Skip to content
BY 4.0 license Open Access Published by De Gruyter February 8, 2023

Surface plasmon resonance assays for the therapeutic drug monitoring of infliximab indicate clinical relevance of anti-infliximab antibody binding properties

  • Melina K. Grasmeier ORCID logo , Susanne Weber ORCID logo , Matthias Treiber ORCID logo , Markus A. Thaler ORCID logo and Peter B. Luppa ORCID logo EMAIL logo

Abstract

Objectives

The therapeutic antibody infliximab (IFX) has improved the life quality of numerous autoinflammatory disease patients. However, IFX can trigger the generation of anti-drug antibodies (ADA), whose optimal evaluation and management are currently subject of controversial discussions. We present two novel surface plasmon resonance (SPR) biosensor assays for therapeutic drug monitoring of IFX and characterization of ADA and investigated the diagnostic value of ADA binding properties.

Methods

IFX and ADA were quantified via developed SPR biosensor assays (IFXmon and ADAmon, respectively) and diagnostics-approved ELISA in sera from inflammatory bowel disease patients. Pre-analytic ADA enrichment with magnetic beads enabled analytical drug tolerance of the ADAmon assay. The dissociation ratio (DissR) as an index for ADA:IFX binding stability was calculated from the SPR sensorgrams of ADA quantification runs.

Results

IFX levels determined by IFXmon assay and ELISA showed high agreement, whereas ADA quantification concordance between ADAmon assay and ELISA was poor. In patients, DissR was predominantly constant over time and differed significantly between therapy outcomes. A DissR cut-off of 1.524 indicated undetectable IFX levels with 71.4% sensitivity and 88.9% specificity. Additionally, the SPR reference surface was exploited as serum-individual negative control to check result plausibility within multi-sample run sequences.

Conclusions

Overall, both SPR biosensor assays exhibited reliable quantitative performance with accuracies superior to their ELISA counterparts and precision inferior to ELISA only for ADAmon. DissR presented itself as promising ADA binding parameter and could contribute to both earlier and more tailored therapeutic decisions.

Introduction

Inflammatory bowel diseases (IBD), encompassing Crohn’s disease and ulcerative colitis, affect up to 1.3% of the Western population [1]. The characteristic progressive, immune-mediated and chronic inflammation of the intestines is driven in various intestinal cell types by the cytokine tumor necrosis factor (TNF) [2]. TNF antagonists – therapeutic proteins that bind TNF and block its pleiotropic downstream signaling pathways – have become a breakthrough in IBD therapy [3]. In contrast to small molecule immunomodulators, therapeutic proteins exhibit a known mechanism of action, more specific binding and reduced toxicity. TNF inhibition does not cure IBD, but can modify the course of IBD towards remission, in contrast to merely stopping its progression, and has thus enabled setting more ambitious goals in IBD therapy [4].

Infliximab (IFX, originator trade name Remicade®) is a therapeutic antibody (tAb) and was the first TNF antagonist to be approved for therapy of IBD and diverse rheumatic disorders [3, 5]. By 2019, over three million patients have received treatment with the originator substance Remicade®, making IFX one of the most prescribed biologics [3]. IFX is a first-generation, chimeric tAb comprising variable domains of murine origin fused to human constant domains. While fully human TNF antagonists and other gut-specific biologics are available for IBD therapy now, IFX is still widely prescribed. One reason is the approval of seven IFX biosimilars, making its prescription more economic [6], [7], [8]. Additionally, the longtime clinical experience with IFX and its safety are well documented [9]. Nevertheless, all tAb can trigger the emergence of anti-drug antibodies (ADA), which occurs in up to 65% of IFX-treated patients [10, 11]. ADA may block the paratope of the tAb and neutralize its function, accelerate its clearance and correlate with adverse effects up to anaphylactic reactions [12]. Overall, these ADA-related effects often result in treatment failure and subsequently enforce switching to alternative medication.

In order to take adequate therapeutic consequences, the physician needs to identify the reason for therapy failure or anticipate it. To do so, both drug and ADA levels are assessed, summarized by the term therapeutic drug monitoring (TDM). Most commonly, enzyme-linked immunosorbent assays (ELISA) serve to quantify ADA, but homogeneous mobility shift assays (HMSA), radioimmunoassays and reporter gene assays exist as well [13]. The heterogeneity of these assays translates to the generated results: Comparability of different methods is impaired by differing sensitivities, unequal calibrators and arbitrary units [14], [15], [16]. Lack of a gold standard method for ADA quantification hinders consensus guideline-based decision-making on how to proceed when ADA emerge. These analytical drawbacks might be alleviated by evaluating ADA concentration dynamics rather than absolute concentrations [17]. Nevertheless, not only the concentration, but also the quality of ADA likely determines their impact on therapy [13]. Investigations of ADA epitopes, for example, have shown that ADA are practically always neutralizing TNF antagonist function [18], [19], [20]. Hence, the use of analyzing neutralizing ADA in routine diagnostics is questionable [12, 18]. In contrast, the binding characteristics of ADA-TNF antagonist interactions are largely unexplored [21, 22]. Yet, ADA binding properties may be clinically relevant, as they likely differ between patients and may thus provide a new piece of information on the hazard that ADA exert. If ADA binding properties would be set at their first emergence in the patient, they could even be predictive of therapy success. However, binding properties are not accessible through currently approved TDM assays.

Surface plasmon resonance (SPR) spectroscopy can be utilized to both quantify and characterize analytes. Indeed, a SPR method for IFX and anti-IFX ADA analysis has recently been introduced and validated by Beeg et al., demonstrating its promising analytical potential [21, 23]. Here, we present two combinable SPR-based biosensors: (i) A sensor for the quantification of IFX in diluted serum (“IFXmon”) and (ii) a sensor for the quantification of pre-purified ADA that allows for simultaneous determination of ADA binding stability (“ADAmon”). In contrast to previous work, we established a simple and quick pre-analytical protocol for ADA purification, which reduces sample complexity and enables drug-tolerant ADA quantification, even in the presence of supratherapeutic IFX trough concentrations. As a proof-of-concept study, sera of IBD patients under IFX therapy were analyzed by the developed SPR biosensor assays and diagnostics-approved ELISA. Finally, the clinical relevance of ADA binding stability evaluation by SPR represented by the dissociation ratio (DissR) was investigated.

Materials and methods

Patients and sera

Within the patients’ routine TDM protocol, 159 sera of 45 adult IBD patients under IFX (Remicade®, Remsima® or Inflectra®) maintenance, treated between 2016 and 2022 either at a gastroenterologic outpatient clinic (Peter Langmann and Monika Weikert, Karlstadt, Germany) or at a tertiary care center (Klinikum rechts der Isar, Munich, Germany), were included in this study (Supplementary Tables S1, S2). 55 healthy volunteer sera served as negative controls. Aliquots were stored at −80 °C until analyzed. The study was approved by the Ethics Committee of the TUM medical faculty (289/19S) and conducted in accordance with the Helsinki Declaration. Written informed consent was obtained from all participants.

