Combining Cytokine-Related Biomarkers to Better Define Tumor Necrosis Factor-α Antagonist Response in Inflammatory Bowel Disease: An Observational Cohort Study

Abstract

INTRODUCTION:

Interleukin-13 receptor alpha 2 (IL13RA2), triggering receptor expressed on myeloid cells-1 (TREM-1), and oncostatin M (OSM) may be associated with response to tumor necrosis factor-α antagonists (TNFAs) in inflammatory bowel disease. We aimed to assess the direction of association between TNFA-induced clinical remission and IL13RA2 and TREM-1, respectively, and assess the value of combining biomarkers for identifying nonresponders.

METHODS:

Plasma samples from a retrospective inflammatory bowel disease cohort were collected before TNFA start. Clinical remission at 1-year, surgery, hospitalization, adverse drug events, and TNFA discontinuation were assessed. IL13RA2 and TREM-1 concentrations were compared between those with and without 1-year clinical remission. OSM data were obtained from our previous cohort. Where significant, TREM-1 and IL23RA2 thresholds associated with clinical remission at 1-year were assessed using a receiver operating characteristic analysis. Significant biomarkers were combined using a linear discriminant analysis. The performance characteristics were assessed for individual biomarkers and biomarker combinations.

RESULTS:

In Crohn's disease (CD) (n = 95) and ulcerative colitis (UC) (n = 53), higher IL13RA2 concentrations, but not TREM-1, were found among those not achieving TNFA-associated clinical remission at 1-year (IL13RA2, CD, P < 0.0001; UC, P = 0.0003). IL13RA2 thresholds, 4.554 ng/mL (CD) and 6.117 ng/mL (UC) separated those with and without clinical remission at 1-year (CD, area under the receiver-operating characteristic curve = 0.80, 95% CI = 0.71–0.90, P < 0.0001; UC, area under the receiver-operating characteristic curve = 0.79, 95% CI = 0.66–0.91, P = 0.0005). In CD, combining IL13RA2 and OSM concentrations enhanced prediction accuracy compared with either biomarker alone and increased the identification of other important clinical outcomes.

DISCUSSION:

IL13RA2, but not TREM-1, was associated with TNFA response. In CD, its prediction accuracy improves when combined with OSM.

INTRODUCTION

Inflammatory bowel disease (IBD), encompassing both Crohn's disease (CD) and ulcerative colitis (UC), are a set of autoimmune diseases that are characterized by inappropriate inflammatory responses in the gastrointestinal tract (1). Disease progression in both UC and CD is associated with an increased risk of hospitalization and the need for surgical intervention (2). For this reason, early diagnosis and proper treatment are important in managing IBD.

The introduction of biologics, such as tumor necrosis factor-α antagonists (TNFAs), has revolutionized the treatment of IBD (3). Targeting cytokine pathways involved in aberrant inflammatory responses promotes intestinal healing, avoids intestinal surgery, and overall improves patient quality of life. Unfortunately, there remains a high degree of interindividual variation in the response to TNFAs, with high rates of nonresponse and loss of response reported over time (4). Some hypothesize that, in individuals with poor TNFA response, intestinal inflammation may relate to the activation of non-TNF-related immune pathways (4). Clinically, variable drug responses have prompted an interest in identifying predictive biomarkers for personalizing treatment strategies and for guiding clinical decision-making around drug selection (4).

Recently, levels of oncostatin-M (OSM), triggering receptor expressed on myeloid cells-1 (TREM-1), and the interleukin 13 receptor subunit alpha-2 (IL13RA2) have been linked to TNFA response (5–9). These cytokine-related molecules contribute to chronic inflammation, epithelial damage, and dysfunction (5,8,10–12). Intestinal gene expression of all 3, as well as plasma concentrations of OSM and TREM-1, have been variably linked to TNFA-induced remission or nonremission in IBD cohorts. To date, small sample sizes and variation in the type of samples assessed (plasma, whole blood, intestinal tissue) have led to conflicting data around the direction of association and the detection of these potential biomarkers in blood. Furthermore, there are little mechanistic data addressing why expression of these cytokine pathway-related players is linked to TNFA response.

