Abstract
Background
In patients with inflammatory bowel disease (IBD), multimorbidity with obesity and type 2 diabetes is common and increasing. Glucagon-like peptide 1 (GLP-1) receptor agonists are increasingly being prescribed for patients with IBD, yet their impact on patients with IBD is largely unknown. We aimed to assess the impact of GLP-1 receptor agonists on the course of IBD.
Methods
We identified all IBD patients prescribed GLP-1 receptor agonists at a large academic healthcare network between 2009 and 2023. We analyzed demographics and IBD characteristics in the year pre- and post–GLP-1 receptor agonist prescription and matched them to non-IBD controls. Our primary outcome was IBD exacerbation in the year following GLP-1 receptor agonist initiation, measured as a composite of IBD-related hospitalization, corticosteroid prescription, medication escalation or changes, or IBD-related surgery. Secondary outcomes included change in metabolic risk factors.
Results
Overall, 224 patients met inclusion criteria. At GLP-1 receptor agonist initiation, the median age was 54 years, 63% were female, 77% were White, and median BMI was 33.2 kg/m2. Compared to the 12-month period prior to GLP-1 receptor agonist initiation, in the 12 months post–GLP-1 receptor agonist initiation, there was no change in rates of IBD exacerbation, IBD-related hospitalization, steroids prescription, medication escalation or changes, or IBD-related surgery. There was a significant decrease in BMI in the year following GLP-1 receptor agonist initiation (median BMI 33.5 vs 31.6 kg/m2, P < .01), with rates of decrease comparable to non-IBD matched controls.
Conclusions
In patients with IBD, GLP-1 receptor agonists are effective for weight loss and associated with few episodes of disease exacerbation.
Keywords: IBD, GLP-1 receptor agonists, obesity, healthcare maintenance
Key Messages.
GLP-1 receptor agonists are effective weight-lowering drugs and decrease gut inflammation in preclinical trials, yet their safety and efficacy among patients with IBD were previously unknown.
Our study demonstrates that GLP-1 receptor agonists carry no increased risk of IBD exacerbation in the year following GLP-1 receptor agonist initiation, and are effective for weight loss among IBD patients.
Given these findings, GLP-1 receptor agonists can be comfortably prescribed for IBD patients.
Introduction
Inflammatory bowel disease (IBD), including Crohn’s Disease (CD) and Ulcerative Colitis (UC), is chronic immune-mediated inflammatory diseases of the gastrointestinal tract affecting millions of individuals worldwide.1 The precise etiology of IBD is incompletely understood and likely includes genetic susceptibility, environmental factors, and immunological abnormalities.2,3 Globally, the incidence and compound prevalence of IBD is increasing. As patients with IBD age, there is increasing multimorbidity with conditions such as obesity and diabetes.4 The impact of obesity on IBD clinical outcomes has been examined. A recent study demonstrated decreased responsiveness to advanced IBD therapies among patients with increased visceral adiposity.5 Methods adopted to target obesity in IBD patients include lifestyle modifications, dietary modifications, as well as pharmacotherapy. Despite promising results of anti-obesity pharmacotherapy in non-IBD patients, studies have demonstrated low utilization of anti-obesity pharmacotherapy in IBD patients.6
Glucagon-like peptide 1 (GLP-1) is a peptide produced by intestinal L-cells, a type of enteroendocrine cell (EEC) primarily found in the small intestine and colon.7 The secretion of GLP-1 has a number of effects; the most important of these are (1) increasing insulin secretion, (2) decreasing glucagon secretion, and (3) delaying gastric emptying.8,9 Glucagon-like peptide 1 receptor agonists are a relatively new class of glucose-lowering medications originally approved for the treatment of type 2 diabetes mellitus. Remarkably, the class of medication has demonstrated cardioprotective as well as weight-lowering properties.10 Weight loss is particularly relevant for patients with IBD, given that obesity is a proinflammatory state which may worsen the course of IBD.9,11 GLP-1 receptor agonists likely activate several pathways to exert their beneficial effects. In the gut, GLP-1 receptor agonists may reduce inflammation by binding to GLP-1 receptors on intestinal intraepithelial lymphocytes.12 Animal models have demonstrated improvement in gut inflammation in mice with GLP-1 receptor agonists,13,14 but studies in humans are limited. Recently presented data demonstrated that GLP-1 receptor agonists are well tolerated in IBD patients, and may reduce levels of C-reactive protein (CRP).15,16
Despite their widespread use, the safety and efficacy of GLP-1 receptor agonism in IBD patients are incompletely understood. To our knowledge, only a single study has evaluated GLP-1 receptor agonists compared to other anti-diabetic medications in a cohort of diabetic IBD patients.17 The impact of GLP-1 receptor agonists on IBD outcomes in nondiabetic IBD patients has not been previously evaluated. Furthermore, the comparative effectiveness of GLP-1 receptor agonists in terms of weight loss among IBD patients compared to non-IBD patients has not been previously studied. Due to alterations in the gut epithelial lining among IBD patients, it is possible for decreased efficacy of GLP-1 receptor agonists among IBD patients. The aim of the present study was to (1) determine the safety and impact of GLP-1 receptor agonists on IBD outcomes in both diabetic and nondiabetic IBD patients and (2) to examine the effectiveness of GLP-1 receptor agonists on weight loss in IBD patients compared to non-IBD controls.
