The IRB formally waived the requirement for individual informed consent as the study involved a retrospective review of medical records and all data were anonymized to ensure confidentiality and privacy.
RESULTS
Demographic and Clinical Characteristics of Crohn’s Disease and Ulcerative Colitis Patients
The age distribution shows that most Crohn’s and UC patients were aged 31–45 (45.5%, 44.4%) and were commonly diagnosed between the ages of 18 and 30 years (52.2%, 57.1%), with disease duration mainly between 5 and 10 years (38.2%, 43.2%). The majority had a BMI of 18.5–24.99 (40.7%, 36%). Males predominated in Crohn’s (56.3%) and females in UC (55.6%). Nearly all were Saudi nationals (97.9%, 98.7%). More Crohn’s patients were single (59.5%), while UC patients were mostly married (55.1%). A family history of IBD was reported in CD Patients 11.5% and UC 8.9%, with higher smoking rates in CD (14.8% vs. 7.7%). Comorbidities were more prevalent in UC (20.9%, 34.1%), mainly primary sclerosing cholangitis (11.1%, 14.9%), psoriasis (8.2%, –), and hypertension (7.6%, 9.2%). Surgical history was more frequent in Crohn’s (45%, 3.4%), mainly resections (70.6%), versus UC, where IPAA was the primary procedure (61.5%). The median number of surgeries was 1 (1–2) for Crohn’s disease. Hospitalizations were common, occurring in 81.3% (66.5%) of cases, with a median of 2 episodes (range, 1–4) in both groups. Medication adherence was high (91.9%, 88%). Crohn’s most often involved ileocolonic disease (64.9%), with penetrating phenotype (38.6%) and perianal disease (40.9%), whereas UC predominantly showed extensive colitis (E3: 63.3%) followed by left-sided colitis (E2: 29.3%), Table 1.
Table 1.
Demographic and Clinical Characteristics of Crohn’s and Ulcerative Colitis Disease Patients
| CD | UC | Family history of IBD | CD | UC | Presence of Comorbidity | CD | UC | |
|---|---|---|---|---|---|---|---|---|
| Age, years | Yes | 70 (11.5%) | 35 (8.9%) | Yes | 124 (20.9%) | 130 (34.1%) | ||
| <18 | 29 (4.8%) | 9 (2.3%) | No | 537 (88.5%) | 357 (91.1%) | No | 470 (79.1%) | 251 (65.9%) |
| 18 – 30 | 240 (39.5%) | 118 (30.1%) | Smoking | N-Miss | 13 | 11 | ||
| 31 – 45 | 276 (45.5%) | 174 (44.4%) | Active smoker | 90 (14.8%) | 30 (7.7%) | Comorbidities | ||
| 46 – 60 | 53 (8.7%) | 72 (18.4%) | Ex-smoker | 35 (5.8%) | 7 (1.8%) | Autoimmune Hepatitis | 1 (0.6%) | 1 (0.5%) |
| >60 | 9 (1.5%) | 19 (4.8%) | Non-smoker | 482 (79.4%) | 355 (90.5%) | RA | 3 (1.8%) | 1 (0.5%) |
| Age at diagnosis, years | Alcohol use | Psoriasis | 14 (8.2%) | 8 (4.1%) | ||||
| <18 | 161 (26.5%) | 70 (17.9%) | Active alcohol intake | 4 (0.7%) | 0 (0%) | Ankylosing spondylitis | 9 (5.2%) | 6 (3.1%) |
| 18 – 30 | 317 (52.2%) | 224 (57.1%) | Ex-alcohol intake | 12 (2.1%) | 2 (0.5%) | Multiple sclerosis | 1 (0.6%) | |
| 31 – 45 | 103 (17%) | 82 (20.9%) | Never Drink | 569 (97.2%) | 365 (99.5%) | Celiac disease | 7 (4.1%) | 3 (1.5%) |
| 46 – 60 | 21 (3.5) | 13 (3.3%) | N-Miss | 22 | 25 | Obesity | 4 (2.4%) | 11 (5.6%) |
| >60 | 5 (0.8%) | 3 (0.8%) | Adherence to Medication | Metabolic syndrome | 1 (0.6%) | |||
| Duration of the disease, years | Yes | 534 (91.9%) | 322 (88%) | (NAFLD) | 4 (2.4%) | 3 (1.5%) | ||
| <5 | 144 (23.7%) | 73 (18.6%) | No | 47 (8.1%) | 44 (12%) | PSC | 19 (11.1%) | 29 (14.9%) |
| 5 – 10 | 232 (38.2%) | 169 (43.2%) | N-Miss | 26 | 26 | Peptic ulcers | 1 (0.6%) | 2 (1%) |
| 11 – 15 | 154 (25.4%) | 84 (21.4%) | History of Hospitalization | Irritable bowel syndrome (IBS) | 11 (6.4%) | 7 (3.6%) | ||
| > 5 | 77 (12.7%) | 66 (16.8%) | Yes | 473 (81.3%) | 242 (66.5%) | Gastroesophageal reflux disease (GERD) | 3 (1.8%) | 4 (2%) |
| BMI | No | 109 (18.7%) | 122 (33.5%) | HBV&HCV | 2 (1.2%) | 1 (0.5%) | ||
| < 8.5 | 93 (15.3%) | 53 (13.5%) | N-Miss | 25 | 28 | Type 1 diabetes | 2 (1.2%) | 3 (1.5%) |
| 18.5 – 24.99 | 247 (40.7%) | 141 (36%) | Number of Hospitalizations | Type 2 diabetes | 9 (5.3%) | 18 (9.2%) | ||
| 25 – 29.99 | 160 (26.4%) | 111 (28.3%) | Median (Q1-Q3) | 2 (1-4) | 2 (1-4) | Osteoporosis | 8 (4.7%) | 5 (2.6%) |
| 30 – 34.99 | 74 (12.2%) | 63 (16.1%) | IBD-related Hospitalization | Hypothyroidism | 6 (3.6%) | 18 (9.2%) | ||
| 35 – 40 | 25 (4.1%) | 18 (4.6%) | Median (Q1-Q3) | 2 (1-3) | 1 (1-3) | Hyperthyroidism | 1 (0.6%) | 1 (0.5%) |
| > 40 | 8 (1.3%) | 6 (1.5%) | History of Surgery | Dyslipidemia | 6 (3.5%) | 11 (5.6%) | ||
| Sex | Yes | 268 (45%) | 13 (3.4%) | Hypertension | 13 (7.6%) | 18 (9.2%) | ||
| Male | 342 (56.3%) | 174 (44.4%) | No | 328 (55%) | 364 (96.6%) | CAD | 5 (2.9%) | 4 (2%) |
| Female | 265 (43.7%) | 218 (55.6%) | N-Miss | 11 | 15 | Depression | 10 (5.8%) | 9 (4.6%) |
| Nationality | Number of Surgeries | Anxiety disorders | 4 (2.4%) | 13 (6.6%) | ||||
| Saudi | 594 (97.9%) | 387 (98.7%) | Median (Q1-Q3) | 1 (1-2) | 1 (1-1) | Bipolar disorder | 1 (0.6%) | |
| Non-Saudi | 13 (2.1%) | 5 (1.3%) | Type of Surgery | Tuberculosis (relevant in the context of biologic therapies) | 2 (1.2%) | |||
| Marital status | Strictureplasty | 4 (1.4%) | 2 (15.4%) | Recurrent urinary tract infections | 3 (1.8%) | 1 (0.5%) | ||
| Single | 361 (59.5%) | 176 (44.9%) | Resection | 208 (70.6%) | 1 (7.7%) | Clostridioides difficile infections | 4 (2.4%) | 6 (3.1%) |
| Married | 246 (40.5%) | 216 (55.1%) | Colectomy | 13 (4.4%) | 2 (15.4%) | Chronic kidney disease | 2 (1.2%) | 2 (1%) |
| Residency | Proctocolectomy with ileostomy | 11 (3.7%) | 8 (61.5%) | Kidney stones | 2 (1.2%) | 2 (1%) | ||
| Inside Riyadh | 295 (49.6%) | 216 (57%) | Fistula repair | 35 (11.8%) | 242 (66.5%) | Asthma | 11 (6.4%) | 9 (4.6%) |
| Outside Riyadh | 300 (50.4%) | 163 (43%) | Seton insertion | 24 (8.1%) | 122 (33.5%) | Lymphoma | 1 (0.6%) | |
| N-Miss | 12 | 13 | Disease location and extension (CD) | Perianal Disease (p) | ||||
| Education Level | CD Ileal (L1) | 147 (24.3%) | Yes | 248 (40.9%) | ||||
| Less than High school degree | 44 (7.6%) | 29 (8%) | CD Colonic (L2) | 64 (10.5%) | No | 359 (59.1%) | ||
| High school degree | 193 (33.1%) | 114 (31.3%) | CD Ileocolonic (L3) | 394 (64.9%) | Disease location and extension (UC) | |||
| Bachelor degree | 314 (54%) | 203 (55.8%) | CD Upper GI (L4) | UC Proctitis (E1) | 29 (7.4%) | |||
| Post Graduate degree | 16 (2.7%) | 8 (2.2%) | Yes | 22 (3.6%) | UC Left side colitis (E2) | 115 (29.3%) | ||
| Diploma | 15 (2.6%) | 10 (2.7%) | No | 583 (96.1%) | UC Extensive colitis (E3) | 248 (63.3%) | ||
| N-Miss | 25 | 0 | N/A | 2 (0.3%) | ||||
| Employment status | Disease Behavior and Phenotype CD | |||||||
| Employed | 300 (51.5%) | 181 (49.6%) | Non-Penetrating and Non-stricturing (B1) | 181 (29.8%) | ||||
| Nonemployed | 221 (38%) | 157 (43%) | Stricturing (B2), inflammatory | 91 (15%) | ||||
| Student | 61 (10.5%) | 27 (7.4%) | Stricturing (B2), Fibrostenotic | 101 (16.6%) | ||||
| N-Miss | 25 | 27 | Penetrating (B3) | 234 (38.6%) | ||||
*N-Miss=number of patients with missing data for the variable. Percentages are calculated from non-missing cases
Characteristics of Drugs Used Among Crohn’s Disease Patients
Infliximab was the most commonly used biologic (n = 413), predominantly as first-line therapy (81.4%) and to a lesser extent as second-line therapy (16.9%). Among those who discontinued treatment (n = 180), the main reasons were primary failure (32.1%), secondary failure (22.2%), and allergic reactions (11.1%). Adalimumab was the second most frequently used biologic (n = 261), mainly as first-line (64.7%) or second-line therapy (33%). Among the 131 patients who discontinued, primary failure was the most common reason (50.4%), followed by secondary failure (15.2%). Vedolizumab was used by 71 patients, most often as third- or fourth-line therapy (50.7%) and less frequently as first- (22.5%) or second-line (26.8%) therapy. Of the 34 discontinuations, 44.1% were due to primary failure and 17.6% postsurgery. Ustekinumab was used by 109 patients, largely as second- or third-line therapy (38.5% and 43.1%, respectively). Of the 20 discontinuations, 40% were due to primary failure, 20% due to surgery, and 10% due to intolerance. The use of certolizumab (n = 3) and upadacitinib (n = 10) was rare, with limited discontinuation data available.
