Biological drugs (monoclonal antibodies), directed against specific molecules, including infliximab and adalimumab, have significantly changed the treatment of chronic inflammatory bowel and rheumatic diseases, where tumor necrosis factor α plays a pivotal role. Infliximab and adalimumab belong to the group of tumor necrosis factor α inhibitors. Treatment with biological drugs is successful in about 60–70% of patients and is mostly associated with the remission of the disease. In addition to the original biological drugs, biosimilars are also available, which show comparability in both safety and efficacy of the treatment, with certain structural differences (e.g. glycosylation). All biological drugs are immunogenic, resulting in a loss of response to the treatment due to the formation of antibodies directed against the biological drugs (anti-drug antibodies). Neutralizing antibodies directed against idiotopes at tumor necrosis factor α binding site and non-neutralizing antibodies directed against other parts of the antibody are formed. By monitoring the biological drug and anti-drug antibody levels, clinicians can successfully optimize treatment. An additional challenge in the therapeutic drug monitoring of biological drugs is the choice of methods, which vary extremely. The largest differences are in the principles and methods of anti-drug antibody quantification. In addition to the problems associated with biological treatment, patients with rheumatoid arthritis are at increased risk of developing atherosclerosis and early development of cardiovascular disease compared to the group without this disease. Two main cytokines, interleukin 1β and tumor necrosis factor α, are involved in pathological processes in both atherosclerosis and rheumatoid arthritis. Within the hypotheses, we assumed that by introducing an in-house enzyme-linked immunosorbent assay to detect levels of the biological drug (infliximab and adalimumab) and specific anti-drug antibodies, we would solve the problem of cross-reactivity between original/biosimilar biological drugs and their antibodies. We also assumed that by using a methodological algorithm, a combination of detection methods, it is possible to determine the most appropriate levels of biological drug and anti-drug antibodies, which are also clinically important. The cross-reactivity between biological drugs (original and biosimilar biological drugs) and antibodies against them was investigated using in-house enzyme-linked immunosorbent assays. A biosimilar biological drug, Remsima®, is used as a standard in in-house enzyme-linked immunosorbent assays for the detection of infliximab and as an antigen and detection antibody in the bridging enzyme-linked immunosorbent assay for the detection of antibodies to infliximab. The cross-reactivity between the original infliximab Remicade®/biosimilar infliximab Remsima® and their antibodies was investigated in two parts. First, we analyzed consecutively collected samples from Remsima®/Remicade®-treated patients with one of the chronic rheumatic or inflammatory bowel diseases. The results of the in-house method were compared with the results of the commercial assay, using Remicade® in the reagents. By calculating the correlations between the results of both methods, we showed that the detection antibody directed against Remicade® cross-reacted with both Remsima® and Remicade®. In the second part, samples from patients treated with Remsima® or Remicade® with undetectable levels of the drug were analyzed with in-house and commercial bridging enzyme-linked immunosorbent assays in which Remicade® is bound on the plate. The correlation coefficients between the results of different methods showed a high degree of agreement and demonstrate cross-reactivity between the original/biosimilar infliximabs and their antibodies. We also showed cross-reactivity between original/biosimilar adalimumabs and antibodies against them in two parts. In the first part, samples with known concentrations of Humira®, Amgevita®, and Imraldi® were prepared and analyzed in the in-house enzyme-linked immunosorbent assay for the detection of adalimumab using Humira® as a standard and detection antibody directed against Humira®. We showed that by testing samples with a known concentration of three adalimumabs with the in-house enzyme-linked immunosorbent assay for adalimumab detection, the concentrations of Humira®, Amgevita®, and Imraldi® were determined accurately and precisely. In the second part, we tested samples from patients with anti-adalimumab antibodies and demonstrated cross-reactivity of antibodies with Humira®, Amgevita®, and Imraldi®. These three drugs were used as antigens in the bridging enzyme-linked immunosorbent assay. The results showed that antibodies against adalimumab cross-reacted with Humira®, Amgevita®, and Imraldi®.