Reagents and materials

Unless stated otherwise, chemicals were purchased from Carl Roth (Karlsruhe, Germany), Thermo Fisher Scientific (Waltham, USA) and Merck (Darmstadt, Germany). Remicade® was from Janssen Biologics (Leiden, Netherlands). Further antibodies were: Monoclonal human anti-IFX IgG (ADA calibrator; #HCA233, Bio-Rad Laboratories, Hercules, USA); mouse anti-human (msαhu) IgG (#209-005-082, Jackson ImmunoResearch, West Grove, USA) and donkey anti-goat (dkαgt) IgG (#705-005-147, Jackson ImmunoResearch). Human TNF (#10602-HNAE) was purchased from Sino Biological (Beijing, China), human transferrin (hTf; #T3309) from Merck and APEX™ Antibody Labeling Kit Alexa 488 from Thermo. Dynabeads® M-280 Tosylactivated were from Life Technologies (Carlsbad, USA) and CM5 sensor chips from Cytiva (Marlborough, USA).

Biosensor surfaces

Two SPR biosensor assays were developed on a Biacore X100 (Cytiva) comprising two microfluidic flow cells (Fc): IFXmon featuring TNF immobilized on Fc2 and ADAmon carrying IFX on Fc2 (Supplementary Figure S1). For referencing, a control ligand was immobilized on Fc1 (IFXmon: hTf; ADAmon: dkαgt IgG). All ligands were immobilized at 25 °C with a flow rate of 10 μL⋅min−1 and PBS as running buffer using a target level wizard (target: 5000 RU). First, the carboxymethylated dextran matrix on the CM5 chip was activated with EDC/NHS (200 mM EDC-HCl and 50 mM NHS) for 420 s (Table 1). Second, ligand in 10 mM sodium acetate, pH 4.5 (TNF: 75 μg⋅mL−1, hTf: 75 μg⋅mL−1, IFX: 20 μg⋅mL−1 or dkαgt IgG: 25 μg⋅mL−1) was injected. Third, coupled ligand was stabilized by a 30-s pulse of EDC/NHS. Fourth, unreacted NHS esters were saturated with 1 M ethanolamine, pH 8.5, for 420 s.

Table 1:

Solutions for SPR analyses.

Solution Compositiona
IFXmon running buffer 0.25% (wt/vol) casein in PBS
ADAmon running buffer 0.1% (wt/vol) casein in PBS
msαhu IgG 50 μg⋅mL−1 msαhu IgG in 0.25% (wt/vol) casein in PBS
Regeneration solution 1 10 mM glycine, 10 mM EDTA, 0.1% (wt/vol) CHAPS, 0.1% (wt/vol) Zwittergent®, 0.1% (vol/vol) tween 20, 0.1% (vol/vol) tween 80
pH adjusted to 1.5
Regeneration solution 2 30 mM glycine, 30 mM ethanolamine, 150 mM KSCN, 600 mM MgCl2, 600 mM GuHCl, 300 mM urea
Regeneration solution 3 10 mM glycine, 150 mM NaCl, 10% (vol/vol) glycerol
pH adjusted to 2.0
Regeneration solution 4 10 mM NaOH
  1. aRegeneration solution optimization was achieved with an adapted version of the multivariate cocktail approach from [24]. All solutions except for msαhu IgG were filtered (0.22 µm).

ADA purification (pulldown)

Pre-analytic acidification followed by neutralization in presence of excess immobilized or labeled ligand is well-known to enable analytical drug tolerance. Direct acidification and neutralization of diluted serum followed by extremely time-critical injection over the biosensor was not reproducible with our Biacore X100 due to an instrument-related time lag [23]. Hence, we included an acidification procedure into a simple pre-analytic protocol for ADA enrichment from serum (Supplementary Figure S2).

IFX-coupled magnetic beads were prepared according to the manufacturer’s instructions. 4 mg of IFX-coupled beads were resuspended in 100 µL serum and 550 µL PBS and incubated at 37 °C for 10 min under overhead rotation. In order to dissociate serum IFX:ADA complexes, reaction batches were subjected to pre-analytic acidification by adding 500 µL 10 mM glycine, pH 1.5 to the bead suspension to achieve a pH of 2.5, followed by 5-min incubation at 37 °C while rotating. After neutralization to pH 7.0 with 150 µL 1 M Tris-HCl, pH 8.0, the beads were incubated for 60 min at 37 °C under rotation, allowing bead-bound IFX to outcompete free IFX during re-association of ADA:IFX complexes. After washing twice with 1 mL PBS, ADA were eluted through a 5-min incubation with 100 µL regeneration solution 3 (Table 1) at room temperature. Eluates were immediately neutralized in fresh reaction vials containing 8 µL 1 M Tris-HCl, pH 8.0.

IFXmon and ADAmon assays

All analyses were performed at 25 °C with a flow rate of 10 μL⋅min−1 using PBS-casein as running buffer (Table 1). Samples were injected simultaneously over Fc2 and Fc1 for 300 s, followed by 300 s dissociation monitoring (Supplementary Figure S3).

For IFXmon, sera diluted 1:50 with running buffer were injected and bound IFX was enhanced by injecting msαhu IgG (Table 1) for 300 s, followed by dissociation for 300 s and subsequent regeneration of the sensor surface by 25-s and 12-s pulses of regeneration solutions 1 and 2 (Table 1), respectively. For ADAmon analyses, 90 µL eluate were mixed with 10 µL 1% (wt/vol) casein in PBS. The sensor surface was regenerated by 25-s and 12-s injections of regeneration solution 3 and 4 (Table 1), respectively. Efficacy of surface regeneration for both biosensor assays was demonstrated (Supplementary Figure S4).

Biosensor calibration and signal referencing

Concentrations or equivalents in the following refer to undiluted serum. For IFXmon calibration, Remicade® was spiked into blank serum matrix (pool of 55 healthy control sera) diluted 1:50 in running buffer. A hyperbolic model was fitted over the msαhu IgG binding responses of seven different Remicade® concentrations (0, 0.5, 1.0, 3.0, 5.0, 15.0, 125 μg⋅mL−1). To control surface functionality, binding of 7.0 μg⋅mL−1 Remicade® was recorded thrice per run. Msαhu IgG binding responses were double-referenced: The Fc2-Fc1 sample binding signal was corrected for the prior Fc2-Fc1 blank serum matrix signal, which was repeatedly measured throughout each run, to correct for sensor age-dependent, non-specific binding.