Materials and Methods
Study Design and Population
We performed a before-and-after observational study of adult patients prescribed a GLP-1 receptor agonist. The NYU Langone Health electronic health record (EHR) system, consisting of data from January 1, 2009 to October 1, 2023, was queried to create a population of adult patients (>18 years old) with a history of IBD identified as having at least one instance of ICD10 code K50 or K51. To identify the reference cohort (N = 422), patients with IBD were further assessed for at least one instance of the following medications—semaglutide, liraglutide, tirzepatide, wegovy, ozempic, rybelsus, trulicity, dulaglutide, victoza, byetta, exenatide, bydureon, and mounjaro. Patients were considered to be utilizing a GLP-1 receptor agonist if the medication remained active in the electronic medical record medication list during the duration of the study, which was manually confirmed in individual patient notes. To identify non-IBD controls, the NYU Langone Health EHR system was queried for adult patients (>18 years old) who were prescribed GLP-1 receptor agonists without a diagnosis of IBD (no ICD10 code of K50 or K51). Controls were matched for age, sex, and presence of diabetes. Individual charts were reviewed to confirm IBD diagnoses and GLP-1 receptor agonist prescribing. Patients were excluded if they did not have outpatient follow-up within the NYU Langone Health system.
Data Collection
Baseline demographic information including sex, age, and medical history were collected. Disease-specific information including the extent of the disease by Montreal classification, disease duration, previous IBD-related surgery, previous IBD-related hospitalization, and previous IBD medication use were collected. Medication use, laboratory values, and body mass index (BMI) were collected at the time of the first GLP-1 receptor agonist prescription, as well as 6-month and 12-month post-prescription.
To provide an internal control for the IBD cohort, charts were reviewed for the 12-month period prior to GLP-1 receptor agonist prescription to collect information on IBD medication use, IBD-related hospitalizations, IBD-related surgery, and corticosteroid use. For non-IBD controls, weight and laboratory values were collected at the time of GLP-1 receptor agonist prescription, as well as 6-month and 12-month post-prescription. Values were considered present at the time of GLP-1 receptor agonist prescription if they were coded 30 days before or after GLP-1 receptor agonist prescription. Laboratory values were considered in the year prior to GLP-1 receptor agonist prescription if they were entered 180-365 days prior to GLP-1 receptor agonist prescription. Values were considered present 6 months post–GLP-1 receptor agonist prescription if they were entered 90-180 days post–GLP-1 receptor agonist prescription; and 1-year post–GLP-1 receptor agonist prescription if they were entered 181-365 days post–GLP-1 receptor agonist prescription. For patients with less than 12 months of GLP-1 receptor agonist prescription by January 1, 2024; data were used for the last entry prior to January 1, 2024.
Outcomes
The primary outcome was IBD exacerbation in the 12-month post–GLP-1 receptor agonist prescription. IBD exacerbation was defined as any of the following: (1) IBD-related surgery; (2) IBD-related hospitalization; (3) corticosteroid prescription, and (4) escalation from nonadvanced IBD therapy to advanced IBD therapy OR change within advanced IBD therapy. Advanced IBD therapy included the following medications or their biosimilars: infliximab, adalimumab, certolizumab, vedolizumab, ustekinumab, tofacitinib, upadacitinib, risankizumab, ozanimod, and golimumab. Rates of IBD exacerbation in the 12 months post–GLP-1 receptor agonist prescription were compared to rates of IBD exacerbation in the 12 months prior to GLP-1 receptor agonist prescription. Secondary analyses included examining each component individually, in addition to changes in metabolic risk factors and laboratory values over the 12-month period post–GLP-1 receptor agonist prescription.
The efficacy of GLP-1 receptor agonists among IBD patients was compared to non-IBD controls by comparing changes in BMI and laboratory values in the 12 months post–GLP-1 receptor agonist prescription among the cohorts.
Statistical Analysis
All continuous data were reported as medians with interquartile ranges (IQR), and categorical variables as frequencies (%). Among IBD patients, disease activity, as well as BMI and other lab values were compared at the time of GLP-1 receptor agonist prescription to 12 months post-prescription using Wilcoxon signed-rank tests for continuous variables and McNemar or McNemar-Bowker’s tests for categorical variables. Patients with less than 12 months of follow-up were analyzed from GLP-1 receptor agonist prescription to the latest follow-up date. Subgroup analyses were performed stratifying by months of follow-up (≥12 or <12 months) and by the presence of diabetes.
Baseline demographics and clinical factors were also compared among IBD cases and non-IBD controls using Wilcoxon Rank-Sum tests for continuous variables and Pearson’s Chi-Squared or Fisher’s Exact tests for categorical variables. Further, BMI and lab values were compared at the time of GLP-1 receptor agonist prescription to 12 months post, stratifying by the presence of IBD using Wilcoxon signed-rank tests.
Multiple linear regression models were conducted to determine the effect of IBD presence on changes in BMI, hemoglobin A1c, total cholesterol, LDL, and HDL, adjusting for race, smoking status, and presence of comorbidities. For the models assessing change in BMI and hemoglobin A1c, baseline BMI and hemoglobin A1c were also adjusted for, respectively. Adjusted regression coefficients (β) and corresponding 95% CIs were reported. For all analyses, a P-value < .05 was considered statistically significant. All statistical analyses were performed using Stata/BE 18.0 (StataCorp, College Station, TX, USA).
Ethical Considerations
Patient medical records were reviewed in order to obtain data for this study. Given that no patient identifiers were collected and the research carried minimal risk of harm to patients, informed consent was not obtained. This study was approved by the NYU Grossman School of Medicine Institutional Review Board.
Results
Study Population
In total, 224 patients with IBD were prescribed a GLP-1 receptor agonist and met the inclusion criteria. At the first GLP-1 receptor agonist prescription, the median age was 54 years (IQR 42-63), 36.6% were male, 77.7% were white, 63% had diabetes, and the median BMI was 33.2 kg/m2 (IQR 29.3-37.4; Table 1). 43.3% had UC, median IBD disease duration was 9.9 years (IQR 6.3-22.6), 43.3% were on an advanced IBD therapy, 29.9% had prior IBD-related hospitalization, and 17% had prior IBD-related surgery. Of GLP-1 receptor agonists, 66.1% of patients received semaglutide. IBD disease phenotype and GLP-1 receptor agonist characteristics are outlined in Table 1.
