CD40 as a therapeutic target in Sjögren’s syndrome

KEYWORDS:

• Primary Sjögren’s syndrome
• CD40
• CD40L
• monoclonal antibody
• co-stimulation

1. Introduction

Primary Sjögren’s syndrome (pSS) is a chronic autoimmune disorder characterized by sicca symptoms and affects ~0.06% of the population [1]. pSS is more common in females (9:1) and owing to B cell hyperactivity carries an increased risk of lymphoma. Currently, evidence-based treatment options for pSS are limited. Symptomatic management comprises tear and salivary substitutes. There are currently no biological or synthetic disease-modifying therapies approved for the treatment of pSS.

CD40 is a transmembrane type I glycoprotein consisting of 277 amino acids. It belongs to the tumor necrosis factor (TNF) gene superfamily and behaves as a co-stimulatory molecule found on numerous cells, including B cells, monocytes, antigen presenting cells, endothelial, smooth muscle cells, fibroblasts, and platelets [2]. Its ligand, CD40L–CD154, is a type II transmembrane protein which exists in soluble (sCD40L) or membrane-bound form. It consists of 260 amino acids and is found on activated T cells, B cells, platelets, endothelial, epithelial, and smooth muscle cells [3,4].

In this article, we will first outline the role of CD40–CD40L in the pathogenesis of pSS. We will then review published data on CD40–CD40L blockade in animal models of autoimmune diseases and in clinical trials.

2. Theoretical basis

Interaction between CD40 and CD40L causes several immunological responses including cellular stress-related gene expression and B cell activation (immunoglobulin class switching, germinal center formation, and cytokine production) [4,5]. CD40–CD40L interaction also plays a role in T cell activation, up-regulation of surface molecules relevant to inflammatory responses in chronic inflammatory disorders and expression of cytokines. These include interleukin (IL)-2, IL-12, TNF-α, interferon (IFN)-α [3,4,6], as well as IL-1 and IL-6, which have been implicated in the pathogenesis of pSS and fatigue. Considering the crucial role of CD40–CD40L as co-stimulatory molecules between B cells and T cells, there is already interest in the effects of CD40–CD40L antagonists in other autoimmune conditions (e.g. psoriasis, lupus nephritis, systemic lupus erythematosis [SLE], and rheumatoid arthritis [RA]) [2,6–8].

2.1. Dysregulated expression of CD40–CD40L in pSS

In an in vitro study using cultured salivary gland (SG) cells, significantly higher CD40 expression was found in cell lines from patients with pSS compared to controls. Furthermore, CD40L staining was detected in 30–50% of infiltrating lymphocytes in SG cells from pSS patients [9].

A case–control study found that expression of CD40L on activated CD4⁺ T cells in females with pSS was significantly higher than in healthy controls (HC) [4]. Increased levels of circulating sCD40L in pSS compared to HCs have also been reported [3].

2.2. Targeting CD40 in the treatment of pSS

The CD40–CD40L co-stimulatory pathway and its downstream effects may be inhibited in several ways; targeting CD40 directly, targeting CD40L, or targeting the CD40–CD40L signaling pathway. As the downstream effector pathways following CD40–CD40L interactions are also utilized by other biological pathways, therapies targeting the downstream signaling molecules triggered by CD40–CD40L interaction may also affect other molecular pathways. Therefore, in this review, we will focus on therapies targeting either CD40 or CD40L.

To date, biological therapies that have been developed to target either CD40 or CD40L have been monoclonal antibodies (mAb). mAb are laboratory-manufactured immunoglobulins with the ability to bind specific molecules or receptors where they exert their therapeutic effect. Fully human mAb are more attractive than their chimeric and ‘humanized’ counterparts, at least theoretically because they are less likely to elicit an anti-drug immune response which may reduce therapeutic efficacy, as well as result in adverse effects.

2.3. CD40–CD40L blockade – animal models

2.3.1. Anti-CD40L

Several studies have demonstrated prevention of autoimmunity initiation, slowing of autoimmune disease progression, reduced leukocytic infiltration, and end tissue damage through blockade of CD40L in murine models [3]. SLE, like pSS, is characterized by over-activation of both B and T cells. Anti-CD40L therapy has pronounced effects on B cell activation through sustained co-stimulatory blockade in murine models of lupus [10], and in lupus-prone mice, anti-CD40L treatment improved renal disease, and survival [6]. In experimental allergic encephalomyelitis, CD40L blockade reduced inflammation and tissue damage in brain tissue [3]. In mice with collagen-induced arthritis, anti-CD40L agent blocked joint inflammation development, production of anti-collagen antibodies, inflammatory cell infiltration in synovial tissue, and erosions of cartilage and bone [11].

2.3.2. Anti-CD40

In characterization studies with ASKP1240 (an antagonistic fully human anti-CD40 IgG4 mAb), weekly administration of 10 mg/kg in monkeys completely suppressed both cell-mediated and antibody immune responses, providing promising implications for potential efficacy in autoimmune disease, including pSS [12]. Anti-CD40 mAbs also prolonged renal allograft survival in cynomolgus monkeys [13]. It is noteworthy that agonistic anti-CD40 mAb administered to mice led to systemic and local inflammatory disease in the form of colitis, induced through TNF-α, IFN-γ, IL-12, and IL-23 mediated immune pathways [14].

2.4. Efficacy

Agents directed at CD40L have been shown to reduce anti-DNA antibodies, increase complement levels, and reduce glomerular inflammation in SLE [6]. BG9588, a humanized anti-CD40L mAb, reduced frequency of IgG-secreting cells by 50–90% in patients with lupus nephritis. However, IDEC-131, another anti-CD40L mAb, was not shown to significantly improve SLE disease activity index (SLEDAI) scores in SLE [6].