Our study aimed to confirm the plasma-based measurability of TREM-1 and IL13RA2 and direction of association between TNFA-induced clinical remission at 1-year and individual pretreatment plasma concentrations of these 2 receptors. We further aimed to assess whether a combination of these cytokine-associated biomarkers, including OSM, improved the prediction accuracy of TNFA-induced clinical remission at 1-year and other important clinical outcomes compared with any biomarker alone.

METHODS

Participant and study procedures

A retrospective cohort study was conducted in adults 18 years of age or older with a clinical, endoscopic, radiographic and/or histopathological diagnosis of CD or UC seen as part of the Western University Personalized Medicine Program in London, Ontario Canada between January 1, 2012, and December 31, 2023. All participants were exposed to either infliximab or adalimumab, collectively known as TNFAs. All participants provided a blood sample before TNFA exposure and were followed for 1 year or until the time of TNFA discontinuation. Participant enrollment including inclusion and exclusion criteria, study procedures, blood sample and data collection during this period were previously reported in addition to the outcomes of clinical remission at 1-year (primary outcome), the need for surgical intervention, use of rescue glucocorticoids, hospitalization, TNFA discontinuation, or TNF-related adverse events (13). Participants were further excluded if their blood sample was no longer available.

Plasma biomarker quantification

Plasma OSM quantification in all participants was performed as previously described (13). OSM data were obtained from our previously published cohort (n = 100). Additional participants seen between 2020 and 2023 were included herein (n = 48) and their plasma OSM concentrations were measured using the previously published protocol. Plasma concentrations of IL13RA2 and TREM-1 for all participants were quantified using an enzyme-linked immunosorbent assay (ELISA) following the manufacturer's protocol (human IL13RA2 ELISA kit; ThermoFisher, Carlsbad, CA; human TREM-1 ELISA kit; ThermoFisher). The lower limits of detection for IL13RA2 and TREM-1 were 0.41 ng/mL and 65.8 pg/mL respectively. All standards and patient samples were measured in duplicate. To quantify plasma IL13RA2, samples were diluted up to 8-fold. TREM-1 quantification was performed with samples diluted up to 4-fold.

Statistical analyses

Statistical analyses were completed with GraphPad Prism version 10.4.1. (GraphPad Software Inc., San Diego, CA). A significance threshold of P < 0.05 was used. Demographic characteristics are presented and separated by clinical remission status at 1-year (Table 1). Categorical variables were analyzed with a Fisher exact test and reported as frequency distributions. Continuous variables were analyzed with a Student t-test and reported as averages and ranges with standard deviations.

Table 1.

Demographic characteristics by disease activity

Variable	Clinical remission 1-yr (n = 99)a	Active disease 1-yr (n = 49)a	P value
Age, yr, (mean, range)	41.2 (18–78)	37.6 (18–65)	0.18
Female sex (%)	53 (53.5)	22 (44.5)	0.38
Weight, kg, (mean ± SD)	81.9 ± 20.3	74.7 ± 16.2	0.052
Crohn's disease (%)	66 (66.7)	29 (59.2)	0.47
Ileal (%)	19 (28.8)	5 (17.2)	0.23
Ileo-colonic (%)	35 (53.0)	17 (58.6)	0.61
Colonic (%)	12 (18.2)	8 (27.6)	0.53
Ulcerative colitis (%)	33 (33.3)	20 (40.8)	0.47
Pan-colitis (%)	18 (54.5)	15 (75.0)	0.14
Left-sided colitis (%)	15 (45.5)	3 (15.0)	0.02
Proctitis (%)	0 (0)	1 (5.0)	0.72b
Smoking history (%)	24 (24.2)	6 (8.6)	0.13
Infliximab exposure (%)	57 (57.6)	28 (57.1)	>0.99
Combination therapy (%)c	55 (55.6)	17 (34.7)	0.02
Disease duration, yr, (mean ± SD)	9.0 ± 9.9	8.4 ± 8.9	0.88

Bold values represent the statistically significant variables.