Table 1.
Population baseline characteristics at GLP-1 receptor agonist prescription, N = 224.
| Variables |
N (%) 224 (100.0) |
|---|---|
| Demographic characteristics at GLP-1 receptor agonist prescription | |
| Age, median (IQR) | 54 (42-63) |
| Female | 142 (63.4) |
| Race | |
| White | 174 (77.7) |
| Black | 15 (6.7) |
| Asian | 7 (3.1) |
| Other/unknown | 28 (12.5) |
| Smoking status | |
| Never smoker | 122 (54.5) |
| Current or former smoker | 55 (24.6) |
| Unknown | 47 (21.0) |
| BMI, median (IQR) | 33.2 (29.3-37.4) |
| Disease characteristics at diagnosis | |
| IBD disease duration in years, median (IQR) | 9.9 (6.3-22.6) |
| IBD subtype | |
| Ulcerative colitis (UC) | 97 (43.3) |
| UC disease extent | |
| E1 proctitis | 18 (18.6) |
| E2 left-sided colitis | 27 (27.8) |
| E3 extensive colitis | 48 (49.5) |
| Unspecified | 4 (4.1) |
| Crohn’s disease (CD) | 100 (44.6) |
| CD disease location | |
| Ileal | 24 (24.0) |
| Colon | 37 (37.0) |
| Ileocolonic | 35 (35.0) |
| Isolated upper GI disease | 0 (0) |
| Cutaneous Crohn’s with no luminal involvement | 3 (3.0) |
| Unspecified | 1 (1.0) |
| CD disease behavior | |
| Nonstricturing and nonpenetrating | 66 (66.0) |
| Stricturing | 8 (8.0) |
| Penetrating/fistulizing | 24 (24.0) |
| Unspecified | 2 (2.0) |
| Perianal disease | |
| No | 83 (83.0) |
| Yes | 17 (17.0) |
| Inflammatory bowel disease unclassified | 27 (12.1) |
| Prior clinical history | |
| Diabetes | 83 (37.1) |
| Hypertension | 99 (44.2) |
| Coronary artery disease | 30 (13.4) |
| Hyperlipidemia | 100 (44.6) |
| Prior IBD-related surgery | 38 (17.0) |
| Prior IBD-related hospitalization | 67 (29.9) |
| GLP-1 receptor agonist characteristics | |
| Type of GLP-1 receptor agonist | |
| Semaglutide | 148 (66.1) |
| Liraglutide | 47 (21.0) |
| Dulaglutide | 16 (7.1) |
| Tirzepatide | 12 (5.4) |
| Exenatide | 1 (0.5) |
IQR, interquartile range.
Rates of IBD Exacerbation
In the year prior to the first GLP-1 receptor agonist prescription, 34.4% experienced an IBD exacerbation (8.9% IBD-related hospitalization, 22.3% corticosteroid prescription, 1.3% IBD-related surgery, and 16.1% medication change or escalation). In the year post–GLP-1 receptor agonist prescription, 30.4% had an IBD exacerbation (7.6% IBD-related hospitalization, 18.3% corticosteroid prescription, 1.8% IBD-related surgery, and 12.5% medication change or escalation). Compared to the year prior to GLP-1 receptor agonist prescription, there was no significant difference in rates IBD exacerbation post–GLP-1 receptor agonist prescription (P = .36) or any of its components (Table 2). Subgroup analysis was performed for patients with less than 12 months of GLP-1 receptor agonist with similar findings (Supplementary Table 1).
Table 2.
Comparison of IBD disease activity and lab values at the time of GLP-1 receptor agonist prescription and post-prescription (5-12 months), N = 224.
| Variables | Time of GLP-1 receptor agonist prescription N (%) 224 (100.0) |
12 months post–GLP-1 receptor agonist prescription or latest follow-up N (%) 224 (100.0) |
P-value* |
|---|---|---|---|
| IBD disease activity 1 | |||
| IBD-related hospitalization in prior year (N = 224) | |||
| No | 204 (91.1) | 207 (92.4) | .70 |
| Yes | 20 (8.9) | 17 (7.6) | |
| Steroid prescription in prior year (N = 224) | |||
| No | 174 (77.7) | 183 (81.7) | .27 |
| Yes | 50 (22.3) | 41 (18.3) | |
| Surgery in prior year (N = 224) | |||
| No | 221 (98.7) | 220 (98.2) | 1.00 |
| Yes | 3 (1.3) | 4 (1.8) | |
| Any IBD-related hospitalization, surgery, steroid prescription, change or escalation in advanced therapy in prior year (N = 224) | |||
| No | 147 (65.6) | 156 (69.6) | .36 |
| Yes | 77 (34.4) | 68 (30.4) | |
| Change or escalation in advanced therapies from prior year (N = 224) | |||
| No change or escalation | 188 (83.9) | 196 (87.5) | .34 |
| Change or escalation | 36 (16.1) | 28 (12.5) | |
| Lab values2, median (IQR) | |||
| Albumin (g/dL, N = 90) | 4.3 (4.0-4.5) | 4.3 (4.1-4.5) | .33 |
| CRP (mg/L, N = 28) | 6.9 (3.1-11.4) | 5.5 (2.8-11.6) | .45 |
| HCT (%, N = 80) | 40.9 (37.8-44.5) | 40.4 (38.7-44.3) | .34 |
Bolded values indicate statistical significance (P < .05).
