Abstract
The connective tissue disorders comprise a number of related conditions that include systemic lupus erythematosus (SLE) and the antiphospholipid (Hughes) syndrome, scleroderma, myositis and Sjögren’s syndrome. They are characterized by autoantibody production and other immune-mediated dysfunction. There are common clinical and serological features with some patients having multiple overlapping connective tissue disorders. The latest advances include new approaches to therapy, including more focused utilization of existing therapies and the introduction of biological therapies in SLE, more precise protocols for assessment of severe disease manifestations such as in interstitial lung disease and pulmonary artery hypertension in scleroderma, new antibodies for disease characterization in myositis and new approaches to patient assessment in Sjögren’s syndrome. B cells have a critical role in most, if not all of these disorders such that B-cell depletion or suppression of B-cell activating cytokines improves disease in many patients. In particular, the introduction of rituximab, a monoclonal antibody targeting the CD20 molecule on B cells, into clinical practice for rheumatoid arthritis and B-cell lymphoma has been a key driver of experimental approaches to therapy in connective tissue disorders. Genetic studies also suggest a role for the innate immune system in disease pathogenesis, suggesting further future targets for biological therapies over the next few years.
Keywords
Systemic lupus erythematosus
Systemic lupus erythematosus (SLE) typically affects women (circa 9:1 female:male ratio) in the 16–55 age group [Ballou et al. 1982]. It is commoner among patients of African/Afro-Caribbean, South Asian or Chinese/other Asian ethnic groups. It ranges from a mild to a life-threatening or fatal disorder [Petri et al. 2012a]. Active disease is associated with raised anti-dsDNA antibodies and reduced levels of complement C3 and/or C4 levels. Current therapy is, in fact, very effective for most patients. In mild to moderate disease, hydroxychloroquine is now commonly used as steroid-sparing therapy for all patients. For patients with active disease, corticosteroids are at the heart of current therapy, often combined with steroid sparing agents such as azathioprine, methotrexate, leflunomide, ciclosporin or mycophenolate. In patients with severe or life-threatening renal, nervous system or pulmonary disease, oral or pulsed intravenous cyclophosphamide has been typically used over the past 20 years [Boumpas et al. 1992]. As a consequence, the survival rate for SLE has steadily increased. In a review of this topic [Urowitz et al. 2008] the authors point out that since the 1950s, SLE has changed from a disease with a 50% mortality at 5 years to one of over 90% 5-year survival.
The challenge, in recent years, has been less about finding drugs with greater efficacy than corticosteroids and cyclophosphamide and more about developing drug regimes with lower toxicity. Corticosteroids, in particular, have been associated with higher rates of infection, cardiovascular and bone disease [Ruiz-Irastorza et al. 2012; Petri et al 2012b]. Cyclophosphamide is associated with increased infection risk and with infertility [Hickman and Gordon, 2011].
This challenge is now being met following a series of clinical trials over the past few years exploring corticosteroid/cyclophosphamide sparing agents to treat SLE, especially the more severe forms of SLE such as lupus nephritis.
Mycophenolate mofetil
The American College of Rheumatology (ACR) and the European League against Rheumatism (EULAR) in collaboration with the European Renal Association–European Dialysis and Transplant Association have recently brought out guidelines for the management of lupus nephritis [Hahn et al. 2012; Bertsias et al. 2012]. Both guidelines use renal biopsy as a starting point for therapeutic regimens that include either mycophenolate or ‘low-dose’ cyclophosphamide regimes. These guidelines are broadly similar. To summarize, in patients with mild renal disease (class I) and in patients with advanced kidney damage without significant inflammation (class VI), immunosuppression is not generally indicated. In class II disease with proteinuria (mesangioproliferative lupus nephritis) the EULAR guidelines are more inclined to recommend therapy (low to moderate dose glucocorticoids with or without azathioprine) than the ACR guidelines. In class III–IV lupus nephritis (focal or diffuse inflammation with proliferation), or class V (membranous nephritis) a study of 370 patients by the Aspreva Lupus Management Study Group demonstrated equivalent efficacy of mycophenolate mofetil (MMF) 3 g/day compared with intravenous cyclophosphamide (monthly pulses of 0.5–1 g/m2) for 6 months [Appel et al. 2009]. Tapered doses of prednisolone were used in both arms of the study. The Euro-Lupus Nephritis Trial group demonstrated that, in general, shorter courses of lower dose cyclophosphamide given more frequently (e.g. 500 mg intravenously every 2 weeks for a total of six doses) are as effective as traditional ‘high-dose’ (500—1000 mg/m2 monthly for 6 months) National Institutes of Health regimes [Houssiau et al. 2002]. The ACR guidelines suggest that lower doses of MMF may be required in Asian patients and that African-American and Hispanic patients may respond less well to cyclophosphamide than other groups.
