Pharmacological strategies for mitigating anti-TNF biologic immunogenicity in rheumatoid arthritis patients

Abstract

Tumor necrosis factor alpha (TNFα) inhibitors are a mainstay of treatment for rheumatoid arthritis (RA) patients after failed responses to conventional disease-modifying antirheumatic drugs (DMARDs). Despite the clinical efficacy of TNFα inhibitors (TNFi), many RA patients experience TNFi treatment failure due to the development of anti-drug antibodies (ADAs) that can neutralize drug levels and lead to RA disease relapse. Methotrexate (MTX) therapy with concomitant TNFα inhibitors decreases the risk of TNFi immunogenicity, but additional and/or alternative strategies are needed to reduce MTX-associated toxicities and to further increase its potency for preventing TNFα inhibitor immunogenicity. In this review, we highlight the limitations of MTX for mitigating TNFα inhibitor immunogenicity, and we discuss potential alternative pharmacological targets for decreasing the risk of immunogenicity during TNFα inhibitor therapy based on the key kinases, second messengers, and shared signaling mechanisms of lymphocyte receptor signaling.

BACKGROUND

Rheumatoid arthritis (RA) is the most common type of chronic inflammatory disease affecting the joints. The goals of RA treatment are to achieve disease remission by attenuating joint inflammation and damage [1]. Low-dose methotrexate (MTX, 5–25 mg/week) is the first-line therapy for RA [2], whereas biological disease-modifying antirheumatic drugs (DMARDs), such as tumor necrosis factor alpha (TNFα) inhibitors are effective in reducing inflammation after failed or inadequate responses to conventional DMARDs [3]. Currently, five TNFα inhibitors (TNFi) have been approved for RA treatment: adalimumab, infliximab, etanercept, golimumab and certolizumab pegol [4]. To overcome the high-cost limitation associated with TNFi therapy and to increase access to these agents, several biosimilars have been introduced that exhibit similar efficacy as the original TNFi but are more cost-effective [5]. Despite the remarkable improvement in disease outcomes, a significant fraction of patients experience TNFi treatment failure due to the development of immune responses to the biologic [6]. Interestingly, it has been demonstrated that concomitant MTX decreases the risk of TNFi immunogenicity [7]. Nevertheless, additional and/or alternative strategies are needed to reduce toxicities associated with MTX therapy and to further decrease the frequency of TNFi immunogenicity. In this review we highlight the limitations of strategies currently used to mitigate TNFi immunogenicity, and we discuss potential alternative pharmacological targets for mitigating the risk of immunogenicity during TNFi biologic therapy.

Immunogenicity to TNFα inhibitors leads to treatment failure.

Immunogenicity is a common limitation associated with the clinical use of biologics or protein-based therapies from non-human and human sources. Biologics induce immune responses that can lead to the development of anti-drug antibodies (ADAs), which can have several clinical consequences ranging from RA disease relapse to severe life-threatening conditions [8]. Generally, ADA development involves a T cell-dependent antibody response that is initiated by the uptake and presentation of the biologic, in the form of antigenic peptides, on the major histocompatibility complex (MHC) class II molecules expressed on the cell surface of antigen-presenting cells (APCs), such as dendritic cells. Two essential steps during antigen sensitization involve T cell and B cell activation. Naïve T cells are activated upon antigen recognition by the T cell receptor (TCR, signal 1) and CD28 costimulation by B7 molecules (signal 2) [9]. Activated T cells can stimulate the differentiation of B cells to antibody-secreting plasma cells upon T cell receptor recognition of an antigen presented on the MHC II molecules of B cells [10]. Repeated antigen exposure can lead to B cell IgG isotype switching [11], where IgG1 and IgG4 ADAs are common to TNFi biologics [12,13].