For ADAmon, ADA calibrator was spiked into blank serum matrix, incubated for 30 min at 37 °C and ADA were purified as described above. For calibration, a hyperbolic model was fitted over the ADA calibrator binding responses for six different ADA calibrator concentrations (0, 0.5, 1.0, 2.0, 5.0, 15.0 μg⋅mL−1). In the following, ADA calibrator concentrations are indicated in µg⋅mL−1, while concentrations of patient ADA relative to the ADA calibrator signal are given in equivalents, µgEq⋅mL−1. To control surface functionality, 2.5⋅μg mL−1 ADA was injected thrice in each run. Fc2-Fc1 binding response was calculated to correct for non-specific binding. The single-referencing approach was justified in case of the ADA biosensor by the higher sample purity and the absence of significant baseline drifts.

Assessment of ADA binding stability

Standard kinetic analyses requiring absolute analyte concentrations are impeded by the polyclonal nature of ADA. Hence, we defined the dissociation ratio (DissR) to estimate ADA:IFX complex stability from the concentration-independent dissociation phase of SPR sensorgrams: DissR=Dissearly/Disslate, where Dissearly and Disslate represent binding signals in the early and the late dissociation phase (415 and 795 s after cycle start, respectively) [25]. A DissR value close to 1 indicates high ADA:IFX stability.

ELISA

IFX and/or ADA levels were determined in all patient sera at the MVZ Medizinisches Labor Oldenburg GmbH (Oldenburg, Germany) employing the IDKmonitor® Infliximab drug level ELISA and IDKmonitor® Infliximab total ADA ELISA (both from Immundiagnostik, Bensheim, Germany), respectively. The manufacturer indicates the limits of quantification (LOQ) as 0.6 μg⋅mL−1 and 10 AU⋅mL−1 for the IFX and ADA ELISA, respectively. For better comparability with ADAmon results, the original ELISA data in AU⋅mL−1 were referenced to ADA calibrator µgEq⋅mL−1. For this, ELISA data were recalculated with a calibration curve consisting of ELISA measurements of blank serum matrix spiked with six different ADA calibrator (HCA233) concentrations, as described in the supplement (Supplementary Figure S5, Supplementary Material 2).

Statistical analysis

Statistical analyses were performed with GraphPad Prism 8.0 (GraphPad Software Inc., San Diego, USA), R version 4.0.3 (R Foundation for Statistical Computing, Vienna, Austria), and RStudio version 1.3.1093 (RStudio PBC, Boston, USA). p-Values were considered significant if <0.05.

Results

Validation of IFX quantification

IFXmon validation is described in the supplement, since a similar assay has been published previously (Supplementary Figures S6, S7; Supplementary Table S3) [23]. IFX concentrations of 84 sera (11, 52 and 21 from Remicade®-, Inflectra®-and Remsima®-treated patients, respectively) from 15 IBD patients were analyzed in duplicates with IFXmon. Method comparison with IDKmonitor® Infliximab drug level ELISA indicated that IFXmon yields minimally higher values (Figure 1). At 95% level, the Passing-Bablok regression line had a slope of 1.040 (0.961–1.113) and an intercept of 0.003 (−0.251–0.511). When constricting Passing-Bablok regression to subtherapeutic IFX concentrations ≤3.0 μg⋅mL−1, the slope and intercept were 1.029 (0.745–1.803) and 0.009 (−0.295–0.235), respectively. IFX detection concordance between the assays was 100%, with n=6 IFX-negative sera. Limits of agreement were −2.18 and 2.93 μg⋅mL−1 for all observed IFX concentrations. IFXmon was similarly sensitive (limit of detection [LOD]: 0.6 μg⋅mL−1, LOQ: 0.9 μg⋅mL−1) as compared to the ELISA. Overall, these data suggest interchangeability of IFXmon and the ELISA.

Figure 1: 
Method comparison regression analysis of IFXmon vs. ELISA. (A) Comparison of IFX concentrations in n=84 patient sera measured with IFXmon and ELISA. Passing-Bablok regression (dark blue line) is depicted with 95% confidence as light gray-shaded area. Slope and intercept (with 95% confidence intervals) of Passing-Bablok regression were 1.040 (0.961–1.113) and 0.003 (−0.215–0.511), respectively. (B) Bland–Altman analysis for absolute differences in IFX concentration between IFXmon and ELISA vs. method mean. Orange lines denote the limits of agreement at 95% confidence.
Figure 1:

Method comparison regression analysis of IFXmon vs. ELISA. (A) Comparison of IFX concentrations in n=84 patient sera measured with IFXmon and ELISA. Passing-Bablok regression (dark blue line) is depicted with 95% confidence as light gray-shaded area. Slope and intercept (with 95% confidence intervals) of Passing-Bablok regression were 1.040 (0.961–1.113) and 0.003 (−0.215–0.511), respectively. (B) Bland–Altman analysis for absolute differences in IFX concentration between IFXmon and ELISA vs. method mean. Orange lines denote the limits of agreement at 95% confidence.

Validation of ADA quantification

The pre-analytical ADA pulldown procedure facilitated analytical IFX tolerance of the ADAmon assay, as demonstrated by experiments with Alexa Fluor 488 dye-labeled ADA calibrator (ADA488), and is described in detail in the supplement. In brief: ADA488 yields ranged from 16 to 21%, independently of both ADA488 concentration and presence of 1.7–17-fold molar excess of IFX (Supplementary Figure S8A, B). Without pre-analytical acidification, ADA488 yields in presence of IFX were reduced drastically (Supplementary Figure S8B).

LOD and LOQ of the ADAmon assay were calculated as 0.14 µgEq⋅mL−1 and 0.30 µgEq⋅mL−1, respectively (Supplementary Figures S9, S10). Due to the additional sample processing, accuracy and precision of ADAmon were slightly inferior to IFXmon (Table 2). ADA concentrations were analyzed with ADAmon in 129 sera of 44 IBD patients and compared to IDKmonitor® Infliximab total ADA ELISA results. Our data indicate poor correlation of ADA concentrations between ADAmon and ELISA results (R2=0.267, p<0.0001, Figure 2A), with absolute values differing extremely: all ADAmon results were higher as compared to ELISA results, with factors ranging from 3 to 1,557 (median: 36.5; interquartile range: 12.5–84). In 89 (69.6%) sera, ADA detection was concordant between ADAmon and ELISA (Figure 2B). 26 (20.3%) sera were ELISA-positive only, as opposed to 13 (10.1%) being ADAmon-positive only.

Table 2:

Accuracy and precision of ADA quantification with ADAmon and comparison to IFXmon.