Among immunomodulators, azathioprine was the most commonly used (n = 428), predominantly as a first-line agent (95.3%). Discontinuations (n = 178) were most often due to intolerance (27.5%) or unclear reasons (30.9%). Methotrexate was used by 40 patients, mainly as second-line therapy (60%); of 25 discontinuations, 36% were due to unclear reasons, 24% to patient choice, and 20% to primary failure. 6-mercaptopurine was used less frequently (n = 15), primarily as second-line therapy, with discontinuations mainly due to intolerance, unclear reasons, or patient choice, Table S1.
Table S1.
Characteristics of drugs used among Crohn’s and Ulcerative Colitis disease patients
| Drugs | CD | UC | Adalimumab | Vedolizumab | Ustekinumab | Certolizumab | |||||
|---|---|---|---|---|---|---|---|---|---|---|---|
| Infliximab | |||||||||||
| Number of patients | N=413 | N=138 | N=261 | 78 | N=71 | N=58 | N=109 | N=56 | N=3 | 3 | |
| 1st Line | 336 (81.4%) | 93 (67.5%) | 169 (64.7%) | 63 (80.8%) | 16 (22.5%) | 29 (50%) | 5 (4.6%) | 2 (3.6%) | 0 (0%) | 1 (33.3%) | |
| 2nd Line | 70 (16.9%) | 42 (30.4%) | 86 (33%) | 15 (19.2%) | 19 (26.8%) | 20 (34.5%) | 42 (38.5%) | 25 (44.6%) | 0 (0%) | 2 (66.7%) | |
| 3rd Line | 7 (1.7%) | 1 (0.7%) | 4 (1.5%) | - | 28 (39.4%) | 8 (13.8%) | 47 (43.1%) | 24 (42.9%) | 3 (100%) | - | |
| 4th Line | - | 2 (1.4%) | 2 (0.8%) | - | 7 (9.9%) | 1 (1.7%) | 15 (13.8%) | 4 (7.1%) | - | - | |
| 5th Line | - | - | - | - | 1 (1.4%) | - | - | 1 (1.8%) | - | - | |
| Reason for Discontinuation | N=180 | N=71 | N=131 | N=55 | N=34 | N=28 | N=20 | N=18 | N=2 | N=2 | |
| Primary Medication Failure | 58 (32.1%) | 30 (42.3%) | 66 (50.4%) | 26 (47.2%) | 15 (44.1%) | 15 (53.5%) | 8 (40%) | 11 (60.9%) | - | - | |
| Secondary Medication Failure | 40 (22.2%) | 15 (21.1%) | 20 (15.2%) | 9 (16.4%) | 2 (5.9%) | 1 (3.6%) | 1 (5%) | 2 (11.1%) | 1 (50%) | - | |
| Medication Unavailable | 3 (1.7%) | 2 (1.5%) | 2 (3.8%) | 1 (3.6%) | 1 (5.6%) | - | - | ||||
| EIM | (AS, Sacroilietis ) | - | - | - | - | 1 (2.9%) | - | - | 1 (5.6%) | - | 1 (50%) |
| Peripheral Arthritis | 1 (0.6%) | - | - | - | - | 1 (3.6%) | - | - | - | - | |
| Psoriasis | 7 (3.9%) | 2 (2.8%) | 1 (0.8%) | 3 (5.4%) | 1 (2.9%) | - | - | - | - | - | |
| Allergic Reaction to the Drug | 20 (11.1%) | 9 (12.7%) | 1 (0.8%) | 1 (1.8%) | - | - | 1 (5%) | 1 (5.6%) | - | - | |
| Active Infection ( Collection, Abscess) | 6 (3.3%) | - | 1 (0.8%) | - | 2 (5.9%) | - | - | - | - | - | |
| Patient Underwent Surgery | 2 (1.1%) | - | 5 (3.8%) | - | 6 (17.7%) | 3 (10.7%) | 4 (20%) | - | - | - | |
| Patient Lost to Follow-up | 1 (0.6%) | 1 (1.4%) | 1 (0.8%) | - | - | - | - | ||||
| Patient Decided Not to Continue | 12 (6.7%) | 1 (1.4%) | 7 (5.3%) | 4 (7.3%) | 2 (5.9%) | 3 (10.7%) | 2 (10%) | - | - | - | |
| Medication (High) Cost | - | - | 1 (0.8%) | - | - | - | - | - | - | - | |
| Pregnancy | 1 (0.6%) | - | - | - | - | - | - | - | - | - | |
| No Clear Reason (from diffirent hospital) | 24 (13.3%) | 8 (11.3%) | 21 (16%) | 7 (12.7%) | 5 (14.7%) | 1 (3.6%) | 2 (10%) | 1 (5.6%) | 1 (50%) | 1 (50%) | |
| Patient Preference ( IV, PO, Subcut) | 1 (0.6%) | 1 (1.4%) | - | - | - | 1 (3.6%) | - | 1 (5.6%) | - | - | |
| Patient Intolerance (Nausea and Vomiting) | 4 (2.2%) | 4 (5.6%) | 5 (3.8%) | 3 (5.4%) | - | 2 (7.1%) | 2 (10%) | - | - | - | |
|
| |||||||||||
| Drugs | Tofacitinib | Updacitinib | MTX | 6-MP | Azathioprine | ||||||
|
| |||||||||||
| Number of patients | - | N=13 | N=10 | N=17 | N=40 | N=11 | N=15 | N=13 | N=428 | N=216 | |
| 1st Line | - | 1 (7.6%) | 0 (0%) | 1 (5.9%) | 14 (35%) | 7 (63.6%) | 4 (26.7%) | 4 (30.8%) | 408 (95.3%) | 208 (96.3%) | |
| 2nd Line | - | 2 (15.4%) | 1 (10%) | 2 (11.8%) | 24 (60%) | 4 (36.4%) | 10 (66.6%) | 9 (69.2%) | 20 (4.7%) | 8 (3.7%) | |
| 3rd Line | - | 4 (30.8%) | 1 (10%) | 7 (41.1%) | 2 (5%) | - | 1 (6.7%) | - | - | - | |
| 4th Line | - | 4 (30.8%) | 6 (60%) | 3 (17.6%) | - | - | - | - | - | - | |
| 5th Line | - | 2 (15.4%) | 2 (20%) | 2 (11.8%) | - | - | - | - | - | - | |
| 6th Line | - | - | - | 2 (11.8%) | - | - | - | - | - | ||
| Reason for Discontinuation | - | N=5 | N=2 | N=3 | N=25 | N=9 | N=6 | N=8 | N=178 | N=92 | |
| Primary Medication Failure | - | 3 (60%) | - | 2 (66.7%) | 5 (20%) | 1 (11.1%) | 1 (16.7%) | - | 15 (8.4%) | 10 (10.9%) | |
| Secondary Medication Failure | - | 1 (20%) | - | - | - | - | - | - | 4 (2.2%) | - | |
| Medication Unavailable | - | - | - | - | - | - | - | - | - | - | |
| EIM | (AS, Sacroilietis ) | - | - | - | - | - | - | - | - | - | - |
| Peripheral Arthritis | - | - | - | - | - | - | - | - | - | - | |
| Psoriasis | - | - | - | - | - | - | - | 1 (0.6%) | - | ||
| Allergic Reaction to the Drug | - | - | - | - | 1 (4%) | - | - | - | 6 (3.4%) | - | |
| Active Infection ( Collection, Abscess ) | - | - | - | - | 1 (4%) | 1 (11.1%) | - | - | 9 (5.1%) | 1 (1.1%) | |
| Patient Underwent Surgery | - | - | 1 (50%) | - | - | - | - | - | 11 (6.2%) | 1 (1.1%) | |
| Patient Lost to Follow-up | - | - | - | - | - | - | - | - | 2 (1.1%) | 1 (1.1%) | |
| Patient Decided Not to Continue | - | 1 (20%) | - | - | 6 (24%) | 3 (33.3%) | 1 (16.7%) | 23 (12.9%) | 15 (16.3%) | ||
| Medication (High) Cost | - | - | - | - | - | - | - | - | - | - | |
| Pregnancy | - | - | 1 (50%) | - | - | - | - | - | 2 (1.1%) | 1 (1.1%) | |
| No Clear Reason (from different hospital) | - | - | - | 1 (33.3%) | 9 (36%) | 3 (33.3%) | 1 (16.7%) | 4 (50%) | 55 (30.9%) | 33 (35.8%) | |
| Patient Preference ( IV, PO, Subcut) | - | - | - | - | - | - | - | 1 (12.5%) | 1 (0.6%) | - | |
| Patient Intolerance (Nausea and Vomiting) | - | - | - | - | 3 (12%) | 1 (11.1%) | 3 (49.9%) | 3 (37.5%) | 49 (27.5%) | 30 (32.6%) | |
Characteristics of Drugs used Among Ulcerative Colitis Patients