ADA generation can be affected by host and drug related factors, including the frequency and duration of the treatment regimen [14,15]. Among patients on TNFi therapy, most develop antibodies within the first year of treatment resulting in loss of clinical response and discontinuation of treatment [16–18]. Since ADAs play such an important role in the prediction of treatment efficacy and failure, monitoring of ADAs and drug levels is key for attaining disease remission. Several methods of ADA detection have been developed [19], many of which are limited with regards to detecting broad immunogenicity [20,21]. Nevertheless, several association studies investigating ADAs have attempted to identify biomarkers that predict TNFi immunogenicity. Recently, anti-adalimumab levels were found to be negatively correlated with anti-hinge antibodies [22]. Furthermore, anti-TROVE2 antibody levels and baseline BAFF levels have been identified as a biomarker capable of independently predicting TNFi immunogenicity [23,24]. Despite these positive contributions, the prediction of TNFi immunogenicity is challenging and requires larger prospective studies for validation. Therefore, there remains an urgent need to identify and develop prevention strategies to attenuate the risk of TNFi immunogenicity in RA patients.

MTX is the standard of care for TNFi immunogenicity, but additional or better options are needed for optimal RA treatment.

Concomitant administration of low-dose MTX with TNFi biologics decreases the risk or incidence of TNFi ADAs, rescues TNFi drug levels, and decreases the risk of TNFi therapy discontinuation [25–29]. Interestingly, protection against TNFi immunogenicity by MTX is more pronounced compared to other DMARDs or corticosteroids in RA patients [30]. This underscores the need to better understand the molecular mechanisms of biologic immunogenicity for achieving optimal RA efficacy while minimizing the risk of toxicities that can adversely impact the quality of life of RA patients.

MTX is a folic acid antimetabolite used for the treatment of hematologic malignancies that potently inhibits dihydrofolate reductase (DHFR) when used at high doses. DHFR inhibition leads to diminished de novo purine and pyrimidine biosynthesis via preventing the synthesis of tetrahydrofolate [31]. However, because MTX is used at much lower doses for the treatment of RA, it is unlikely that this mechanism provides protection against ADAs [32]. Leucovorin, also known as folinic acid, is a reduced folate that rescues from MTX toxicities during leukemia therapy by bypassing DHFR and directly restoring the conversion of deoxyuridine monophosphate (dUMP) to deoxythymidine monophosphate (dTMP) by thymidylate synthase [32]. Consistent with the effect of MTX on immunogenicity being independent of DHFR inhibition, clinical studies in RA patients receiving folinic acid supplementation have shown that DHFR inhibition rescue protects from MTX toxicities without compromising RA efficacy [33]. Rather, MTX (i.e., MTX polyglutamates) has been shown to inhibit 5-aminoimidazole-4-carboxamide ribonucleotide (AICAR) transformylase, resulting in increased intracellular AICAR levels and thereby extracellular adenosine [34]. Several studies have validated the role of adenosine on the anti-inflammatory effects of MTX, including those demonstrating that the effects of MTX are dependent on A2A and A3 adenosine receptors and cell surface CD39 and CD73 enzymes, which convert adenine nucleotides to adenosine [33]. Preclinical studies support that the effect of MTX on extracellular adenosine is essential for protection against TNFi immunogenicity, where BAFF transgenic mice overexpressing CD73 and CD39 were protected from TNFi ADAs by MTX, whereas wild-type mice were not [35]. Based on studies elucidating the effect of MTX on extracellular adenosine levels, there has been a call to identify strategies that can further enhance or improve protection against TNFi immunogenicity [36]. Furthermore, while MTX is suitable for protecting against TNFi immunogenicity, its safety profile can lead to concerns among RA patients and thereby MTX discontinuation [37]. Common toxicities with low-dose MTX include gastrointestinal, hepatic, and hematologic adverse events [38]. MTX therapy may also require lifestyle modifications including changes in contraception use and alcohol consumption habits due to MTX teratogenicity and risk of hepatoxicity, respectively [39]. Altogether, because of the need for enhanced MTX efficacy and decreased toxicities, there remains a need to identify alternative pharmacological targets that can safely mitigate TNFi biologic immunogenicity.

PHARMACOLOGICAL STRATEGIES FOR MITIGATING BIOLOGIC IMMUNOGENICITY

Strategies for mitigating biologic immunogenicity require attenuating antigen-specific CD4⁺ T and B cell activation.