ADAmon IFXmon
c(ADA), µg⋅mL−1 n Mean, µg⋅mL−1 SDa, µg⋅mL−1 CVb, % Accuracyc, % Precisionc, % c(IFX), µg⋅mL−1 CVb, % Accuracy, % Precision, %
0.25 9 0.274 0.074 27.0 109.6 73.0 1.0 8.8 97.3 91.2
0.5 9 0.508 0.056 11.1 101.6 88.9
1.0 8 0.999 0.132 13.2 99.9 86.8 5.0 6.3 91.5 93.7
2.5 8 2.708 0.422 15.6 108.3 84.4
5.0 8 5.357 0.793 14.8 107.1 85.2 30.0 4.9 98.1 95.1
Summary 99.9–109.6% 73.0–88.9% Summary: 91.5–98.1% 91.2–95.1%
  1. aSD, standard deviation; bCV, coefficient of variation. cADAmon accuracy and precision were determined on three different days and two different sensor chips. Precision is given as 100% – CV. More detailed data on IFXmon precision and accuracy are contained in the Supplemental Material (Supplemental Table S3).

Figure 2: 
Method comparison of ADAmon and ELISA for ADA quantification and detection. (A) Linear regression analysis of ADA quantification using ADAmon and ELISA for unique analyses of 127 sera from 43 patients. Two out of the 129 sera measured by ADAmon were excluded (W-IFX-15b: outlier, MRI-IFX-22g: no ELISA data for ADA available). The 95% confidence band of the regression is represented by light gray shading. (B) ADA detection rate comparison between ADAmon and ELISA. Concordance of ADAmon and ELISA is highlighted in gray.
Figure 2:

Method comparison of ADAmon and ELISA for ADA quantification and detection. (A) Linear regression analysis of ADA quantification using ADAmon and ELISA for unique analyses of 127 sera from 43 patients. Two out of the 129 sera measured by ADAmon were excluded (W-IFX-15b: outlier, MRI-IFX-22g: no ELISA data for ADA available). The 95% confidence band of the regression is represented by light gray shading. (B) ADA detection rate comparison between ADAmon and ELISA. Concordance of ADAmon and ELISA is highlighted in gray.

Validation of DissR

Investigating ADA-related information besides quantity may be worthwhile, since ADA concentrations are challenging to interpret and to harmonize. Hence, the real-time binding information offered by ADAmon was exploited to characterize the patient-individual ADA binding properties (Supplementary Figure S11). In theory, DissR should indicate ADA:IFX stability independently of ADA concentration and inter-serum matrix effects, but co-purified non-specific binders might confound DissR. Therefore, the influence of these potential confounding factors was tested (Supplementary Figure S12). A slight trend towards smaller DissR variability for higher ADA concentrations was observed, but the coefficient of variation was overall very small (<1%). In relation to DissR variability of different ADA-positive sera, concentration-related variability and serum matrix-related variability were negligible.

Diagnostic implications of ADA binding properties

Of note, ADAmon-only positive sera had a median DissR of 2.117 with 10 sera above and 3 sera below the overall median DissR (1.835, Figure 3). For ADAmon and ELISA double-positive sera, median DissR was 1.758, with 17 sera above and 23 sera below overall median. The difference of DissR medians between these two groups was at the borderline of significance (p=0.052), similarly to Fisher’s exact test of the distribution of low-DissR and high-DissR sera between the groups (p=0.054). Our data indicate that in the low DissR group, ADA quantification by ELISA correlates with DissR over the entire ADA concentration range (R2=0.581, p<0.0001), while ADAmon measures ADA concentrations more independently of DissR (R2=0.292, p=0.004; Supplementary Figure S13).

Figure 3: 
DissR comparison between ADA detected by ADAmon only and ADA detected by both methods. The dotted line at 1.835 denotes the overall median DissR, which separates the data points into a low-DissR (orange circles) and high-DissR group (dark blue squares).
Figure 3:

DissR comparison between ADA detected by ADAmon only and ADA detected by both methods. The dotted line at 1.835 denotes the overall median DissR, which separates the data points into a low-DissR (orange circles) and high-DissR group (dark blue squares).

Additionally, DissR and ADA concentration dynamics over time were evaluated in all patients with n≥2 ADA-positive sera (Figure 4A). The median time interval between data points was 14.8 weeks, with a range from 4.9 (W-IFX-20) to 96.1 (MRI-IFX-22) weeks. In most patients, DissR was relatively stable over time. However, a very obvious change in DissR was observed in three patients, two of which had to discontinue therapy due to secondary loss of response (LOR). Overall, most LOR patients had tendentially low DissR and high ADA concentrations.

Figure 4: 
Evaluation of the diagnostic relevance of DissR. (A) Temporal development of DissR and ADA concentrations determined by ADAmon in 14 patients with n≥2 ADAmon-positive sera alongside their remission status. Symbol color indicates the chronology of the TDM visits for each patient, at which the respective analyzed serum was collected. Minimum, median and maximal DissR and ADA concentrations are marked by black bars for each patient, respectively. (B) Correlation of IFX concentrations with DissR for 52 ADA-positive sera from 26 different patients with available DissR values and IFX concentrations (determined by ELISA). Different symbols (and colors) indicate the remission status of the patients at their last TDM visit. The 26 overrepresentation-corrected sera, which represent the last available serum of each patient, are shown as filled symbols. Of note, the conspicuously high-IFX LOR serum (MRI-IFX-25) was mistakenly drawn 3 weeks after the prior IFX dose and does therefore not represent a trough concentration. (C) DissR boxplots grouped by known outcome for all sera. (D) DissR boxplots grouped by known outcome, considering overrepresentation-corrected sera only. Median DissR differed significantly between remission (blue squares) and LOR (orange circles; C: p=0.0086, D: p=0.0014). (E) ROC analysis of all 52 sera from (B). A DissR cut-off of 1.524 indicated undetectable IFX concentration (<0.6 μg⋅mL−1) with 71.4% sensitivity and 88.9% specificity (AUC=0.825).
Figure 4:

Evaluation of the diagnostic relevance of DissR. (A) Temporal development of DissR and ADA concentrations determined by ADAmon in 14 patients with n≥2 ADAmon-positive sera alongside their remission status. Symbol color indicates the chronology of the TDM visits for each patient, at which the respective analyzed serum was collected. Minimum, median and maximal DissR and ADA concentrations are marked by black bars for each patient, respectively. (B) Correlation of IFX concentrations with DissR for 52 ADA-positive sera from 26 different patients with available DissR values and IFX concentrations (determined by ELISA). Different symbols (and colors) indicate the remission status of the patients at their last TDM visit. The 26 overrepresentation-corrected sera, which represent the last available serum of each patient, are shown as filled symbols. Of note, the conspicuously high-IFX LOR serum (MRI-IFX-25) was mistakenly drawn 3 weeks after the prior IFX dose and does therefore not represent a trough concentration. (C) DissR boxplots grouped by known outcome for all sera. (D) DissR boxplots grouped by known outcome, considering overrepresentation-corrected sera only. Median DissR differed significantly between remission (blue squares) and LOR (orange circles; C: p=0.0086, D: p=0.0014). (E) ROC analysis of all 52 sera from (B). A DissR cut-off of 1.524 indicated undetectable IFX concentration (<0.6 μg⋅mL−1) with 71.4% sensitivity and 88.9% specificity (AUC=0.825).

