Chapter 16 Drugs for inflammation and joint disease

A vast burden of human disease involves the process of inflammation and the engagement of an immune response, often entirely appropriately as in the case of defence against infection. By contrast, a number of organ-specific and multi-system rheumatic diseases are characterised by a primary abnormality within the immune response, requiring treatments that modify or suppress it. This chapter reviews the process of inflammation, drugs in current use and those in development that act to modify it, and the management of certain common inflammatory diseases. The chapter covers the following areas:

• Acute inflammation.

• Adaptive immune system.

• Glucocorticoids.

• NSAIDs/aspirin/paracetamol.

• Immunomodulatory drugs.

• Biologics.

• Management of common rheumatic diseases.

Introduction

The immune system is a complex of interrelated genetic, molecular and cellular components that provides defence against invading microorganisms and aberrant native cells, and repairs tissues once the pathogen is eradicated. The central process by which these are achieved is inflammation: the sequence of events by which a pathogen is detected, cells of the immune system are recruited, the pathogen is eliminated and resulting tissue damage repaired.

Inflammation is appropriate as a response to physical damage, microbial infection or malignancy. A number of illnesses result from abnormal activation or prolongation of the immune response. These include allergy (hay fever, asthma), autoimmunity (rheumatoid arthritis (RA), systemic lupus erythematosus (SLE)), and allograft rejection.

Anti-inflammatory drugs, by acting on and modifying the response of the innate immune system to a challenge, are useful in many settings to damp down an over-exuberant or pathologically prolonged inflammatory response. Immunomodulatory agents, which act on components of the adaptive immune response, are important for the treatment of complex autoimmune diseases and in preventing allograft rejection. Many drugs used in the treatment of these diseases have complex mechanisms of action, working on multiple arms of the immune response, and in some cases the principal way in which they exert their effects is not clear. Partly this is due to the complexity of the immune system itself; many components have overlapping functions, leading to redundancy, and many have several apparently unrelated actions.

Research over the last few decades has vastly improved our appreciation of the complexity of the immune system and of the pathogenesis of many autoimmune diseases. Although there does not appear to be one single factor that leads inexorably to the development or perpetuation of any inflammatory disease, certain mediators that play central roles in specific diseases have been identified. The advent of monoclonal antibodies, fusion proteins and other new drug development technology has allowed the manufacture of a rapidly expanding group of new agents (‘biologicals’) that target specific components of the immune response thought to be driving particular diseases. These drugs have dramatically changed treatment paradigms, and hopefully will lead to significant improvements in the future outlook for patients suffering rheumatic disease.

Inflammation

The process of acute inflammation is initiated when resident tissue leucocytes (macrophages or mast cells) detect a challenge, for example pathogenic bacteria, or monosodium urate crystals in the case of gout. This sets off a cascade of intracellular signalling that results in activation of the cell, release of soluble cytokines such as tumour necrosis factor-α (TNFα), interleukin-1 (IL-1) and interleukin-6 (IL-6) and other mediators such as histamine and prostaglandins. IL-1, IL-6 and TNFα stimulate endothelial cells at the site of injury to express cellular adhesion molecules, which attract and bind circulating leucocytes, principally neutrophils, and induce them to leave the circulation and migrate into the affected area. They also have systemic effects such as the development of fever and the production of acute phase proteins including C-reactive protein (CRP).

Neutrophils, along with macrophages, phagocytose the injurious stimulus and destroy it. Neutrophils and macrophages may also cause damage to the surrounding host tissue through the release of digesting enzymes such as matrix metalloproteinases and collagenases. The inflammatory process therefore needs to be halted rapidly once the invading organism has been cleared. This occurs partly because neutrophils have a very short lifespan and die quickly once they have left the circulation, and partly through the release of anti-inflammatory mediators.

Several drugs in current use act on the various stages of this inflammatory process. Antagonists of TNFα, IL-1 and IL-6 are available (see Biologic agents, p. 253). Colchicine, used in the treatment of gout, interferes with neutrophil chemotaxis, thus inhibiting their recruitment to the site of inflammation.