1For patients with <12 months of follow-up post–GLP-1 receptor agonist prescription; hospitalization, surgery, steroid prescriptions, and change or escalation of advanced therapies were analyzed from GLP-1 receptor agonist prescription to the latest follow-up date in year post–GLP-1 receptor agonist initiation.
2Lab values were compared at the time of GLP-1 receptor agonist prescription (~60 days before or after) and 6-12 months post–GLP-1 receptor agonist prescription. Patients with <12 months were assessed at the latest follow-up or reported as missing. Missing values (albumin = 134, CRP = 196, and HCT = 144).
* P-values calculated by Wilcoxon signed-rank tests for continuous variables and McNemar or McNemar-Bowker’s test for categorical variables.
CRP, C-reactive protein; HCT, hematocrit; IBD, inflammatory bowel disease.
Efficacy of GLP-1 Receptor Agonist
Among the IBD cohort, there was a significant decrease in BMI in the 12 months post–GLP-1 receptor agonist (median BMI decrease from 33.5-31.6 kg/m2; P = <0.01; Table 3). Among IBD patients with laboratory values at both GLP-1 receptor agonist initiation and at 12-month follow-up, there was a significant decrease in the level of hemoglobin A1c (n = 44; median decrease 6.5%-6.2%, P = .01). Additionally, there was an increase in levels of total cholesterol (n = 34; increase 169-177 mg/dL, P < .01), low-density lipoprotein (n = 33, increase 79-89 mg/dL, P < .01) and high-density lipoprotein (n = 34, increase 52-58 mg/dL, P = .03; Table 3). When stratified by diabetes status, changes in hemoglobin A1c, total cholesterol, and LDL remained significant only among IBD patients with diabetes, but not among IBD patients without diabetes. Changes in BMI remained significant in both subgroups (Table 4).
Table 3.
Comparison of lab values at the time of GLP-1 receptor agonist prescription and post-prescription (5-12 months) among IBD cases and non-IBD controls, N = 224.
| Variables | IBD cases | Non-IBD controls | Adjusted linear regression1 | ||||||
|---|---|---|---|---|---|---|---|---|---|
| Total | Time of GLP-1 receptor agonist prescription N (%) 224 (100.0) |
12 months post–GLP-1 receptor agonist prescription or latest follow-up N (%) 224 (100.0) |
P-value* | Total | Time of GLP-1 receptor agonist prescription N (%) 224 (100.0) |
12 months post–GLP-1 receptor agonist prescription or latest follow-up N (%) 224 (100.0) |
P-value* | Beta Coefficient (95% CI) | |
| BMI (kg/m2), median (IQR) | N = 115 | 33.5 (29.4-37.5) | 31.6 (28.2-36.2) | <.01 | N = 86 | 36.1 (31.9-41.0) | 34.4 (30.1-39.6) | <.01 | −0.07 (−1.17, 1.02) |
| Hemoglobin A1c (%), median (IQR) | N = 44 | 6.5 (5.9-7.4) | 6.2 (5.4-7.0) | .01 | N = 33 | 6.6 (5.9-8.0) | 6.3 (5.7-7.6) | .03 | −0.13 (−0.60, 0.34) |
| Total cholesterol (mg/dL), median (IQR) | N = 34 | 169 (130-189) | 177 (151-207) | <.01 | N = 31 | 153 (125-192) | 151 (127-193) | .85 | 13.1 (−1.20, 27.48) |
| LDL (mg/dL), median (IQR) | N = 33 | 79 (58-108) | 89 (69-111) | <.01 | N = 28 | 71 (54-98) | 69 (62-104) | .20 | 4.62 (−7.02, 16.27) |
| HDL (mg/dL), median (IQR) | N = 34 | 52 (40-60) | 58 (44-66) | .03 | N = 31 | 45 (42-61) | 46 (39-58) | .69 | 6.83 (−0.48, 14.14) |
BMI and lab values were compared at time of GLP-1 receptor agonist prescription (~60 days before or after) and 6-12 months post–GLP-1 receptor agonist prescription. Patients with <12 months were assessed at latest follow-up or reported as missing. Missing Values (IBD cases: BMI = 109, hemoglobin A1c = 180, total cholesterol = 190, LDL = 191, HDL = 190. Non-IBD controls: BMI = 138, hemoglobin A1c = 191, total cholesterol = 193, LDL = 196, HDL = 193).
Bolded values indicate statistical significance (P < .05).
1Association of IBD presence (IBD cases compared to non-IBD controls) and change in BMI, hemoglobin A1c, total cholesterol, LDL, and HDL from time of GLP-1 receptor agonist to 12 months post. Adjusted for race, smoking, and presence of comorbidities. Models with change in BMI and change in hemoglobin A1c additionally adjusted for baseline BMI and baseline hemoglobin A1c, respectively.
* P-values calculated by Wilcoxon signed-rank tests for continuous variables.
BMI, body mass index; HDL, high-density lipoprotein; LDL, low-density lipoprotein.
Table 4.
Comparison of IBD disease activity and lab values at time of GLP-1 receptor agonist prescription and post-prescription (5-12 months) stratified by diabetes presence.