Infliximab was used by 138 patients, primarily as first-line therapy (67.5%). Among those who discontinued, the main reasons were primary failure (42.3%) and secondary failure (21.1%). Adalimumab was used by 78 patients, mainly as first-line therapy (80.8%). Of the discontinuations, primary failure was most common (47.2%). Vedolizumab was used by 58 patients, with half receiving it as first-line therapy (50%). Discontinuations (n = 31) were mainly due to medication failure (53.5%). Ustekinumab was used by 56 patients, predominantly as second- (44.6%) or third-line therapy (42.9%). Among those who discontinued, primary failure accounted for 60.9%. Use of less common agents was limited. Certolizumab (n = 3), tofacitinib (n = 13), and upadacitinib (n = 17) were primarily prescribed as third- or later-line therapies. Discontinuations in these groups were few but were predominantly due to primary medication failure, with occasional cases attributed to intolerance, extraintestinal manifestations, or unclear reasons. Among immunomodulators, azathioprine was the most commonly used (n = 216), predominantly as first-line therapy (96.3%). Discontinuations (n = 92) were often due to intolerance (32.6%) or primary failure (10.9%). Methotrexate was used by 11 patients, mainly as first-line therapy (63.6%), with discontinuations linked to unclear reasons (33.3%) or intolerance (11.1%). 6-mercaptopurine was used by 13 patients, primarily as second-line therapy (69.2%), and was discontinued mainly for unclear reasons (50%) or intolerance (37.5%), Table S1.
Switching Patterns Among Biologic Treatments from First to Second Drug
In Crohn’s disease, the predominant first-to-second line switches were from infliximab to adalimumab (23.5%) and ustekinumab (8.0%), and from adalimumab to infliximab (36.1%) or ustekinumab (8.9%). Transitions from vedolizumab were infrequent, most commonly to infliximab (4.2%) or adalimumab (3.0%). Other biologics were only rarely utilized in the secondary setting.
In ulcerative colitis, infliximab was most frequently switched to ustekinumab (18.3%), adalimumab (15.1%), or vedolizumab (14.0%). Adalimumab was most commonly switched to infliximab (36.5%), followed by ustekinumab (9.5%) and vedolizumab (9.5%). Patients discontinuing vedolizumab were predominantly switched to infliximab (37.9%). Transitions involving certolizumab, tofacitinib, or upadacitinib were uncommon, Table 2.
Table 2.
Switching Patterns among Biologic Treatments from First to Second Drugs
| Primary Drug | Infliximab | Adalimumab | Vedolizumab | Ustekinumab | |||||
|---|---|---|---|---|---|---|---|---|---|
| Number of Patients | 336 | 93 | 169 | 63 | 16 | 29 | 5 | 2 | 1 |
| Secondary drug | |||||||||
| Infliximab | - | - | 61 (36.1%) | 23 (36.5%) | 3 (18.8%) | 11 (37.9%) | 0 (0%) | 1 (50%) | 1 (100%) |
| Adalimumab | 79 (23.5%) | 14 (15.1%) | - | - | 3 (18.8%) | 1 (3.4%) | 0 (0%) | 0 (0%) | 0 (0%) |
| Vedolizumab | 14 (4.2%) | 13 (14%) | 5 (3%) | 6 (9.5%) | - | - | 0 (0%) | 0 (0%) | 0 (0%) |
| Ustekinumab | 27 (8%) | 17 (18.3%) | 15 (8.9%) | 6 (9.5%) | 0 (0%) | 1 (3.4%) | - | - | |
| Certolizumab | 0 (0%) | 0 (0%) | 0 (0%) | 2 (3.2%) | 0 (0%) | 0 (0%) | 0 (0%) | 0 (0%) | - |
| Tofacitinib | 0 (0%) | 1 (1.1%) | 0 (0%) | 1 (1.6%) | 0 (0%) | 0 (0%) | 0 (0%) | 0 (0%) | 0 (0%) |
| Upadacitinib | 1 (0.3%) | 0 (0%) | 0 (0%) | 2 (3.2%) | 0 (0%) | 0 (0%) | 0 (0%) | 0 (0%) | 0 (0%) |
Switching Patterns Among Biologic Treatments from Second to Third Drug
In Crohn’s disease, the major switching patterns were infliximab to ustekinumab (25.7%) or vedolizumab (14.3%), and adalimumab to ustekinumab (23.3%) or vedolizumab (18.6%). Patients discontinuing vedolizumab were most frequently transitioned to ustekinumab (36.8%) or infliximab (15.8%). Other biologics, such as certolizumab and upadacitinib, were rarely employed as third-line agents.
In ulcerative colitis, infliximab was most often switched to ustekinumab (23.8%), with smaller proportions transitioning to upadacitinib (9.5%) or vedolizumab (9.5%). Adalimumab was most frequently switched to ustekinumab (40.0%), and vedolizumab was also primarily transitioned to ustekinumab (40.0%). Switching to tofacitinib and other agents occurred only sporadically, Table S2.
Table S2.
Switching Patterns Among Biologic Treatments from Second to Third drug
| Third Drug | Infliximab | Adalimumab | Vedolizumab | Ustekinumab | Upadacitinib | Certolizumab | Tofacitinib | |||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Number of Patient | 70 | 42 | 86 | 15 | 19 | 20 | 42 | 1 | 2 | 2 | 2 | |
| Fourth drug | ||||||||||||
| Infliximab | - | - | 2 (2.3%) | 0 (0%) | 3 (15.8%) | 1 (5%) | 2 (4.8%) | 0 (0%) | 0 (0%) | 0 (0%) | 0 (0%) | 0 (0%) |
| Adalimumab | 0 (0%) | 0 (0%) | - | - | 1 (5.3%) | 0 (0%) | 1 (2.4%) | 0 (0%) | 0 (0%) | 0 (0%) | 0 (0%) | 0 (0%) |
| Vedolizumab | 10 (14.3%) | 4 (9.5%) | 16 (18.6%) | 3 (20%) | - | - | 1 (2.4%) | 0 (0%) | 0 (0%) | 0 (0%) | 0 (0%) | 1 (50%) |
| Ustekinumab | 18 (25.7%) | 10 (23.8%) | 20 (23.3%) | 6 (40%) | 7 (36.8%) | 8 (40%) | - | - | 0 (0%) | 0 (0%) | 0 (0%) | 0 (0%) |
| Certolizumab | 1 (1.4%) | 0 (0%) | 0 (0%) | 0 (0%) | 2 (10.5%) | 0 (0%) | 0 (0%) | 0 (0%) | 0 (0%) | 0 (0%) | - | 0 (0%) |
| Tofacitinib | 0 (0%) | 1 (2.4%) | 0 (0%) | 0 (0%) | 0 (0%) | 0 (0%) | 0 (0%) | 2 (8%) | 0 (0%) | - | 1 (50%) | - |
| Upadacitinib | 0 (0%) | 4 (9.5%) | 1 (1.2%) | 0 (0%) | 0 (0%) | 1 (5%) | 0 (0%) | 2 (8%) | - | 0 (0%) | 0 (0%) | 0 (0%) |
Inferential Among Biological Therapies Regarding Persistence Time Before Switching
Crohn’s disease
Switching biologics from the first to the second drug among Crohn’s disease patients reveals that patients initially treated with Vedolizumab tend to switch to a second biologic treatment sooner than those initially treated with Infliximab or Adalimumab. The median persistence time, representing the point at which 50% of patients switched treatments, was 69 months for Infliximab, longer than 60 months observed for Adalimumab. The Breslow (P = 0.007) and Tarone–Ware tests (P = 0.018) were statistically significant. The log-rank test was not statistically significant (P = 0.091), Figure 1.