Patients whose condition fails to respond to the initial regime can be tried with the alternative (MMF in those who have failed to respond to cyclophosphamide and vice versa). In refractory patients, rituximab, ciclosporin, tacrolimus or other therapies can be considered. For maintenance, either mycophenolate or azathioprine are recommended as preferred therapies [Dooley et al. 2011; Houssiau et al. 2010].
The two sets of guidelines have also highlighted other aspects of good practice such as the initial use of three intravenous methylprednisolone pulses and good control of hypertension (e.g. with angiotensin-converting enzyme (ACE) inhibitors or angiotensin receptor blockers) and other cardiovascular risk factors.
Other important issues to pay close attention to include adjunctive use of hydroxychloroquine and statin therapy if appropriate; the use of aspirin/anticoagulation when there is a thrombosis risk; annual and other immunization with nonlive vaccines; consideration of annual measurement of immunoglobulins; the use of bone protection agents, particularly in patients on corticosteroids; appropriate sun protection and annual cervical smear tests for female patients. Vitamin D supplementation may also reduce disease activity in SLE [Bello et al. 2012].
Belimumab
It has been known for over 10 years that high levels of circulating B-lymphocyte stimulator (BlyS also known as B-cell activating factor or BAFF) are seen in patients with SLE and correlate with immunoglobulin G (IgG) levels, suggesting that inhibition of BlyS could be therapeutic [Cheema et al. 2001; Zhang et al. 2001].
Belimumab is a human IgG1 lambda monoclonal antibody that binds soluble BlyS and inhibits its function. There have been two large-scale phase III studies of belimumab in SLE; the BLISS-52 and BLISS-76 studies [Manzi et al. 2012]. In BLISS-52, 865 patients with active SLE were treated with standard care along with 10 mg/kg belimumab, 1 mg/kg intravenous belimumab or placebo at 0, 14 and 28 days and then every 28 days until week 48 with review at week 52. More patients in the belimumab groups achieved an SLE Responder Index (SRI) response than in the placebo group. For belimumab 1 mg/kg the number of responders was 148/288 (51%) [odds ratio (OR) 1.55; 95% confidence interval (CI) 1.10–2.19; p = 0.0129], for the 10 mg/kg dose 167/290 (58%) (OR 1.83; 95% CI 1.30–2.59; p = 0.0006) and for placebo 125/287 (44%) at week 52 [Navarra et al. 2011]. Belimumab at both doses was well tolerated, it reduced the flare rate, improved serologic markers in patients with serologically active disease and reduced corticosteroid use. Adverse events were similar to those of the placebo group. In BLISS-76 a similar trial design in 819 patients generated SRI response rates for belimumab 1 mg/kg, 10 mg/kg and placebo at 52 weeks of 40.6%, 43.2% and 33.5% (p = 0.017 for 10 mg/kg versus placebo and p = 0.089 for 1 mg/kg versus placebo). Again the rate of flares was less in the belimumab groups with no increase in adverse event rate [Furie et al. 2011]. These studies excluded patients with active lupus nephritis or severe central nervous system involvement. Patients with higher disease activity clinically or serologically had greater benefit [van Vollenhoven et al. 2012]. The US Food and Drug Administration (FDA) approved belimumab for the therapy of SLE in 2011. Belimumab is also approved in Europe and Canada, although the National Institute for Health and Clinical Excellence in the UK is not currently recommending its use on cost-effectiveness grounds.
Rituximab
Rituximab is a chimeric monoclonal antibody that depletes CD20+ B cells and has been effective therapy in B-cell non-Hodgkin lymphoma, rheumatoid arthritis and antineutrophil cytoplasmic antibody-positive vasculitis. In the EXPLORER study, of 257 patients with moderately active SLE assessed, 88 received placebo and 169 received 1 g rituximab (1:2 ratio) at week 0, 2, 24 and 26 with evaluation at week 52. Intravenous methylprednisolone 100 mg was given with each infusion. Oral prednisolone was tapered in both groups according to a preset protocol and background therapy with azathioprine, mycophenolate or methotrexate was continued. The study failed to demonstrate benefit for rituximab as additive therapy in this situation [Merrill et al. 2011].