| Variables | IBD cases with no diabetes | IBD cases with diabetes | ||||||
|---|---|---|---|---|---|---|---|---|
| Total | Time of GLP-1 receptor agonist Prescription N (%) 141 (100.0) |
12 months post–GLP-1 receptor agonist prescription or latest follow-up N (%) 141 (100.0) |
P-value* | Total | Time of GLP-1 receptor agonist Prescription N (%) 83 (100.0) |
12 months post–GLP-1 receptor agonist prescription or latest follow-up N (%) 83 (100.0) |
P-value* | |
| IBD disease activity 1 | ||||||||
| IBD-related hospitalization in prior year | ||||||||
| No | N = 141 | 133 (94.3) | 133 (94.3) | 1.00 | N = 83 | 71 (85.5) | 74 (89.2) | .61 |
| Yes | 8 (5.7) | 8 (5.7) | 12 (14.5) | 9 (10.8) | ||||
| Steroid prescription in prior year | ||||||||
| No | N = 141 | 104 (73.8) | 116 (82.3) | .08 | N = 83 | 70 (84.3) | 67 (80.7) | .58 |
| Yes | 37 (26.2) | 25 (17.7) | 13 (15.7) | 16 (19.3) | ||||
| Surgery in prior year | ||||||||
| No | N = 141 | 138 (97.9) | 140 (99.3) | .63 | N = 83 | 83 (100.0) | 80 (96.4) | .25 |
| Yes | 3 (2.1) | 1 (0.7) | 0 (0.00) | 3 (3.6) | ||||
| Any IBD-related hospitalization, surgery, steroid prescription, change or escalation in advanced therapy in prior year | ||||||||
| No | N = 141 | 89 (63.1) | 98 (69.5) | .26 | N = 83 | 58 (69.9) | 58 (69.9) | 1.00 |
| Yes | 52 (36.9) | 43 (30.5) | 25 (30.1) | 25 (30.1) | ||||
| Change or escalation in advanced therapies from prior year | ||||||||
| No change or escalation | N = 141 | 117 (83.0) | 121 (85.8) | .61 | N = 83 | 71 (85.5) | 75 (90.4) | .50 |
| Change or escalation | 24 (17.0) | 20 (14.2) | 12 (14.5) | 8 (9.6) | ||||
| BMI and Lab Values2, Median (IQR) | ||||||||
| BMI (kg/m2) | N = 77 | 33.5 (29.9-37.0) | 31.5 (27.1-35.8) | <.01 | N = 38 | 33.2 (29.3-38.4) | 32.6 (28.6-37.5) | .04 |
| Albumin (g/dL) | N = 47 | 4.3 (4.1-4.5) | 4.3 (4.1-4.6) | .91 | N = 43 | 4.2 (3.9-4.5) | 4.3 (4.1-4.5) | .24 |
| CRP (mg/L) | N = 20 | 6.4 (2.3-10.6) | 7.0 (2.5-13.3) | .06 | N = 8 | 8.8 (6.4-22.5) | 4.3 (3.1-11.4) | .28 |
| Hemoglobin A1c (%) | N = 13 | 5.4 (5.0-5.8) | 5.3 (5.0-5.5) | .54 | N = 31 | 7.0 (6.3-8.1) | 6.6 (6.1-7.6) | .02 |
| Total cholesterol (mg/dL) | N = 11 | 184 (175-197) | 183 (177-207) | .10 | N = 23 | 139 (119-189) | 155 (137-207) | <.01 |
| LDL (mg/dL) | N = 11 | 115 (95-125) | 104 (89-125) | .88 | N = 22 | 66 (47-87) | 77 (59-96) | <.01 |
| HDL (mg/dL) | N = 11 | 54 (38-59) | 59 (44-70) | .05 | N = 23 | 51 (40-61) | 53 (39-66) | .21 |
| HCT (%) | N = 42 | 41.4 (38.6-44.0) | 40.3 (38.2-44.4) | .96 | N = 38 | 39.3 (37.4-45.1) | 40.8 (39.2-44.2) | .16 |
Bolded values indicate statistical significance (P < .05).
1For patients with <12 months of follow-up post–GLP-1 receptor agonist prescription; hospitalization, surgery, steroid prescriptions, and change or escalation of advanced therapies were analyzed from GLP-1 receptor agonist prescription to latest follow-up date in year post–GLP-1 receptor agonist initiation.
2BMI and lab values were compared at time of GLP-1 receptor agonist prescription (~60 days before or after) and 6-12 months post–GLP-1 receptor agonist prescription. Patients with <12 months were assessed at latest follow-up or reported as missing. Missing values (no diabetes: BMI = 64, albumin = 94, CRP = 121, hemoglobin A1c = 128, total cholesterol = 130, LDL = 130, HDL = 130, HCT = 99, diabetes: BMI = 45, albumin = 40, CRP = 75, hemoglobin A1c = 52, total cholesterol = 60, LDL = 61, HDL = 60, HCT = 45).
* P-values calculated by Wilcoxon signed-rank tests for continuous variables and McNemar or McNemar-Bowker’s test for categorical variables.
BMI, body mass index; CRP, C-reactive protein; HCT, hematocrit; HDL, high-density lipoprotein; LDL, low-density lipoprotein.
Non-IBD controls were matched for age, sex, and presence of diabetes. Baseline characteristics of non-IBD matched controls can be found in Supplementary Table 2. Among non-IBD controls, there was a significant decrease in BMI in the 12 months post–GLP-1 receptor agonist prescription (median BMI decrease 36.1-34.4 kg/m2, P < .01). There was no significant change in levels of cholesterol, LDL, or HDL among controls (Table 3; Figure 1). When controls were stratified for the presence of diabetes, similar findings were found (Supplementary Tables 3 and 4).
Figure 1.
Changes in lab values from GLP-1 receptor agonist initiation to 12 months post among IBD cases and non-IBD controls.
When directly comparing the change in BMI and hemoglobin A1c between IBD patients and non-IBD patients, there was no significant difference (BMI: −1.0 kg/m2 vs −1.2 kg/m2, ns; hemoglobin A1c −0.4% vs −0.2%, ns). Similar findings were present when cases and controls were stratified by baseline BMI as a categorical variable of obese (BMI > 30.0) vs nonobese (BMI < 30.0; Tables 5 and 6). Additionally, changes in lab values were directly compared between IBD patients and non-IBD controls while stratifying by the presence of diabetes. The only change in cholesterol among patients with diabetes was significantly different between the cohorts (Supplementary Figures 1 and 2).