Figure 1.
Kaplan–Meier Curve for Switching of Biologics from First to Second among Patients with Crohn's Disease. Patients initially treated with infliximab (median persistence 69 months) and adalimumab (60 months) demonstrated longer persistence compared to vedolizumab (26 months). The Breslow (P = 0.007) and Tarone–Ware (P = 0.018) tests indicated significant differences, while the log-rank test was not significant (P = 0.091)
Switching from the second to the third biologic drug in CD patients shows no significant difference in the time to switch to a third biologic between patients who were initially treated with Infliximab, Adalimumab, Vedolizumab, or Ustekinumab. None of the inferential statistical tests used to analyze the curve (log-rank, Breslow, and Tarone–Ware) were statistically significant, Figure S1 (61.2KB, tif) .
Switching from the third to the fourth biologic drug among Crohn’s disease patients reveals that patients initially treated with vedolizumab tend to switch to a fourth biologic treatment sooner than those initially treated with Ustekinumab. The median persistence time for Vedolizumab was 26 months. All inferential statistical tests used to analyze the curve (log-rank, Breslow, and Tarone–Ware) were statistically significant (P = 0.026, P = 0.021, and P = 0.022, respectively), Figure 2.
Figure 2.
Kaplan–Meier Curve for Switching of Biologics From Third to Fourth Drug among Patients with Crohn's Disease. Vedolizumab (median persistence 26 months) was associated with significantly shorter persistence compared to ustekinumab (median persistence 51 months). All survival tests were statistically significant (log-rank P = 0.026; Breslow P = 0.021; Tarone–Ware P = 0.022)
Ulcerative colitis
Switching from the first to the second biologic drug among UC patients reveals that patients initially treated with Adalimumab tend to switch to a second biologic treatment sooner than those initially treated with Infliximab or Vedolizumab. The median persistence time for Infliximab was 47 months, exceeding that of Vedolizumab at 38 months and Adalimumab at 23 months. All inferential statistical tests used to analyze the curve were statistically significant: log-rank test (P = 0.019), Breslow test (P = 0.016), and Tarone–Ware test (P = 0.016), Figure 3.
Figure 3.
Kaplan–Meier Curve for Switching of Biologics From First to Second among Patients with Ulcerative Colitis. Median persistence was longest for infliximab (47 months), followed by vedolizumab (38 months) and adalimumab (23 months). All survival tests indicated significant differences (log-rank P = 0.019; Breslow P = 0.016; Tarone–Ware P = 0.016)
Switching from the second to the third line of treatment among ulcerative colitis patients reveals no significant difference in time to switch to a third-line biologic therapy among patients treated with Infliximab, Adalimumab, Vedolizumab, or Ustekinumab. The median persistence times for Adalimumab, Infliximab, and Vedolizumab were 35, 26, and 24 months, respectively, Figure S2 (55.3KB, tif) .
Switching from the third to the fourth line of treatment shows no significant difference in time to switch to a fourth biologic treatment between patients who were initially treated with Vedolizumab, Ustekinumab, Tofacitinib, or Upadacitinib. Vedolizumab had a median persistence time of 54 months, while ustekinumab had a median persistence time of 51 months, Figure S3 (52.9KB, tif) .
Regression Results
The Cox regression analysis revealed that the only statistically significant risk factor for switching to a secondary biologic drug among Crohn’s disease patients was a family history of IBD, with a hazard ratio (HR) of 1.459 (95% CI: 1.013–2.102, P = 0.042). None of the other studied demographic variables showed statistically significant associations (P > 0.05), Table S3.
Table S3.
Cox Regression Analyzing Risk Factors for Switching to secondary biologic drug in Crohn’s disease patients
| Variables | Hazard Ratio | 95% Confidence Interval | P | |
|---|---|---|---|---|
|
| ||||
| Lower | Upper | |||
| Age | ||||
| <18 | 0.609 | 0.172 | 2.161 | 0.443 |
| 18 – 30 | 0.806 | 0.297 | 2.19 | 0.672 |
| 31 – 45 | 0.677 | 0.249 | 1.84 | 0.445 |
| 46 – 60 | 0.725 | 0.245 | 2.143 | 0.561 |
| >60 | Reference | |||
| Age at diagnosis | ||||
| <18 | 1.432 | 0.349 | 5.869 | 0.618 |
| 18 – 30 | 1.647 | 0.407 | 6.667 | 0.484 |
| 31 – 45 | 1.88 | 0.454 | 7.782 | 0.384 |
| 46 – 60 | 1.763 | 0.374 | 8.306 | 0.474 |
| >60 | Reference | |||
| Duration of the disease, years | ||||
| <5 | 1.106 | 0.703 | 1.742 | 0.663 |
| 5 – 10 | 0.877 | 0.577 | 1.334 | 0.54 |
| 11 – 15 | 0.726 | 0.463 | 1.138 | 0.163 |
| >15 | Reference | |||
| BMI | ||||
| <8.5 | 0.631 | 0.246 | 1.621 | 0.339 |
| 18.5 – 24.99 | 0.505 | 0.203 | 1.257 | 0.142 |
| 25 – 29.99 | 0.522 | 0.209 | 1.305 | 0.164 |
| 30 – 34.99 | 1.039 | 0.405 | 2.666 | 0.937 |
| 35 – 40 | 0.491 | 0.166 | 1.452 | 0.198 |
| >40 | Reference | |||
| Sex | ||||
| Male | 1.019 | 0.768 | 1.352 | 0.898 |
| Female | Reference | |||
| Nationality | ||||
| Saudi | 1.19 | 0.439 | 3.226 | 0.733 |
| Non-Saudi | Reference | |||
| Marital status | ||||
| Single | 1.129 | 0.856 | 1.489 | 0.391 |
| Married | Reference | |||
| Residency | ||||
| Inside Riyadh | 1.257 | 0.962 | 1.643 | 0.094 |
| Outside Riyadh | Reference | |||
| Education Level | ||||
| Less than High school degree | 1.146 | 0.444 | 2.956 | 0.778 |
| High school degree | 1.15 | 0.518 | 2.553 | 0.732 |
| Bachelor degree | 1.073 | 0.5 | 2.305 | 0.856 |
| Post Graduate degree | 0.751 | 0.25 | 2.257 | 0.61 |
| Diploma | Reference | |||
| Employment status | ||||
| Employed | 1.239 | 0.685 | 2.241 | 0.479 |
| Non-employed | 1.341 | 0.783 | 2.296 | 0.286 |
| Student | Reference | |||
| Family history of IBD | ||||
| Yes | 1.459 | 1.013 | 2.102 | 0.042 |
| No | Reference | |||
| Smoking | ||||
| Active smoker | 0.895 | 0.609 | 1.316 | 0.574 |
| Ex-smoker | 0.876 | 0.479 | 1.602 | 0.666 |
| Non-smoker | Reference | |||
| Alcohol use | ||||
| Active alcohol intake | 1.25 | 0.384 | 4.069 | 0.712 |
| Ex-alcohol intake | 1.333 | 0.532 | 3.341 | 0.539 |
| Never Drink | Reference | |||
| Presence of Comorbidity | ||||
| Yes | 0.936 | 0.671 | 1.304 | 0.695 |
| No | Reference | |||
| History of Surgery | ||||
| Yes | 1.151 | 0.847 | 1.564 | 0.369 |
| No | Reference | |||
| History of Hospitalization | ||||
| Yes | 1.257 | 0.831 | 1.901 | 0.279 |
| No | Reference | |||
| Adherence to Medication | ||||
| Yes | 0.727 | 0.438 | 1.207 | 0.218 |
| No | Reference | |||
| Disease location and extention (CD) | ||||
| CD Ileal (L1) | 0.329 | 0.039 | 2.781 | 0.307 |
| CD Colonic (L2) | 0.444 | 0.051 | 3.883 | 0.463 |
| CD Ileocolonic (L3) | 0.426 | 0.051 | 3.553 | 0.43 |
| Isolated CD Upper GI (L4) | Reference | |||
| CD Upper GI (L4) | ||||
| Yes | 0.686 | 0.082 | 5.714 | 0.727 |
| No | Reference | |||
| Disease Behavior and Phenotype | ||||
| Non Penetrating and Non stricturing (B1) | 0.706 | 0.482 | 1.034 | 0.074 |
| Stricturing (B2), inflammatory | 0.856 | 0.552 | 1.326 | 0.485 |
| Stricturing (B2), Fibrostenotic | 0.998 | 0.704 | 1.415 | 0.992 |
| Penetrating (B3) | Reference | |||
| Perianal Disease (p) | ||||
| Yes | 0.908 | 0.694 | 1.189 | 0.483 |
| No | Reference | |||
The Cox regression analysis for switching to a secondary biologic drug among UC patients revealed that the 18–30 age group had a significantly higher risk of switching, with an HR of 5.939 (95% CI: 1.093–32.265, P = 0.039). Additionally, marital status was a significant factor, with single patients having a lower risk of switching than married patients, as evidenced by an HR of 0.496 (95% CI: 0.276–0.890, P = 0.019), Table S4.