In the second major trial of rituximab in SLE, the Lupus Nephritis Assessment with rituximab Study (LUNAR), 144 patients with active class III or class IV lupus nephritis were randomized to receive 1 g rituximab (n = 72) or placebo (n = 72) at weeks 0, 2, 24 and 26 with the primary outcome an assessment of renal status at week 52. Intravenous methylprednisolone 1 g was given prior to the first infusion and within 3 days and a further 100 mg was given prior to the subsequent infusions. Oral prednisolone up to 60 mg was given daily up to day 16 and then tapered. Mycophenolate 1.5 g/day increased up to a maximum of 3 g/day in divided doses was also given. The overall renal response rates were 56.9% in the patients receiving rituximab and 45.8% among those receiving placebo (p = 0.18). Eight patients in the placebo group required cyclophosphamide rescue therapy compared with none of the rituximab-treated patients. The rituximab-treated group had a greater reduction in anti-dsDNA antibody levels (p = 0.007). Rituximab also had a greater beneficial effect in patients of African-American or Hispanic ethnicity. Nevertheless, as in the EXPLORER study, the primary outcome was not met [Rovin et al. 2012].
In both the EXPLORER and LUNAR studies, the widely held expert view is that the relatively effective background therapy with corticosteroids obscured the underlying value of rituximab as a nonsteroid therapy for SLE and lupus nephritis and that further studies designed to reflect this will demonstrate this in the future [Weidenbusch et al. 2012; Isenberg, 2012; Gregersen and Jayne, 2012; Pepper et al. 2009]. Another consideration is that although the effect size was comparable to that observed in the belimumab studies, the smaller numbers of participants in the rituximab studies made it more difficult to achieve the primary outcomes.
Furthermore, there are some data that regimes for lupus nephritis that include rituximab with pulsed steroids followed by maintenance therapy with MMF and no further steroid use could even replace corticosteroid-based regimes in the future [Lightstone, 2012].
Antiphospholipid (Hughes) syndrome
In the antiphospholipid antibody (Hughes) syndrome (APS), antibodies directed against phospholipids lead to an increased risk of arterial and venous thrombosis, recurrent pregnancy loss or other obstetric features [Harris et al. 1986]. It can occur as a syndrome in its own right or in association with SLE. The antibodies can be detected directly in the form of antiphospholipid/anticardiolipin or anti-ß2 glycoprotein (a cofactor for phospholipids) antibodies by enzyme-linked immunosorbent assay or indirectly through the lupus anticoagulant clotting test typically using the dilute Russell viper’s venom test to demonstrate an in vitro (antibody) inhibitor of the clotting system.
Treatment depends on the nature of the clinical and laboratory presentation. In patients with the antibodies but without a history of thrombosis, the risk–benefit analysis for taking low-dose aspirin is unclear [Erkan et al. 2007]. A positive lupus anticoagulant test is most closely linked with thrombosis risk and adverse obstetric outcome [Galli et al. 2003; Lockshin et al. 2012] and patients with a positive result for all three antibody tests are at greatest risk [Pengo et al. 2012]. Anecdotally, clopidogrel might be used in patients who are allergic to aspirin.
Once a patient has had a thrombosis, however, the consensus would be to anticoagulate them for life due to the high risk of further thrombosis [Khamashta et al. 1995; Crowther et al. 2003; Finazzi et al. 2005]. Current practice would distinguish between patients with a single venous thrombosis for whom an international normalized ratio (INR) of 2–3 may be sufficient and patients with an arterial thrombosis or recurrent venous thromboses for whom a higher target INR of 3–4 would be more appropriate [Ruiz-Irastorza et al. 2007].
In patients who are intolerant of, or resistant to, warfarin [Scoble et al. 2011], low molecular weight heparin (LMWH) can be used [Vargas-Hitos et al. 2011]. Long-term use of LMWH is associated with osteoporosis and thrombocytopenia.