Many leucocytes, including mast cells and macrophages, as well as endothelial cells, synthesise pro-inflammatory eicosanoids and platelet-activating factor (PAF) (Fig. 16.1). These are 20-carbon unsaturated fatty acids derived from phospholipid substrates in the plasma membrane by the enzymes phospholipase A₂, cyclo-oxygenase (COX) and lipo-oxygenase (which are induced by IL-1). The prostaglandins, thromboxanes and leukotrienes have diverse pro-inflammatory roles. Leukotrienes promote the activation and accumulation of leucocytes at sites of inflammation. Prostaglandins induce vasodilatation of the microcirculation and are important in pain signalling from locally inflamed tissue. Platelet-activating factor and thromboxane A₂ affect the coagulation and fibrinolytic cascades. Non-steroidal anti-inflammatory drugs (NSAIDs), including aspirin, inhibit COX and hence prostaglandin and thromboxane synthesis. Glucocorticoids act by inducing the synthesis of lipocortin-1, a polypeptide that inhibits phospholipase A₂, and thereby exert a broad anti-inflammatory effect. The leukotriene receptor antagonists montelukast and zafirlukast cause bronchodilatation and are used to treat asthma.

Fig. 16.1 Synthesis of eicosanoids and platelet-activating factor, and drugs acting on this pathway. The enzymes phospholipase A₂, cyclo-oxygenase and lipo-oxygenase are involved in the synthesis of prostaglandins, thromboxanes, leukotrienes and platelet-activating factor, all of which are derived from phospholipid substrates in plasma membranes. Glucocorticoids induce lipocortin 1, which inhibits phospholipase A_2. NSAIDs inhibit cyclo-oxygenases 1 and 2, preventing formation of prostaglandins and thromboxanes.

The adaptive immune response

The adaptive immune response, although integrated into the process of inflammation, becomes active at later stages. Its key properties are (1) specificity: each B and T lymphocyte recognises a single specific peptide sequence; and (2) memory: when an invading pathogen has been recognised once, a small number of specific cells remain dormant within the lymph tissue for many years. If that pathogen is detected again, a very rapid response is mounted to eradicate it before the development of clinical symptoms.

An adaptive immune response is initiated when a helper T cell recognises a peptide antigen presented on the surface of an antigen-presenting cell (APC) and is activated (Fig. 16.2). The activated helper T cell is then able to activate other types of T cell and B cells. This results in the proliferation of adaptive cellular effectors, the generation and release of antibodies by plasma cells and the production of a range of cytokines by the participating leucocytes. On occasion, amplification loops may become self-perpetuating, leading to chronic autoimmune disease.

Fig. 16.2 Three general mechanisms of action of glucocorticoids and the glucocorticoid receptor in the inhibition of inflammation. TNFα, tumour necrosis factor α; HSP, heat-shock protein; mRNA, messenger RNA; P, phosphate. The three mechanisms are non-genomic activation, DNA-dependent regulation, and protein interference mechanisms (e.g. NF-KB elements). Black arrows denote activation, the red line inhibition, the red dashed arrow repression, and the red X lack of product (i.e. no mRNA).

Many immunomodulatory drugs, such as the calcineurin inhibitors, seek to break these loops by inhibiting lymphocyte proliferation. Other newer approaches target specific components of the immune system. For example, rituximab binds CD20, a cell surface molecule found only on B lymphocytes and not memory cells. It is used to treat diseases in which pathogenic autoantibody production is prominent, such as rheumatoid arthritis and SLE. Abatacept blocks co-stimulatory signals, which are required when a helper T cell is activated, by recognising bound antigen presented by an APC. This is a central process in the pathogenesis of rheumatoid arthritis, and abatacept is licensed to treat this.

Pharmacological manipulation of inflammatory mediators

Glucocorticoids

Glucocorticoids (GCs) are among the most widely prescribed anti-inflammatory drugs, due to their profound efficacy, long history of use and familiarity. Besides their anti-inflammatory actions they exert effects on carbohydrate, protein and lipid metabolism, some of which contribute to their substantial adverse effect profile.

GCs are used in acute and chronic settings to treat inflammatory conditions ranging from asthma and allergy, to prevention of allograft rejection, to rheumatoid arthritis and systemic vasculitis. Because of their rapid onset of action (less than 24 h) and potency, they remain the preferred acute treatment for severe inflammatory disease in a wide variety of settings despite their adverse effects. In chronic disease the aim now is to minimise cumulative exposure by using other immunomodulatory ‘steroid-sparing’ drugs, and proactively to manage potential long-term side-effects such as osteoporosis.

