Only recently have studies examined the durability of anti-TNF agents and their pharmacokinetic (pk) profiles, despite their having being approved since 1998 in adults and 2006 in pediatric patients. Most studies have examined infliximab (IFX), with evolving literature for the other anti-TNF agents, including adalimumab, certolizumab pegol (CZP), and a paucity of data with golimumab.

Although the response to IFX induction is highly successful in 75–90% of pediatric IBD patients, maintenance of a sustained and durable remission can be more challenging [14, 15]. In the REACH trial, only 60% of pediatric CD patients who responded to induction were in remission at 1 year, and half of these patients required dose modification after losing response [14]. In a meta-analysis of adult IBD patients on IFX, 23–46% required dose escalation and 5–13% discontinued the drug at 1 year [16]. Using TDM, one can better understand the etiology of primary nonresponse and secondary loss of response, and TDM may augment clinical management by increasing the likelihood of sustained response to therapy.

In 2003, initial studies found higher serum IFX concentrations to be correlated with longer duration of response [17]. It was reported in 2006 that detectable serum IFX concentrations were associated with a higher rate of clinical remission, endoscopic improvement, and lower CRP values in CD patients [18]. Other studies also support that detectable IFX concentrations are predictive of a sustained response in CD patients [19]. In UC, the data is just as strong, with detectable IFX concentrations associated with higher remission rates, endoscopic improvement, and a significant decrease in colectomy risk (55% vs. 7%, OR 9.3; 95% CI 2.9–29.9; p < 0.001) [20]. In the post hoc analysis of the ACT trials, higher IFX concentrations in UC patients were associated with an increased likelihood of achieving clinical remission and mucosal healing with increasing quartiles of IFX levels [21]. Patients with drug levels in the third or fourth quartile had remission rates at week 30 closer to 60% as compared to those in the second quartile whose remission rates were 25%. Similarly, studies have found that higher adalimumab concentrations correspond to mucosal healing and clinical remission, higher CZP concentrations in CD patients are associated with endoscopic remission and response, and higher golimumab concentrations were associated with clinical remission [22–24].

The minimum anti-TNF trough concentration associated with improved outcomes remains debatable. Murthy et al. demonstrated that an IFX concentration of >2 μg/mL in UC patients was associated with a higher rate of corticosteroid-free remission, compared to a trough concentration of <2 μg/mL (69% vs. 16%; p < 0.001) [25]. A trough concentration >3 μg/mL during IFX maintenance therapy has been shown by Vande Casteele et al. to be independently associated with a lower CRP and has been proposed as a cutoff to improve outcomes [26]. More studies suggest that yet even higher IFX trough drug concentrations at week 14, the time of the first maintenance dose, are associated with better 1-year efficacy outcomes (≥3.5 μg/mL to >5 μg/mL) [27, 28].

Differing cutoffs have also been suggested for adalimumab concentrations. Velayos et al. found that an adalimumab concentration of >5 μg/mL was associated with decreased CRP level; Yarur et al. confirmed this association [29, 30]. Karmiris et al. suggested an even higher therapeutic threshold of >8 mg/ml [31]. For CZP, in the post hoc analysis of the WELCOME trial, evaluating induction therapy of CZP in 203 patients, remission rates were higher among patients whose CZP concentration fell within the two highest quartiles during induction at weeks 0, 2, 4, and 6 (27.5 to 33.8 μg/ml and ≥33.8 μg/ml, respectively); thus, a CZP concentration of >27.5 μg/ml has been proposed for clinical use [32]. For golimumab, patients with drug concentrations in the highest quartile with a concentration of >3.1 μg/ml had higher rates of clinical remission at 30 and 54 weeks when compared to the lower quartiles [24].

Despite a high primary response rate to the anti-TNF agents, two-thirds of patients losing response do so within the first year [16]. The loss of response to anti-TNF agents is most often due to an individual’s unique physiologic profile driven by drug clearance – factors that influence drug clearance include a low serum albumin concentration, high baseline CRP levels, large body size, male sex, and high degree of systemic inflammation [33]. Additionally, the development of antidrug antibodies (ADAs), referred to as immunogenicity, remains a significant driver of loss of response. It should be noted that non-chimeric anti-TNF therapies have the same issues with ADA formation as chimeric anti-TNF agents [34]. The presence of ADA increases the clearance of the drug, resulting in lower drug concentrations. This, in turn, results in shorter duration of response, which has been demonstrated in multiple studies [17, 18, 22, 33, 35–38]. In a prospective study of patients receiving IFX therapy, ATI development preceded clinical loss of response in over half of patients [39]. Similar results have been reported with adalimumab, with 20% of patients developing anti-adalimumab antibodies, which predicted biochemical and clinical loss of response [40]. Another study also confirmed the association of anti-adalimumab antibodies with increased markers of inflammation and with clinical indices indicating increased disease activity [41]. Antibodies to certolizumab were also found to be associated with reduced remission rates through week 26 in the PRECISE-2 trial (71 vs 62%), with similar findings in the WELCOME trial [42, 43]. In addition to the negative effect ADAs have on efficacy, they also increase toxicity, with the example of anti-infliximab-antibodies (ATIs) being associated with infusion reactions [36]. Additionally, a recent pediatric study found that the presence of ATIs was a predictor of lower IFX concentrations and a higher risk of surgery [44].

Recent data demonstrate that ATIs may be transient. Vande Casteele et al. retrospectively found that in 28% of patients’ ATIs disappeared over time, whereas they were sustained in 72% of patients [38]. It is unclear whether lower concentrations of ATI may be overcome by IFX dose escalation. However, Vande Casteele’s study suggested that higher ATI concentrations of >9.1 U/ml were less likely to be overcome, with a likelihood ratio of 3.6 of failure [38], and thus such patients should be changed to another anti-TNF therapy.

