Development of anti-rituximab antibodies in rituximab-treated patients: Related parameters & consequences

The targeting of cancerous cells and the strengthening of anti-tumour immune mechanisms are among the strategies that are frequently considered in cancer immunotherapy1. Monoclonal antibodies (mAbs) exhibit promising therapeutic potentials in cancer immunotherapy and treatment of autoimmune diseases as they bind specifically to antigenic targets. The therapeutic effects of mAbs are exerted through a number of mechanisms such as the killing of target cells, receptor-ligand inhibition and receptor blocking23456.

The clinical application of a mAb has been challenged by a number of problems, especially its immunogenicity. The administration of a non-humanized mAb to humans may stimulate the production of antibodies to some regions of that mAb such as fragment of antigen binding, fragment of crystallizable and complementarity-determining regions (CDRs)27. Further, fully human mAbs can contain epitopes in their CDRs which may cause an antibody response through the network of idiotypes/anti-idiotypes2. The produced anti-drug antibody (ADA) limits the binding of mAb to target antigens and promotes its clearance largely through hepatic and splenic macrophages27. In addition, the ADA may interfere with immunodiagnostic techniques leading to false results and incorrect diagnosis, and therefore, inappropriate treatment8.

Several factors can affect the immunogenicity of therapeutic mAbs such as protein structure, doses, treatment programme, patient co-medication, immune status of the patients, genetic predisposition of the patients, underlying disease and, age and gender of the patients91011121314. Antibodies targeting cell membrane-linked molecules may have a higher risk of immunogenicity compared with antibodies targeting soluble molecules215. This phenomenon may be attributed to the antigen internalization into target cells and subsequently its processing and presentation to patient’s specific T lymphocytes, which then enable B-cells to produce high-affinity ADA215. When a target antigen is present on the cell membrane, mAbs bind to the target antigen and are quickly internalized along with the target antigen, leading to rapid uptake of mAbs into the cell16. Interestingly, the disappearance of CD20 and internalization of CD20-rituximab have been reported in some rituximab-treated patients with CLL1718. The internalized mAb which then acts as an antigen, is processed and eventually presented to T cells through interaction between the T cell receptor and the major histocompatibility complex II-antigen complex on antigen-presenting cells (APC), resulting in ADA production through a T-cell-dependent manner. In these circumstances, the Th cell-derived cytokines help B cells to produce high-affinity ADA from various isotypes, such as IgG and IgE19. The mAb-related epitopes may directly cross-link the surface immunoglobulins of the specific B-cells, resulting in the production of anti-drug IgM in a T-cell-independent manner1920. As there are different ADA isotypes, these may also cause various side effects in their recipients, such as allergic reactions, serum sickness and renal failure1920.

Rituximab is a human/murine chimeric mAb that is composed of the human kappa and IgG1 constant regions connected to the murine light- and heavy-chain variable parts, respectively21. Rituximab specifically binds to the CD20 marker that is expressed on the B lymphocytes and exerts its cytotoxicity through induction of the apoptosis, complement activation, and antibody-dependent cell-mediated cytotoxicity222324252627. Rituximab is used for the treatment of some malignancies such as CD20⁺ B-cell non-Hodgkin’s lymphoma and chronic lymphocytic leukaemia, autoimmune disorders associated with the presence of autoantibodies, including systemic lupus erythematosus (SLE) and rheumatoid arthritis (RA), or improvement of the graft survival2829.

The characterization of parameters affecting the production of anti-mAb antibodies can be used to design strategies to reduce the immunogenicity of a mAb and promote its efficacy in humans. There are many studies on the rituximab immunogenicity and its sequels, but it is necessary to provide a comprehensive description of this subject. In this review, a comprehensive insight regarding the host and therapeutic programme-related parameters that influence the development of the anti-rituximab antibody (ARA) are provided and novel insights to reduce ARA-associated adverse events and enhancement of drug efficacy are suggested.