Table 5.
BMI Changes at time of GLP-1 receptor agonist prescription and post-prescription (5-12 months) stratified by baseline BMI.
| IBD cases | Non-IBD matched controls | |||||||
|---|---|---|---|---|---|---|---|---|
| Total | Time of GLP-1 receptor agonist prescription Median (IQR) |
12 months post–GLP-1 receptor agonist prescription or latest follow-up Median (IQR) |
P-value* | Total | Time of GLP-1 receptor agonist prescription Median (IQR) |
12 months post–GLP-1 receptor agonist prescription or latest follow-up Median (IQR) |
P-value* | |
| Baseline BMI | ||||||||
| Not obese (BMI < 30.0) | N = 34 | 28.0 (25.3-28.9) | 26.7 (23.3-29.1) | .12 | N = 12 | 26.4 (25.0-27.7) | 25.6 (24.2-27.7) | .20 |
| Obese (BMI ≥ 30.0) | N = 81 | 35.8 (32.6-39.1) | 34.3 (31.2-37.3) | <.01 | N = 74 | 36.8 (34.2-42.0) | 36.1 (31.8-39.8) | <.01 |
BMI, body mass index.
* P-values calculated by Wilcoxon rank-sum tests. Bolded values indicate statistical significance (P < .05).
Table 6.
Median change in BMI at time of GLP-1 receptor agonist prescription and post-prescription (5-12 months) stratified by baseline BMI.
| IBD cases | Non-IBD matched controls | ||||
|---|---|---|---|---|---|
| Total | Median (IQR) | Total | Median (IQR) | P-value* | |
| Baseline BMI | |||||
| Not obese (BMI < 30.0) | N = 34 | −0.9 (−3.1 to 1.0) | N = 12 | −1.0 (−2.0 to 0.8) | .95 |
| Obese (BMI ≥ 30.0) | N = 81 | −1.1 (−4.3 to 0.5) | N = 74 | −1.2 (−3.1 to 0.5) | .98 |
* P-values calculated by Wilcoxon rank-sum tests.
BMI, body mass index
Discussion
In this study, we investigated the impact of GLP-1 receptor agonism in patients with IBD. Among an IBD cohort comprised of diabetics and nondiabetics, we found that GLP-1 receptor agonists were not associated with major adverse IBD outcomes. Furthermore, we found that GLP-1 receptor agonists were effective and associated with significant decreases in BMI and hemoglobin A1c. These changes were not significantly different from those observed in a control group without IBD who initiated a GLP-1 receptor agonist during the same time period.
By positively impacting weight and hyperglycemia, GLP-1 receptor agonists may help reduce systemic inflammation. Data have demonstrated that they exert anti-inflammatory effects through direct action on immune cells and metabolic pathways.12 In the colon, GLP-1 is thought to play an anti-inflammatory role through its action on the cAMP/NF-κB pathway, which helps to regulate levels of proinflammatory cytokines.18 The interplay between disease, inflammation, and GLP-1 has been well-documented in animal studies; in one mouse model of IBD, the authors found a significant drop in colonic levels of GLP-1 secondary to the destruction of L-cells.19 Similar findings have been observed in human studies: intestinal biopsies from IBD patients show downregulation of GLP-1 mRNA relative to healthy controls.20 Conversely, the use of GLP-1 receptor agonists has been associated with reduced levels of inflammatory markers including CRP, TNF-alpha, and malondialdehyde, a marker of oxidative stress.21
While the above data suggest that part of the pathophysiology of IBD is mediated through decreased GLP-1 expression, and that supplementing GLP-1 levels or simulating its effect may help to ameliorate or improve the course of disease, there are relatively scarce clinical data that bear this theory out in real-life. One recent case report describes the complete remission of symptoms in a UC patient who was treated with the GLP-1 receptor agonist liraglutide.22 Meanwhile, in a larger-scale study of Danish IBD registries, IBD patients with diabetes who were prescribed either a GLP-1 receptor agonist or DPP-4 inhibitor had lower rates of adverse events, including hospitalization and the need for oral steroids, than diabetic patients prescribed other diabetes medications.17 While there is more research assessing DPP-4 inhibitors and IBD outcomes, they offer conflicting findings concerning whether their use is associated with new development of IBD.23,24 In this study, we found that IBD patients who initiated GLP-1 receptor agonists had no change in rates of IBD decompensation in the year post–GLP-1 receptor agonist initiation. The current study was not powered to assess for improvement in IBD management although our data demonstrated trends in improvement in CRP. One recently presented study demonstrated a significant decrease in CRP among IBD patients initiated on GLP-1 receptor agonists, with no significant change in fecal calprotectin levels.15,16 Larger prospective studies are necessary.
While they offer significant benefits, GLP-1 receptor agonists also have a range of side effects. Nausea is the most common adverse effect, seen in 25%-60% of patients, followed by vomiting and diarrhea.8,9,25 Of note, several of these adverse effects align with common symptoms in IBD, such as abdominal pain, diarrhea, abdominal cramps, and generalized fatigue. In addition, GLP-1 receptor agonists have been observed to alter the pharmacodynamics of other medications, raising concern that they may affect the impact of IBD medications, though to date there is no compelling evidence to suggest these interactions exist or, if they do, exert significant impact.26 In this study, we found no increased rate of IBD exacerbation in the first year post-initiation of a GLP-1 receptor agonist, suggesting these medications may be used safely in patients with IBD.
