Current Role of Monoclonal Antibody Therapy in Pediatric IBD: A Special Focus on Therapeutic Drug Monitoring and Treat-to-Target Strategies

In the last two decades, biologicals have become essential in treating children and adolescents with inflammatory bowel disease. TNF-α inhibitors (infliximab, adalimumab and golimumab) are preferentially used. Recent studies suggest that early application of TNF-α inhibitors is beneficial to inducing disease remission and preventing complications such as development of penetrating ulcers and fistulas. However, treatment failure occurs in about one third of pediatric patients. Particularly, children and adolescents differ in drug clearance, emphasizing the importance of pharmacokinetic drug monitoring in the pediatric setting. Here, current data on the choice and effectiveness of biologicals and therapeutic drug monitoring strategies are reviewed.


Introduction
The use of biologicals in pediatric patients with inflammatory bowel disease (IBD) has widely increased since their introduction [1]. Biologicals have substantially improved the disease course for many pediatric patients suffering from Crohn's disease (CD) and ulcerative colitis (UC). Current guidelines recommend the use of TNF-α inhibitors in pediatric IBD patients with high disease activity or in those who do not respond to other therapeutic strategies [2,3]. Infliximab and adalimumab were approved for the treatment of pediatric Crohn's disease in 2006 and 2012, respectively, and later for pediatric UC also. These compounds are thus the most recommended. As a result, the data on the use of biologics in pediatric and adolescent patients with IBD have increased in number significantly in recent years. Biologic agents are effective and safe, and they are one of the most used medication classes in pediatric IBD [4].
In a large German cohort (CEDATA), adalimumab was the most commonly used biologic when therapy with infliximab failed. According to Cozijnsen and colleagues, this approach was effective in a small retrospective study [5]. Infliximab, adalimumab and golimumab bind sTNF and mTNF, while etanercept only binds soluble TNFu [6]. The reason for the use of adalimumab after infliximab treatment failure may be that other biological agents lack official approval in childhood and are used off-label despite being available for treatment in adults. However, information on treatment success exists in children for alternative biologicals such as golimumab, certolizumab (both TNF-α inhibitors), vedolizumab, an α4β7-integrin blocker, and ustekinumab, an IL-12/IL-23 blocker (see Figure 1 for an overview of currently used biologicals in pediatric IBD). Approximately 54% of 42 pediatric IBD patients treated with vedolizumab entered clinical remission within 14 weeks, even when prior treatment with anti-TNF-α drugs had failed, according to a retrospective multicenter study [7]. Another large multicenter study demonstrated ustekinumab to be safe and effective in the treatment of Crohn's disease [8]. Pediatric patients with UC achieved deep according to a retrospective multicenter study [7]. Another large multicenter study demonstrated ustekinumab to be safe and effective in the treatment of Crohn's disease [8]. Pediatric patients with UC achieved deep mucosal remission when treated with ustekinumab, even when a relapse occurred under therapy with infliximab and vedolizumab [9]. Similarly, golimumab resulted in clinical remission [10]. Ninety percent of TNF-α naïve patients remained steroid-free compared to 50% of patients who did not respond to other biologic therapies [10]. Therefore, European guidelines recommend treatment with infliximab or adalimumab as an effective option for pediatric patients with moderate-tosevere Crohn's disease. After the failure of anti-TNF-α or other treatments, ustekinumab and vedolizumab are recommended [2,3].