Mode of action

GCs, being lipophilic, diffuse across the cell membrane and bind the cytosolic glucocorticoid receptor (GR) (Fig. 16.3). Receptor polymorphisms influence the strength of the receptor interaction, and represent one source of variation in sensitivity to exogenous steroids. Once bound, the GC-GR complex translocates to the nucleus where it acts in at least two ways to alter gene transcription:

1. The GC-GR complex binds to the glucocorticoid response element within target gene promoters, increasing transcription of various anti-inflammatory genes. These include I-κB, which inhibits the activation of nuclear factor (NF)-κB, and the cytokines IL-4, IL-10, IL-13 and transforming growth factor (TGF)β, which have immunosuppressive and anti-inflammatory activity.

2. The GC-GR complex interferes with the binding of the transcription factors activating protein (AP)-1 and NF-κB to their response elements. This action decreases the transcription of a range of pro-inflammatory mediators. These include IL-1β and TNFα; IL-2, which stimulates T-cell proliferation; a range of chemokines and cellular adhesion molecules; metalloproteinases; COX-2; and inducible nitric oxide synthase.

3. GCs increase synthesis of the polypeptide lipocortin-1, which inhibits phospholipase A₂ and thereby the synthesis of prostaglandins, thromboxane A₂ and PAF (see Fig. 16.1).

Fig. 16.3 Mechanisms of glucocorticoid anti-inflammatory effects. Glucocorticoids exert pleiotropic effects on cellular metabolism. They increase transcription of a number of genes encoding anti-inflammatory proteins and decrease transcription of pro-inflammatory genes. AP-1, activating protein-1; COX, cyclo-oxygenase; GR, glucocorticoid receptor; GRE, glucocorticoid response element; IL, interleukin; iNOS, inducible nitric oxide synthase; NF, nuclear factor; TGF, transforming growth factor.

At the cellular level, GCs reduce the numbers of circulating lymphocytes, eosinophils and monocytes. This is maximal 4–6 hours after administration and is achieved by a combination of apoptosis induction and inhibition of proliferation. Chronic administration of GC is associated with a neutrophilia caused by release of neutrophils from the bone marrow and reduced adherence to vascular walls.

In inflammatory disease, the choice of GC preparation will reflect the site and the extent of inflamed tissue, e.g. oral or parenteral for systemic disease, inhaled in asthma, topical in cutaneous, ocular, oral or rectal disease. Different corticosteroid preparations, their pharmacokinetics, modes of delivery and adverse effects are discussed elsewhere, except for the management of steroid-induced osteoporosis, which is found in the section on management of rheumatoid arthritis at the end of this chapter.

Non-steroidal anti-inflammatory drugs (NSAIDs)

NSAIDs are an extremely widely prescribed group of drugs that are mainly used for their analgesic effects. They possess a single common mode of action: inhibition of cyclo-oxygenase, thereby reducing prostaglandin synthesis. This is also the mode of action of paracetamol (acetaminophen) and aspirin.

Recently concern has arisen over the effect of traditional NSAIDs and COX-2 inhibitors on the cardiovascular system, with analysis of the VIGOR ¹ study showing that rofecoxib in particular increases the risk of myocardial infarction (rofecoxib has since been withdrawn from use). While they retain an important role in the treatment of acute gout, inflammatory arthritis, ankylosing spondylitis and dysmenorrhea, long-term prescription should only be undertaken following a full discussion with the patient regarding the balance of risks and benefits.

Mode of action

NSAIDs inhibit cyclo-oxygenase (COX), which catalyses the synthesis of prostaglandins and thromboxane from arachidonic acid (see Fig. 16.1). There are two isoforms of COX:

• COX-1 is constitutively² expressed in most cell types.

• COX-2 is induced when inflammatory cells (fibroblasts, endothelial cells and macrophages) are activated by cytokines such as IL-1β and TNFα. It is also often upregulated in cancer cells, and is constitutively expressed in the kidney and brain.

Both isoforms generate prostaglandins during an inflammatory response, while COX-1 activity is required for prostaglandin synthesis for tissue homeostasis. Thus, many NSAID adverse effects are due to reductions in beneficial prostaglandin production, for example in the gastric mucosa (PGI₂ and PGE₂) and renal medulla. Recognition of this led to the development of COX-2 selective drugs.