A randomized control trial using anti-CD40 mAbs in patients with RA on methotrexate resulted in moderate improvements in American College of Rheumatology (ACR) criteria for ACR20 responses and biomarker levels compared to placebo [8]. Furthermore, CFZ533, a fully human IgG1 anti-CD40 mAb, administered intravenously in patients with pSS led to improvement in several disease assessment indices, including European League Against Rheumatism (EULAR) Sjögren’s Syndrome Disease Activity Index (ESSDAI), EULAR Sjögren’s syndrome Patient Reported Index (ESSPRI) and led to reduced germinal center-related serum biomarker, CXCL13 [15]. This study reinforces CD40–CD40L blockade as a good candidate for future therapy in pSS.

3. Safety

3.1. Targeting CD40L

Thromboembolic (TE) event risk has been associated with anti-CD40L antibody administration [6]. These studies involved BG9588 or IDEC-131 (also a humanized anti-CD40L mAb) in humans and ATTC 5C8.33 (IgG mAb) in monkeys [2,6]. Co-administration of heparin with ATTC 5C8.33 resulted in fewer thromboembolic events. Trials using BG9588 and IDEC-131 stopped prematurely because of TE events (including myocardial infarction in a 20- and 40-year old receiving BG9588 [2]). The exact mechanism of TE formation with these agents is not clear; blockage of CD40L on platelets causing platelet instability has been suggested. The most frequently occurring non-severe adverse events (AE) using BG9588 in 28 patients with lupus nephritis were headache, fatigue, chest pain, and pharyngeal pain [2]. Two patients also developed worsening renal function.

3.2. Targeting CD40

ASKP1240 was developed in response to thromboembolic events in trials involving anti-CD40L mAbs [12]. It was theorized that blocking CD40 alone would leave CD40L platelet-regulating function intact. ASKP1240 was administered to 12 cynomolgus monkeys at weekly doses of 1–100 mg/kg for 4 weeks [12]. There was no reduction in circulating B cells or any AEs, including TE. In vitro platelet studies by the same author found that ASKP1240 was not pro-thromboembolic [12].

In the first-in-human phase 1, placebo-controlled randomized, double-blinded ascending single-dose study, ASKP1240 had good safety and tolerability in 72 healthy subjects [13] who were administered single escalating doses (0.1–10 mg/kg) of intravenous ASKP1240. Mean vital sign values and serum biochemistry remained within normal range. There were no hematological dose-related AEs or consistent changes in total and peripheral lymphocyte counts. Most AEs were mild and included gastrointestinal (diarrhea, vomiting, nausea), respiratory (cough, upper respiratory infection, rhinorrhea), skin (dermatitis, itch, site hematoma), and other symptoms (headache, dizziness); 7.1% of patients developed antibodies to ASKP1240. There was no occurrence of TE events or cytokine release syndrome. It should be noted, however, that while 90% of pSS patients are female, in this study 92% of subjects receiving active drug were male.

Other studies have demonstrated favorable safety and tolerability outcomes using ASKP1240. 49 patients with moderate-to-severe plaque psoriasis received ASKP1240 [7]. There were no significant reactions, cytokine release or TE events. However, 8.5% of patients had evidence of raised alanine transaminase (ALT) >3 times upper limit of normal during the study.

Lucatumumab, another fully human recombinant IgG1 anti-CD40 mAb, underwent safety and tolerability assessments in 28 patients with multiple myeloma [5]. Low rates of toxicity were observed at doses below the maximum tolerated dose. Transient decreases in peripheral CD19⁺ B cells occurred. Infusion-related reactions were common but managed with infusion adjustments and premedication prophylaxis. Most reactions decreased over repeated exposure. Most AEs were mild-moderate (anemia, chills, nausea, headache, pyrexia, and hypotension).

There is little evidence of the safety of anti-CD40 agents in pSS specifically. However, data from a recently reported trial involving CFZ533, given to 29 patients with pSS in a phase II study, showed CFZ533 to be safe and well tolerated [15]. Patients received subcutaneous or intravenous CFZ533 in doses of 3 and 10 mg/kg, respectively. Most AEs were mild to moderate with only one serious AE, which was not related to CFZ533.

As some studies involving agonistic anti-CD40 antibodies have shown promising results in the treatment of solid and hematological cancers in murine models and humans, it would be prudent to assess cancer risk associated with long-term treatment with antagonistic anti-CD40 antibodies. Furthermore, as one of the main modulators of T cell function, the CD40 system probably plays a role in viral and intra/extra-cellular organism infections. Although mild infections, such as upper respiratory tract infection, frequently appear as AEs in trials blocking this pathway, they are yet to be proven as drug-related.

4. Conclusions

Dysregulated expression of CD40–CD40L in pSS provides a strong theoretical basis for pursuing blockage of this co-stimulatory pathway. Anti-CD40L agents, although showing promising therapeutic responses, are hampered by their pro-thrombotic nature. Exploration with concomitant use of anti-coagulants may be considered. Anti-CD40 agents, on the other hand, demonstrate more reassuring safety profiles with promising therapeutic benefits. This is welcomed in a time of limited evidence-based therapeutic options for patients with pSS. Further studies are required to understand optimal dosing, risk of hepatoxicity, pharmacokinetics; including the fate of bound anti-CD40 monoclonal antibodies in vivo, and the impact of seroconversion to anti-CD40 agents. Any beneficial effect of anti-CD40–CD40L blockade on lymphoma risk in pSS remains to be seen.

Declaration of interest

The authors have no relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript. This includes employment, consultancies, honoraria, stock ownership or options, expert testimony, grants or patents received or pending, or royalties.

Reviewer disclosures

Peer reviewers on this manuscript have no relevant financial or other relationships to disclose.

Additional information

Funding

This paper is not funded.