Early and Effective Use of TNF-α Blockers Prevents Disease Progression and Disease Complications
Evidence suggests that early treatment with biological agents is favorable for pediatric CD patients [1,28,29]. Early application of biological agents significantly prevented treatment failure [1]. In 2020, the first randomized direct comparison of first-line infliximab with exclusive enteral nutrition or corticosteroids as first-line treatment in pediatric patients with moderate-to-severe Crohn's disease was provided by Jongsma and colleagues [50]. Of the patients treated with first-line anti-TNF-α therapy, a significantly higher percentage accomplished clinical and endoscopic remission [50]. In addition, first-line TNF-α-blocker therapy needed less dose escalation while achieving mucosal healing [50]. Consistent with this, data from the CEDATA registry study showed that first-line infliximab therapy led to a higher rate of clinical remission in the short term compared to conventional therapy with biologics, which led to endoscopic remission in another study [1,50]. In addition, Jongsma and colleagues found that the probability of continued clinical remission at week 52 with monotherapy of azathioprine was higher in children who received infliximab as first-line therapy to induce remission [50]. Comparable results were shown for adalimumab [72]. In Crohn's disease, early and effective use of TNF-α blockers also prevented the development of disease complications, for example, strictures or penetrating ulcerations and disease progression [28]. A Korean study revealed that early anti-TNF-α medication led to a lower risk of surgery during disease progression [30]. In another study, early admission of biologics significantly reduced the risk of penetrating complications but not stenosing complications [31].
A large cohort study showed that unrelated to the IBD subtype, the administration of biologicals a short time after diagnosis (<120 days) is connected to fewer glucocorticoids being needed [32]. Another large cohort study of pediatric patients with IBD demonstrated that the early treatment with TNF-α blockers was superior to immunomodulators in achieving remission within three months [29]. In a retrospective Canadian study, earlier initiation of anti-TNF-α treatment in patients with Crohn's disease and ulcerative colitis was more common in adolescents and was associated with higher PCDAI/PUCAI and lower serum albumin levels at diagnosis [33].
The incidence of extraintestinal manifestation in IBD throughout treatment with biologicals ranges from 14% to 25% [1,78]. However, the study with the longer follow-up reported a higher incidence of extraintestinal manifestations [1]. In the large German multicenter cohort, first-line prescription significantly reduced the incidence of extraintestinal manifestations longitudinally [1]. These patients had the highest disease activity due to systemic inflammation before treatment, so the findings are promising. Moreover, the incidence of extraintestinal manifestations was reduced from about 27% to 25% immediately after treatment initiation, with a further reduction to 17% at six months [1].

Occurrence and Frequency of Adverse Treatment Events
In several studies evaluating different biologicals, adverse treatment events occurred in around 46% of patients within all IBD subtypes [1,79]. Immediate infusion reactions (11%) and a psoriasis-like rash (11%) were reported [34,79]. Symptoms of infusion reactions include dyspnea, coughing, cyanosis and vomiting [19]. Minor infections were reported in 15.4% of patients [34].
Several studies have reported varying rates of skin complications due to biological treatment ranging from 13% to 39%, with the most recent study reporting 17% [1,36,37]. Even for golimumab, severe skin reactions were the reason for discontinuation in a case study of adults with CD [80]. Dolinger et al. recommend switching to ustekinumab in the event of skin reactions on TNF-α inhibitor therapy. In the study by Nuti et al., a psoriasislike rash was observed in 11% of patients treated with infliximab or adalimumab [34]. The risk of skin adverse events appears to be increased only with adalimumab and not with infliximab in patients with IBD and juvenile idiopathic arthritis [38]. There appears to be no association between higher drug concentrations and increased adverse event rates [39].
Vedolizumab and ustekinumab showed a good safety profile [79,81]. The most common adverse events were respiratory tract infections (33%) with vedolizumab [79] and infusion reactions with ustekinumab [10]. Additional adverse events related to vedolizumab were headaches (4%) and myalgia (3%), while only 1% of patients discontinued treatment due to adverse events in a multicenter cohort study [75]. More serious adverse events are rarely observed and are described in case reports. The currently available evidence suggests that treatment with TNF-α inhibitors is associated with a very low risk of developing malignancies. These almost exclusively occurred when combining TNF-α blockers with azathioprine in male patients [35].