One recent development is of new oral anticoagulants such as apixaban (a factor Xa inhibitor), dabigatran (a thrombin inhibitor) and rivaroxaban (a factor Xa inhibitor) [Garcia et al. 2010; Cohen and Machin, 2010]. They do not require time-intensive monitoring and dose adjustment but they are not easily reversible if bleeding problems arise. There is a planned clinical trial of rivaroxaban in APS funded by Arthritis Research UK that will address equivalence to warfarin in an APS population (http://www.ucl.ac.uk/ctu/researchareas/other/othertrials).
There is no evidence in most ‘straightforward’ cases of APS that immunosuppression with corticosteroids or disease-modifying drugs have any beneficial effect. Hydroxychloroquine does appear to have a beneficial effect on thrombosis risk in APS, at least in patients with SLE [Petri, 1996]. Statins also have antithrombotic effects and may be sensible adjunctive therapy [Jajoria et al. 2009].
The potential exception to this is in rare patients (about 1% of APS cases) with the catastrophic APS in which a diffuse thrombotic microvasculopathy affecting the lungs, brain, heart, kidney, skin and gastrointestinal tract is associated with a high mortality rate of 30–50%. The mortality rate has improved in recent years with the use of full anticoagulation, intravenous steroids, with or without cyclophosphamide, antibiotics for any sepsis and the use of plasma exchange and intravenous immunoglobulins (IVIGs) [Cervera et al. 2009]. Some recent papers have also described the use of rituximab with beneficial effect in resistant cases [Khattri et al. 2012] and a clinical trial is also ongoing [ClinicalTrials.gov identifier: NCT00537290].
Warfarin is associated with a high risk of fetal malformation in the first trimester of pregnancy and with fetal bleeding thereafter. The standard of care, therefore, for women with antiphospholipid antibodies and recurrent pregnancy loss is the combination of low-dose aspirin plus prophylactic doses of heparin (unfractionated or low molecular weight) [American College of Obstetricians and Gynaecologists Committee on Practice Bulletins – Obstetrics, 2011; Empson et al. 2005; Mak et al. 2010; Danza et al. 2012]. This has improved the rate of live births very substantially to around 75% in some of the above studies. Some studies have suggested, however, that heparin does not confer additional benefit to aspirin alone [Laskin et al. 2009], but this is not the current consensus.
The role of prednisolone in improving the outcome of obstetric APS has been controversial. Initial studies using high-dose prednisolone did not identify any benefit and an increased risk of hypertensive complications and diabetes mellitus [Laskin et al. 1997]. More recent studies of low-dose prednisolone, for example 10 mg/day during the first trimester, have suggested benefit in women with recurrent pregnancy loss in whom conventional approaches have not been effective [Bramham et al. 2011]. As a result the role of prednisolone is being reevaluated. IVIG is another therapy that can be tried, although evidence for effectiveness is limited.
Scleroderma
The principle abnormality in scleroderma is of skin and internal organ thickening (fibrosis). The spectrum of the disorder ranges from more localized disease such as the CREST syndrome (Calcinosis, Raynaud’s, oEsophageal dysfunction, Sclerodactyly, Telangiectasia) to widespread skin and internal organ fibrosis. The more serious and potentially life-shortening complications of the disease are pulmonary arterial hypertension (PAH), interstitial lung disease (ILD) and kidney involvement. There are no specific therapies for the disease in general but over the past few years there have been significant improvements in the overall management of each of these more serious components.
Pulmonary arterial hypertension
Pulmonary arterial hypertension (PAH) is a major cause of death in scleroderma. Prior to specific therapies the 5-year survival rate of patients with scleroderma and PAH was 10% compared with 80% for patients without PAH [Fischer et al. 2012]. It is now over 10 years since the key studies of bosentan, an endothelin receptor antagonist, demonstrated improvement in haemodynamic parameters and symptoms in PAH [Rubin et al. 2002]. Among the FDA-approved therapies it is recommended that calcium channel blockade, bosentan, ambrisentan, sildanefil and tadalafil are used for less severely ill patients while intravenous iloprost, teprostinil and epoprostenol are used for more severely affected patients [Walker and Pope, 2011, 2012].
One key advance in the management of this condition has been the introduction of regular annual or alternate year [in low-risk patients with stable disease >10 years with no dyspnoea, a carbon monoxide diffusing capacity (DLCO) > 70% and a forced vital capacity % (FVC%):DLCO% ratio <1.6] screening with echocardiography and lung function tests to identify early asymptomatic PAH [Pope et al. 2012]. A disproportionate fall in the DLCO, particularly to less than 50% or a FVC%:DLCO% ratio of over 1.6 is suggestive of PAH rather than ILD. Echocardiographic estimation of pulmonary systolic arterial pressure of over 35–45 mmHg is another indicator of potential PAH requiring confirmation by right heart catheterization.
