Development and validation of rapid assays to quantify infliximab trough levels and antibodies towards infliximab.

Crohn’s disease and ulcerative colitis, the two main subtypes of inflammatory bowel diseases (IBD), are chronic, inflammatory diseases of the gastrointestinal tract that have a relapsing-remitting character. They may affect up to 0.5% of the Western population and inflict a lifelong burden to the patient and healthcare system. The treatment of these incurable diseases dramatically improved upon the introduction of biological drugs, such as infliximab, increasing patients’ quality of life and reducing hospitalizations and surgery. Unfortunately, not all IBD patients benefit equally from these drugs; some patients do not respond at all to treatment, while a considerable group of patients experiences loss of treatment response over time. The latter is often due to the recognition of infliximab as a non-self protein, triggering the immune system to develop anti-infliximab antibodies (ATI) that neutralize the effect of infliximab and increase its elimination.

Recently, therapeutic drug monitoring (TDM) of infliximab and ATI was introduced to guide clinical decision-making in patients with IBD since low infliximab trough concentrations, which is the drug concentration just before the next dose administration, and the presence of ATI were associated with loss of treatment response. However, the current available infliximab and ATI monitoring assays exhibit large inter-assay variability and often lack sensitivity and specificity. In addition, the assay speed and need for specialized laboratory facilities along with experienced personnel hampers implementation in the clinic. Altogether, this may lead to a less efficient or wrongly adjusted treatment of these patient wasting financial resources. The objectives of this PhD project were therefore to improve the current available assays and explore novel assay platforms to facilitate the implementation of TDM in the clinic.

Application of monoclonal antibodies in the infliximab and ATI ELISA

The first objective was to generate a large number of monoclonal anti-infliximab antibodies (MA-IFX) that could be applied in the infliximab and ATI assays. Monoclonal antibodies offer the advantage to bind with high specificity and affinity to only one particular binding site of their target, thereby avoiding potential cross-reactivity with seemingly similar proteins. In addition, they are easily reproduced enabling assay standardization and harmonization. Out of 55 generated monoclonal antibodies, MA-IFX6B7 was selected to replace the polyclonal antibody (pAb) as detection antibody in the infliximab ELISA (Chapter II). The MA-based infliximab ELISA specifically detects infliximab and no other biological drugs like adalimumab and golimumab or a human IgG-mixture. Multiple comparison studies were performed with the novel MA-based infliximab ELISA, demonstrating an excellent agreement with other (commercially) available assays. Next, we replaced the pAb calibrator in the ATI ELISA by MA-IFX10F9, which can serve as a universal ATI calibrator to harmonize the ATI assays (Chapter III). Meanwhile the ATI ELISA was optimized to improve its sensitivity (4-fold) and specificity. As a result, the assay detected ATI earlier in almost half of analyzed ATI positive patients, while false positive ATI measurements were eliminated.

Development and evaluation of drug-tolerant anti-infliximab antibody assays

Conventional assays cannot detect ATI in the presence of drug (referred to as drug-sensitive ATI assays). Therefore, our second objective was to develop an ATI ELISA that does allow detection of ATI in the presence of infliximab, which is also called a drug-tolerant ATI assay. Two approaches were pursued. The first approach encompassed the addition of a sample pretreatment step, a simple add-on to the protocol of the ATI ELISA, which effectively improved the drug tolerance up to 2.5 µg/mL infliximab (Chapter IV). The second approach (“ACE approach”) included two acidification steps and removal of excess infliximab, resulting in a drug tolerance up to 10 µg/mL infliximab. Both assays markedly increased the detection of ATI in patient samples containing detectable concentrations of infliximab. In a post-hoc analysis of a subpopulation of the TAXIT trial (Chapter V), the ACE drug-tolerant assay detected ATI in 41% of patients throughout the trial (at screening, after optimization and at the end of the study), while ATI in the drug-sensitive assay disappeared upon optimization in all patients.

We could, however, not assign much clinical relevance to the drug-tolerant ACE ELISA in the TAXIT trial. Detected ATI in the drug-tolerant assay, but not in the drug-sensitive assay, were typically of low concentration and had a high chance of disappearing over time, while ATI, detected in the drug-tolerant and drug-sensitive assay, were typically of high concentration and remained detectable over time. In Chapter VI, the drug-tolerant ACE ELISA was applied in a prospective observational study in patients with ulcerative colitis who started infliximab induction treatment and did not yet receive biological therapy before. Apart from an increased detection of ATI using the drug-tolerant assay, ATI detection in the drug-tolerant assay always preceded later ATI detection in the drug-sensitive assay, but the latter did not occur in all patients. In two of these patients, long-term follow-up identified ATI-related adverse events that maybe could have been prevented by early identification of ATI.

Development and validation of rapid infliximab assays

Finally, our aim was to transfer the infliximab ELISA to alternative platforms with potential for rapid and on site infliximab monitoring. In this perspective, we evaluated a fiber optic-surface plasmon resonance (FO-SPR) platform and a lateral flow-based assay (LFA) for quantitative determination of infliximab. In FO-SPR, the reflection of light on a thin gold layer is directly linked with the number of molecular interactions on the gold interface. A FO-SPR-based bioassay was established using the antibody pair MA-IFX20G2/MA-IFX3D5 allowing quantification of infliximab in the ng/mL range, within 20 minutes. Analysis of patient samples revealed an excellent agreement of FO-SPR with the infliximab ELISA. In co-development with R-Biopharm AG (Darmstadt, Germany), we also explored the applicability of LFA for infliximab detection using MA-IFX6B7, ultimately resulting in a CE-marked assay that was validated in Chapter VI. The principle of LFA relies on thin layer chromatography combined with ELISA; the most well-known example of a LFA is the pregnancy test. In the LFA, that takes 20 minutes to complete, the presence of infliximab leads to the development of a red band at the height of the test line, where it binds to immobilized MA-IFX6B7. Analysis of more than 200 patient samples in parallel with ELISA revealed an excellent agreement between both methods.

In conclusion, this PhD project presents novel assays with excellent analytical performance for the quantitative measurement of infliximab and ATI. First, the introduction of MA-IFX in the monitoring assays eliminated specificity and reproducibility issues inherent to the first generation of assays. On top, these MA-IFX also support assay harmonization as they can be continuously produced. Second, drug-tolerant assays were developed that allow detection of ATI in the presence of drug. So far, however, we could only assign little clinical relevance to these assays. Finally, alternative infliximab monitoring assays were developed with shortened time-to-result and increased accessibility, which will facilitate the implementation of infliximab monitoring in the clinic. Summarizing, it is our hope that these novel assays will contribute to a better disease management and ultimately lead to an increased quality of life in patients with inflammatory diseases.