Therapeutic Drug Monitoring to Optimize the Treatment Strategy and Maintain the Efficacy of Biological Agents
Primary and secondary treatment failures with anti-TNF-α drugs are common and challenging in daily clinical practice. Approximately 10% to 30% of adult patients experience primary non-response [82], while 20% to 50% develop secondary loss of response during biological therapy [82]. A large retrospective registry study of adults with ulcerative colitis showed that around 50% of the patients had a suboptimal response to anti-TNF-α agents, leading to dose escalation or treatment discontinuation [83]. The main causes of primary non-response or loss of response are low trough levels or anti-drug antibodies, respectively [2]. For preventing treatment failure and also for following a treat-to-target strategy, optimal dosing is important for achieving not only clinical remission but also mucosal healing as one of the most important long-term goals [84]. Short-term goals that reflect adequate therapy management include the normalization of inflammation markers in the serum and feces [84].
A meta-analysis by the ESPGHAN-IBD working group suggests that a higher dose per kilogram of body weight may be appropriate in younger IBD patients, as they often have lower trough levels in the early phase of therapy induction [51]. Indeed, a retrospective British study demonstrated that children with very early onset IBD received the increased dose of 10 mg/kg body weight [40]. The remission rate in these patients was 62%; otherwise, the course of very young children with IBD treated with biologics appears to be similar to that of older patients [40,41]. In particular, studies in pediatric patients with ulcerative colitis and Crohn's disease have shown that higher infliximab trough levels after induction predict remission one year after infliximab administration [52,53] (see also Table 2).
A large UK prospective observational study (PANTS) of approximately 1000 children and adults with IBD tried to identify clinical and pharmacokinetic factors that might predict primary non-response at week 14, non-remission at week 54, and adverse events leading to drug discontinuation [42]. In the multivariable regression analysis, the only factor independently associated with a primary non-response was low drug trough levels of infliximab and adalimumab at week 14 [42]. Approximately 63% of the patients developed anti-drug antibodies to infliximab and 29% to adalimumab [42]. For both drugs, suboptimal drug concentrations at week 14 predicted immunogenicity and the development of neutralizing anti-drug antibodies predicted subsequent low drug concentrations [42]. A further important finding was that a combination immunomodulatory therapy (thiopurine or methotrexate) reduced the risk of anti-drug antibody development for infliximab and adalimumab [42].
Therapeutic drug monitoring (TDM) is advocated to assess trough levels and/or neutralizing anti-drug antibodies (ADA) to optimize the treatment strategy and maintain the efficacy of biological agents. TDM can either be reactive or proactive. In reactive TDM, drug concentrations and/or the occurrence of ADAs are assessed in the serum in case of persistent or recurrent flares of IBD. Reactive TDM can streamline the management of primary non-response and secondary loss of response. If the drug concentration appears to be subtherapeutic, the dose may either be increased or the interval between the doses reduced. If the ADA titer is low, adding an immunomodulator to the biologic treatment should be considered. In case of a high ADA titer, the biological agent may be switched to another biologic in the same class or a different class.
Proactive TDM signifies the assessment of drugs' trough levels during remission to ensure effective therapy, prevent a disease relapse by maintaining adequate drug levels, and reduce the formation of ADA. Proactive TDM increases clinical remission and the durability of the response to a biological agent [69]. In recent years, guidelines and consensus statements have been published on the emerging topic of TDM for adults [85][86][87][88][89][90] and children [2,3]. These guidelines recommend that reactive TDM to guide treatment in patients with biologicals is more cost-effective than empiric dose escalation.
In recent years, randomized control trials such as the Pediatric Crohn's Disease Adalimumab Level-based Optimization Treatment (PAILOT) trial [69] and the NOR-DRUM B study have suggested the utility of proactive TDM [91]. Proactive drug monitoring of adalimumab in the randomized PAILOT trial was associated with significantly higher rates of corticoid-free remission and lower inflammatory markers [69]. Infliximab trough levels greater than 10 mg/mL are generally associated with remission and higher rates of perianal fistula healing in pediatric IBD patients [54,92]. Yarur and colleagues recommend in adults a treat-to-target strategy until adequate infliximab levels are achieved [92]. Of note, current data indicate that an infliximab or adalimumab therapy should generally not be discontinued unless drug levels are greater than 10 µg/mL [92].