The DETECT study (http://www.med.umich.edu/scleroderma/pdf/1-DETECT%20ACR%20poster,%20D3%2013%2010%2008%20(5-column).pdf) is currently ongoing to further evaluate screening for PAH in scleroderma. Brain natriuretic peptide (BNP) or N-terminal BNP are being studied for their potential as screening tools for PAH but are not sensitive enough for routine clinical use yet [Cavagna et al. 2010; Chung et al. 2010]. Riociguat, a soluble guanylate cyclase stimulator is a potential therapy of interest in PAH [Schermuly et al. 2011].
Interstitial lung disease
ILD is another major cause of mortality. There have been two randomized controlled trials of cyclophosphamide in the treatment of scleroderma-associated ILD [Hoyles et al. 2006; Tashkin et al. 2006]. Tashkin and colleagues evaluated 145 patients with scleroderma and ILD in the USA, randomized to either oral cyclophosphamide up to 2 mg/kg/day or placebo. There was a statistically significant but small beneficial effect at 12 and 24 months in favour of cyclophosphamide on FVC of 2.53% (CI 0.28–4.79%) and some modest effect on symptoms and quality of life. A further study, Scleroderma Lung Study-2 (SLS-2), will compare cyclophosphamide with MMF over 2 years. The second trial [Hoyles et al. 2006] involved 45 patients in the UK who were randomized to receive either prednisolone and six infusions of cyclophosphamide at monthly intervals followed by maintenance therapy with azathioprine, or to receive placebo formulations of each of these. At 12 months there was no significant difference between the groups, although a trend towards improvement in FVC (p = 0.08) and six withdrawals in the placebo group compared with three in the active group. Nevertheless, these trials do not suggest major benefit for a potentially toxic therapy. A meta-analysis of 69 patients in the literature who received MMF demonstrated similar results [Tzouvelekis et al. 2012].
In the absence of highly effective therapy one recent approach has been to try and stratify patients with more extensive disease for whom intensive therapy might be more likely to be worthwhile [Goh et al. 2008]. Goh and colleagues reviewed a cohort of 215 patients with scleroderma ILD for whom high-resolution computed tomography (HRCT) and pulmonary function tests (PFT) data were available. A total of 75 (35%) died. An HRCT with over 20% lung involvement by ILD (ground glass or reticulation) was identified as a threshold for increased mortality. Patients with lung disease of less than 20% were classified as having limited ILD with a better prognosis and those with over 20% as having extensive ILD with a worse prognosis. In those patients whose HRCT could not be clearly classified in this way, the FVC was utilized such that those with an FVC of at least 70% were classified into the limited ILD group and those with an FVC less than 70% were classified as having more extensive disease. The logic of this approach is that systemic sclerosis (SSc)-associated ILD often corresponds to nonspecific interstitial pneumonia (inflammation > fibrosis) rather than the usual interstitial pneumonia (fibrosis > inflammation) in the majority of cases [Mouthon et al. 2007] and therefore individuals with more extensive ILD would receive cyclophosphamide therapy whereas MMF therapy may be more appropriate for patients with limited scleroderma-related ILD. Although they are only present in a small number of patients (around 3%) the presence of antibodies against the U11/U12 RNP complex may indicate ILD in patients with scleroderma [Fertig et al. 2009].
It is also important to remember simple treatable contributors to lung pathology, such as reflux disease as well as optimizing other general health issues such as good nutrition [Baron et al. 2009] and treating causes of iron deficiency anaemia, for example due to gastric antral vascular ectasia [Ingraham et al. 2010]. Development of a cough can be due to dryness (secondary Sjögren’s syndrome), fibrosis, laryngeal or gastro-oesophageal reflux or more serious general pathology (e.g. lung cancer) [Colaci et al. 2013].