It is well-known that IFX levels correlate with therapy success [26, 27]. Subtherapeutic IFX indicates increased risk for LOR and may necessitate a change of the therapy regimen. Hence, the association of DissR with IFX levels was analyzed (Figure 4B) and DissR was compared between different therapy outcomes (Figure 4C and D). Of note, IFX concentrations determined by ELISA were used for these evaluations to increase sample size, since ELISA data were complete for these sera. Since DissR remained relatively stable over time in most patients, DissR from the serum collected at the latest available TDM visit was selected in order to obtain a better reflection of therapy outcome. Clusters of patients depending on therapy success can be identified: LOR patients exhibited lower IFX concentrations and low DissR, while patients in remission had a broader spectrum of IFX concentrations and significantly higher DissR (p=0.0053 and p=0.0055 for all and for corrected sera, respectively; Figure 4B–D). ROC analysis showed that in our cohort, DissR ≤1.524 indicated undetectable IFX with 71.4% sensitivity and 88.9% specificity (AUC=0.825, Figure 4E).

Employing the reference surface in SPR for sample-individual plausibility testing

The dual flow cell system of our SPR setup may be utilized for testing the specific binding plausibility of individual sera. For the 26 sera that were ADAmon-negative and ELISA-positive (Figure 2B), Fc1 binding was additionally analyzed in detail. The mean and standard deviation (SD) of the Fc1 binding signal of all other eluates determined in the same run as the respective sample were calculated. Two out of the 26 ELISA-positive but ADAmon-negative sera deviated by ≥3 SD from the average Fc1 binding signal in the respective run and were selected as critically odd (Supplementary Material 2, Supplementary Figure S14).

Discussion

This study presents two SPR-based biosensors with promising potential for diagnostic application in TDM: IFXmon for IFX quantification and ADAmon for quantification and characterization of ADA. Apart from the recent realization that ADA concentration dynamics may be more informative than absolute concentrations, we report now that DissR is a new, robust indicator of ADA binding properties [17]. To the best of our knowledge, this is the first comparison of a drug-tolerant SPR assay for ADA quantification with a drug-tolerant ELISA and the largest SPR study to decipher ADA:IFX stability [21, 23, 28]. Furthermore, the high number of sera per patient in the present cohort allows unique insights into the dynamics of analytes and ADA characteristics.

Precision and accuracy of IFXmon were slightly superior to the commercial ELISA (Supplementary Table S3). Despite the fundamentally different measurement principles, IFXmon results agreed well with the ELISA. Of note, 90.6% of samples with concentrations ≤7 μg⋅mL−1 (therapeutic range: 3–7 μg⋅mL−1) exhibited differences ≤1.5 μg⋅mL−1, which is not therapy-decisive. As expected, IFX biosimilars were quantified reliably using Remicade® for calibration [29], [30], [31]. High method concordance for IFX quantification between SPR and ELISA has also been reported by Beeg et al. [21, 23]. The ADAmon showed better accuracy, but slightly lower precision as compared to the diagnostic-approved ELISA (80.0–118.2% accuracy and 91.2–93.9% precision [32]). Most ADA concentrations determined by ADAmon in IBD sera were in the semi-quantitative range and correlation with ELISA was poor, whereby ELISA strongly underestimated ADA compared to ADAmon (3- to 1557-fold). As drug tolerance was tested in detail only for calibrator-spiked sera, the observed differences between ADAmon and ELISA might be caused by drug sensitivity of ADAmon in authentic patient sera. This could, however, be excluded when looking at patient sera with different IFX levels (Supplementary Figure S15). Beeg et al., who directly quantified ADA in diluted sera using a drug-tolerant SPR method and compared their results to a drug-sensitive ELISA, reported that both methods did not correlate [21, 23]. In accordance with our data, they observed that ELISA determined absolute ADA concentrations between 7- and 490-fold lower as compared to their SPR method [21]. Poor method concordance between established assays for ADA quantification has also been described by other groups. These discrepancies have not only been reported between different methodologies, such as ELISA and HMSA, but even between similar ELISA formats and are partly owed to the use of different calibrators [14], [15], [16, 33, 34]. Method comparability is also hampered by unit heterogeneity (e.g., ng⋅mL−1, µg⋅mL−1, AU ⋅mL−1), whereby arbitrary units are often used to underline the relative nature of indicated ADA quantities [35].

Indeed, ADA quantification is always relative to the affinity of the employed ADA calibrator and hence has to be viewed as semi-quantitative in general [33]. Accordingly, it is difficult to compare sensitivities between different methods. Our data suggest slightly higher ADA detection sensitivity for the ELISA as compared to ADAmon. However, the median ADA concentration of the 26 ELISA-only positive samples was 0.002 µgEq⋅mL−1 (21.7 AU⋅mL−1), which is not far above the LOQ (10 AU⋅mL−1) and constitutes a diagnostic gray area. Furthermore, ADA detection and quantification by ELISA were more confounded by ADA:IFX stability represented by DissR than ADAmon. Median DissR of ADAmon-only positive sera was higher than median DissR of double-positive sera, implying that ELISA misses faster-dissociating ADA. This finding is congruent with results reported by Beeg et al. (Figure 3, p=0.052; Beeg: p<0.001) [21]. A mechanistic explanation can be found in the high number of washing steps in ELISA, in which fast-dissociating ADA are more likely to be lost than slower-dissociating ADA.

Our evaluations furthermore suggest that DissR is usually stable over time and that high ADA:IFX stability is associated with therapy failure and undetectable IFX trough levels. Determination of DissR at first ADA emergence is easily possible with the ADAmon assay and may be of prognostic value. Interestingly, LOR patient MRI-IFX-35 had a high DissR, but also very high ADA, indicating that high ADA concentration can counterbalance faster dissociation (Figure 4A). It is important to mention that in our study, therapeutic decisions, including the evaluation of therapy outcomes, were based on ELISA results. Patient MRI-IFX-22 was ADA-negative in 3 out of the 5 included visits by ELISA, but had consistently low IFX concentrations (Supplementary Material 2). Since ADA levels were bordering LOQ, the patient was maintained on IFX. In contrast, ADAmon detected ADA in all except the first sample from MRI-IFX-22. It is therefore possible that the patient was falsely treated like a high-clearance patient and kept on IFX, because ADA were missed by ELISA. Therefore, the remission classification of this patient has to be interpreted with caution. It must also be considered as confounding factor that all patients in remission may still develop LOR in the future, especially as the considered therapy durations differ.