^aBased on the Harvey-Bradshaw Index for Crohn’s disease and the partial Mayo score for ulcerative colitis.

^bThe Haldane correction was used to calculate the P value.

^cUse of one of methotrexate or azathioprine.

IL13RA2 and TREM-1 concentrations from pretreatment plasma samples were used to test the association between biomarker concentrations and clinical remission after 1 year of TNFA treatment. Disease activity at 1-year was assessed using the Harvey Bradshaw Index (14) for CD, where remission was defined as a score less than 5 and the partial Mayo score (15) for UC, where remission was defined as a score less than 2.In addition, participants were recorded as having active disease at 1-year if TNFA treatment was discontinued before 1-year due to persistent disease activity. If TNFA treatment was discontinued before1-year due to an adverse drug event, participants were assigned to the disease activity group (remission or active disease at 1-year) based on their clinical disease activity score at the time of discontinuation. A Mann-Whitney U test was used to compare median IL13RA2 and TREM-1 plasma concentrations respectively between the active disease and clinical remission cohorts. Similarly, quartile coefficients of dispersion were used to report the degree of interpatient variability. Separate analyses were conducted for participants with UC and CD.

A receiver operating characteristic and Youden index analysis was used to identify threshold biomarker concentrations associated with TNFA-induced clinical remission at 1-year. Area under the curve with 95% confidence intervals (95% CI), sensitivity and specificity are reported. OSM threshold concentrations for UC and CD were previously published by Guo et al (2022) (13) and were included in the combined biomarker analysis. Biomarkers with a statistically significant threshold value were combined using a linear discriminant (LD) analysis to create a weighted-score (LD score) for the prediction of TNFA-induced clinical-remission (16). A receiver operating characteristic and Youden index analysis was further used to identify the LD score threshold that maximized sensitivity and specificity for the identification of TNFA-induced clinical remission, with Area under the curve with 95% CI, sensitivity and specificity additionally reported. Other performance characteristics such as positive and negative predictive values, and accuracies are reported for the individual biomarkers and for the biomarker combination. The incidence of secondary outcomes (hospitalization, adverse drug event, use of rescue glucocorticoids, surgical intervention, and TNFA discontinuation) stratified by biomarkers with statistically significant concentration thresholds, as appropriate, were tested using a Fisher exact test and a multiple variable logistic regression adjusting for the covariates of age, sex, weight, TNFA type, dose, and use of combination therapy. Disease location was also adjusted for in the UC cohort. This was expressed as an odds ratio with 95% CI.

A further multivariable logistic regression was used to assess the association between clinical remission at 1-year with each of TREM-1, IL13RA2 and biomarker combinations (represented as LD scores). Adjusting covariates included age, sex, weight, TNFA dose, use of combination therapy, and drug type. For the UC cohort, disease location was also adjusted for.

RESULTS

Participant study flow is shown in Figure 1. A total of 148 participants were included in the final analyses (UC = 53, CD = 95). Table 1 highlights the baseline characteristics of included participants. A higher proportion of participants achieving TNFA-induced remission at 1-year had left-sided UC and received combination therapy with an immunomodulator (methotrexate or azathioprine). Seven participants experienced adverse drug events that led to treatment discontinuation before 1 year despite achieving clinical remission with TNFA treatment. Plasma OSM concentrations were previously measured and published for 100 participants and are included in the herein analyses (13). Median plasma IL13RA2 and TREM-1 concentrations for those with TNFA-induced clinical remission at 1-year and those without are shown in Figure 2. Both IL13RA2 and TREM-1 were detectable in participant plasma samples with large interpatient variability (IL13RA2, quartile coefficient of dispersion, QCD: CD, active = 97.0%, remission = 90.1%; UC, active = 84.8%, remission = 92.9%; TREM-1, QCD: CD, active = 80.8%, remission = 87.6%; UC, active = 81.3%, remission = 82.9%). In both UC and CD, significantly higher median IL13RA2 concentrations were observed in participants with active disease at 1-year compared with those achieving remission (Figure 2a). Median TREM-1 plasma concentrations were not significantly different between groups for both CD and UC participants (Figure 2b).