As higher infliximab levels after induction were associated with clinical remission [55], proactive drug monitoring in the induction phase of infliximab was associated with optimal trough levels at week 52 and clinical remission in pediatric IBD patients [56]. In particular, early response and drug monitoring during induction appear to predict response rates, possibly due to higher drug clearance in children and an association with higher cytokine levels at diagnosis [57,58]. Higher drug clearance was associated with hypoalbuminemia, high CRP, higher BMI, male sex and anti-drug antibodies [59,81,93]. A small Spanish study points out that proactive drug monitoring during maintenance is favorable in order to maintain long term clinical response and showed response rates of 92.8% after three years in pediatric patients with Crohn's disease [43].
In addition, in adults, a high initial serum TNF-α and a severe inflammation with extensive mucosal involvement leads to increased drug consumption [94] and fecal loss [95], while younger age (<10 years) is attributed to different pharmacokinetics in children compared to adults [51,96]. This has led to the revised recommendation of an intensified infliximab treatment (10 mg/kg body weight at weeks 0, 1 and 4) to achieve remission in cases of an acute severe colitis by the ESPGHAN in 2018 [97].
In a small Spanish cohort of pediatric patients with Crohn's disease, proactive drug monitoring (measurement of trough levels) prevented loss of response to infliximab and adalimumab due to antibodies [43]. Anti-drug antibodies are associated with loss of response to infliximab [60]. In patients who have already developed anti-drug antibodies, dose escalation of the biological drug suppressed anti-drug antibodies in the subsequent study [43,45]. Another approach to suppress anti-drug antibodies is to combine the biologic with an immunomodulator which is supported by the evidence for infliximab [46] but not for adalimumab [70] in pediatric patients with Crohn's disease. Patients receiving infliximab as a second-line treatment for failed therapy benefit significantly from combination therapy with immunomodulators [20]. In general, combination therapy increased the likelihood of continuing infliximab at two years [20].
For several of the above-mentioned reasons, the dose of the biological drug does not necessarily correspond to the determined drug trough levels [54]. Therefore, a Bayesian calculation model applied to drug concentrations represents a new approach to optimize treatment response to biologics in IBD by incorporating several individual parameters that affect drug clearance, such as sex, hypoalbuminemia, and fecal loss [54]. It predicts the treatment response to optimize dosing [98]. With the implementation of three trough-level measurements, the model was able to predict drug concentrations and thus be helpful for therapy adjustments [54]. Precision dosing showed better remission and response rates in adults compared to traditional dosing regimens [61]. Sufficient models for children in clinical practice have yet to be determined due to the large amount of data needed to test the robustness and identify an appropriate computational model to predict individual drug concentrations. However, data on TDM in pediatric IBD are emerging and allow for recommendations for treatment monitoring (summarized in Table 2). Notably, to date, the superiority of proactive TDM has not been consistently demonstrated in randomized controlled trials [99]. ≥10.1 µg/mL-12 µg/mL (*) to prevent treatment failure [42] In case of loss of response->new induction dose or weekly application [71,100] Golimumab >0.97 µg/mL at week 14 for clinical response in UC [101] Ustekinumab ≈6.6 µg/mL at week 8 (associated with steroid-free remission week 52) in all IBD subtypes [10] Vedolizumab >37 µg/mL before 3rd infusion and >20 µg/mL before 4th infusion to achieve steroid free-clinical remission in UC and CD [77] >30 µg/mL in week 2 (*) for endoscopic remission, clinical remission in CD and UC [102] <30 kg: >7 µg/mL for steroid free and EEN-free remission in all IBD subtypes [75] >30 kg: ≥11.5 µg/mL for clinical and biochemical remission in CD and UC (*) [76] Note that not all studies performed cut-off tests for trough levels and some studies did not find an association between trough levels and disease outcome (e.g., [9,103]), (*) = adult studies.
