Rheumatoid arthritis (RA) is a chronic multisystem disease of unknown cause. Although there are a variety of systemic manifestations, the characteristic feature of established RA is persistent inflammatory synovitis, usually involving peripheral joints in a symmetric distribution. The potential of the synovial inflammation to cause cartilage damage and bone erosions and subsequent changes in joint integrity is the hallmark of the disease. Despite its destructive potential, the course of RA can be quite variable. Some patients may experience only a mild oligoar-ticular illness of brief duration with minimal joint damage, but most will have a relentless progressive polyarthritis with marked functional impairment.
Epidemiology and Genetics
The prevalence of RA is ~0.8% of the population (range 0.3-2.1%); women are affected approximately three times more often than men. The prevalence increases with age, and sex differences diminish in the older age group. RA is seen throughout the world and affects all races. However, the incidence and severity seem to be less in rural sub-Saharan Africa and in Caribbean blacks. The onset is most frequent during the fourth and fifth decades of life, with 80% of all patients developing the disease between the ages of 35 and 50.The incidence of RA is more than six times greater in 60- to 64-year-old women compared to 18- to 29-year-old women. Recent data indicate that the incidence of RA may be diminishing. Moreover, disease severity appears to be declining, although it is uncertain whether this reflects more aggressive therapeutic interventions.
Family studies indicate a genetic predisposition. For example, severe RA is found at approximately four times the expected rate in first-degree relatives of individuals with disease associated with the presence of the autoantibody, rheumatoid factor; ~10% of patients with RA will have an affected first-degree relative. Moreover, monozygotic twins are at least four times more likely to be concordant for RA than dizygotic twins, who have a similar risk of developing RA as nontwin siblings. Only 15-20% of monozygotic twins are concordant for RA, however, implying that factors other than genetics play an important etiopathogenic role. Despite this, genetic factors are thought to explain ~60% of the disease susceptibility of RA. Of note, the highest risk for concordance of RA is noted in twins who have two HLA-DRB1 alleles known to be associated with RA. The class II major histocompatibility complex allele HLA-DR4 (DRß1*0401) and related alleles are known to be major genetic risk factors for RA. Early studies showed that as many as 70% of patients with RA express HLA-DR4 compared with 28% of control individuals. This association is particularly strong for individuals who develop
RA associated with antibodies to cyclic citrullinated polypeptides (CCP). An association with HLA-DR4 has been noted in many populations, but not all. In some populations, including Israeli Jews, Asian Indians, and Yakima Indians of North America, however, there is no association between the development of RA and HLA-DR4. In these individuals, there is an association between RA and the closely related HLA-DR1 (DRß1*0101). The term shared epitope has been used to denote the HLA-ß1 alleles that appear to convey increased risk for RA because they have similar amino acids in the third hypervariable region of the peptide binding cleft of the molecule. It has been estimated that the risk of developing RA in a person with DRß1*0401 or the closely related DRß1*0404 is 1 in 35 and 1 in 20, respectively, whereas the presence of both alleles puts persons at an even greater risk. In certain groups of patients, there does not appear to be a clear association between HLA-DR4-related epitopes and RA. Thus, nearly 75% of African-American RA patients do not have this genetic element. Moreover, there is an association with HLA-DR10 (DRß1*1001) in Spanish and Italian patients, with HLA-DR9 (DRß1*0901) in Chileans, and with HLA-DR3 (DRß1*0301) in Arab populations.
Additional genes in the HLA-D complex may also convey altered susceptibility to RA. These include portions of the HLA region outside of the coding regions of HLA-DR molecules that increase risk. In addition, certain HLA-DR alleles, including HLA-DR5 (DRß1*1101), HLA-DR2 (DRß1*1501), HLA-DR3 (DRß1*0301), and HLA-DR7 (DRß1*0701), may protect against the development of RA in that they tend to be found at lower frequency in RA patients than in controls. A recent classification of HLA-ß1 alleles based on the sequence of the third hypervariable region, encoding a portion of the peptide binding cleft, has established a hierarchy of disease susceptibility. Alleles of a group that contained a lysine at position 71 conveyed the highest risk, whereas alleles of a group that contained an arginine at position 71 also conveyed increased risk compared to all other HLF-ß1 alleles. It has been estimated that HLA genes contribute about one-third of the genetic susceptibility to RA. Thus, genes outside the HLA complex also contribute. Recent analyses have identified PTPN22, a phosphatase involved in antigen receptor signaling in lymphocytes, FcRL3, a molecule involved in regulating B cell activation, PADI4, an enzyme involved in conversion of citrulline to arginine in proteins, and CTLA4, a molecule involved in regulation of T cell activation, as susceptibility genes for RA, in at least some populations. Except for PADI4, these genes also appear to convey risk for other autoimmune diseases.