We have shown that we can use therapeutic drug monitoring of biological drugs as well as antibodies against them in patients treated with biosimilars and patients treated with original drugs, and thus achieve equally reliable and reproducible results with financial efficiency solving the potential problem of cross-reactivity. The results of the cross-reactivity and method comparison studies were combined, and the obtained results were the basis for setting up a methodological algorithm. As part of the methodological algorithm, in the first step, the level of the biological drug is determined in all samples, and in the next step antibodies against biological drugs are determined only in samples with undetectable drug level. All methods for determining antibodies to biological drugs used during the experimental part of the doctoral thesis have their advantages and disadvantages. The bridging enzyme-linked immunosorbent assay, otherwise the most commonly used method due to its methodology, detects both non-neutralizing and neutralizing antibodies, but does not detect immunoglobulins group G4 due to their bi-specificity. The reporter gene assay is financially unfavorable due to its methodology where transfected cells are used. Therefore, as part of our dissertation, we developed a competitive enzyme-linked immunosorbent assay that mimics the methodology of a functional cell reporter gene assay. The results of the competitive enzyme-linked immunosorbent assay were first compared with the results of in-house bridging enzyme-linked immunosorbent assay and reporter gene assay. The correlation coefficients showed an excellent agreement between the results. Antibodies against biological drugs were detected in up to 19% more samples using a competitive enzyme-linked immunosorbent assay. These findings gave us a basis for the two-step determination of antibodies against biological drugs. Thus, samples with undetectable concentrations of the biological drug are first analyzed with the bridging enzyme-linked immunosorbent assay. If no antibodies are determined with the bridging enzyme-linked immunosorbent assay, the samples are further analyzed with the competitive enzyme-linked immunosorbent assay. Using a methodological algorithm, we maintain a well-established and widely accepted practice of determining antibodies to biological drugs using a clinically evaluated bridging enzyme-linked immunosorbent assay. By additionally testing samples with competitive enzyme-linked immunosorbent assay, antibodies are detected in a larger proportion of the samples and, therefore, earlier than with bridging enzyme-linked immunosorbent assay.
We further hypothesized and showed that biological drugs and their antibodies modulate the inflammatory response of tumor necrosis factor α-, interleukin 1β- and serum amyloid A-stimulated human coronary artery endothelial cells in culture. Human coronary artery endothelial cells were incubated with tumor necrosis factor α, interleukin 1β, serum amyloid A, adalimumab, infliximab, anti-infliximab antibodies and combinations thereof. Anti-infliximab antibodies were isolated by affinity chromatography from serum samples of two patients. In cell supernatants, the protein levels of atherogenic and anti-atherogenic analytes were measured using the enzyme-linked immunosorbent assay and multiplex assay. The expression of messenger ribonucleic acid was determined by real-time quantitative polymerase chain reaction. The roles of the individual above-mentioned stimulators have been previously demonstrated, while the synergy between them was investigated during the doctoral dissertation, along with the effects of tumor necrosis factor α inhibitors and their antibodies. Our results showed that the analytes analyzed in cell supernatants were divided into two groups. Analytes from the first group, interleukin 6, interleukin 8, granulocyte-macrophage colony-stimulating factor and growth-regulated oncogene α, were slightly responsive to serum amyloid A and tumor necrosis factor α and highly responsive to interleukin 1β. A synergistic effect upon the stimulation with tumor necrosis factor α, interleukin 1β and serum amyloid A was observed in their released levels. All mentioned analytes are associated with the occurrence of atherosclerosis. From the second group of analytes, where a synergistic effect on stimulation with tumor necrosis factor α, interleukin1β and serum amyloid A was not observed, vascular cell adhesion molecule-1 and monocyte chemotactic protein-1 were the most elevated. These two analytes are also associated with the development of atherosclerosis. In the experiments using infliximab and adalimumab concentrations considered clinically relevant in the serum of patients, our results showed an inhibition of the inflammatory effect of tumor necrosis factor α and the potential involvement of infliximab and adalimumab in the reduction of endothelial activation. The appearance of antibodies against biological drugs in patients can, if they are present for a long time, affect both the failure of treatment of the primary disease and the development of atherosclerosis. The results of the doctoral dissertation significantly contribute to the knowledge of cross-reactivity between biological drugs (original and biosimilar) and antibodies against them, and show the development of stepwise detection of anti-drug antibodies using methodological algorithm in daily routine practice. The results of cell experiments show synergistic effects of tumor necrosis factor α, interleukin 1β, and serum amyloid A on human coronary artery endothelial cells by measuring various molecules in their supernatants involved in atherosclerosis, and show the involvement of adalimumab, infliximab and anti-infliximab antibodies in the inflammatory processes of human coronary artery endothelial cells. The results suggest the intertwining of tumor necrosis factor α signaling pathways in the pathogenesis of rheumatic diseases and in atherosclerosis.
|
