Older Adults With Inflammatory Bowel Disease Are at Higher Risk of Developing Antibodies to Infliximab

Introduction

Older adults are the fastest-growing subpopulation of patients with inflammatory bowel disease (IBD) and are forecasted to represent approximately one-third of the IBD patient population within the next decade.¹ Despite this, limited data exist regarding the optimal pharmacologic management of older adults with IBD.

Infliximab (IFX), an anti-tumor necrosis factor (anti-TNF) biologic, was the first approved therapy for moderate to severe ulcerative colitis (UC) and Crohn’s disease (CD), and remains a mainstay of therapy for the treatment of IBD.² Despite its efficacy, ~10% of individuals may develop neutralizing antibodies to the medication.^2,3 However, this may be mitigated by the addition of an immunomodulator (6-mercaptoupurine, azathioprine, methotrexate), or with higher levels of IFX.⁴

Older adults, however, may be less likely to receive combination therapy (with immunomodulator) or higher doses of IFX due to safety concerns. As such, in a nationwide Danish cohort study, older adults with new-onset IBD were 71% less likely to receive biologics at 5 years postdiagnosis as compared with those with adult-onset IBD.⁵

Additionally, aging-related declines in the immune response (immunosenescence) likely allay clinician fears about the development of antidrug antibodies among older adults. However, a single study of 110 individuals found that higher rates of anti-TNF antibodies were observed in older adults as compared with younger adults with IBD (odds ratio [OR], 2.9; 95% confidence interval [CI], 1.09‐9.12).⁶ Therefore, in order to further what is known, we looked to assess (1) whether older adults were less likely to receive escalated doses of IFX and (2) whether older adults were more likely to develop antibody to IFX (ATI) when adjusting for dose and IBD subtype, among a large cohort of individuals with IBD.

Methods

We used a large commercial laboratory database (Prometheus Laboratories) that included data on therapeutic drug monitoring of IFX using anti-TNF antibody levels. We extracted IFX dosing as well as ATI for all individuals who were tested with this drug assay from 2015 to 2021. The concentrations of IFX and ATI were determined using drug-tolerant homogeneous mobility shift assays.⁷

Data were extracted from the laboratory information system without any personal identifying information. Our primary outcome was the presence of ATI (defined as an antibody titer >3.1 U/mL). The International Classification of Diseases–Ninth Revision and –Tenth Revision codes from patients’ test requisition forms were used to identify the included cohort of patients with diagnoses of UC and CD. Frequencies were recorded as categorical variables and chi-square analysis was used to assess univariable associations between IFX dose (<10 mg/kg every 8 weeks vs ≥10 mg/kg every 8 weeks), age (<60 years of age vs ≥60 years of age), and IBD subtype (UC vs CD), and the presence of ATI. Multivariable logistic regression was then used to assess the independent effect of IFX dose, age, and IBD subtype on the presence of ATI. Age was stratified by <60 or ≥60 years to capture both the population of younger adults with IBD who are aging as well as individuals who are newly diagnosed with IBD (based on the bimodal peak of incident IBD); however, as a sensitivity analysis, we also included age as a continuous variable. Further, although the presence of ATI alone may impact clinical decision making (eg, adjust infliximab dosing, closely monitor drug level), we conducted an additional sensitivity analysis in which our outcome included both the presence of ATI as well as an undetectable drug level (indicating the complete neutralization of IFX). Last, given that one individual can account for several samples, we conducted an additional sensitivity analysis in which we only considered the first sample from each individual to capture a unique population of patients.

Results

Overall, there were 22 197 specimens, with 14 791 having CD, 6050 having UC, and 1356 having an unknown IBD subtype. In total, 2612 samples were obtained from individuals ≥60 years of age, 17 030 samples were obtained from individuals <60 years of age, and 2555 samples were missing associated ages.

In the entire cohort, 3028 (13.6%) specimens were noted to have ATI. When stratified by age, 18.1% (n = 473 of 2612) of individuals ≥60 years of age developed ATI as compared with 15.0% (n = 2555 of 17 030) among individuals <60 years of age (P < .01) (Figure 1). Overall, older individuals were also less likely to receive escalated IFX dosing as compared with younger individuals with IBD (38.4% vs 49.7%; P < .01). Further, when substratifying by age and IFX dosing, older as compared with younger adults with IBD were more likely to develop ATI when treated with IFX doses <10 mg/kg every 8 weeks (22.8% vs 16.2%; P < .01); however, similar rates of ATI were observed with IFX doses ≥10 mg/kg every 8 weeks (9.9% if <60 years of age vs 10.6% if ≥60 years of age; P = .48).

Figure 1.

Percentage of sample with antibodies to infliximab, stratified by age in decades.

On multivariable analysis, when adjusting for all variables included, ≥60 years of age (adjusted OR, 1.35; 95% CI, 1.20-1.51), IFX dose <10 mg/kg every 8 weeks (adjusted OR, 1.89; 95% CI, 1.85-2.04), and having UC as compared with CD (adjusted OR, 1.44; 95% CI, 1.33-1.57) were independently associated with the development of ATI (Table 1).

Table 1.