Table S4.
Cox Regression Analyzing Risk Factors for switching to secondary biologic drug in Ulcerative colitis patients
| Variables | Hazard Ratio | 95% Confidence Interval | P | |
|---|---|---|---|---|
|
| ||||
| Lower | Upper | |||
| Age | ||||
| <18 | 1.772 | 0.137 | 22.873 | 0.661 |
| 18 – 30 | 5.939 | 1.093 | 32.265 | 0.039* |
| 31 – 45 | 3.916 | 0.754 | 20.355 | 0.104 |
| 46 – 60 | 3.51 | 0.682 | 18.053 | 0.133 |
| >60 | Reference | |||
| Age at diagnosis | ||||
| <8 | 0.67 | 0.074 | 6.056 | 0.722 |
| 18 – 30 | 0.631 | 0.073 | 5.44 | 0.675 |
| 31 – 45 | 0.523 | 0.057 | 4.787 | 0.566 |
| >60 | Reference | |||
| Duration of the disease, years | ||||
| <5 | 1.282 | 0.6 | 2.739 | 0.522 |
| 5 – 10 | 0.656 | 0.335 | 1.286 | 0.219 |
| 11 – 15 | 0.701 | 0.322 | 1.526 | 0.37 |
| >5 | Reference | |||
| BMI | ||||
| <18.5 | 0.335 | 0.036 | 3.089 | 0.335 |
| 18.5 – 24.99 | 0.211 | 0.024 | 1.843 | 0.16 |
| 25 – 29.99 | 0.19 | 0.021 | 1.671 | 0.134 |
| 30 – 34.99 | 0.193 | 0.021 | 1.761 | 0.145 |
| 35 – 40 | 0.134 | 0.012 | 1.528 | 0.106 |
| >40 | Reference | |||
| Sex | ||||
| Male | 1.65 | 0.978 | 2.782 | 0.06 |
| Female | Reference | |||
| Nationality | ||||
| Saudi | 0.495 | 0.097 | 2.511 | 0.396 |
| Non-Saudi | Reference | |||
| Marital status | ||||
| Single | 0.496 | 0.276 | 0.89 | 0.019* |
| Married | Reference | |||
| Residency | ||||
| Inside Riyadh | 1.144 | 0.71 | 1.842 | 0.581 |
| Outside Riyadh | Reference | |||
| Education Level | ||||
| Less than High school degree | 0.382 | 0.054 | 2.7 | 0.335 |
| High school degree | 0.645 | 0.149 | 2.799 | 0.559 |
| Bachelor degree | 0.417 | 0.102 | 1.712 | 0.225 |
| Post Graduate degree | 0.463 | 0.052 | 4.1 | 0.489 |
| Diploma | Reference | |||
| Employment status | ||||
| Employed | 0.772 | 0.296 | 2.016 | 0.598 |
| Non-employed | 0.828 | 0.345 | 1.983 | 0.671 |
| Student | Reference | |||
| Family history of IBD | ||||
| Yes | 1.16 | 0.515 | 2.609 | 0.72 |
| No | Reference | |||
| Smoking | ||||
| Active smoker | 1.033 | 0.456 | 2.342 | 0.938 |
| Ex-smoker | 0.425 | 0.083 | 2.174 | 0.304 |
| Non-smoker | Reference | |||
| Presence of Comorbidity | ||||
| Yes | 1.315 | 0.802 | 2.158 | 0.278 |
| No | Reference | |||
| History of Surgery | ||||
| Yes | 0.538 | 0.205 | 1.414 | 0.209 |
| No | Reference | |||
| History of Hospitalization | ||||
| Yes | 1.31 | 0.735 | 2.336 | 0.359 |
| No | Reference | |||
| Adherence to Medication | ||||
| Yes | 0.53 | 0.25 | 1.123 | 0.098 |
| No | Reference | |||
| Disease location and extention (UC) | ||||
| UC Proctitis (E1) | 0.36 | 0.072 | 1.81 | 0.215 |
| UC Left side colitis (E2) | 0.78 | 0.467 | 1.3 | 0.34 |
| UC Extensive colitis (E3) | Reference | |||
DISCUSSION
Our study involved 607 CD and 392 UC patients, investigating primarily medication use and switching patterns among them separately. Medication adherence was high in both groups, with 91.9% in CD and 88% in UC. In both cohorts, infliximab and adalimumab were the most commonly used biologics, with primary failure being the primary reason for discontinuation. In CD patients, Infliximab was commonly switched to Adalimumab (23.5%) and Ustekinumab (8%). This pattern of switching from infliximab to adalimumab is a common and evidence-supported strategy. Meta-analyses and guidelines suggest that a substantial proportion of patients can achieve clinical remission with a second anti-TNF agent.[23] Ustekinumab and vedolizumab are also options after anti-TNF failure. Still, adalimumab is often preferred as the next step due to its similar mechanism of action and established efficacy in this setting.[24]
In contrast to UC patients on Infliximab, switches were primarily to Ustekinumab (18.3%), Adalimumab (15.1%), and Vedolizumab (14%), and switching to another anti-TNF agent, such as Adalimumab, is less effective. Instead, guidelines recommend switching to a biologic with a different mechanism of action, like ustekinumab or tofacitinib. These agents have demonstrated greater efficacy than vedolizumab or adalimumab in this context. This is supported by network meta-analyses showing that, among patients with prior anti-TNF exposure, ustekinumab and tofacitinib are superior to vedolizumab for inducing clinical remission, and adalimumab is not preferred.[24] In addition to these international guidelines, regional data from Saudi Arabia and the Middle East have also reported on biologic use and switching outcomes, which complement our findings.[16,17,18] Persistence times varied, with CD patients on Infliximab lasting longer (69 months) before switching to Adalimumab (60 months), while UC patients on Infliximab lasted 47 months compared to Adalimumab’s 23 months. Risk factors for switching included a family history of IBD in CD patients (HR 1.459). For UC patients, being aged 18–30 years (HR 5.939) increased the risk of switching, while being single decreased it (HR 0.496) compared to married patients.
Primary nonresponse (PNR) to anti-TNF therapy is often defined as the failure to achieve clinical improvement within 14 weeks of treatment initiation[24,25,26,27,28] affecting 10–40% of patients with IBD.[29,30,31] PNR may result from pharmacokinetic factors, such as increased drug clearance or intestinal drug loss, or pharmacodynamic factors, such as persistent disease activity despite therapeutic drug levels. Pharmacodynamic PNR may occur due to mechanisms such as blocked TNF binding, noninflammatory complications (e.g., abscesses or infections), or underlying diseases driven by other inflammatory mediators. Low albumin levels, commonly seen in severe IBD like acute ulcerative colitis (UC), correlate with reduced infliximab levels and poor response.[32,33,34,35] In contrast, loss of response (LOR) occurs when patients initially respond to anti-TNF therapy but progressively lose efficacy over time. LOR affects up to 50% of patients, with an annual incidence of 5–20%, varying due to differing definitions across studies. These definitions include worsening symptoms, dose escalation, or increased inflammation markers. LOR may stem from low serum drug concentrations or anti-drug antibodies (ADAbs) that lower TNF-binding capacity or drug levels. Additionally, a shift in disease pathophysiology, such as the involvement of alternative cytokine pathways, may contribute to LOR.[36,37,38,39,40]
In patients with suspected PNR or LOR to anti-TNF therapy, guidelines recommend a comprehensive evaluation to identify the underlying cause, as this will inform therapeutic strategies. The primary step is to determine whether symptom exacerbation results from an actual increase in IBD activity or an alternative etiology. Differential diagnoses for symptom escalation that require exclusion include GI infections, IBS, SIBO, and BAM, which are frequently observed in individuals with CD, characterized by extensive ileal involvement or prior ileal resection. The next step involves assessing the extent of disease-associated inflammatory activity.[24,25,26,27,28]
The application of Therapeutic Drug Monitoring (TDM) facilitates decisions regarding intra-class switching or transitioning to an alternative anti-TNF agent. In cases where ADAbs have developed but the patient previously responded to an anti-TNF, transitioning to a different anti-TNF remains a viable option as ADAbs are specific to individual therapeutic molecules (biosimilars being equivalent to the reference anti-TNF in this context). This strategy represents an effective alternative. Evidence suggests that intra-class switching to a different anti-TNF following LOR is feasible for a subset of patients, with TDM playing a critical role in identifying suitable candidates.[7] Additionally, limited studies have shown clinical efficacy with a third or even a fourth anti-TNF in patients with CD, who have failed two or more prior anti-TNF therapies.[41,42,43] However, with the introduction of agents with alternative mechanisms of action, this strategy is now less commonly utilized. It may be considered primarily for patients with extra-intestinal manifestations (EIMs), particularly when subtherapeutic drug levels and elevated ADAbs are observed. This factor was incorporated into our study, prompting an inferential analysis of transitions between first- to second-, second- to third-, and even third- to fourth-line treatments. Additionally, we assessed the treatment line associated with the highest persistence, as previously detailed.