Genetic risk factors do not fully account for the incidence of RA, suggesting that environmental factors also play a role in the etiology of the disease. This is emphasized by epidemiologic studies in Africa that have indicated that climate and urbanization have a major impact on the incidence and severity of RA in groups of similar genetic background. Smoking has clearly been identified as a risk for RA in persons expressing an HLA-β1 susceptibility allele. Such persons have an increased risk to develop severe RA associated with antibodies to CCP.
Etiology
The cause of RA remains unknown. It has been suggested that RA might be a manifestation of the response to an infectious agent in a genetically susceptible host. Because of the worldwide distribution of RA, it has been hypothesized that if an infectious agent is involved, the organism must be ubiquitous. A number of possible causative agents have been suggested, including Mycoplasma, Epstein-Barr virus (EBV), cytomegalovirus, parvovirus, and rubella virus, but convincing evidence that these or other infectious agents cause RA has not emerged. The process by which an infectious agent might cause chronic inflammatory arthritis with a characteristic distribution also remains unknown.
Pathology and Pathogenesis
Microvascular injury and an increase in the number of synovial lining cells appear to be the earliest lesions in rheumatoid synovitis. The nature of the insult causing this response is not known. Subsequently, an increased number of synovial lining cells is seen along with perivascular infiltration with mononuclear cells. Before the onset of clinical symptoms, the perivascular infiltrate is predominantly composed of myeloid cells, whereas in symptomatic arthritis, T cells can also be found, although their number does not appear to correlate with symptoms. As the process continues, the synovium becomes edematous and protrudes into the joint cavity as villous projections.
Light-microscopic examination discloses a characteristic constellation of features, which include hyperplasia and hypertrophy of the synovial lining cells; focal or segmental vascular changes, including microvascular injury, thrombosis, and neovascularization; edema; and infiltration with mononuclear cells, often collected into aggregates around small blood vessels (Fig. 5-1). The endothelial cells of the rheumatoid synovium have the appearance of high endothelial venules of lymphoid organs that have been altered by cytokine exposure to facilitate entry of cells into tissue. Rheumatoid synovial endothelial cells express increased amounts of various adhesion molecules involved in this process. Although this pathologic picture is typical of RA, it can also be seen in a variety of other chronic inflammatory arthritides. The mononuclear cell collections are variable in composition and size. The predominant infiltrating cell is the T lymphocyte. CD4+ T cells predominate over CD8+ T cells and are frequently found in close proximity to HLA-DR+ macrophages and dendritic cells.
FIGURE 5-1
Histology of rheumatoid synovitis. A. The characteristic features of rheumatoid inflammation with hyperplasia of the lining layer (arrow) and mononuclear infiltrates in the sublining layer (double arrow). B. A higher magnification of the largely CD4+ T cell infiltrate around postcapillary venules (arrow).
Increased numbers of a separate population of T cells expressing the γδ form of the T cell receptor have also been found in the synovium, although they remain a minor population there and their role in RA has not been delineated. The major population of T cells in the rheumatoid synovium is composed of CD4+ memory T cells that form the majority of cells aggregated around postcapillary venules. Scattered throughout the tissue are CD8+ T cells. Both populations express the early activation antigen, CD69. Besides the accumulation of T cells, rheumatoid synovitis is also characterized by the infiltration ofvariable numbers ofB cells and antibody-producing plasma cells. In advanced disease, structures similar to germinal centers of secondary lymphoid organs may be observed in the synovium, but these are observed only in a small fraction of patients. Both polyclonal immunoglobulin and the autoantibody rheumatoid factor are produced within the synovial tissue, which leads to the local formation of immune complexes. Antibodies to synovial tissue components may also contribute to inflammation. Recent evidence suggests that antibodies to CCP, which are generated within the synovium, may contribute to RA synovitis. Increased numbers of activated mast cells are also found in the rheumatoid synovium. Local release of the contents of their granules may contribute to inflammation. Finally, the synovial fibroblasts in RA manifest evidence of activation in that they produce a number of enzymes such as collagenase and cathepsins that can degrade components of the articular matrix. These activated fibroblasts are particularly prominent in the lining layer and at the interface with bone and cartilage.
Osteoclasts are also prominent at sites of bone erosion. Activated mesenchymal stromal cells, similar to those found in normal bone marrow, can also be found in the rheumatoid synovium.