To achieve analytical drug tolerance, a simple pre-analytic ADA purification method was established, including a crucial temporary acidification step. This protocol eliminates potential re-binding with serum IFX and generates a timely stable sample. Additionally, it yields acceptably pure and concentrated ADA, enabling reliable binding analyses. Even though drug tolerance has been shown to enable earlier ADA detection and the detection of transient ADA, no clinical value has yet been found for this information [15, 36], [37], [38], [39]. However, satisfactory answers to the question whether earlier therapeutic reactions to ADA are clinically relevant, require more prospective studies [4041]. Our data suggest a more pronounced benefit of earlier ADA detection enabled by drug tolerance, in combination with DissR assessment: Since DissR was mostly stable upon first ADA detection and was correlated with clinical outcome, low DissR may alert the physician at an earlier time point to adjust therapy, whereas ADA concentrations are more inconclusive.

The small number of patients with confirmed LOR (n=10) limits the robustness of these prognostic data and their interpretation demands caution. In addition, ADA pulldown eluates contain enriched, but not pure ADA. Hence, in the SPR dissociation phase, unknown co-purified serum components (e.g., other ADA antigens if ADA were polyreactive or proteins binding to the IgG Fc region) could possibly dissociate from surface-bound ADA faster than ADA:IFX dissociates and thereby confound DissR (and ADA quantification). We have analyzed inter-individual serum matrix effects, but these analyses do not cover the complete list of potentially confounding components of IBD sera. Therefore, the promising results regarding the clinical relevance of ADA binding properties implied by our study require confirmation in a larger study cohort with a higher number of ADAmon-positive sera. The comparison of ADA concentrations determined by the ADAmon biosensor with other assays and studies is complicated by the poor harmonization of ADA quantifying assays and only meaningful if referenced to the same calibrator. Here, unlike ELISA, SPR biosensors offer a plausibility test for individual sera. Matrix-related peculiarities may cause erroneous TDM results and hence wrong therapy decisions. Using these biosensors, which also feature a reference channel (Fc1) to detect anomalous non-specific surface binding, sera causing analytical interferences and necessitating repeated blood draws could be identified more easily. Finally, SPR methods have not been established in medical laboratories so far. Thus, implementing the developed SPR assays is more cumbersome than introducing, for example, new ELISA formats, and requires advanced training for personnel and purchasing of specialized equipment. In the current assay formats with Biacore X100, the IFXmon and ADAmon biosensors are less time-efficient compared to ELISA (Supplementary Figure S1) and the purification procedure has not been automated so far. The material cost for one IFXmon and ADAmon serum analysis was 5 € and 23 €, respectively, which is comparable to ELISA. High-throughput SPR devices capable of assay multiplexing could probably outplay ELISA in time and cost efficiency.

In conclusion, this study gives new insights on the comparison of ELISA and SPR biosensor assays for IFX and ADA analytics. The presented assay principles can be easily transferred to other biologics and multiple sensors may be combined in SPR devices with more flow cells to boost time and cost efficiency. ELISA and SPR likely assess different, DissR-dependent ADA populations. As long as clinical implications of ADA binding properties remain unclear, ADA quantities must hence be regarded with caution. Since DissR offers a complement to the quantitative information on ADA, its predictive potential with respect to LOR warrants further research. If our results can be confirmed by other studies, proactive drug monitoring combined with DissR analysis could accelerate therapeutic decision-making. For example, the presence of low-DissR ADA could alert the physician to monitor more frequently and increase IFX dose, even if ADA concentration is low and IFX trough concentration is therapeutically significant. By this means, the efficiency of IFX therapy might be increased for the benefit of the patient.


Corresponding author: Prof. Dr. med. Peter B. Luppa, Institute of Clinical Chemistry and Pathobiochemistry, Klinikum rechts der Isar der Technischen Universität München, Ismaninger Str. 22, 81675 Munich, Germany, Phone: +49 89 4140 4759, E-mail:

Funding source: Stiftung für Pathobiochemie und Molekulare Diagnostik

Award Identifier / Grant number: ADAmon

Acknowledgments

We thank Christine Grubmüller for excellent technical assistance and Anna Langmann for diligent assistance in IFXmon analyses [42]. Peter Langmann kindly provided IBD patient sera and Remicade®. We are indebted to all patients for enabling this work.

  1. Research funding: This work was supported by the Stiftung für Pathobiochemie und Molekulare Diagnostik (Bonn) within the project “Monitoring TNF inhibitor serum concentrations and characterization of anti-drug antibodies with surface plasmon resonance for patient-tailored therapy of inflammatory autoimmune diseases”.

  2. Author contributions: All authors have accepted responsibility for the entire content of this manuscript and approved its submission.

  3. Competing interests: Authors state no conflict of interest.

  4. Informed consent: Informed consent was obtained from all individuals included in this study.

  5. Ethical approval: Research involving human subjects complied with all relevant national regulations, institutional policies and is in accordance with the tenets of the Helsinki Declaration (as revised in 2013), and has been approved by the Ethics Committee of the TUM medical faculty (approval no.: 289/19 S).

References

1. Ng, SC, Shi, HY, Hamidi, N, Underwood, FE, Tang, W, Benchimol, EI, et al.. Worldwide incidence and prevalence of inflammatory bowel disease in the 21st century: a systematic review of population-based studies. Lancet 2017;390:2769–78. https://doi.org/10.1016/s0140-6736(17)32448-0.Search in Google Scholar

2. Friedrich, M, Pohin, M, Powrie, F. Cytokine networks in the pathophysiology of inflammatory bowel disease. Immunity 2019;50:992–1006. https://doi.org/10.1016/j.immuni.2019.03.017.Search in Google Scholar PubMed

3. Melsheimer, R, Geldhof, A, Apaolaza, I, Schaible, T. Remicade® (infliximab): 20 years of contributions to science and medicine. Biologics 2019;13:139–78. https://doi.org/10.2147/btt.s207246.Search in Google Scholar PubMed PubMed Central

4. Pineton de Chambrun, G, Blanc, P, Peyrin-Biroulet, L. Current evidence supporting mucosal healing and deep remission as important treatment goals for inflammatory bowel disease. Expet Rev Gastroenterol Hepatol 2016;10:915–27. https://doi.org/10.1586/17474124.2016.1174064.Search in Google Scholar PubMed