The possible parameters influencing anti-rituximab antibody production

The effect of number of rituximab infusions on anti-rituximab antibody formation: The number of rituximab infusions may be associated with the production of ARA. In one study, the ARA was detected in about 33 per cent of the rituximab-treated multiple sclerosis (MS) patients30. Among patients with relapsing-remitting MS (RRMS) and primary progressive MS (PPMS), a negative correlation was also observed between the infusion numbers of rituximab and ARA positivity30. Furthermore, the ARA was not detectable in rituximab-administered patients with relapsed mantle cell lymphoma shortly after the second and third course of treatment31. There was also a significant association between serum concentrations of rituximab, serum ARA titre, B-cell count and clinical responses31. The mentioned studies display an inverse correlation between number of rituximab infusions and risk of ARA development. The diminished count of B-cells or immune tolerance to rituximab may be considered as possible reasons for decreased ARA titre in subsequent administrations in rituximab-treated patients. Nevertheless, there was no significant relationship between the injection numbers of rituximab and development of the ARA in rituximab-treated patients with lymphoma or leukaemia in one of our previous studies32. There was also no significant relationship between the number of infusions and ADA concentrations in infliximab-treated patients with Crohn’s disease7.

The impact of age and gender on anti-rituximab antibody formation: The relationship between age and ARA production has rarely been studied. In one study performed on infliximab-administered patients with Crohn’s disease, the ARA was detected in 2.7 per cent and 11 per cent of children and adults, respectively33. In another study, the ARA was detected in 37 per cent of rituximab-administered patients with RRMS and 26 per cent of rituximab-treated patients with progressive forms of MS. However, association was found between ARA production and the age or gender of MS patients30.

The effects of disease type on anti-rituximab antibody formation: It has been demonstrated that the disease type influences the immunogenicity of rituximab in mAb-administered subjects. Therefore, variable ARA positivity was reported in rituximab-treated patients with different diseases. The development of ARA was reported in about 2.7 per cent of patients with non-Hodgkin’s lymphoma34353637, in <4 per cent patients with diffuse large B-cell lymphoma383940 and in 19.8 per cent patients with follicular B-cell lymphoma41. It seems that rituximab exhibits differential immunogenicity in different types of B-cell lymphoma (Table I)4243. In a study44 on 166 rituximab-treated patients with relapsed low-grade or follicular lymphoma, ARA was detected only in one patient on day 50. Also, no association was observed between the ARA seropositivity and laboratory or clinical abnormalities. Similarly, in 11 rituximab-administered patients with relapsed B-cell lymphoma, no patients were found to develop ARA34. Moreover, ARA was not quantifiable in 15 rituximab-administered patients with B-cell lymphoma36. In our recent study, the development of the ARA was found in four out of 32 (12.5%) rituximab-treated patients with lymphoma or leukemia32. It was also observed that the chemotherapy may influence the development of the ARA in patients with lymphoma or leukemia32.

Regarding autoimmune diseases, ARA was detected in 7 - 37 per cent of patients with RRMS30,44,45, 26.5 per cent patients with PPMS, 1.8 - 21.7 per cent of patients with RA, 16.6 - 50 per cent patients with SLE4647, 27 per cent patients with Sjögren’s syndrome48 and in 18.18 per cent patients with Pemphigus vulgaris49, who were treated with rituximab (Table II). The development of ARA was also indicated in 21 per cent of rituximab-treated patients with Crohn’s disease56. It is obvious that the rate of ADA positivity was higher in rituximab-administered patients with autoimmune diseases compared to patients with lymphoma. Therefore, in active autoimmune diseases, a mAb tends to exhibit greater immunogenicity, regardless of the type of disease. However, the results from an investigation suggest a greater rate of ARA positivity in patients with RRMS than patients with PPMS (37 vs. 26%). The reason for this differential immunogenicity of rituximab in patients with various patterns of MS is not clear. However, higher intensity of immune responses during relapsing stages may influence this parameter30.

Some immunological disorders such as defects in the effector T-cell-mediated anti-tumour immune response and hyper-activation of regulatory T cells have been reported in patients with malignancies5758596061. Therefore, lower immunogenicity of a mAb in malignant patients, such as B-cell malignancy may be due to the general immunosuppression that dampens the B-cells responsible for ADA production1162. Since mAbs against B cell-related markers suppress the B cells responsible for ADA production, it may be postulated that mAbs against B cell-related markers would be inherently less immunogenic than other therapeutic mAbs63.