Figure 1.

Participant study flow. IBD, inflammatory bowel disease; TNFA, tumor necrosis factor-α antagonists.

Figure 2.

Plasma IL13RA2 (a) and TREM-1 (b) concentrations among CD and UC participants with TNFA-induced clinical remission at 1-year vs those without. Data are presented in a boxplot, whereby median values (thick horizontal line), 25th and 75th percentiles (box outlined) and 5–95% confidence intervals (whiskers) are shown; CD, Crohn's disease; IL13RA2, Interleukin-13 receptor alpha 2; ns, nonsignificant; TNFA, tumor necrosis factor-α antagonists; TREM-1, Triggering Receptor Expressed on Myeloid cells-1; UC, ulcerative colitis.

For participants with CD, a threshold plasma IL13RA2 concentration of 4.554 ng/mL (area under the receiver-operating characteristic curve [AUROC] = 0.80, 95% CI = 0.71–0.90, P < 0.0001) separated those achieving clinical remission at 1-year from those with persistently active disease (Figure 3a) with a sensitivity of 83% and specificity of 65%. For patients with UC, a threshold plasma IL13RA2 concentration of 6.117 ng/mL (AUROC = 0.787, 95% CI = 0.66–0.91, P = 0.0005) separated those achieving clinical remission at 1-year from those with persistently active disease (sensitivity = 80%, specificity = 70%) (Figure 3b). Other plasma IL13RA2 performance characteristics are reported in Table 2. A threshold plasma concentration value could not be generated for TREM-1 in either CD or UC participants given there was no significant difference in median TREM-1 plasma concentration between those achieving clinical remission at 1-year and those who did not (Supplementary Figure 1A and 1B, http://links.lww.com/CTG/B433). (CD: AUROC = 0.515, 95% CI = 0.39–0.63, P = 0.81, UC: AUROC = 0.560, 95% CI = 0.39–0.73, P = 0.47). As a result, no further statistical analyses were completed for TREM-1.

Figure 3.

Receiver operator characteristic analysis for the mean Interleukin-13 receptor alpha 2 concentrations for participants with Crohn's disease (panel a) and ulcerative colitis (panel b) with and without 1-year clinical remission on tumor necrosis factor-α antagonists therapy. 95% CI, 95% confidence interval; AUC, area under the curve.

Table 2.

Biomarker performance characteristics

Characteristic	Crohn's disease			Ulcerative colitis
	IL13RA2	OSM	Combined	IL13RA2	OSM	Combined
Positive predictive value	0.51	0.72	0.79	0.62	0.86	0.70
Negative predictive value	0.90	0.91	0.90	0.85	0.79	0.87
Accuracy	0.71	0.83	0.86	0.74	0.81	0.79

IL13RA2, Interleukin 13 receptor alpha 2; OSM, oncostatin M.

Combined = based on integration of IL13RA2 and OSM into the weighted linear discriminant score.

On multivariable analysis, being below the plasma IL13RA2 threshold remained a significant predictor of clinical remission at 1-year, adjusting for age, sex, weight, TNFA dose and type, and use of combination therapy (Supplementary Tables 1 and 2, http://links.lww.com/CTG/B433). Further to this, for participants with CD, plasma IL13RA2 concentrations above the identified threshold were associated with TNFA discontinuation, use of rescue glucocorticoids, and the occurrence of TNFA associated adverse drug events adjusting for defined covariates (Table 3). For participants with UC, being above the IL13RA2 threshold was not associated with any secondary outcome (Supplementary Table 3, http://links.lww.com/CTG/B433).

Table 3.