If a patient fails to achieve an adequate therapeutic response with an anti-TNF agent despite having therapeutic or supratherapeutic drug levels, selecting an agent from a different therapeutic class is a viable strategy. Agents with alternative mechanisms of action, such as vedolizumab, ustekinumab, and tofacitinib (approved solely for UC), may be considered.[27,28,29,30,31] The efficacy of therapy switching is contingent on the type of treatment and prior therapies. Singh et al.[44] reported that patients with PNR to anti-TNF agents were less likely to respond to second-line non-TNF biologics compared to those who discontinued treatment due to intolerance. Furthermore, patients with PNR were less likely to respond to second-line ustekinumab than those with LOR, although no significant difference was observed with vedolizumab. These variations may be attributed to the pharmacokinetics and pharmacodynamics of anti-TNFs in patients with PNR. These findings are consistent with previous real-world studies that have reported similar switching patterns, treatment responses, and biologic persistence outcomes among patients with inflammatory bowel disease, who were previously exposed to anti-TNF therapy.[45,46]
Biologic-naïve patients typically exhibit better responses to therapy than those with prior anti-TNF exposure. Post-hoc analyses from the GEMINI studies demonstrated that vedolizumab had higher response and remission rates in biologic-naïve CD patients, whereas a meta-analysis indicated that ustekinumab was effective in anti-TNF-experienced patients, with the exception of those with PNR.[42,43] Non-anti-TNF biologics like Ustekinumab and vedolizumab demonstrate efficacy irrespective of prior anti-TNF immunogenicity, and some address IBD-associated EIMs.[47,48] However, PNR cases may indicate severe disease, which complicates treatment outcomes.[49] In our study, vedolizumab and Ustekinumab were predominantly used as third-line treatments, compared to their use as first-, second-, or fourth-line therapies. Consequently, their use was primarily observed in patients who had been previously exposed to TNF inhibitors, supporting previous literature findings. In our KM analysis, ustekinumab demonstrated superior persistence as a third-line therapy in CD, whereas no significant differences in persistence were observed among agents in UC.
In our investigation of drug persistence within our patient cohort, we initially calculated the proportion of patients transitioning from one biological to another. We observed results consistent with previously reported data, indicating higher discontinuation rates for TNF-α inhibitors compared to other biologic classes.[37,47,50] Additionally, Kaplan–Meier analysis was employed to evaluate survival across different biological classes. Interestingly, we found that infliximab had the highest persistence, followed by adalimumab and vedolizumab, when used as the first line, before shifting in either UC or CD. Additionally, Ustekinumab had higher persistence than Vedolizumab when used as the third line before moving to the fourth line in CD. However, previous literature has not found a significant difference in persistence times between biologic types in either CD or UC classes.[37,47,50] This was also the case in Mosli’s study, conducted in Saudi Arabia.[18] However, the discrepancy between our findings could be attributed to different initial proportions of use for each type of drug, which may have made the patient naive to TNF.
Last, we analyzed shifting from second to third and even from third to fourth-line drugs. Biologic switching over time predominantly occurs due to treatment failure or intolerance.[49] As described in the Methods, our use of “switching” encompasses both intra-class switches and inter-class swaps, ensuring consistency with the existing literature. To explore predictors of treatment failure, we assessed factors influencing biological switching. Using multiple logistic regression analyses, we identified specific predictors. In CD, a family history of IBD significantly increased the likelihood of switching (HR 1.459). For UC, patients aged 18–30 exhibited a markedly higher switching risk (HR 5.939), while single marital status reduced this risk (HR 0.496) compared to married patients. In UC, we observed that being single was associated with a lower hazard of switching, whereas age 18–30 years showed a higher hazard. The marital-status finding likely reflects behavioral and management factors rather than a biological effect. Married patients may engage more frequently with care and pursue treat-to-target escalation—particularly around family planning—leading clinicians to recognize partial response earlier and switch sooner. Conversely, single patients may persist longer on the index agent or discontinue therapy without switching, which can artefactually lower the measured switching hazard in standard Cox models. We therefore interpret the “single” association as a marker of healthcare-seeking and treatment-pathway differences, and we note the potential for residual confounding (e.g., socioeconomic status, disease-severity proxies not captured, pregnancy-related decision making) and competing risks.
The higher switching risk among younger adults is compatible with more proactive escalation in this group and possibly a more inflammatory phenotype; however, the confidence interval was wide, indicating imprecision and the need for cautious interpretation. In sensitivity analyses that additionally adjusted for the index biologic and calendar period, results were directionally similar. Taken together, these predictors should be viewed as markers that help flag patients who may enter earlier escalation pathways rather than definitive biological risk factors for loss of response.[24,26,39,51]
Additionally, hypoalbuminemia in this group can accelerate infliximab clearance and contribute to early treatment failure.[32,35,39,51] It is also possible that these patterns are shaped in part by unmeasured confounding—such as disease extent, biomarker severity, or healthcare access—that is difficult to fully account for in observational studies.[19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40] Still, Mosli’s study regression differed from ours as follows: They identified predictors of drug changes in IBD patients with a history of VTE and active smoking influencing changes across all subtypes.[18] For CD, diagnosis during early adulthood and the presence of EIMs were associated with drug changes, whereas in UC, selecting vedolizumab as the first biologic, male gender, and a history of VTE were significant factors. These findings highlight high-risk markers and drug resistance predictors. The discrepancies with our regression findings could stem from differences in statistical methods, sample characteristics, or unmeasured confounders.
To our knowledge, our study represents the largest and most comprehensive analysis of its kind in the Gulf region. With a robust analytical approach and a larger sample size compared to similar studies, we assessed switching patterns up to the fourth line of treatment. In contrast, most prior research has focused on persistence with first-line therapies. Our work also highlighted the potential role of biologics beyond TNF inhibitors, paving the way for comparisons with established agents such as infliximab and adalimumab. Such insights are crucial, as previously discussed, for understanding the utility of these drugs in non-naïve patients and their potential benefit in PNR or LOR cases.
However, our study is not without limitations. First, the retrospective design limits generalizability as causality cannot be established, and controlling for confounding factors is challenging despite efforts to address this through robust analyses, including logistic regression and Kaplan–Meier analysis. We applied various tests (log-rank, Breslow, and Tarone–Ware), each selected for its sensitivity to the timing of events. This was important due to the differing rates of events associated with each drug. Second, additional data were needed to investigate the precise causes of transitions, including antibody levels, immunomodulator use, and their influence on switching, but such information was unavailable. Third, a limitation of our analysis is the small number of patients exposed to certain biologics, such as certolizumab (n = 3) and tofacitinib (n = 2), which restricts the reliability of subgroup inferences. These findings should therefore be interpreted with caution. Fourth, conducting the study at a single center restricts its generalizability as we could not assess patients across diverse regions with varying treatment protocols and demographics, which could influence our study’s hypotheses.
Finally, the demographic data in our study, with over 97% of patients being Saudi nationals residing in Riyadh, the capital of Saudi Arabia, and with negligible alcohol consumption, reflect the cultural features of IBD in Saudis. All the same, most patients in Gulf countries share regional features, including a rising prevalence of IBD and low alcohol consumption, supporting the broader relevance of our results across the Gulf countries.
In conclusion, our study examined medication use and switching patterns among patients with CD and UC. In both cohorts, infliximab and adalimumab were the most common biologics, with primary failure being the primary reason for discontinuation. In CD patients, Infliximab was commonly switched to Adalimumab and Ustekinumab, while in UC patients, switches were primarily to Adalimumab, Vedolizumab, and Ustekinumab. Persistence times varied, with CD patients on Infliximab lasting longer before switching compared to Adalimumab, with UC patients showing similar patterns. Risk factors for switching included a family history of IBD in CD patients, while for UC patients, being aged 18–30 years increased the risk, and being single decreased the risk compared to married patients. These findings carry important implications for clinical practice in the Gulf region. They highlight the need for the earlier adoption of non-TNF biologics (ustekinumab and vedolizumab) in younger UC patients following infliximab failure, the integration of therapeutic drug monitoring to distinguish between pharmacokinetic and pharmacodynamic failure, thereby minimizing unnecessary switching, and the development of cost-conscious treatment algorithms that optimize the use of biosimilars in resource-limited settings. Future research should focus on multicenter validation and the impact of newer biologics. These findings support the development of personalized treatment algorithms and underscore the importance of targeted monitoring for high-risk patients to optimize therapeutic outcomes.