The rheumatoid synovium is characterized by the presence of a number of secreted products of activated lymphocytes, macrophages, and fibroblasts. The local production of these cytokines and chemokines appears to account for many of the pathologic and clinical manifestations of RA. These effector molecules include those that are derived from T lymphocytes, those originating from activated myeloid cells, and those secreted by other cell types in the synovium, such as fibroblasts and endothelial cells. The activity of these chemokines and cytokines appears to account for many of the features of rheumatoid synovitis, including the synovial tissue inflammation, synovial fluid inflammation, synovial proliferation, and cartilage and bone damage, as well as the systemic manifestations of RA. In addition to the production of effector molecules that propagate the inflammatory process, local factors are produced that tend to slow the inflammation, including specific inhibitors of cytokine action and additional cytokines, such as transforming growth factor ß(TGF-ß), which inhibits many of the features of rheumatoid synovitis including T cell activation and proliferation, B cell differentiation, and migration of cells into the inflammatory site, and may be involved in generating a population of regulatory T cells, as a means to control inflammation.
These findings have suggested that the propagation of RA is an immunologically mediated event, although the original initiating stimulus has not been characterized. One view is that the inflammatory process in the tissue is driven by the CD4+ T cells infiltrating the synovium. Evidence for this includes (1) the predominance of CD4+ T cells in the synovium; (2) the increase in soluble interleukin (IL)-2 receptors, a product of activated T cells, in blood and synovial fluid of patients with active RA; and (3) amelioration of the disease by removal of T cells by thoracic duct drainage or peripheral lympha-pheresis or suppression of their proliferation or function by drugs, such as cyclosporine, leflunomide, or nondepleting monoclonal antibodies to CD4, or inhibitors of T cell activation, such as the T cell co-stimulation competitor, CTLA-4-Ig (abatacept). In addition, the association of RA with certain HLA-DR alleles, whose only known functions are to shape the repertoire of CD4+ T cells during ontogeny in the thymus and bind and present antigenic peptides to CD4+ T cells in the periphery, strongly implies a role for CD4+ T cells in the pathogenesis of the disease. Within the rheumatoid synovium, the CD4+ T cells differentiate predominantly into TH1-like effector cells producing the proinflammatory cytokine interferon (IFN)^ and appear to be deficient in differentiation into TH2-like effector cells capable of producing the anti-inflammatory cytokine IL-4. As a result of the ongoing secretion of IFN-γ without the regulatory influences of IL-4, macrophages are activated to produce the proinflammatory cytokines IL-1 and tumor necrosis factor (TNF) and also increase expression of HLA molecules. Direct contact between activated T cells and myeloid cells may also lead to the production of proinflammatory cytokines by the latter. Moreover, T lymphocytes express surface molecules such as CD154 (CD40 ligand) and also produce a variety of cytokines that promote B cell proliferation and differentiation into antibody-forming cells and therefore may also promote local B cell stimulation. The resultant production of immunoglobulin and rheumatoid factor can lead to immune-complex formation with consequent complement activation and exacerbation of the inflammatory process by the production of the anaphylatoxins, C3a and C5a, and the chemotactic factor C5a. In addition, antibodies may be produced to other self-antigens, such as CCP, that can contribute to disease pathogenesis. The tissue inflammation is reminiscent of chronic inflammatory responses to persistent microorganisms, although it has become clear that the number of T cells producing cytokines such as IFN-γ is less than is found in typical delayed-type hypersensitivity reactions, perhaps owing to the large amount of reactive oxygen species produced locally in the synovium that can dampen T cell function or the action of local regulatory cells. It remains unclear whether the persistent T cell activity represents a response to a persistent exogenous antigen or to altered autoantigens such as collagen, immunoglobulin, one of the heat shock proteins, or CCP. Alternatively, it could represent persistent responsiveness to activated autologous cells such as might occur as a result of EBV infection or persistent response to a foreign antigen or superantigen in the synovial tissue. Finally, rheumatoid inflammation could reflect persistent stimulation of T cells by synovial-derived antigens that cross-react with determinants introduced during antecedent exposure to foreign antigens or infectious microorganisms. The important contribution of B lymphocytes to the chronic inflammatory process has been emphasized by the observation that treatment with a monoclonal antibody to the B cell marker, CD20 (rituximab), caused prompt depletion of B lymphocytes, a decline in serum rheumatoid factor titers, and a partial amelioration of signs and symptoms of inflammation.