Our study was limited in number of patients with laboratory values at the time of GLP-1 receptor agonist initiation as well as 1 year later. As such, changes identified such as a decrease in hemoglobin A1c and cholesterol changes must be interpreted with caution. It is noteworthy that among the subgroup of IBD patients with laboratory values at GLP-1 receptor agonist initiation and 1 year later, there was a significant increase in levels of total cholesterol, LDL, and HDL. When stratified by the presence of diabetes, these findings were only present among the IBD-diabetic cohort. These findings were present in a small subgroup, and the explanation for these findings is unclear. Larger well-designed studies in non-IBD patients have found either no change or improvement in lipid profiles with the use of GLP-1 receptor agonists.27 We hypothesize that these findings in such a small subgroup may be related to improved IBD and diabetes management which may lead to improved nutritional status and perhaps liberalized diet, something that is reflected in elevated cholesterol levels. Larger studies among IBD-diabetic patients are necessary to further elucidate this finding.
The strengths of the current study include one of the first evaluations of GLP-1 receptor agonists in nondiabetic IBD patients to assess safety and impact on IBD outcomes. Additionally, to our knowledge, it is the first study to assess the efficacy of GLP-1 receptor agonists compared to non-IBD controls. The limitations of this study include the retrospective nature, which precludes controlling for all confounders and relies on information documented in the EHR. This is especially true when assuming medication adherence. Furthermore, in assessing IBD disease outcomes in patients in the year post–GLP-1 receptor agonist prescription, many of our patients did not have comprehensive disease assessment (imaging, colonoscopy, and biomarkers) during the short interval of 1 year. Furthermore, when evaluating changes in lab values, only small numbers had values at both time points. A component of selection bias may be present and may impact our results. Larger prospective trials are necessary to control for these variables.
Overall, in this observational study, the use of GLP-1 receptor agonists in patients with IBD was safe from an IBD perspective, and effective from a weight loss and glycemic control perspective.
Supplementary Data
Supplementary data is available at Inflammatory Bowel Diseases online.
Contributor Information
Irving Levine, Division of Gastroenterology, Department of Medicine, NYU Langone Health, Manhattan, NY 10016, USA.
Shaina Sekhri, Division of Gastroenterology, Department of Medicine, NYU Langone Health, Manhattan, NY 10016, USA.
William Schreiber-Stainthorp, Department of Medicine, NYU Langone Health, Manhattan, NY 10016, USA.
Brandon Locke, New York University, Manhattan, NY 10003, USA.
Olivia Delau, Inflammatory Bowel Disease Center, NYU Langone Health, New York, NY 10016, USA.
Mohamed Elhawary, Inflammatory Bowel Disease Center, NYU Langone Health, New York, NY 10016, USA.
Krutika Pandit, NYU Grossman School of Medicine, Manhattan, NY 10016, USA.
Xucong Meng, NYU Grossman School of Medicine, Manhattan, NY 10016, USA.
Jordan Axelrad, Inflammatory Bowel Disease Center, NYU Langone Health, New York, NY 10016, USA.
Funding
J.A. has received support from the Crohn’s and Colitis Foundation, the Judith Stewart Colton Center for Autoimmunity, and the NIH NIDDK Diseases (K23DK124570).
Conflicts of Interest
J.A. has received research grants from BioFire Diagnostics and Genentech; and consultancy fees, advisory board member, or honorarium from bioMérieux, Adiso, Addvie, Bristol Myers Squibb, AbbVie, Pfizer, Fresnius, Ferring, and Janssen.
References
- 1. Kaplan GG. The global burden of IBD: from 2015 to 2025. Nat Rev Gastroenterol Hepatol. 2015;12(12):720-727. doi: https://doi.org/ 10.1038/nrgastro.2015.150 [DOI] [PubMed] [Google Scholar]
- 2. Ponder A, Long MD. A clinical review of recent findings in the epidemiology of inflammatory bowel disease. Clin Epidemiol. 2013;5(Jul):237-247. doi: https://doi.org/ 10.2147/CLEP.S33961 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3. Satsangi J, Silverberg MS, Vermeire S, Colombel JF. The Montreal classification of inflammatory bowel disease: controversies, consensus, and implications. Gut. 2006;55(6):749-753. doi: https://doi.org/ 10.1136/gut.2005.082909 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4. Wallace E, Salisbury C, Guthrie B, Lewis C, Fahey T, Smith SM. Managing patients with multimorbidity in primary care. BMJ. 2015;350(jan20 2):h176. doi: https://doi.org/ 10.1136/bmj.h176 [DOI] [PubMed] [Google Scholar]
- 5. Yarur AJ, Bruss A, Moosreiner A, et al. Higher intra-abdominal visceral adipose tissue mass is associated with lower rates of clinical and endoscopic remission in patients with inflammatory bowel diseases initiating biologic therapy: results of the constellation study. Gastroenterology. 2023;165(4):963-975.e5. doi: https://doi.org/ 10.1053/j.gastro.2023.06.036 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6. Elangovan A, Shah R, Ali SMJ, Katz J, Cooper GS. High burden of obesity and low rates of weight loss pharmacotherapy in inflammatory bowel disease: 10-year trend. Crohns Colitis 360. 2023;5(2):otad007. doi: https://doi.org/ 10.1093/crocol/otad007 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7. Smith NK, Hackett TA, Galli A, Flynn CR. GLP-1: Molecular mechanisms and outcomes of a complex signaling system. Neurochem Int. 2019;128(Sep):94-105. doi: https://doi.org/ 10.1016/j.neuint.2019.04.010 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8. Meier JJ. GLP-1 receptor agonists for individualized treatment of type 2 diabetes mellitus. Nat Rev Endocrinol. 2012;8(12):728-742. doi: https://doi.org/ 10.1038/nrendo.2012.140 [DOI] [PubMed] [Google Scholar]