Multivariable analysis of factors associated with development of antibodies to infliximab

	Odds ratio (95% CI)
Age ≥60 y	1.35 (1.20-1.51)
Infliximab dose <10 mg/kg every 8 wk	1.89 (1.85-2.04)
IBD subtype (ulcerative colitis)	1.44 (1.33-1.57)

Abbreviations: CI, confidence interval; IBD, inflammatory bowel disease.

On sensitivity analysis, when considering age as a continuous variable, and when adjusting our outcome to only included samples with ATI and an undetectable drug level, we observed similar results. Notably, age (adjusted OR, 1.01; 95% CI, 1.00-1.01), UC as compared with CD (adjusted OR, 1.55; 95% CI, 1.40-1.72), and IFX dose <10 mg/kg every 8 weeks (adjusted OR, 2.74; 95% CI, 2.45-3.06), were independently associated with immune response. Analogously, when considering only the first sample received from the unique population of 15 455 individuals, similar results were seen (age: adjusted OR, 1.003; 95% CI, 1.00-1.01; UC as compared with CD: adjusted OR, 1.41; 95% CI, 1.28-1.56; IFX dose <10 mg/kg every 8 weeks: adjusted OR, 1.86; 95% CI, 1.68-2.05).

Discussion

In this large retrospective cohort of individuals with IBD, we found that older adults were significantly more likely to develop ATI as well as receive to a lower dose of IFX as compared with younger individuals with IBD. Further, we found that individuals with UC as compared with those with CD, as well as those who received lower doses of IFX, were at significantly higher risk to develop ATI.

Immunosenescence, defined as the decline of immune system function with age, is often discussed as a consideration when initiating advanced therapies among older adults with IBD. However, prior data from a study of 110 individuals found that in antithesis of what might be expected, older adults with IBD were more likely to develop antibodies to anti-TNF therapy. In our study, we observed similar results, with older adults having 35% higher odds of developing ATI, even when adjusting for IFX dosing and IBD subtype. Further, when only including one sample per individual, and when adjusting our outcome to only include those samples in which both ATI and an undetectable drug concentration were present, we found increasing age to be independently associated with development of our outcome. It refers to the finding that for every 1-year increase in an individual’s age, their risk of developing ATI (antibodies to infliximab) increase by 0.3% and 0.6%- corresponding to the prior statements.

Developing ATI may be mitigated by combination therapy and higher IFX dosing. However, this is rarely used among older adults with IBD, despite prior data showing that combination therapy can be safely used in this cohort.⁸ Additionally, in our study we see that older adults are less likely to receive escalated doses of IFX. This is likely due to safety concerns, though current data suggest that higher IFX levels are not associated with an increased risk of adverse events.⁹

In order to shift this trajectory, there is a need for continued education regarding the risk of ATI in older adults with IBD, as well as the risk of adverse outcomes with undertreatment of disease. In support of this, recent work from Cheng et al¹⁰ has shown that anti-TNF therapy does not increase the risk of an adverse event among older adults with moderate-severe UC, and that older individuals on placebo actually had a higher number of serious adverse events as compared with those on anti-TNF therapy.

Strengths of this study include its size, as we had more than 20 000 samples, and were able to adjust for IFX dosing, making our data generalizable. Further, we performed additional sensitivity analyses adjusting our outcome to include those with the presence of ATI and an undetectable drug level, as well as assessing outcomes from a unique population of individuals, ensuring the robustness of our findings. However, limitations include the lack of additional confounders such as albumin, disease activity, and concomitant immunomodulator use. Additionally, we were unable to assess for HLA-DQA1*05 presence, which may in part explain the different rates of ATI observed between older and younger adults (if individuals with older-onset IBD were more likely to have this allele); however, this needs to be assessed in future follow-up studies.

In summary, we observed that older adults were significantly more likely to develop ATI and less likely to receive escalated doses of IFX as compared with younger adults with IBD. Thus, when initiating IFX to treat older adults with IBD, particularly among those with UC, consideration should be given to upfront combination therapy, higher IFX dosing, or proactive drug monitoring to minimize the likelihood of ATI development. Further, it is imperative to ensure that dosing in older adults is based on factors such as ongoing endoscopic disease activity and albumin level, and not underdosed solely based on chronological age alone. Future studies should validate these results and provide additional data pertaining to the characteristics of older adults who are at risk for the development of anti-TNF antibodies.

Author Contribution

All authors have made substantial contributions to all of the following: (1) the conception and design of the article and interpretation of the relevant literature, (2) drafting the article or revising it critically for important intellectual content, (3) final approval of the version to be submitted.

Funding

No study funding.

Conflicts of Interest

K.E.L.: none declared. D.H. has received consulting fees from AbbVie, BMS, Fresenius Kabi, Janssen, Pfizer, Prometheus, Takeda, and UCB; has received grant/research support from Janssen and Pfizer; and has served on the advisory committee/board for AbbVie, BMS, Fresenius Kabi, Janssen, Pfizer, Prometheus, and Takeda. T.D. is an employee of Prometheus Laboratories.

A.S.F. has received consulting fees from BMS, AbbVie, and Douglas; and grant support from the National Institute on Aging (R03 AG078927-02), via an ACG Career Development Award, and from the Crohn’s and Colitis Foundation.