The association of the B-cell number with anti-rituximab antibody formation: The results of a study on rituximab-administered patients with MS indicated that there was a powerful association between ADA positivity and higher B-cell count. The ARA titre and positivity were greater in patients with lower B-cell depletion30. Similarly, ARA development was associated with reduced B-cell depletion in rituximab-treated patients with SLE46. The aforementioned studies clearly indicate an inverse correlation between serum levels of ARA and the circulating number of B-cells. However, no association was found between ARA titres with circulating B cell numbers, mAb-related harmful events, or clinical response rate in ARA-positive patients with RA64.

The effects of rituximab types on anti-rituximab antibody formation: The results of a study on RA patients showed that the ARA was similarly developed in patients treated with rituximab biosimilar forms such as rituximab-Pfizer (PF-05280586), rituximab-EU and rituximab-US65. In another study, it was also demonstrated that the immunogenicity of biosimilar GP2013, rituximab-EU and rituximab-US was similar in patients with active RA53. Moreover, similar immunogenicity was reported for CT-P10 (a rituximab biosimilar) and rituximab in RA patients5266. Similar immunogenicity was also reported for Kikuzubam (Rituximab biosimilar) and MabThera (Rituximab)22. Collectively, the findings from different studies summarized in Tables I and II indicate that in most circumstances, the rituximab and its biosimilar drugs exhibit similar immunogenicity in patients with lymphoma or autoimmune diseases. However, a discrepancy was observed in some situations considering the immunogenicity of rituximab and its biosimilar (Tables I and II).

The association between patients’s genetic profile and anti-rituximab antibody formation: The genetic background is an essential patient-related parameter affecting the antigenicity of a biological therapeutic agent67. The polymorphisms in human leukocyte antigen (HLA) are related to the likelihood of ADA formation. For instance, the HLA-DR1 locus was associated with a greater prevalence of the ADA against infliximab in patients with Crohn’s disease68. Around 81 per cent of ADA-positive patients displayed DRβ S13 residue, in comparison with 50 per cent of ADA-negative patients68. To date, there is no way to distinguish the producers of ARA from non-producers of ARA before treatment with a rituximab. As immune response genes, especially human leukocyte antigen (HLA)-related genes play a fundamental role in the induction of antibody response to a given antigen69, the clarification of the association between the HLA genes and development of ARA needs to be considered in future studies. If the association of some HLA genes with the development of ARA is confirmed, then the risk of drug immunogenicity may be predictable prior to rituximab treatment.

The association between the rituximab efficacy and anti-rituximab antibody development

The ADA development may have important clinical consequences in patients with autoimmune and malignant diseases treated with mAbs. However, no significant association was reported between the ARA positivity and serum drug levels in rituximab-treated patients with lymphoma (Table I). Furthermore, ARA positivity did not significantly influence the drug efficacy in rituximab-administered patients with lymphoma (Table I).

The association of the ARA positivity and drug levels and efficacy was also reported in a number of autoimmune disorders with inconsistent results (Table II). For example, no significant difference was found between ADA-positive and ADA-negative MS patients concerning the efficacy of rituximab3070. The results from an investigation revealed that the ADA positivity did not influence the drug level and efficiency in rituximab-treated patients with SLE71. The results from a multinational study indicate that ARA-positive patients with RA exhibit lower drug levels compared with ARA-negative patients, but without influencing the drug efficacy53. However, it was reported that ARA-positive patients with SLE exhibit lower drug levels along with lower drug efficacy compared to ARA-negative patients46. In addition, the ARA positivity suggestively influences the treatment efficacy in rituximab-treated patients with Pemphigus negatively49. Furthermore, higher titre of ARA was accompanied by higher disease activity at baseline in rituximab-administered patients with SLE46. The presence of ADA may reduce the serum levels of administrated mAb. The RA patients who were positive for ADA had lower levels of administrated mAb and higher clearance as compared to patients who were negative for ADA65.