Because CD4⁺ T and B cells are essential for the development of biologic immunogenicity, suppressing their activation and proliferation during biologic exposure is essential for attenuating drug-induced immunogenicity. Unlike immunosuppressive agents used for the prevention of graft-versus-host-disease (e.g., tacrolimus and cyclosporine A or CsA), optimal immunosuppressive strategies will have minimal toxicities, limit the risk of neutropenia and infections, and will not attenuate RA efficacy. Below we review pharmacological strategies that can potentially modulate CD4⁺ T cell and/or B cell receptor signaling for attenuating biologic immunogenicity (Figure 1). While cytotoxic antiproliferative agents, such as azathioprine and cyclophosphamide, have been previously used for immunosuppression [40]; these agents will not be considered due to concerns regarding their toxicity and adverse effects.

Figure 1. There are several pharmacological targets of lymphocyte signaling that can be inhibited for the prevention of TNFi immunogenicity.

B (BCR) and T cell receptor (TCR) ligation leads to a cascade of signaling events culminating in the activation of transcription factors that drive lymphocyte activation and function. TCR signaling is initiated by co-receptor (e.g., CD4)-associated lymphocyte-specific protein tyrosine kinase (LCK) phosphorylation of tyrosine residues located on the CD3-ζ subunits of the TCR. Zeta-chain-associated protein kinase-70 (ZAP-70) docks to the phosphorylated sites, phosphorylates linker for the activation of T cells (LAT), resulting in activation of phospholipase C-ɣ (PLC-ɣ) and generation of inositol triphosphate (IP3) and diacylglycerol (DAG). Similarly, BCR crosslinking leads to phosphorylation of CD79 and activation of spleen tyrosine kinase (SYK). SYK phosphorylates several targets, including Bruton’s tyrosine kinase (BTK), which results in the generation of secondary messengers, similar to TCR signaling. IP3 and DAG generated during lymphocyte receptor signaling results in activation of calcineurin, protein kinase C (PKC), and mitogen-activated protein (MAP) kinases, which initiate the transcription of genes involved in lymphocyte activation via nuclear factor of activated T cells (NFAT), nuclear factor kappa B (NF-κb), and activator protein-1 (AP-1) transcription factors. Adenosine signaling can mitigate proinflammatory responses by increasing cyclic 3′,5′-adenosine monophosphate (cAMP) levels and activating protein kinase A (PKA), which phosphorylates cAMP responsive element binding protein (CREB). Activated CREB can lead to anti-inflammatory signaling by modulating NF-κB-driven transcription. Potential pharmacological targets for the prevention of biologic immunogenicity are indicated in red font. Created with BioRender.com

T cell receptor (TCR) and B cell receptor (BCR) signaling: potential therapeutic targets.

Naïve T cell activation requires antigen-specific signaling through the TCR and secondary costimulatory signaling through CD28 from a professional APC. TCR cross-linking and co-stimulation leads to a cascade of intracellular signaling involving several enzymes and transcription factors that promote cytokine secretion and proliferation. Briefly (Figure 1), TCR signaling leads to phosphorylation of tyrosine residues located in the cytosolic domains of the CD3-ζ subunits of the TCR by lymphocyte-specific protein tyrosine kinase (LCK). These phosphorylated residues are docked by zeta-chain-associated protein kinase-70 (ZAP-70), which phosphorylates linker for the activation of T cells (LAT), resulting in activation of phospholipase C-ɣ (PLC-ɣ) and generation of inositol triphosphate (IP3) and diacylglycerol (DAG). These second messengers lead to the activation of protein kinase C (PKC)-Ɵ, mitogen-activated protein (MAP) kinases, and Ca²⁺-calcineurin which collectively initiate the transcription of genes involved in T cell activation via nuclear factor of activated T cells (NFAT), activator protein-1 (AP-1), and nuclear factor kappa B (NF-κb) transcription factors. Similarly, B cell activation signaling via the B cell receptor (BCR) involves several kinases, including Src family kinases (e.g., Lyn, Fyn), spleen tyrosine kinase (SYK), Bruton tyrosine kinase (BTK), AKT, and phosphoinositide 3-kinases (PI3K) [41]. BCR crosslinking leads to Lyn-dependent PI3K activation, which phosphorylates phosphatidylinositol 4,5-bisphophate (PIP2) to phosphatidylinositol-3,4,5-triphosphate (PIP3) [41]. PIP3 recruits BTK and AKT to the plasma membrane [42]. The secondary messengers DAG and IP3 are produced from a signaling complex of SYK, BTK, and PLC-γ resulting in PKC-β and calcineurin activation [41]. Activated AKT phosphorylates Forkhead Box class O (Foxo) transcription factors, which are involved in regulating apoptosis [43]. Both NF-κB and NFAT transcription factors are important regulators of BCR-induced gene expression changes during B cell activation [44]. Based on the key kinases, second messengers, and shared signaling mechanisms between lymphocyte receptor signaling, it is plausible that strategies mitigating key known signaling events during T/B cell activation can attenuate unwanted drug-induced immunogenicity.