Conflicts of interest
There are no conflicts of interest.
Kaplan-Meier Curve for Switching of Biologics From Second to Third drug among patients with Crohn's Disease. Median persistence times were 54 months for infliximab, 49 months for adalimumab, 36 months for vedolizumab, and 32 months for ustekinumab, with no statistically significant differences across groups (log-rank P > 0.05)
Kaplan-Meier Curve for Switching of Biologics From Second to Third drug among patients with Ulcerative Colitis. Second to third biologic: Median persistence times were 35 months for adalimumab, 26 months for infliximab, 24 months for vedolizumab, and 22 months for ustekinumab. No significant survival differences were observed (log-rank P > 0.05)
Kaplan-Meier Curve for Switching of Biologics from Third to Fourth drug among patients with Ulcerative Colitis. Third to fourth biologic: Vedolizumab (median 54 months) and ustekinumab (51 months) had the longest persistence compared with tofacitinib and upadacitinib, with no statistically significant survival differences across groups (log-rank P > 0.05)
Funding Statement
Nil.
REFERENCES
- 1.Gomollón F, Dignass A, Annese V, Tilg H, Van Assche G, Lindsay JO, et al. 3rd European evidence-based consensus on the diagnosis and management of Crohn's disease 2016: Part 1: Diagnosis and medical management. J Crohns Colitis. 2017;11:3–25. doi: 10.1093/ecco-jcc/jjw168. [DOI] [PubMed] [Google Scholar]
- 2.Atreya R, Neurath MF, Siegmund B. Personalizing treatment in IBD: Hype or reality in 2020?Can we predict response to anti-TNF? Front Med (Lausanne) 2020;7:517. doi: 10.3389/fmed.2020.00517. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Eder P, Katulska K, Lykowska-Szuber L, Stawczyk-Eder K, Krela-Kaźmierczak I, Klimczak K, et al. Magnetic resonance enterographic predictors of one-year outcome in ileal and ileocolonic Crohn's disease treated with anti-tumor necrosis factor antibodies. Sci Rep. 2015;5:10223. doi: 10.1038/srep10223. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Sochał M, Krzywdzińska M, Gabryel M, Małecka-Panas E, Cichoż-Lach H, Garlicki A, et al. A simple index to predict the efficiency of adalimumab treatment in Crohn's disease with a limited duration of therapy. Pol Arch Intern Med. 2020;130:524–31. doi: 10.20452/pamw.15507. [DOI] [PubMed] [Google Scholar]
- 5.Stidham RW, Lee TC, Higgins PDR, Deshpande AR, Sussman DA, Singal AG, et al. Systematic review with network meta-analysis: The efficacy of anti-TNF agents for the treatment of Crohn's disease. Aliment Pharmacol Ther. 2014;39:1349–62. doi: 10.1111/apt.12749. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Sochał M, Gabryel M, Krzywdzińska M, Białas M, Fichna J, Małecka-Panas E, et al. Efficiency and safety of one-year anti-TNF-α treatment in Crohn's disease: A Polish single-centre experience. Prz Gastroenterol. 2014;9:33–9. doi: 10.5114/pg.2019.90079. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Gisbert JP, Marín AC, Chaparro M. Systematic review with meta-analysis: The efficacy of a second anti-TNF in patients with inflammatory bowel disease whose previous anti-TNF treatment has failed. Aliment Pharmacol Ther. 2015;41:613–23. doi: 10.1111/apt.13083. [DOI] [PubMed] [Google Scholar]
- 8.Ma C, Fedorak RN, Kaplan GG, Dieleman LA, Devlin SM, Stern N, et al. Systematic review: The short-term and long-term efficacy of adalimumab following discontinuation of infliximab. Aliment Pharmacol Ther. 2014;39:760–70. doi: 10.1111/j.1365-2036.2009.04101.x. [DOI] [PubMed] [Google Scholar]
- 9.Gisbert JP, Chaparro M. Predictors of primary response to biologic treatment in patients with inflammatory bowel disease: From basic science to clinical practice. J Crohns Colitis. 2020;14:694–709. doi: 10.1093/ecco-jcc/jjz195. [DOI] [PubMed] [Google Scholar]
- 10.Gisbert JP, Chaparro M. Primary failure to an anti-TNF agent in inflammatory bowel disease: Switch or swap? J Clin Med. 2021;10:5318. doi: 10.3390/jcm10225318. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Gisbert JP, Domènech E. Ustekinumab in the treatment of Crohn's disease. Gastroenterol Hepatol. 2017;40:688–98. doi: 10.1016/j.gastrohep.2017.08.006. [DOI] [PubMed] [Google Scholar]
- 12.Gisbert JP, Domènech E. Vedolizumab in the treatment of Crohn's disease. Gastroenterol Hepatol. 2015;38:338–48. doi: 10.1016/j.gastrohep.2014.12.003. [DOI] [PubMed] [Google Scholar]
- 13.Panés J, Gisbert JP. Efficacy of tofacitinib in the treatment of ulcerative colitis. Gastroenterol Hepatol. 2019;42:403–12. doi: 10.1016/j.gastrohep.2019.03.002. [DOI] [PubMed] [Google Scholar]
- 14.Barré A, Colombel JF, Ungaro R. Review article: Predictors of response to vedolizumab and ustekinumab in inflammatory bowel disease. Aliment Pharmacol Ther. 2018;47:896–905. doi: 10.1111/apt.14550. [DOI] [PubMed] [Google Scholar]
- 15.Campochiaro C, Ramoni V, Pucci A, Tomelleri A, Galli E, Cocchiara E, et al. Failure of first anti-TNF agent in Takayasu's arteritis: To switch or to swap? Clin Exp Rheumatol. 2021;39((Suppl 129)):129–34. doi: 10.55563/clinexprheumatol/1xi8ag. [DOI] [PubMed] [Google Scholar]
- 16.Alharbi O, Alkhathlan M, Almadi M, Alharbi A, Azzam N, Aljebreen A, et al. Predictors of anti-TNF failure in inflammatory bowel disease: A Saudi tertiary center experience. J Clin Med. 2022;11:4157. doi: 10.3390/jcm11144157. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Al Sulais E, Mosli M, Abou Rached A, Sanai FM, AlAmeel T, AlHarthi R, et al. Effectiveness of biologic therapies in Middle Eastern patients with inflammatory bowel disease (ENROLL study) Inflamm Bowel Dis. 2024;30:907–16. [Google Scholar]
- 18.Mosli M. A retrospective observational study of patterns of biologic drug change in inflammatory bowel disease. Inflamm Intest Dis. 2024;9:71–84. doi: 10.1159/000538250. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Cuschieri S. The STROBE guidelines. Saudi J Anaesth. 2019;13((Suppl 1)):S31–4. doi: 10.4103/sja.SJA_543_18. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Mahlich J, Manceur AM, Nishioka D, Sruamsiri R, Botteman MF, Sruamsiri R, et al. Persistence with biologic therapy and associated costs of patients with inflammatory bowel disease: A German retrospective claims data analysis. Expert Rev Pharmacoecon Outcomes Res. 2021;21:1085–94. [Google Scholar]
- 21.In J, Lee DK. Survival analysis: Part I—analysis of time-to-event. Korean J Anesthesiol. 2018;71:182–91. doi: 10.4097/kja.d.18.00067. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Ilker E. Choosing statistical tests for survival analysis. Biom Biostat Int J. 2018;7:249–52. [Google Scholar]
- 23.Feuerstein JD, Ho EY, Shmidt E, Singh H, Falck-Ytter Y, Sultan S, et al. AGA clinical practice guidelines on the medical management of moderate to severe luminal and perianal fistulizing Crohn's disease. Gastroenterology. 2021;160:2496–508. doi: 10.1053/j.gastro.2021.04.022. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Feuerstein JD, Isaacs KL, Schneider Y, Siddique SM, Falck-Ytter Y, Singh S, et al. AGA clinical practice guidelines on the management of moderate to severe ulcerative colitis. Gastroenterology. 2020;158:1450–61. doi: 10.1053/j.gastro.2020.01.006. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Harbord M, Eliakim R, Bettenworth D, Karmiris K, Katsanos K, Kopylov U, et al. Third European evidence-based consensus on diagnosis and management of ulcerative colitis. Part 2: Current management. J Crohns Colitis. 2017;11:769–84. doi: 10.1093/ecco-jcc/jjx009. [DOI] [PubMed] [Google Scholar]
- 26.Magro F, Gionchetti P, Eliakim R, Ardizzone S, Armuzzi A, Acosta MB, et al. Third European evidence-based consensus on diagnosis and management of ulcerative colitis. Part 1: Definitions, diagnosis, extra-intestinal manifestations, pregnancy, cancer surveillance, surgery, and ileo-anal pouch disorders. J Crohns Colitis. 2017;11:649–70. doi: 10.1093/ecco-jcc/jjx008. [DOI] [PubMed] [Google Scholar]