Surgical interventions and partial bowel resection for Crohn's disease still represent a rescue option. However, compared to the beginning of the 2000's, these procedures are less frequent, especially in patients responding to TNF-α inhibitors [68]. It is well known, especially in pediatric Crohn's disease patients, that the postoperative recurrence risk after surgery is substantial. In a pediatric series, clinical recurrence rates after partial intestinal resection were 17% at 1 year, 38% at 3 years and 60% at 5 years [104]. Therefore, a postoperative remission-maintaining therapy should be used after surgically induced remission in children, as recommended by the ECCO/ESPGHAN expert committee [2,105]. Thiopurine is recommended as the first choice for postoperative relapse prophylaxis in IFX-naïve patients and anti-TNF-α antibodies in high-risk cases. While in pediatric IBD, randomized controlled trials on this topic are lacking, supporting data for the postoperative use of anti-TNF-α therapy to reduce the risk of recurrence at the anastomoses was reported by three RCTs conducted in adult patients with ileocolonic resections and primary anastomoses [106][107][108]. A recent German study reported a reduced endoscopic recurrence after ileocecal resection in children receiving preoperative TNF-α inhibitors [49].
In summary, the quality and efficacy of treatment in pediatric IBD appear to have improved, as children with Crohn's disease suffer fewer relapses in the last five years than 10-15 years ago [47]. Early treatment with infliximab or adalimumab should be considered if patients are at high risk of a poor outcome, e.g., Crohn's with persistently high disease activity despite adequate induction therapy, extensive or pan-enteric manifestation, deep colonic ulcerations, marked growth retardation, severe perianal involvement, radiologically or endoscopically proven structures, the occurrence of fistulas, intestinal perforations, inflammatory conglomerates and/or abscesses, and CMV infections [2]. Similar features in ulcerative colitis qualify for a TNFα inhibitor as pancolitis, extensive and deep colonic ulcerations, the early need for (recurrent) steroid therapy, and recurrent infections with Clostridioides difficile or CMV [3,97].
Additional recommendations will soon further refine biological therapy strategies; for example, trials are comparing longer dosing intervals in children in remission on TNF-α blockers [109], which would further improve the quality of life of children. One study even showed that discontinuation of biologics could be considered if endoscopic and histologic remission occurs in children with ulcerative colitis on TNF-α blockers [48]. Individualized medicine, considering pharmacogenetic and pharmacogenomic aspects, is expected to lead to further advances in treatment. For example, a study of response to infliximab found that a variant in the FCGR3A gene was associated with a decreased response to infliximab with lower levels and higher anti-IFX antibody concentrations [62]. HLA polymorphisms (G allele of rs2395185 and the C allele of rs2097432) were associated with reduced long-term response in adults but not children with CD to anti-TNF-α medication [44]. So pharmacological models might have to take different polymorphisms in children and adults into account.

Conclusions
TNF-α blockers are a safe and efficient way to treat IBD with high disease activity in children and adolescents. Infliximab and adalimumab are efficient in achieving clinical and mucosal remission. However, as treatment failure still occurs, therapeutic drug monitoring and exclusion of the formation of anti-drug antibodies are helpful for further treatment management. For both infliximab and adalimumab, drug concentrations to achieve different treatment goals are available. Therapeutic drug monitoring involves a proactive and a reactive strategy, yet further prospective RCTs are still needed to pose recommendations for which one to prefer. For other monoclonal antibodies, such as vedolizumab and ustekinumab, favorable drug concentrations are mostly derived from adult studies.

Conflicts of Interest:
The authors declare no conflict of interest.