Renal crisis
Scleroderma renal crisis describes the development of new and severe hypertension in association with renal failure in patients with scleroderma. It is usually but not exclusively associated with diffuse disease and patients with anti-RNA polymerase antibodies [Denton et al. 2009; Khanna and Denton, 2010]. Prior to the introduction of ACE inhibition the outcome was poor. Even so, the renal outcome is variable with the best outlook occurring if it is treated early and aggressively [Batal et al. 2009, 2010]. Management involves strict monitoring, tight control of blood pressure, inpatient management with intravenous prostacyclin in the event of microangiopathic haemolytic anaemia or hypertensive retinopathy.
Digital ulceration and Raynaud’s
Severe Raynaud’s phenomenon, particularly with digital ulceration, is a major clinical problem in SSc. Calcium channel blockers are typically used as effective first-line medical therapy. ACE inhibition was commonly used second line until clinical trials demonstrated that although this may be effective therapy in scleroderma renal crisis it has little or no benefit in preventing digital ulceration [Gliddon et al. 2007]. α Blockers, angiotensin receptor inhibitors (e.g. losartan), serotonin uptake inhibitors (fluoxetine), moxisylyte and other oral therapies may be worth trying in particular patients [Henness and Wigley, 2007; Stewart and Morling, 2012]. Patients who fail to respond to oral therapy have traditionally been treated with 3–5-day or longer courses of iloprost [Henness and Wigley, 2007; Walker and Pope, 2011]. Sildenafil has also shown to have potential benefit in small studies [Brueckner et al. 2010]. Two randomized controlled trials have demonstrated that bosentan reduces the development of new ulcers, although it does not speed healing of existing ulcers [Korn et al. 2004; Matucci-Cerinic et al. 2011].
Myositis
Polymyositis and dermatomyositis (DM) are the two major forms of idiopathic inflammatory myopathies (IIMs) [Miller, 2012]. Inclusion body myositis (IBM) is a third important subgroup of IIM. IBM is associated histologically with fewer inflammatory changes and with the presence of basophilic granular inclusions, rimmed vacuoles and eosinophilic cytoplasmic inclusions. Involvement of the long flexors of the hands and a particularly high risk of falls might suggest IBM. DM has also been linked to malignancy, particularly in older patients [Callen, 1994].
Corticosteroids are the usual first-line therapy, for example, prednisolone 1 mg/kg/day initially at least for 4–12 weeks and then tapered after a month by 10–20% of the daily dose monthly to the lowest possible maintenance dose. There are small, randomized trials of conventional disease-modifying antirheumatic drugs, such as methotrexate, azathioprine, ciclosporin, tacrolimus, cyclophosphamide or mycophenolate [Gordon et al. 2012; Aggarwal and Oddis, 2012] (http://www.controlled-trials.com/ISRCTN40085050), demonstrating only modest effects. IBM is less likely to respond and in a patient who is not responding to immunosuppressant therapy the diagnosis of IBM should be carefully considered, if need be with a second muscle biopsy. Motor neurone disease is another differential to consider.
In patients resistant to corticosteroids, IVIG has been identified as a promising second-line therapy [Wang et al. 2012]. Dastmalchi and colleagues reported on a 14-week open-label trial with infliximab in 13 refractory patients showing no benefits but more adverse events [Dastmalchi et al. 2008]. In addition, some biological therapies, particularly rituximab, may have a role, although larger studies are needed [Dalakas, 2010]. The Rituximab in Myositis (RIM) study presented in ACR 2011 failed to meet the primary and secondary end points but a large number of patients met the International Myositis Assessment and Clinical Studies Group (IMACS) definition of improvement.
Myositis is associated with long-term disability in 80% of patients and has a standardized mortality ratio three times higher than the general population [Marie, 2012]. Screening early and often for ILD in particular with lung function tests and HRCT is advisable.
One of the challenges with rare diseases is the difficulty of validating assessment tools that can be used to assess improvement in clinical trials. International collaborations such as IMACS and the Paediatric Rheumatology International Trials Organisation have been working to develop individual measures and have also proposed preliminary definitions of improvement [Rider et al. 2011; Miller, 2012].
In the absence of major therapeutic advances the key recent development in myositis has been the use of novel autoantibodies to classify patients and as potential prognostic indicators (Table 1).
Autoantibodies in myositis.
CADM, clinically amyopathic dermatomyositis; CI, confidence interval; DM, dermatomyositis; ILD, interstitial lung disease; JDM, juvenile dermatomyositis; OR, odds ratio.