Multi-variable regression for the effect of cytokine biomarkers on secondary outcomes in CD (n = 95)

Outcomea	Odds ratio (95% CI)	Multivariate P value	Univariate P value
TNFA discontinuation
IL13RA2 (above)	4.76 (1.50–17.50)	0.01	0.08
Combined (LD score) (above)	7.24 (2.10–29.69)	0.003	0.006
Use of rescue glucocorticoids
IL13RA2	3.21 (1.15–9.72)	0.03	0.05
Combined (LD score)	8.25 (2.64–29.48)	0.0005	0.0005
Hospitalizations
IL13RA2	2.31 (0.79–7.22)	0.13	0.21
Combined (LD score)	5.12 (1.62–17.78)	0.007	0.01
Surgery
IL13RA2	2.06 (0.59–8.02)	0.27	0.41
Combined (LD score)	5.34 (1.41–23.47)	0.02	0.02
Adverse drug events
IL13RA2	4.07 (1.32–14.25)	0.02	0.08
Combined (LD score)	2.84 (0.87–10.20)	0.09	0.18

Bold values represent the statistically significant variables.

CI, confidence interval; IL13RA2, interleukin 13 receptor alpha 2; LD, linear discriminant; TNFA, tumor necrosis factor-α antagonists.

^aAdjusted for age, sex, weight, TNFA type, dose, and use of combination therapy.

An LD analysis was used to combine individual biomarkers, with statistically significant thresholds, into a combined and weighted score (LD score). Plasma IL13RA2 and OSM concentrations derived from the entire cohort were included. The equations generated to compute LD scores for CD (equation 1) and UC (equation 2), respectively, are as follows:

$$\text{LD}\hspace{0.17em}\text{score}=0.001648\times \text{IL}13\text{RA}2\hspace{0.17em}\text{concentration}+0.000630\times \text{OSM}\hspace{0.17em}\text{concentration}.$$ (1)

$$\text{LD}\hspace{0.17em}\text{score}=0.000470\times \text{IL}13\text{RA}2\hspace{0.17em}\text{concentration}+0.001429\times \text{OSM}\hspace{0.17em}\text{concentration}.$$ (2)

For participants with CD, an LD score of 0.19 separated participants with and without clinical remission at 1-year (AUROC = 0.82, 95% CI = 0.71–0.93, P < 0.0001) with a sensitivity of 76% and a specificity of 91% (Supplementary Figure 2A, http://links.lww.com/CTG/B433). For participants with UC, an LD score of 0.136 separated participants with and without clinical remission at 1-year (AUROC = 0.86, 95% CI = 0.74–0.97, P < 0.0001) with a sensitivity of 80% and a specificity of 79% (Supplementary Figure 2B, http://links.lww.com/CTG/B433). Other performance characteristics for the combined biomarkers are reported in Table 2.

On multivariable analysis, being below the LD score threshold remained a significant predictor of clinical remission at 1-year, adjusting for age, sex, weight, TNFA dose and type, and use of combination therapy (Supplementary Tables 4 and 5, http://links.lww.com/CTG/B433). For participants with CD, LD scores above the identified threshold were associated with TNFA discontinuation, use of rescue glucocorticoids, hospitalization, and surgical intervention adjusting for defined covariates (Table 3). For participants with UC, being above the LD score threshold was associated with use of glucocorticoid rescue therapy adjusting for defined covariates (Supplementary Table 3, http://links.lww.com/CTG/B433).

DISCUSSION

Our study is the first to show that IL13RA2 is detectable in plasma in a large IBD cohort. It also confirms that high plasma IL13RA2 concentrations are associated with a lack of clinical remission at 1-year with TNFA therapy in the largest CD and UC cohort to date. IL13RA2 remains an independent predictor of clinical disease remission at 1-year, when adjusting for other clinical variables. The direction of association reported herein is consistent with the limited data reported in other smaller cohort studies. Arijs et al (2009) reported higher intestine mucosal IL13RA2 gene expression in UC patients (n = 9) who were deemed TNFA nonresponders vs TNFA responders (n = 8) (9). Similarly, Verstockt et al (2019a) reported high mucosal IL13RA2 gene expression in TNFA nonresponders (n = 23) vs TNFA responders (n = 28) with CD (8). A major difference between our study and others is the detection of IL13RA2 in plasma rather than intestine mucosal biopsy.