NFAT inhibition can attenuate drug-induced immunogenicity

Tacrolimus (FK506) and CsA prevent allograft rejection via inhibiting the enzyme activity of calcineurin, which is required for activation of NFAT family members during TCR signaling and activation [45]. NFAT regulates the expression of several genes involved in T cell activation and Th2 cell differentiation, including IL-2, IL-4, and IFN-γ [46]. Consistent with NFAT playing a role in the development of biologic immunogenicity, a genome-wide association study (GWAS) of asparaginase immunogenicity identified an association between a polymorphism in NFATC2 (rs6021191) and the risk of developing asparaginase immunogenicity, where asparaginase is a chemotherapeutic biologic (i.e., enzyme) used for the treatment of pediatric acute lymphoblastic leukemia [47]. The identified NFAT variant results in a gain-of-function phenotype, and, in agreement with the clinical discovery, a subsequent preclinical study using wild type and NFATC2-deficient mice demonstrated that genetic NFATC2 inhibition or pharmacological inhibition of NFAT using the 11R-VIVIT peptide resulted in attenuated anti-drug antibody and Th2 cytokine levels [48]. 11R-VIVIT inhibits NFAT activation via attenuating calcineurin-mediated NFAT dephosphorylation but does not inhibit calcineurin enzyme activity, unlike CsA and tacrolimus [49]. The peptide NFAT inhibitor has also demonstrated efficacy in lowering arthritis disease severity [50], but there are limitations associated with the delivery of the peptide and its protection from proteolytic degradation. Among clinically available pharmacological agents, only the antiplatelet agent dipyridamole has been reported to inhibit NFAT activation [51]. Dipyridamole has been shown preclinically and clinically to have anti-inflammatory properties [52,53]; however, dipyridamole has multiple pharmacological targets apart from NFAT that may contribute to its ability to modulate immune responses [52]. Therefore, the development of additional small molecule NFAT inhibitors are required to evaluate whether targeting NFAT activation is suitable for protecting against biologic immunogenicity.

Adenosine via the adenosine receptors leads to anti-inflammatory signaling

In addition to inhibiting NFAT, dipyridamole is a known equilibrative nucleoside transporter (ENT1/ENT2) inhibitor, which leads to increased extracellular adenosine by blocking the transport or intracellular reuptake of adenosine [54]. Adenosine signaling via the adenosine A2A receptor increases cyclic 3′,5′-adenosine monophosphate (cAMP) levels and activates PKA, which can phosphorylate and activate cAMP responsive element binding protein (CREB). Phosphorylated CREB can inhibit proinflammatory cytokines by modulating NF-κB-driven transcription (Figure 1) [55]. Consistent with the putative mechanisms of action of dipyridamole and MTX, selective adenosine A2A receptor agonists, such as CGS 21680 and regadenoson also exhibit anti-inflammatory properties [56,57]. Adenosine signaling is a compelling strategy for mitigating undesirable immune responses because the A2A receptor is expressed by various immune cells and because it can attenuate both T and B cell function [55,58–61]. However, adenosine signaling can cause substantial side effects including dizziness, headaches, hypotension, nausea, and tachycardia due to the ubiquitous expression of adenosine receptors and the widespread effect of signaling through these receptors [62]. Therefore, despite many years of research towards the development of adenosine receptor agonist, only regadenoson has been approved by the FDA for clinical use [63].