The knowledge of the presence of ADA is also important in the setting of re-induction of anti-TNF therapies after a prolonged interruption or “drug holiday.” Baert et al. reported that the presence of ATI 2 weeks after the first re-induction dose of IFX was associated with lower response rates and higher rates of infusion reactions [45]. The data suggest that if a patient has discontinued IFX for at least 6 months, it is important to check for the presence of ATIs prior to administering the second induction dose. It remains unclear whether, following a drug holiday, a patient should be re-induced with the standard initial induction regimen (0, 2, 6 weeks) or forego re-induction and resume with every 8-week interval.

The reported rates of ADA are entirely dependent on the specific assay used to measure ADA. Several techniques are available for measuring anti-TNF concentrations and ADA. Thus, comparison of results from different assays should be performed with caution, as there remains no standardization between different assays. Drug concentrations are generally detected sensitively between assay types, yet the detection and accurate quantification of ADAs have been more challenging. First-generation assays, such as the enzyme-linked immunosorbent assay (ELISA), have less clinical utility, given the lower sensitivity for measuring ADAs. Using the ELISA assay, serum anti-TNF drug competes with the ADA detection moiety, and thus when drug is detected in the sample, ADA is unable to be accurately measured. Radioimmune assay (RIA) is sensitive and specific for drug and ADA detection, yet disadvantages include the complexity of the test, prolonged incubation time, expense, and the handling of radioactive materials [46, 47]. The homogenous mobility shift assay (HMSA), using high-performance liquid chromatography, has the advantage of separating and quantifying the drug and antibody concentrations independently, making it feasible to detect ADAs in the presence of anti-TNF drug. ELISA and ELISA-like assays (LabCorp, Esoterix Inc) as well as HMSA assays (Prometheus labs) are currently commercially available for IFX and adalimumab.

Given the negative effects of ADA on therapeutic efficacy, durability, and association with infusion reactions, attempts should be made to reduce the likelihood of ADA formation. Various strategies have been recommended in order to do so, such as the addition of an immunomodulator and even proactive optimization of drug concentrations.

In the ACCENT 1 trial, concomitant immunomodulator use with IFX was associated with lower rates of ATI formation [48]. In another prospective CD cohort, patients who received concomitant immunomodulator therapy had higher IFX concentrations and less likelihood of ATI formation than those not receiving a concomitant immunomodulator (43% vs. 75%; p < 0.01) [17]. A logistic regression analysis further demonstrated that the only significant variable predictive of IFX concentrations was the use of a concomitant immunosuppressive agent (p < 0.001) [17]. The SONIC trial demonstrated that combination therapy of IFX with AZA is superior to IFX monotherapy in achieving clinical remission and mucosal healing [37]. This is potentially due to less formation of antibodies and higher trough levels associated with combination therapy. Likewise, patients receiving an immunomodulator in combination with adalimumab have been noted to have higher drug concentrations than those on monotherapy [30]. Additionally, with golimumab therapy, patients receiving a concomitant immunomodulator had a lower incidence of antibody formation (1.1% vs 3.8% p = 0.01) [24].

Studies also suggest that addition of concomitant immunomodulator can help recapture response for patients with low anti-TNF drug concentrations. Ben Horin et al. reported in a small case series that the addition of an immunomodulator to maintenance infliximab monotherapy increased IFX concentrations and lowered antibody concentrations, improving patient outcomes by restoring clinical response [49]. Other small studies have shown that the addition of a thiopurine in patients losing response to anti-TNF monotherapy was an effective strategy to recapture response [50]. Overall, these studies suggest that concomitant immunomodulator use not only decreases immunogenicity preemptively as suggested by SONIC but may also augment recapture of response in patients with low drug concentrations.

In pediatric patients, particularly in males, the substitution of methotrexate (MTX) for thiopurines may provide a safety advantage, given the rare yet positive association between combination therapy of IFX with thiopurines and malignancy, including hepatosplenic T-cell lymphoma. The efficacy of combining an anti-TNF agent with methotrexate (MTX) has been examined as well. In the rheumatoid arthritis literature, a low dose of 7.5 mg weekly was associated with lower rates of ATI development in IFX-treated patients [51]. In a cohort of pediatric IBD patients, concomitant MTX use for at least 6 months was found to be associated with increase of IFX durability, with those on MTX with IFX for >6 months having a higher 5-year probability of remaining on IFX than those without concomitant MTX use (0.97 ± 0.03 vs. 0.41 ± 0.11, p < 0.001) [52]. The dose of methotrexate varied, with no clear dose threshold identified. In another pediatric study, no clinical benefit in IFX durability or efficacy was found when using very low-dose oral MTX (<10 mg/week) as concomitant therapy in pediatric IBD patients [53]. It has been proposed that a dose of at least 12.5 mg of oral MTX is needed to avoid immunogenicity. The COMMIT trial found that patients on IFX combination therapy with 25 mg of weekly subcutaneous MTX were significantly less likely to develop ATIs and had higher IFX concentrations, yet no clear benefit was found in inducing and maintaining clinical remission [54]. A German group found that concomitant use of MTX with infliximab had a positive effect in the treatment of refractory CD adult patients, using a MTX dose of 20 mg weekly, either parenterally or orally administered [55]. The ideal dose of MTX remains unclear, with suggestions of using a dose of 12.5–15 mg orally weekly.