There are groups who have described an inability to detect soluble IL13RA2 in patient serum or plasma (8,17,18). The examined cohorts included participants with asthma, CD, unspecified atopy, and healthy controls. They concluded that a soluble form of IL13RA2 does not exist. However, in vitro studies have demonstrated that other environmental exposures and molecular processes can lead to the production of a soluble form of IL13RA2 (19,20). Interestingly, O'Toole et al (2008) described a cross-reactivity between their plasma IL13RA2 detection method and other plasma cytokines (TNF-α, IL-17A), limiting their ability to confirm the presence of IL13RA2 in plasma rather than a confirmation of its nonexistence (17). Furthermore, other groups have shown that IL13RA2 is detectable in plasma or serum (21,22). Successful groups used an alternate ELISA test vs those who were unsuccessful. This further suggests that differences in the ability to measure a soluble form of IL13RA2 may depend on the disease state and the sensitivity and specificity of the test used.

Conversely, plasma TREM-1 concentrations were detectable, but not associated with disease remission on TNFA therapy, for participants with CD or UC in our study cohort. To date, there have been mixed reports regarding the presence and direction of association between serum TREM-1 and TNFA drug response (6,23–25). This may be the reason it has not been further evaluated for inclusion in clinical practice as a biomarker of TNFA response.

Our study is the first to combine IL13RA2 and OSM in a weighted score to assess whether this would improve on the accuracy of either test for identifying clinical remission at 1-year on TNFA therapy in IBD. We identified that the combined OSM-IL13RA2 score improved the accuracy and positive predictive value of IL13RA2 and OSM for identifying TNFA clinical remission and was highly associated with other important clinical outcomes such as TNFA discontinuation, the need for surgical intervention, the use of rescue glucocorticoids, and hospitalization. Conversely, the combined score was not more accurate for the identification of TNFA-induced clinical remission at 1-year than OSM in UC. Reasons for differences in these associations are not known but may be linked to differences in UC vs CD pathophysiology and immune system activation. Differences may also be influenced by divergences in clinical decision-making around UC vs CD for example limited intestinal resection in CD may be a more palatable approach, and thus more oft selected strategy, than full proctocolectomy in UC.

Strengths of this study include its relatively large sample size, longitudinal design, and evaluation of other important clinical outcomes. It confirms the presence and direction of association between IL13RA2 and TNFA drug response seen in other IBD cohorts. This strengthens the justification for further evaluating IL13RA2 as a clinically actionable biomarker of TNFA response. Furthermore, the evaluation of biomarker combinations is a novel approach for further testing the utility of TNFA biomarkers.

The main study limitations include the retrospective study design and the absence of a validation cohort to test the weighted combined score. Future studies focused on testing the weighted combined score would bolster support for its further consideration for use in clinical practice. Moreover, post-TNFA treatment biomarker measurements, while not performed in this study, may provide further data to better understand the link between IL13RA2, OSM, and TNFA response.

Of note, participants who experienced adverse drug events, and discontinued treatment before 1 year, were assigned to the disease activity group based on their clinical score at the time of discontinuation. This may have over or underestimated the clinical effect of TNFA therapy; however, only a small subset of participants (3 UC and 4 CD) fell into this category, likely limiting any confounding effect.

Finally, the mechanism underlying these associations remains unclear and is not addressed in this study. Future studies should aim to address these limitations through prospective and mechanistic study designs, comprehensive sample quantification, and external validation efforts.

Ultimately, this study identifies pretreatment plasma IL13RA2, and not TREM-1, as a significant and independent predictor of TNFA response in IBD, with higher pretreatment plasma concentrations associated with a reduced likelihood of achieving 1-year clinical remission in both CD and UC. Furthermore, the combination of pretreatment plasma IL13RA2 and OSM concentrations demonstrate superior predictive accuracy compared with either biomarker alone and identified patients at risk of other important deleterious clinical outcomes with the selection of TNFA therapy. This finding highlights the potential for multibiomarker approaches in optimizing TNFA selection.