PDE inhibitors can suppress immune cell function

In addition to its effects on endogenous adenosine, dipyridamole can also inhibit the aggregation of platelets through non-selective inhibition of phosphodiesterase (PDE) enzyme. PDE enzymes play a role in immune cell function by degrading cAMP and cGMP [64]. PDE inhibitors can thereby suppress immune cell function by increasing cAMP/cGMP and activating PKA/PKG, which phosphorylates CREB [64]. PKA and PKG can also phosphorylate several other targets, including the c-Src tyrosine kinase (Csk) and NFAT, thus PDE inhibition can attenuate lymphocyte signaling via multiple mechanisms [65,66]. The PDE enzyme family includes 11 subfamilies with different substrate specificities [67]. T lymphocytes express PDE1, PDE2, PDE3, PDE4, PDE5, PDE7, and PDE8, whereas B cells express PDE3, PDE4, and PDE7 [64]. PDE inhibition has been demonstrated to decrease T cell activation and IL-2 production [68]. However, while PDE4 inhibition in PBMCs has demonstrated decreased IgE production [69], the effect of PDE inhibition on B cell function is unclear [64].

Dipyridamole inhibits PDE5 [70,71], and similar to the anti-inflammatory properties of dipyridamole, the PDE5 inhibitor sildenafil can decrease cytokine levels in a model of multiple sclerosis [72], inhibit neuroinflammation [73], and prevent experimental autoimmune encephalomyelitis progression [74]. In patients, sildenafil has been shown to increase the suppressive effects of regulatory T cell on effector T cell function [74]. Several PDE inhibitors have been investigated for RA efficacy [64], including the clinically available PDE4 inhibitors roflumilast and apremilast. Altogether, while PDE inhibition is a viable strategy for attenuating biologic immunogenicity and controlling disease progression, the PDE family members required for TNFi immunogenicity are unclear.

Kinases involved in TCR and BCR signaling can modulate proinflammatory responses

Common transcription factors that mediate TCR and BCR signaling include NFAT, NF-κB, and AP-1 (Figure 1) [75]. The signaling that leads to activation of these transcription factor is initiated by receptor-ligand interaction with many common signaling molecules that can be consider as therapeutic targets for mitigating biologic immunogenicity. TCR and BCR signaling both include cleavage of phosphatidylinositol 4, 5 biphosphate (PIP2) by PLCγ to IP3 and DAG, which leads to activation of calcineurin, PKC, and mitogen-activated protein (MAP) kinases. Small molecules have been developed to inhibit IP3-mediated calcium mobilization; however, toxicity concerns have limited their clinical potential [76]. The development of PKC-Ɵ inhibitors is an area of active research with CC-90005 identified as a selective, safe, and in vivo-compatible small molecule that can mitigate IL-2 production and protect from graft versus host disease [77]. However, the lack of a clinically available PKC inhibitor and concerns of compensation by other PKC family members limit the current translational potential of these agents for biologic immunogenicity [78].