Overriding the chronic inflammation in the synovial tissue is an acute inflammatory process in the synovial fluid. The exudative synovial fluid contains more polymorphonuclear leukocytes (PMNLs) than mononuclear cells. A number of mechanisms play a role in stimulating the exudation of synovial fluid. Locally produced antibodies to tissue components and immune complexes can activate complement and generate anaphylatoxins and chemotactic factors. Local production of chemokines and cytokines with chemotactic activity as well as inflammatory mediators such as leukotriene B4 and products of complement activation can attract neutrophils. Moreover, many of these same agents can also stimulate the endothelial cells of postcapillary venules to become more efficient at binding circulating cells. The net result is the enhanced migration of PMNLs into the synovial site. In addition, vasoactive mediators such as histamine produced by the mast cells that infiltrate the rheumatoid synovium may also facilitate the exudation of inflammatory cells into the synovial fluid. Finally, the vasodilatory effects of locally produced prostaglandin E2 may also facilitate entry of inflammatory cells into the inflammatory site. Once in the synovial fluid, the PMNLs can ingest immune complexes, with the resultant production of reactive oxygen metabolites and other inflammatory mediators, further adding to the inflammatory milieu. Locally produced cytokines and chemokines can additionally stimulate PMNLs. The production of large amounts of cyclooxygenase and lipoxygenase pathway products of arachidonic acid metabolism by cells in the synovial fluid and tissue further accentuates the signs and symptoms of inflammation.
The precise mechanism by which bone and cartilage destruction occurs has not been completely resolved. Although the synovial fluid contains a number of enzymes potentially able to degrade cartilage, the majority of destruction occurs in juxtaposition to the inflamed synovium, or pannus, that spreads to cover the articular cartilage. This vascular granulation tissue is composed of proliferating fibroblasts, small blood vessels, and a variable number of mononuclear cells and produces a large amount of degradative enzymes, including collagenase and stromelysin, that may facilitate tissue damage. The cytokines IL-1 and TNF play an important role by stimulating the cells of the pannus to produce collagenase and other neutral proteases. These same two cytokines also activate chondrocytes in situ, stimulating them to produce proteolytic enzymes that can degrade cartilage locally and also inhibiting synthesis of new matrix molecules. Finally, these two cytokines, along with IL-6, may contribute to the local demineralization of bone by contributing to the activation of osteoclasts that accumulate at the site of local bone resorption. Prostaglandin E2 produced by fibroblasts and macrophages may also contribute to bone demineralization. The common final pathway of bone erosion is likely to involve the activation of osteoclasts that are present in large numbers at these sites. Systemic manifestations of RA can be accounted for by release of inflammatory effector molecules from the synovium.These include IL-1,TNF, and IL-6, which account for many of the manifestations of active RA, including malaise, fatigue, and elevated levels of serum acute-phase reactants. The importance of TNF in producing these manifestations is emphasized by the prompt amelioration of symptoms following administration of a monoclonal antibody to TNF or a soluble TNF receptor Ig construct to patients with RA.This is associated with a decrease in production of other proinflam-matory cytokines,including IL-1 and IL-6. Other cytokines may also contribute to the inflammatory milieu, including IL-17. In addition, immune complexes produced within the synovium and entering the circulation may account for other features of the disease, such as systemic vasculitis.
As shown in Fig. 5-2, the pathology of RA evolves over the duration of this chronic disease. The earliest event appears to be a nonspecific inflammatory response initiated by an unknown stimulus and characterized by accumulation of macrophages and other mononuclear cells within the sublining layer of the synovium. The activity of these cells is demonstrated by the increased appearance of macrophage-derived cytokines, including TNF, IL-1ß, and IL-6. Subsequently, activation and differentiation of memory CD4+ T cells is induced, presumably in response to antigenic peptides displayed by a variety of antigen-presenting cells in the synovial tissue. The activated memory T cells are capable of producing cytokines, especially IFN-γ, which amplify and perpetuate the inflammation. The presence of activated T cells expressing CD154 (CD40 ligand) can induce polyclonal B cell stimulation and differentiation of memory B cells and plasma cells that produce autoantibodies locally. The cascade of cytokines produced in the synovium activates a variety of cells in the synovium, bone, and cartilage to produce effector molecules that can cause tissue damage characteristic of chronic inflammation. It is important to emphasize that there is no current way to predict the progress from one stage of inflammation to the next, and once established, each can influence the other. Important features of this model include the following:
(1) the major pathologic events vary with time in this chronic disease; (2) the time required to progress from one step to the next may vary in different patients, and the events, once established, may persist simultaneously;
(3) once established, the major pathogenic events operative in an individual patient may vary at different times;
(4) the process is chronic and reiterative, with successive events stimulating progressive amplification of inflammation; and (5) once memory T cells and B cells have been generated, anti-inflammatory and anti-cytokine therapy may be capable of suppressing disease manifestations but not preventing recrudescence of disease activity once therapy is discontinued.These considerations have important implications with regard to appropriate treatment.