CONFLICTS OF INTEREST

Guarantor of the article: Aze Wilson, MD, PhD.

Specific author contributions: A.W. supervised the study. R.B.K., N.C., J.G., T.P., M.B., R.K., M.S., and A.W. were involved in data acquisition. A.W. was responsible for the study concept and design. A.W. and G.D.C. carried out all data analyses. E.R. and G.D.C. carried out sample analysis. E.R., G.D.C., and A.W. were involved in data interpretation. Statistical analyses were performed by G.D.C., A.W. E.R. drafted the manuscript. Critical revisions were carried out by A.W., G.D.C., N.C., J.G., T.P., M.B., R.K., M.S., K.M., and R.B.K. All authors had full access to all the data. All authors reviewed and approved the final version of this manuscript.

Financial support: This work was supported by the Wolfe Medical Research Chair in Pharmacogenomics (R.B.K.), Canadian Institutes of Health Team Grant: Personalized Health (FRN 178435) (R.B.K.).

Potential competing interests: T.P. has served as a consultant for Bristol Myers Squibb, Celltrion, Eli Lilly and Pfizer recently. He currently serves as an investigator on clinical research studies for Abivax, Alimentiv, AnaptysBio, Eli Lilly, Merck, Morphic Therapeutic, Pfizer/Kanyos Bio, Spyre Therapeutics and Takeda. J.C.G. has received speaking honoraria from AbbVie, Pfizer, Takeda, Celltrion; consulting honoraria from Fresinius Kabi; educational grants from Janssen, AbbVie, Organon. NC has received honoraria for speaking/consulting from AbbVie, Janssen, Takeda, Pfizer, Novartis, Ferring, Pharmascience, Allergan, Lupin, Shire, Fresenius Kabi, Celltrion, BIOJAMP, Organon. M.B. has received advisory board or consultancy from AbbVie, Celltrion, Ferring, Janssen, Novo Nordisk, Pfizer, and Takeda. She has participated in clinical trials with AbbVie, Novo Nordisk, Gilead, Takeda, Janssen, Pfizer and Astra Zeneca. R.K. reports fees for Consulting/Speaking from: AbbVie, Amgen, BMS, Celltrion, Encycle, Innomar, Gilead, Janssen, Jamp, Lilly, Merck, Pendopharm, Pfizer, Roche/Genentech, Alimentiv (formerly Robarts Clinical Trials), Shire, and Takeda Canada outside the submitted work; Research fees from Roche/Genetech. A.W. has received speaking fees from Takeda, Abbvie and Pfizer and consulting fees from Fresenius Kabi, Celltrion and Pfizer.

Ethical considerations: This study was approved by the Western University Health Sciences Research Ethics Board. All methods were performed in accordance with the relevant guidelines and regulations of the Tri-Council Policy Statement.

Study Highlights.

WHAT IS KNOWN

• ✓ Mucosal interleukin 13 receptor-α2 and mucosal and blood-based triggering receptor expressed on myeloid cells-1 and oncostatin-M have been linked to tumor necrosis factor-α antagonist response in small inflammatory bowel disease cohorts.

WHAT IS NEW HERE

• ✓ Interleukin 13 receptor-α2 is detectable in the plasma of patients with IBD and its combination with pretreatment oncostatin-M concentrations, more accurately identifies those with clinical nonremission at 1-year on tumor necrosis factor-α antagonist therapy and other important deleterious outcomes in Crohn’s disease.

Supplementary Material

ct9-17-e00960-s001.docx ^{(95.6KB, docx)}

ABBREVIATIONS:

ADE

adverse drug event

AUROC

area under the receiver operating characteristic curve

CD

crohn's disease

ELISA

enzyme-linked immunosorbent assay

LD

linear discriminant

IBD

inflammatory bowel disease

IL13RA2

interleukin 13 receptor alpha-2

Kg

kilogram

OSM

oncostatin-M

SD

standard deviation

TNFA

tumor necrosis factor alpha antagonist

TREM1

triggering receptor expressed on myeloid cells-1

UC

ulcerative colitis

Yr

year