The AP-1 transcription factor consists of a Fos-Jun heterodimer that is regulated by the MAP kinase family of enzymes, which are involved in vital signal transduction pathways [79]. The MAPK kinase extracellular signal-regulated kinase (ERK1/2) is activated by MAPK/ERK kinase (MEK) subsequent to PLCγ-mediated DAG production and phosphorylates the ETS-like 1 (ELK-1) transcription factor. Phosphorylated Elk-1 induces expression of Fos, thereby contributing to the formation of AP-1, which, along with NFAT, drives IL-2 expression and T cell proliferation [80]. Consistent with the crucial role AP-1 plays in the induction of IL-2, MEK inhibitors have been shown to decrease the secretion of IL-2 from donor CD4⁺ and CD8⁺ T cell in dendritic cell co-cultures [81], and also decrease anti-CD3/CD28-induced T cell proliferation [82]. However, while the MEK inhibitor trametinib attenuates naïve and memory T-cell function in vitro, it paradoxically does not limit the effectiveness of adoptive cell therapy or checkpoint blockade in mice [83]. While preclinical studies also support that MEK inhibition can attenuate antigen-specific B cell activation, antibody production, and Th2 cytokine secretion [84], toxicities of MEK inhibition, which include rash, dermatitis, diarrhea, and fatigue [85], may preclude its use versus MTX or other agents for biologic immunogenicity.

Additional kinases that are critical for initiating TCR signaling include LCK, ZAP-70, and (IL-2 inducible T-cell kinase (ITK) (Figure 1). However, clinical pharmacological agents with high target selectivity and that have suitable safety profiles are not currently available to assess the utility of these targets for preventing biologic immunogenicity [86–88]. In contrast, several kinase inhibitors targeting essential signaling molecules involved in initiating BCR signaling are clinically available for the treatment of hematological malignancies, including inhibitors of BTK, SYK, and PI3K (Figure 1) [89]. However, clinical studies investigating the safety of pharmacologically inhibiting these kinases have indicated that only BTK or SYK inhibition has suitable safety profiles for potentially mitigating biologic immunogenicity [90–92]. Furthermore, both BTK and SYK inhibitors have demonstrated their potential for attenuating inflammatory disease, including to RA [93,94]. Pharmacological BTK inhibition has been demonstrated to block B but not T cell proliferation and modulate proinflammatory cytokine secretion [95,96]. In contrast, SYK inhibition decreases CD4⁺ T and B cell proliferation and proinflammatory cytokine levels [97–100]. Therefore, based on the suitable safety profiles, immunomodulatory effects, and potential benefit of BTK or SYK inhibition on RA disease progression, additional studies should be considered to directly assess their ability to attenuate biologic immunogenicity relative to MTX.

Inhibition of cytokine signaling may also be a potential strategy for preventing biologic immunogenicity, where Janus kinase (JAK) / signal transducer and activator of transcription 1 (STAT1) signaling mediates cytokine responses, including to IL-2 [101]. Consistent with the key role of JAKs and STATs on signal transduction after cytokines induction by transcription factors, the JAK1/3 inhibitor tofacitinib, which is approved for the treatment of RA, has been shown to attenuate T cell proliferation and cytokine secretion [89,102]. Furthermore, tofacitinib has been shown to modulate B cell activation and reduce antibody secretion of peripheral blood B cells [103]. In agreement, clinical tofacitinib studies in RA patients have demonstrated that the JAK inhibitor reduces IgG and RF circulating levels [104]. Importantly, tofacitinib has an acceptable safety profile among RA patients that make the JAK/STAT pathway a potential clinical therapeutic target for improving biologic immunogenicity while limiting the toxicities associated with MTX [105].

CONCLUSIONS AND FUTURE PERSPECTIVES

The success of MTX in reducing TNFi immunogenicity has indicated that strategies leading to adenosine signaling can protect from this toxicity, presumably via attenuating NF-κB. Mechanistically, strategies that can attenuate the transcriptional activity of NFAT and/or AP-1 should enhance the anti-inflammatory properties of MTX. It is also possible that adenosine receptor agonist or inhibitors that increase extracellular adenosine levels may lead to similar effects as MTX or allow for lower MTX doses if used in combination. Ideally, alternative targets will have strong anti-inflammatory effects, limit toxicities, and contribute to RA efficacy. Therefore, pharmacological targets that have high specificity for lymphocyte activation or function may limit adverse drug reactions and enhance efficacy. In accordance, inhibitors of BTK, SYK, or JAK have suitable safety profiles and potent anti-inflammatory properties. Nevertheless, additional preclinical and clinical studies are required to determine the most cost- and treatment-effective strategy for preventing biologic immunogenicity.