5. Knight, DM, Trinh, H, Le, J, Siegel, S, Shealy, D, McDonough, M, et al.. Construction and initial characterization of a mouse-human chimeric anti-TNF antibody. Mol Immunol 1993;30:1443–53. https://doi.org/10.1016/0161-5890(93)90106-l.Search in Google Scholar PubMed

6. Food and Drug, Administration. Biosimilar product information; 2022 [Online]. Available from: https://www.fda.gov/drugs/biosimilars/biosimilar-product-information [Accessed 23 Sep 2022].Search in Google Scholar

7. European Medicines Agency. Medicines (search for “active substance” = “infliximab”); 2022 [Online]. Available from: https://www.ema.europa.eu/en/medicines/search_api_aggregation_ema_active_substance_and_inn_common_name/infliximab [Accessed 23 Sep 2022].Search in Google Scholar

8. Gulacsi, L, Pentek, M, Rencz, F, Brodszky, V, Baji, P, Vegh, Z, et al.. Biosimilars for the management of inflammatory bowel diseases: economic considerations. Curr Med Chem 2019;26:259–69. https://doi.org/10.2174/0929867324666170406112304.Search in Google Scholar PubMed

9. Papamichael, K, Lin, S, Moore, M, Papaioannou, G, Sattler, L, Cheifetz, AS. Infliximab in inflammatory bowel disease. Ther Adv Chronic Dis 2019;10: 2040622319838443. https://doi.org/10.1177/2040622319838443.Search in Google Scholar PubMed PubMed Central

10. Lázár-Molnár, E, Delgado, JC. Implications of monoclonal antibody therapeutics use for clinical laboratory testing. Clin Chem 2019;65:393–405. https://doi.org/10.1373/clinchem.2016.266973.Search in Google Scholar PubMed

11. Vermeire, S, Gils, A, Accossato, P, Lula, S, Marren, A. Immunogenicity of biologics in inflammatory bowel disease. Therap Adv Gastroenterol 2018;11: 1756283X17750355. https://doi.org/10.1177/1756283X17750355.Search in Google Scholar PubMed PubMed Central

12. Atiqi, S, Hooijberg, F, Loeff, FC, Rispens, T, Wolbink, GJ. Immunogenicity of TNF-inhibitors. Front Immunol 2020;11:312. https://doi.org/10.3389/fimmu.2020.00312.Search in Google Scholar PubMed PubMed Central

13. Steenholdt, C, Bendtzen, K, Brynskov, J, Thomsen, OØ, Ainsworth, MA. Clinical implications of measuring drug and anti-drug antibodies by different assays when optimizing infliximab treatment failure in Crohn’s disease: post hoc analysis of a randomized controlled trial. Am J Gastroenterol 2014;109:1055–64. https://doi.org/10.1038/ajg.2014.106.Search in Google Scholar PubMed

14. Imbrechts, M, Van Stappen, T, Compernolle, G, Tops, S, Gils, A. Anti-infliximab antibodies: how to compare old and new data? J Pharm Biomed Anal 2020;177:112842. https://doi.org/10.1016/j.jpba.2019.112842.Search in Google Scholar PubMed

15. Vande Casteele, N. Assays for measurement of TNF antagonists in practice. Frontline Gastroenterol 2017;8:236–42. https://doi.org/10.1136/flgastro-2016-100692.Search in Google Scholar PubMed PubMed Central

16. Lázár-Molnár, E, Delgado, JC. Immunogenicity assessment of tumor necrosis factor antagonists in the clinical laboratory. Clin Chem 2016;62:1186–98. https://doi.org/10.1373/clinchem.2015.242875.Search in Google Scholar PubMed

17. Grasmeier, MK, Langmann, AF, Langmann, P, Treiber, M, Thaler, MA, Luppa, PB. Dynamics of serum concentrations of antibodies to infliximab: a new approach for predicting secondary loss of response in inflammatory bowel diseases. Therap Adv Gastroenterol 2021;14: 17562848211037848. https://doi.org/10.1177/17562848211037849.Search in Google Scholar PubMed PubMed Central

18. van Schie, KA, Hart, MH, de Groot, ER, Kruithof, S, Aarden, LA, Wolbink, GJ, et al.. The antibody response against human and chimeric anti-TNF therapeutic antibodies primarily targets the TNF binding region. Ann Rheum Dis 2015;74:311–4. https://doi.org/10.1136/annrheumdis-2014-206237.Search in Google Scholar PubMed

19. van Schie, KA, Kruithof, S, van Schouwenburg, PA, Vennegoor, A, Killestein, J, Wolbink, G, et al.. Neutralizing capacity of monoclonal and polyclonal anti-natalizumab antibodies: the immune response to antibody therapeutics preferentially targets the antigen-binding site. J Allergy Clin Immunol 2017;139:1035–7.e6. https://doi.org/10.1016/j.jaci.2016.09.014.Search in Google Scholar PubMed

20. van Schouwenburg, PA, van de Stadt, LA, de Jong, RN, van Buren, EEL, Kruithof, S, de Groot, E, et al.. Adalimumab elicits a restricted anti-idiotypic antibody response in autoimmune patients resulting in functional neutralisation. Ann Rheum Dis 2013;72:104–9. https://doi.org/10.1136/annrheumdis-2012-201445.Search in Google Scholar PubMed

21. Beeg, M, Burti, C, Allocati, E, Ciafardini, C, Banzi, R, Nobili, A, et al.. Surface plasmon resonance unveils important pitfalls of enzyme-linked immunoassay for the detection of anti-infliximab antibodies in patients’ sera. Sci Rep 2021;11:14976. https://doi.org/10.1038/s41598-021-94431-x.Search in Google Scholar PubMed PubMed Central

22. Real-Fernández, F, Cimaz, R, Rossi, G, Simonini, G, Giani, T, Pagnini, I, et al.. Surface plasmon resonance-based methodology for anti-adalimumab antibody identification and kinetic characterization. Anal Bioanal Chem 2015;407:7477–85. https://doi.org/10.1007/s00216-015-8915-8.Search in Google Scholar PubMed

23. Beeg, M, Nobili, A, Orsini, B, Rogai, F, Gilardi, D, Fiorino, G, et al.. A surface plasmon resonance-based assay to measure serum concentrations of therapeutic antibodies and anti-drug antibodies. Sci Rep 2019;9:2064. https://doi.org/10.1038/s41598-018-37950-4.Search in Google Scholar PubMed PubMed Central

24. Andersson, K, Hämäläinen, M, Malmqvist, M. Identification and optimization of regeneration conditions for affinity-based biosensor assays. A multivariate cocktail approach. Anal Chem 1999;71:2475–81. https://doi.org/10.1021/ac981271j.Search in Google Scholar PubMed