Primary Sjögren’s syndrome
In primary Sjögren’s syndrome (PSS), focal lymphocytic inflammation of exocrine glands is associated with clinical features of glandular dysfunction, particularly of oral and ocular dryness. The immunological features, apart from these histological features, are of anti-Ro with or without anti-La antibodies found in around 70% of patients. Therapy has been principally symptomatic with a focus on oral and ocular hygiene and topical therapies. Glandular stimulants such as pilocarpine have been shown to be effective in symptomatic improvement in patients with residual glandular function [Vivino et al. 1999].
For the past 10 years the gold standard classification criteria for PSS has been the American-European Consensus Group criteria [Vitali et al. 2002]. Recently, alternative classification criteria, the preliminary ACR criteria, have been proposed [Shiboski et al. 2012; Whitcher et al. 2010]. Neither set of criteria include salivary gland ultrasound. In clinical practice, the presence of hypoechoic areas and other features on ultrasound are frequently seen in PSS [Takagi et al. 2010; Milic et al. 2012]. Magnetic resonance imaging may yield similar results [Regier et al. 2009].
In order to reach the point at which biological therapies can be studied in PSS there has been considerable development over the past decade of patient-related outcome measures such as the Profile of Fatigue – Sjögren’s International Collaborative Clinical Alliance (SICCA) Symptoms Inventory [Bowman et al. 2009] and more recently the EULAR Sjögren’s Patient Reported Index [Seror et al. 2011]; and disease activity measures such as the Sjögren’s Clinical Activity Index [Bowman et al. 2007] and Sjögren’s Syndrome Disease Activity Index [Vitali et al. 2007]. More recently, a collaboration, the EULAR Sjögren’s working group, has developed the EULAR Sjögren’s Syndrome Disease Activity Index [Seror et al. 2010a, 2010b]. Damage measures have also been developed [Vitali et al. 2007; Barry et al. 2008]. Other collaborations include the UK Primary Sjögren’s Syndrome Registry, which has been set up to facilitate clinical and basic science research into Sjögren’s syndrome [Ng et al. 2011]. The SICCA group in the USA has proposed the above alternative criteria sets as well as establishing the SICCA Registry as a repository of clinical and biological samples [Daniels et al. 2009].
Generally, the prognosis for PSS is good unless patients develop lymphoma when it is more variable [Voulgarelis et al. 2012]. Although rituximab has not yet been established as routine therapy for PSS, it is now standard practice to incorporate it in regimes, for example, with CHOP Rituximab-cyclophosphamide, Hydroxydaunorubicin, Oncovin (Vincristine), Prednisolone (R-CHOP) for salivary gland lymphoma. A paper by Theander and colleagues demonstrated that salivary gland biopsy at diagnosis provided important data on long-term lymphoma risk depending on whether B-cell germinal centre (GC) structures were present in the biopsy or not [Theander et al. 2011]. In a long-term cohort of 175 patients, 6 of 43 patients with GCs on salivary gland biopsy developed lymphoma compared with only 1 of 132 without. The positive predictive value for GC was 16% and the negative predictive value was 99%.
Attempts at using biological therapies have been mixed. Antitumour necrosis factor therapy was ineffective in a double-blind randomized trial [Mariette et al. 2004]. In the past few years evidence of B-cell and B-cell cytokine involvement, such as BLyS/BAFF, has been of particular interest [Ittah et al. 2006; Lavie et al. 2007]. A randomized double-blind controlled trial of a single course of rituximab versus placebo in France [Tolerance and Efficacy of rituximab in Sjögren’s syndrome (TEARS)] [ClinicalTrials.gov identifier: NCT00740948] failed to meet its primary endpoint but suggested an approximately 10% benefit over placebo (Professor Xavier Mariette, oral presentation at EULAR Congress 2012). A second trial in the UK of two courses of rituximab against placebo funded by Arthritis Research UK is currently ongoing.
There is a theoretical basis for trying other existing biologics such as belimumab (anti-BlyS/BAFF) or tocilizumab (anti-interleukin-6) as well as tackling other biological pathways [Ng and Bowman, 2011].
Basic science
Looking back over the past decade one of the main themes has been the role of B cells and B-cell cytokines in the pathogenesis of many of the autoimmune connective tissue disorders and the use of anti-B-cell/anti-B-cell cytokine agents in their therapy (see above). Looking forwards, a number of themes are emerging, in particular, the potential role of the innate immune system, particularly of type I interferons and toll-like receptors (TLRs) in the triggering and maintenance of these conditions. A second related theme is the insight that high-throughput genome-wide association screening (GWAS), ‘next-generation’ screening and functional genomics will provide into the pathogenesis of these conditions.