- 27.Terdiman JP, Gruss CB, Heidelbaugh JJ, Sultan S, Falck-Ytter YT, et al. AGA Institute Clinical Guidelines Committee. American Gastroenterological Association Institute guideline on the use of thiopurines, methotrexate, and anti-TNF-α biologic drugs for the induction and maintenance of remission in inflammatory Crohn's disease. Gastroenterology. 2013;145:1459–63. doi: 10.1053/j.gastro.2013.10.047. [DOI] [PubMed] [Google Scholar]
- 28.Torres J, Bonovas S, Doherty G, Kucharzik T, Gisbert JP, Raine T, et al. ECCO guidelines on therapeutics in Crohn's disease: Medical treatment. J Crohns Colitis. 2020;14:4–22. doi: 10.1093/ecco-jcc/jjz180. [DOI] [PubMed] [Google Scholar]
- 29.Kassouri L, Pariente B, Bejan-Angoulvant T, Amiot A, Altwegg R, Roblin X, et al. The outcome of Crohn's disease patients refractory to anti-TNF and either vedolizumab or ustekinumab. Dig Liver Dis. 2020;52:1148–55. doi: 10.1016/j.dld.2020.07.031. [DOI] [PubMed] [Google Scholar]
- 30.Panaccione R, Ghosh S. Optimal use of biologics in the management of Crohn's disease. Ther Adv Gastroenterol. 2010;3:179–89. doi: 10.1177/1756283X09357579. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Alghamdi AG, Aldofaian HS, Alahmari MS, Alibrahim BK, Alshowair MA, Alhamidi HA, et al. New biologics and small molecules in inflammatory bowel disease: An update. Ther Adv Gastroenterol. 2019;12:1756284819853208. doi: 10.1177/1756284819853208. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Brandse JF, van den Brink GR, Wildenberg ME, van der Kleij D, Rispens T, Jansen JM, et al. Loss of infliximab into feces is associated with lack of response to therapy in patients with severe ulcerative colitis. Gastroenterology. 2015;149:350–5.e2. doi: 10.1053/j.gastro.2015.04.016. [DOI] [PubMed] [Google Scholar]
- 33.Fasanmade AA, Adedokun OJ, Blank M, Zhou H, Davis HM, Johanns J, et al. Population pharmacokinetic analysis of infliximab in patients with ulcerative colitis. Eur J Clin Pharmacol. 2009;65:1211–28. doi: 10.1007/s00228-009-0718-4. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Sparrow MP, Papamichael K, Ward MG, Rentsch C, Bredenoord AJ, Hart A, et al. Therapeutic drug monitoring of biologics during induction to prevent primary non-response. J Crohns Colitis. 2020;14:542–56. doi: 10.1093/ecco-jcc/jjz162. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35.Syal G, Berzin TM, Staller K, Pleskow DK, Chuttani R, Sawhney MS, et al. Hypoalbuminemia and bandemia predict failure of infliximab rescue therapy in acute severe ulcerative colitis. Dig Dis Sci. 2021;66:199–205. doi: 10.1007/s10620-020-06177-7. [DOI] [PubMed] [Google Scholar]
- 36.Ben-Horin S, Chowers Y. Review article: Loss of response to anti-TNF treatments in Crohn's disease. Aliment Pharmacol Ther. 2011;33:987–95. doi: 10.1111/j.1365-2036.2011.04612.x. [DOI] [PubMed] [Google Scholar]
- 37.Gisbert JP, Panés J. Loss of response and requirement of infliximab dose intensification in Crohn's disease: A review. Am J Gastroenterol. 2009;104:760–7. doi: 10.1038/ajg.2008.88. [DOI] [PubMed] [Google Scholar]
- 38.Kennedy NA, Heap GA, Green HD, Hamilton B, Bewshea C, Walker GJ, et al. Predictors of anti-TNF treatment failure in anti-TNF-naive patients with active luminal Crohn's disease: A prospective, multicentre, cohort study. Lancet Gastroenterol Hepatol. 2019;4:341–53. doi: 10.1016/S2468-1253(19)30012-3. [DOI] [PubMed] [Google Scholar]
- 39.Lamb CA, Kennedy NA, Raine T, Hendy P, Smith PJ, Limdi JK, et al. British Society of Gastroenterology consensus guidelines on the management of inflammatory bowel disease in adults. Gut. 2019;68((Suppl 3)):s1–106. doi: 10.1136/gutjnl-2019-318484. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 40.Roda G, Jharap B, Neeraj N, Colombel JF. Loss of response to anti-TNFs: Definition, epidemiology, and management. Clin Transl Gastroenterol. 2016;7:e135. doi: 10.1038/ctg.2015.63. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 41.Allez M, Karmiris K, Louis E, Van Assche G, Ben-Horin S, Klein A, et al. The efficacy and safety of a third anti-TNF monoclonal antibody in Crohn's disease after failure of two other anti-TNF antibodies. Aliment Pharmacol Ther. 2010;31:92–101. doi: 10.1111/j.1365-2036.2009.04130.x. [DOI] [PubMed] [Google Scholar]
- 42.De Silva PSA, Nguyen DD, Sauk J, Korzenik J, Yajnik V, Ananthakrishnan AN. Long-term outcome of a third anti-TNF monoclonal antibody after the failure of two prior anti-TNFs in inflammatory bowel disease. Aliment Pharmacol Ther. 2012;36:459–66. doi: 10.1111/j.1365-2036.2012.05214.x. [DOI] [PubMed] [Google Scholar]
- 43.Russi L, Scharl M, Rogler G, Biedermann L. The efficacy and safety of golimumab as third- or fourth-line anti-TNF therapy in patients with refractory Crohn's disease: A case series. Inflamm Intest Dis. 2017;2:131–8. doi: 10.1159/000481400. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 44.Singh S, Murad MH, Fumery M, Dulai PS, Sandborn WJ. Primary non-response to tumor necrosis factor antagonists is associated with inferior response to second-line biologics in patients with inflammatory bowel diseases: A systematic review and meta-analysis. J Crohns Colitis. 2018;12:635–43. doi: 10.1093/ecco-jcc/jjy004. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 45.Kawalec P, Mikrut A, Wiśniewska N, Pilc A. Efficacy and safety of ustekinumab in the induction therapy of TNF-α-refractory Crohn's disease patients: A systematic review and meta-analysis. J Comp Eff Res. 2017;6:601–12. doi: 10.2217/cer-2017-0022. [DOI] [PubMed] [Google Scholar]
- 46.Sands BE, Peyrin-Biroulet L, Loftus EV, Jr, Danese S, Colombel JF, Törüner M, et al. Vedolizumab as induction and maintenance therapy for Crohn's disease in patients naive to or who have failed tumor necrosis factor antagonist therapy. Inflamm Bowel Dis. 2017;23:97–106. doi: 10.1097/MIB.0000000000000979. [DOI] [PubMed] [Google Scholar]
- 47.Dalal RS, Esckilsen S, Syal G, Farah K, Malter LB, Long MD, et al. Comparative long-term drug survival of vedolizumab, adalimumab, and infliximab in biologic-naive patients with ulcerative colitis. Dig Dis Sci. 2023;68:223–32. doi: 10.1007/s10620-022-07472-1. [DOI] [PubMed] [Google Scholar]
- 48.Wu J, Smogorzewski J. Ustekinumab for the treatment of paradoxical skin reactions and cutaneous manifestations of inflammatory bowel diseases. Dermatol Ther. 2021;34:e14883. doi: 10.1111/dth.14883. [DOI] [PubMed] [Google Scholar]
- 49.Marsal J, Kaplan GG. Management of non-response and loss of response to anti-tumor necrosis factor therapy in inflammatory bowel disease. Front Med (Lausanne) 2022;9:949226. doi: 10.3389/fmed.2022.897936. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 50.Singh S, Fumery M, Sandborn WJ, Murad MH. First- and second-line pharmacotherapies for patients with moderate to severely active ulcerative colitis: An updated network meta-analysis. Clin Gastroenterol Hepatol. 2020;18:2179–91.e6. doi: 10.1016/j.cgh.2020.01.008. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 51.Gros B, Kaplan GG. Ulcerative colitis in adults: A review. JAMA. 2023;330:951–65. doi: 10.1001/jama.2023.15389. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Kaplan-Meier Curve for Switching of Biologics From Second to Third drug among patients with Crohn's Disease. Median persistence times were 54 months for infliximab, 49 months for adalimumab, 36 months for vedolizumab, and 32 months for ustekinumab, with no statistically significant differences across groups (log-rank P > 0.05)
Kaplan-Meier Curve for Switching of Biologics From Second to Third drug among patients with Ulcerative Colitis. Second to third biologic: Median persistence times were 35 months for adalimumab, 26 months for infliximab, 24 months for vedolizumab, and 22 months for ustekinumab. No significant survival differences were observed (log-rank P > 0.05)
Kaplan-Meier Curve for Switching of Biologics from Third to Fourth drug among patients with Ulcerative Colitis. Third to fourth biologic: Vedolizumab (median 54 months) and ustekinumab (51 months) had the longest persistence compared with tofacitinib and upadacitinib, with no statistically significant survival differences across groups (log-rank P > 0.05)