25. GE Healthcare Life Sciences. Biacore X100 assay handbook. Uppsala; 2012.Search in Google Scholar

26. Vande Casteele, N, Feagan, BG, Gils, A, Vermeire, S, Khanna, R, Sandborn, WJ, et al.. Therapeutic drug monitoring in inflammatory bowel disease: current state and future perspectives. Curr Gastroenterol Rep 2014;16:378. https://doi.org/10.1007/s11894-014-0378-0.Search in Google Scholar PubMed

27. Brandse, JF, van den Brink, GR, Wildenberg, ME, van der Kleij, D, Rispens, T, Jansen, JM, et al.. Loss of infliximab into feces is associated with lack of response to therapy in patients with severe ulcerative colitis. Gastroenterology 2015;149:350–5.e2. https://doi.org/10.1053/j.gastro.2015.04.016.Search in Google Scholar PubMed

28. Thoren, KL, Pasi, B, Delgado, JC, Wu, AHB, Lynch, KL. Quantitation of infliximab and detection of antidrug antibodies in serum by use of surface plasmon resonance. J Appl Lab Med 2018;2:725–36. https://doi.org/10.1373/jalm.2017.024380.Search in Google Scholar PubMed

29. Afonso, J, de Sousa, HT, Rosa, I, Carvalho, J, Dias, CC, Magro, F. Therapeutic drug monitoring of CT-P13: a comparison of four different immunoassays. Therap Adv Gastroenterol 2017;10:661–71. https://doi.org/10.1177/1756283x17722915.Search in Google Scholar PubMed PubMed Central

30. Magro, F, Rocha, C, Vieira, AI, Sousa, HT, Rosa, I, Lopes, S, et al.. The performance of Remicade®-optimized quantification assays in the assessment of Flixabi® levels. Therap Adv Gastroenterol 2018;11:1756284818796956. https://doi.org/10.1177/1756284818796956.Search in Google Scholar PubMed PubMed Central

31. Neveu, B, Kunst, A, Prosser, C, Robitaille, R. An in vitro comparison of four different immunoassays for the monitoring of Infliximab biosimilars drug levels. Clin Biochem 2020;78:58–62. https://doi.org/10.1016/j.clinbiochem.2020.01.006.Search in Google Scholar PubMed

32. Immundiagnostik, AG. IDKmonitor Infliximab total ADA ELISA: Instruction manual; 2020 [Online]. Available from https://www.immundiagnostik.com/de/testkits/k-9654 [Accessed 23 Sep 2022].Search in Google Scholar

33. Luo, YR, Chakraborty, I, Lazar-Molnar, E, Wu, AHB, Lynch, KL. Development of label-free immunoassays as novel solutions for the measurement of monoclonal antibody drugs and antidrug antibodies. Clin Chem 2020;66:1319–28. https://doi.org/10.1093/clinchem/hvaa179.Search in Google Scholar PubMed

34. Gils, A, Van Stappen, T, Dreesen, E, Storme, R, Vermeire, S, Declerck, PJ. Harmonization of infliximab and anti-infliximab assays facilitates the comparison between originators and biosimilars in clinical samples. Inflamm Bowel Dis 2016;22:969–75. https://doi.org/10.1097/mib.0000000000000709.Search in Google Scholar

35. Papamichael, K, Afif, W, Drobne, D, Dubinsky, MC, Ferrante, M, Irving, PM, et al.. Therapeutic drug monitoring of biologics in inflammatory bowel disease: unmet needs and future perspectives. Lancet Gastroenterol Hepatol 2022;7:171–85. https://doi.org/10.1016/s2468-1253(21)00223-5.Search in Google Scholar

36. Grossberg, LB, Cheifetz, AS, Papamichael, K. Therapeutic drug monitoring of biologics in Crohn’s disease. Gastroenterol Clin N Am 2022;51:299–317. https://doi.org/10.1016/j.gtc.2021.12.007.Search in Google Scholar PubMed

37. Cheifetz, AS, Abreu, MT, Afif, W, Cross, RK, Dubinsky, MC, Loftus, EV, et al.. A comprehensive literature review and expert consensus statement on therapeutic drug monitoring of biologics in inflammatory bowel disease. Am J Gastroenterol 2021;116:2014–25. https://doi.org/10.14309/ajg.0000000000001396.Search in Google Scholar PubMed PubMed Central

38. Mitrev, N, Vande Casteele, N, Seow, CH, Andrews, JM, Connor, SJ, Moore, GT, et al.. Review article: consensus statements on therapeutic drug monitoring of anti-tumour necrosis factor therapy in inflammatory bowel diseases. Aliment Pharmacol Ther 2017;46:1037–53. https://doi.org/10.1111/apt.14368.Search in Google Scholar PubMed

39. Van Stappen, T, Vande Casteele, N, Van Assche, G, Ferrante, M, Vermeire, S, Gils, A. Clinical relevance of detecting anti-infliximab antibodies with a drug-tolerant assay: post hoc analysis of the TAXIT trial. Gut 2018;67:818–26. https://doi.org/10.1136/gutjnl-2016-313071.Search in Google Scholar PubMed

40. Nice, R, Chanchlani, N, Green, H, Bewshea, C, Ahmad, T, Goodhand, JR, et al.. Validating the positivity thresholds of drug-tolerant anti-infliximab and anti-adalimumab antibody assays. Aliment Pharmacol Ther 2021;53:128–37. https://doi.org/10.1111/apt.16135.Search in Google Scholar PubMed

41. van Schouwenburg, PA, Krieckaert, CL, Rispens, T, Aarden, L, Wolbink, GJ, Wouters, D. Long-term measurement of anti-adalimumab using pH-shift-anti-idiotype antigen binding test shows predictive value and transient antibody formation. Ann Rheum Dis 2013;72:1680. https://doi.org/10.1136/annrheumdis-2012-202407.Search in Google Scholar PubMed

42. Langmann, AF. Bestimmung von Infliximab und Anti-Infliximab-Antikörpern im Rahmen des therapeutischen Drug-und Immunogenitäts-Monitoring bei chronisch entzündlichen Darmerkrankungen [Dr. med. thesis]. Munich, Germany: Technische Universität München; 2022.Search in Google Scholar


Supplementary Material

This article contains supplementary material (https://doi.org/10.1515/cclm-2022-0949).


Received: 2022-09-23
Accepted: 2023-01-27
Published Online: 2023-02-08
Published in Print: 2023-06-27

© 2023 the author(s), published by De Gruyter, Berlin/Boston

This work is licensed under the Creative Commons Attribution 4.0 International License.

Downloaded on 5.3.2024 from https://www.degruyter.com/document/doi/10.1515/cclm-2022-0949/html
Scroll to top button