Innate immunity
TLRs are a group of receptor proteins that recognize structurally conserved molecules shared by many microbial pathogens [Mills, 2011]. The name is derived from the Toll gene of Drosophila. TLRs are key molecules that alert the immune system to the presence of foreign microorganisms. They can be triggered by a range of bacterial molecules, including lipopolysaccharides, flagella proteins and also some DNA and RNA species. The latter in particular offer a potential mechanism for autoimmunity, since many human autoantigens are either nucleic acids or nucleic acid binding proteins. TLRs then activate a number of signalling pathways, including the activation of type I interferons [Biggioggero et al. 2010; Pascual et al. 2010; Sozzani et al. 2010].
There has been a substantial amount of work over the past few years exploring the relationships between the innate immune system, TLRs, the type I interferon ‘signature’ [pattern of upregulation of downstream molecules such as the interferon regulatory factor 5 (IRF5) and signal transducer and activator of transcription 4 (STAT4) genes following interferon activation] in SLE [Banchereau and Pascual, 2006], PSS [Nordmark et al. 2012; Brkic et al. 2012], myositis [Suber et al. 2008] and scleroderma [Lafyatis and York, 2009; Higgs et al. 2011]. Sifalimumab, an anti-interferon antibody, is now being studied in SLE [Merrill et al. 2011] and potentially could be investigated in other autoimmune diseases over the next few years.
Genome-wide screening and genomic approaches
In parallel with hypothesis-driven research (e.g. investigating the role of the innate immune system in autoimmune disease), another powerful technique over the past decade has been the use of GWAS to examine the association of common genetic variants with particular diseases. These studies take advantage of several technical developments. The first is the identification of a large number of single-nucleotide polymorphisms (SNPs) covering the entire human genome [Sherry et al. 2001; Gibbs et al. 2003]. By demonstrating statistical linkage between the presence of these variants and a particular disease it has been possible to then identify possible genes in the location of the relevant SNPs that could be involved in disease pathogenesis. The second technical development has been of automated ‘chip’ or ‘microarray’ technology that allows millions of SNPs from thousands of individuals to be examined in a short space of time.
A number of GWAS have been performed to date in SLE and scleroderma. They have perhaps been easier to study than some of the rarer connective tissue disorders due to access to large, established cohorts of patients. A number of SNP linkages have been identified in both of these diseases but the human leukocyte antigen region and the regions of the IRF5 and STAT4 genes have emerged as common areas of interest both in SLE [Moser et al. 2009; Deng and Tsao, 2010; Sandling et al. 2011; Ramos et al. 2011] and scleroderma [Allanore et al. 2011; Gorlova et al. 2011; Radstake et al. 2010; Broen et al. 2012], again supporting the potential role of the type I interferon system as important in the pathogenesis of these connective tissue disorders and as a potential target for therapy. Early studies in PSS have demonstrated similar data [Lessard et al. 2011].
These studies are still at a relatively early stage and it is likely that other potential genes of interest will emerge for hypothesis-driven evaluation over the next decade as will new insights from other next-generation sequencing technologies and functional genomic studies [Cho and Gregersen, 2011; Scofield and Kaufman, 2012].
Conclusion
The past few years have seen significant changes in clinical practice, for example the increasing use of renal biopsy-driven choices of therapy in lupus nephritis and the introduction of biological therapy in SLE. In scleroderma, closer assessment of patients and the introduction of therapies such as bosentan have improved the outlook for patients with PAH. Challenges still remain, however, particularly to find effective therapies for scleroderma, myositis and Sjögren’s syndrome and to reduce the usage of toxic therapies such as corticosteroids and cyclophosphamide wherever possible. With new oral anticoagulants and a broad range of new biological therapies becoming available, the next few years should also be a period of potentially rapid progress in the therapy of these disorders.
Footnotes
Acknowledgements
We would like to thank David D’Cruz, Chris Denton, Caroline Gordon, Patrick Gordon Munther Khamashta, Voon Oong and Sonali Wijetilleka for their advice on content.
Funding
This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.
Conflict of interest statement
The authors declare no conflicts of interest in preparing this article.