Anti-rituximab antibody-related adverse clinical consequences

Immune response-linked adverse events are the most frequent side-effects in mAb-treated patients, which mainly affect the skin and gastrointestinal tract with less frequent manifestations in the liver, endocrine and nervous organs72. The side effects of a mAb may be partly attributed to its immunogenicity. The results from studies summarized in Table 1 indicate that there was no significant association between ARA positivity and expression of adverse events in rituximab-treated patients with lymphoma. Moreover, no significant association was reported between ARA positivity and expression of adverse events in rituximab-treated patients with autoimmune diseases such as MS and RA, and pemphigus vulgaris (Table II). In studies carried out on rituximab-administered MS patients, the presence of ARA was not related to the type or severity of harmful events during the study4445. No significant differences were reported between rituximab-treated RA patients with positive or negative ARA status concerning the serious adverse events47,50-54. No correlation was reported between ARA production and elevated risk of infusion-associated adverse reactions in rituximab-treated patients with pemphigus vulgaris49.

However, the expression of serum sickness events was reported in some rituximab-treated patients with SLE or Sjögren’s syndrome (Table II). Four serum sickness events (3 of the 4 patients were positive for ARA) were observed in the 169 rituximab-administered patients with SLE compared with no event in the 88 placebo-treated subjects55. In another study carried out on the 15 rituximab-administered patients with Sjögren’s syndrome, four out of 15 patients (27%), were positive for ARA, of these three exhibited serum sickness48. The results of a systematic review on 25 studies indicated 33 cases with rituximab-mediated serum sickness73. The expression of the serum sickness occurred mainly after the second dose in the first cycle of infusion73. Further, the rituximab-mediated serum sickness is more prevalent (> 12 times) in patients with autoimmune disorders compared to patients with haematological malignancies74. However, the reasons for higher development of the serum sickness in patients with autoimmune diseases remains to be clarified in the future. Serum sickness has been observed in patients with concomitant presence of hypergammaglobulinemia and rheumatoid factor7374. The concomitant chemotherapy used for treatment of the malignancies may be protective against serum sickness in rituximab-treated subjects74.

The B cells act as efficient APCs, because they express HLA class II molecules75. The internalization of CD20-rituximab was reported in some rituximab-treated patients with CLL1718. Therefore, B cells can present rituximab-derived peptides to specific Th cells in association with the HLA class II molecules. Then, Th cell-derived cytokines help B cells to produce high-affinity ADA from various isotypes, such as IgG and IgE19. The antibody response to rituximab as an antigenic protein can be influenced by numerous parameters including host-related factors (HLA gene, cytokine gene polymorphisms, age, gender, immunosuppression, disease type and concomitant medication). Some other host-related parameters may also affect the ADA development, such as weight, nutrition, psychological stress and smoking. The treatment-related factors such as route of administration, dose, infusion numbers and duration of treatment also influence the ADA production.

The findings presented in this study indicate that the risk of ARA development and ARA-related adverse events is low in rituximab-treated patients with lymphoma. No significant association was reported between the ARA positivity and serum levels and drug efficacy either in rituximab-treated patients with lymphoma (Table I). However, the patients with autoimmune disorders exhibit a greater risk of ARA development and ARA-related adverse events. Therefore, it is required to outline the major criteria to predict the rituximab immunogenicity before starting the drug treatment. In autoimmune diseases, ARA positivity may have no significant impacts on either the drug level or its efficacy30,44,45,47,50, so it may reduce drug levels without influencing its efficacy53, or may reduce the drug efficacy without influencing the levels51, or may reduce both drug level as well as its efficacy4649. The exact evaluation of both the host- and treatment-related parameters, and the characterization of ARA are, hence, necessary to clarify regarding factors influencing the drug levels and its efficacy in rituximab-treated patients with autoimmune disorders. Structural modifications in a drug to decrease its immunogenicity, combinational therapy using an appropriate B-cell modulator, and removal of the ADA may be considered as strategies to increase the efficacy of a mAb.

Moreover, various immunoassay methods such as enzyme-linked immunosorbent assay (ELISA), electrochemiluminescence (ECL) and colorimetric assays were used to detect ARA, with inconsistent results in some cases30. The affinity capture elution-ELISA technique has been introduced as a valid method for the detection of ARA in which the rituximab–ARA complexes were dissociated by adding an acidic reagent to serum49. It has been also reported that the ECL method exhibits more sensitivity than the ELISA method for detection of ARA30. Overall, the standardization of the methods for ARA detection need further consideration.