- 9. Drucker DJ. Mechanisms of action and therapeutic application of glucagon-like peptide-1. Cell Metab. 2018;27(4):740-756. doi: https://doi.org/ 10.1016/j.cmet.2018.03.001 [DOI] [PubMed] [Google Scholar]
- 10. Giugliano D, Scappaticcio L, Longo M, et al. GLP-1 receptor agonists and cardiorenal outcomes in type 2 diabetes: an updated meta-analysis of eight CVOTs. Cardiovasc Diabetol. 2021;20(1):189. doi: https://doi.org/ 10.1186/s12933-021-01366-8 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11. Kim JH, Oh CM, Yoo JH. Obesity and novel management of inflammatory bowel disease. World J Gastroenterol. 2023;29(12):1779-1794. doi: https://doi.org/ 10.3748/wjg.v29.i12.1779 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12. Drucker DJ. The cardiovascular biology of glucagon-like peptide-1. Cell Metab. 2016;24(1):15-30. doi: https://doi.org/ 10.1016/j.cmet.2016.06.009 [DOI] [PubMed] [Google Scholar]
- 13. Anbazhagan AN, Thaqi M, Priyamvada S, et al. GLP-1 nanomedicine alleviates gut inflammation. Nanomedicine. 2017;13(2):659-665. doi: https://doi.org/ 10.1016/j.nano.2016.08.004 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14. Yazbeck R, Howarth GS, Geier MS, Demuth HU, Abbott CA. Inhibiting dipeptidyl peptidase activity partially ameliorates colitis in mice. Front Biosci. 2008;13(May):6850-6858. doi: https://doi.org/ 10.2741/3193 [DOI] [PubMed] [Google Scholar]
- 15. Sehgal P, Lichtenstein GR, Khanna T, et al. Sa1950 Impact of GLP1 agonists on inflammatory biomarkers in patients with inflammatory bowel disease. Gastroenterology. 2024;166:S-591. doi: https://doi.org/ 10.1016/S0016-5085 [DOI] [Google Scholar]
- 16. Sehgal P, Lichtenstein GR, Khanna T, et al. Safety and clinical effectiveness of GLP1 agonists in inflammatory bowel disease patients. Gastroenterology. 2024;166:S-173. doi: https://doi.org/ 10.1016/S0016-5085 [DOI] [PubMed] [Google Scholar]
- 17. Villumsen M, Schelde AB, Jimenez-Solem E, Jess T, Allin KH. GLP-1 based therapies and disease course of inflammatory bowel disease. EClinicalMedicine. 2021;37(Jun):100979. doi: https://doi.org/ 10.1016/j.eclinm.2021.100979 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18. Al-Dwairi A, Alqudah TE, Al-Shboul O, Alqudah M, Mustafa AG, Alfaqih MA. Glucagon-like peptide-1 exerts anti-inflammatory effects on mouse colon smooth muscle cells through the cyclic adenosine monophosphate/nuclear factor-κB pathway in vitro. J Inflamm Res. 2018;11(Mar):95-109. doi: https://doi.org/ 10.2147/jir.S152835 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19. Hunt JE, Holst JJ, Jeppesen PB, Kissow H. GLP-1 and intestinal diseases. Biomedicines. 2021;9(4):383. doi: https://doi.org/ 10.3390/biomedicines9040383 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20. Bang-Berthelsen CH, Holm TL, Pyke C, et al. GLP-1 induces barrier protective expression in brunner’s glands and regulates colonic inflammation. Inflamm Bowel Dis. 2016;22(9):2078-2097. doi: https://doi.org/ 10.1097/MIB.0000000000000847 [DOI] [PubMed] [Google Scholar]
- 21. Bray JJH, Foster-Davies H, Salem A, et al. Glucagon-like peptide-1 receptor agonists improve biomarkers of inflammation and oxidative stress: a systematic review and meta-analysis of randomised controlled trials. Diabetes Obes Metab. 2021;23(8):1806-1822. doi: https://doi.org/ 10.1111/dom.14399 [DOI] [PubMed] [Google Scholar]
- 22. Jeffrey L. A novel use of liraglutide: induction of partial remission in ulcerative colitis and ankylosing spondylitis. Clin Med Rev Case Rep. 2019;6(8):281. doi: https://doi.org/ 10.23937/2378-3656/1410281 [DOI] [Google Scholar]
- 23. Wang T, Yang JY, Buse JB, et al. Dipeptidyl peptidase 4 inhibitors and risk of inflammatory bowel disease: real-world evidence in U.S. adults. Diabetes Care. 2019;42(11):2065-2074. doi: https://doi.org/ 10.2337/dc19-0162 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24. Abrahami D, Douros A, Yin H, et al. Dipeptidyl peptidase-4 inhibitors and incidence of inflammatory bowel disease among patients with type 2 diabetes: population based cohort study. BMJ. 2018;360(Nov):k872. doi: https://doi.org/ 10.1136/bmj.k872 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25. Arvanitakis K, Koufakis T, Popovic D, et al. GLP-1 receptor agonists in obese patients with inflammatory bowel disease: from molecular mechanisms to clinical considerations and practical recommendations for safe and effective use. Curr Obes Rep. 2023;12(2):61-74. doi: https://doi.org/ 10.1007/s13679-023-00506-3 [DOI] [PubMed] [Google Scholar]
- 26. Maselli DB, Camilleri M. Effects of GLP-1 and its analogs on gastric physiology in diabetes mellitus and obesity. Adv Exp Med Biol. 2021;1307:171-192. doi: https://doi.org/ 10.1007/5584_2020_496 [DOI] [PubMed] [Google Scholar]
- 27. Muzurovic E, Mikhailidis DP. Impact of glucagon-like peptide 1 receptor agonists and sodium-glucose transport protein 2 inhibitors on blood pressure and lipid profile. Expert Opin Pharmacother. 2020;21(17):2125-2135. doi: https://doi.org/ 10.1080/14656566.2020.1795132 [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.