NSAIDs can be categorised according to their COX specificity as:

• COX-2-selective compounds (coxibs), which inhibit COX-2 with at least five times greater potency than COX-1. The group includes celecoxib, etoricoxib, lumiracoxib, meloxicam and etodolac.

• Non-COX-2-selective compounds, which comprise all other NSAIDs. These drugs inhibit COX-1 as well as COX-2.

Pharmacokinetics

NSAIDs are absorbed almost completely from the gastrointestinal tract, tend not to undergo first-pass elimination (see p. 87), are highly protein bound and have small volumes of distribution. Their t_½ values in plasma tend to group into short (1–5 h) or long (10–60 h). Differences in t_½ are not necessarily reflected proportionately in duration of effect, as peak and trough drug concentrations at their intended site of action following steady-state dosing are much less than those in plasma. The vast majority of NSAIDs are weak organic acids and localise preferentially in the synovial tissue of inflamed joints (see pH partition hypothesis, p. 80).

Uses

Adverse effects

Gastrointestinal

Dyspepsia is one of the commonest side-effects of NSAIDs. The propensity to gastrointestinal (GI) ulceration may result in occult or overt blood loss. Use of NSAIDs is associated with an approximately four-fold increased incidence of severe gastrointestinal haemorrhage, and such complications account for between 700 and 2000 deaths in the UK each year. In addition, ulceration and stricture of the small intestine can result in anaemia, diarrhoea and malabsorption, similar to Crohn’s disease. The risk of NSAID-induced GI haemorrhage is associated with high doses and prolonged use, age over 65 years, previous history of peptic ulceration, concomitant use of glucocorticoids, anticoagulants or other NSAIDs, heavy smoking and alcohol use, and the presence of Helicobacter pylori infection.

NSAID-associated gastrointestinal disease appears to result from the inhibition of COX–1-mediated production of cytoprotective mucosal prostaglandins, especially PGI₂ and PGE₂, which inhibit acid secretion in the stomach, promote mucus production and enhance mucosal perfusion. Several large randomised controlled trials have investigated the incidence of gastrointestinal adverse effects in traditional NSAIDs compared with coxibs. The VIGOR (rofecoxib versus naproxen), CLASS ³ (celecoxib versus ibuprofen and diclofenac) and TARGET⁴ (lumiracoxib versus ibuprofen and naproxen) studies all indicate that coxib use leads to an approximately 50% reduction of upper gastrointestinal adverse events.

The gastrointestinal toxicity of traditional NSAIDs may be reduced by co-prescription of a proton pump inhibitor, e.g. omeprazole, an H₂-receptor blocker, e.g. ranitidine, or the prostaglandin analogue misoprostol. Proton pump inhibitors are more effective than the other classes of gastroprotective agent and should be considered in all patients with at least one of the above risk factors. In fact, it is now recommended by the UK National Institute for Health and Clinical Effectiveness (NICE) that all patients over 45 years prescribed an NSAID, whether COX-2 selective or not, also receive a proton pump inhibitor.

Cardiovascular

The VIGOR and APPROVE ⁵ trials reported increased thrombotic cardiovascular events in patients treated with rofecoxib, leading to concerns about a class effect of the coxibs. It was suggested that COX-2 selectivity resulted in an imbalance between prostacyclin and thromboxane production, an effect which would not be seen with traditional NSAIDs which inhibited the synthesis of both equally. Subsequent data from the prospective MEDAL⁶ and TARGET trials have not supported a class effect based on COX-2 selectivity. These studies suggest that treatment with either a coxib or an NSAID results in a small increase in cardiovascular risk. The risk is dose-related and rofecoxib, particularly at doses exceeding 50 mg per day, confers the highest cardiovascular risk in the majority of studies. A recent study of more than 1 million patients quantified cardiovascular risk as a composite of coronary death, non-fatal myocardial infarction and fatal and non-fatal stroke, and reported that diclofenac and rofecoxib were associated with the highest cardiovascular risk, while naproxen and perhaps celecoxib at doses ≤ 200 mg per day were the least likely to cause a cardiovascular event.⁷ NICE guidelines recommend that patients with pro-thrombotic risk, coronary artery or cerebrovascular disease should not be prescribed NSAIDs or a coxib. For other patients, treatment decisions should be made on an individual patient basis taking into account both cardiovascular and gastrointestinal risk factors. The medication should be prescribed for the shortest possible time and regularly reviewed.