Introduction

Patients with Sjögren's syndrome (SS) primarily present with sicca symptoms involving the major mucosal surfaces. The main sicca features of this systemic autoimmune disease are xerophthalmia (dry eyes) and xerostomia (dry mouth), which are diagnosed using specific ocular (Rose Bengal or Lissamine Green staining, or the Schirmer test) and oral (salivary flow measurement or parotid scintigraphy) assessments, respectively.1 Patients with SS can present with a broad spectrum of disease, extending from sicca syndrome to systemic manifestations (extraglandular involvement), and the disease can be further complicated by lymphomas. A wide range of analytical features are associated with SS, including cytopenia, hypergammaglobulinemia and a high erythrocyte sedimentation rate (ESR), and cryoglobulins and hypocomplementemia are the main prognostic markers.2 SS is also associated with autoantibody production; antinuclear antibodies are the most frequently detected, whereas anti-Ro/SSA are the most specific for SS.2 Focal lymphocytic infiltration of the exocrine glands, usually determined by minor labial salivary gland biopsy, is the histological hallmark of SS.

At present, no cure exists for primary SS, and no therapy prevents disease progression. The classic therapeutic approach is based on symptomatic treatment of glandular manifestations and broad-spectrum immunosuppression directed against organ-specific extraglandular disease. Over the past decade, SS research has centred on the development of more effective targeted therapies, including various completed and ongoing randomised controlled trials (RCTs), a number of which have shown promising results. These new therapies provide hope for better disease management in patients with SS.3 The objective of this Review is to summarize the reported data on targeted therapies for the treatment of SS, with the aim of providing physicians with a rational therapeutic approach tailored to the different clinical scenarios that patients with SS can present. In the following sections, we deem the highest-to-lowest quality of evidence to be derived from study designs in the following order: RCTs, prospective cohort studies, case–control studies, retrospective studies and case series.

Topical medications—level of evidence

Saliva substitutes

Saliva substitutes are frequently used in clinical practice, owing to the near-complete absence of appreciable adverse events and their beneficial effects on nocturnal oral dryness. A number of studies in SS have evaluated different saliva substitutes (Table 1).4,5,6,7,8,9,10 In one RCT,9 no marked differences in efficacy were observed between three different saliva substitutes and placebo. By contrast, three RCTs in patients with SS found that saliva substitutes considerably improved several oral symptoms, but not salivary flow rates.4,5,6 In a comparison of mucin-containing and carboxymethylcellulose-containing substitutes, neither was shown to provide superior therapeutic benefit, although the former preparation was reportedly preferred by patients.10 Data from uncontrolled studies have shown considerable improvement in xerostomia and salivary flow after application of a hydroxymethylcellulose-containing oral spray,8 and a reduction in the dental plaque burden in patients treated with chlorhexidine mouthrinses.7

Table 1 Studies of topical medications in patients with SS
Full size table

Eye drops

Glucanes

Marked improvements in xerophthalmia compared with baseline measurements were observed for topical eye drops containing glucanes; five studies using sodium hyaluronate11,12,13,14,15 and one using hydroxypropylmethylcellulose16 as the active ingredient (Table 1). Condon et al.14 found that the use of artificial tear drops containing hyaluronate resulted in a greater improvement in dry eye symptoms and ocular tests than saline eye drops. A comparative study showed greater therapeutic benefit for hypotonic than isotonic hyaluronate preparations.11 No marked differences in efficacy were observed between hyaluronate-containing and carboxymethylcellulose-containing eye drops.13

NSAIDs

Limited evidence is available for the use of artificial tear drops containing topical NSAIDs in patients with primary SS. In an uncontrolled comparative study that enrolled 20 patients with SS,17 ocular symptoms improved after treatment with eye drops containing diclofenac or indomethacin, with greater improvement in corneal sensitivity observed for diclofenac. However, a small RCT (32 patients enrolled) found no substantial difference in the resolution of filamentary keratitis at day 28 after initiation of treatment with diclofenac-containing tears compared with saline eye drops.18 Adverse events are the key point to consider when using topical NSAIDs. Guidera et al.19 retrospectively described severe adverse events associated with the use of topical NSAIDs in 18 eyes of 16 patients with different medical conditions, including SS. The adverse events occurred after 2–4 weeks treatment with topical NSAIDs in most patients, and included corneal–scleral melts, perforation, ulceration and severe keratopathy.

Corticosteroids

At present, the use of topical corticosteroids for treatment of ocular symptoms in patients with dry eyes has been analysed in four studies; considerable improvements were demonstrated after treatment with fluorometholone or methylprednisolone, but not loteprednol, in comparison with placebo.20,21,22,23 One of these studies also compared topical corticosteroids and NSAIDs and found that corticosteroids were superior.20 Adverse events associated with prolonged corticosteroid use should be taken into account; Marsh et al.21 reported serious adverse effects (increased intraocular pressure, worsening or development of cataracts) in 14% of patients with SS treated with topical nonpreserved methylprednisolone sodium succinate.

Cyclosporine A

Cyclosporine A has been used to treat autoimmune diseases for more than three decades, and an ophthalmic formulation was approved by the FDA in December 2002 for the treatment of dry eye disease. Improvements have been observed after administration of twice daily topical cyclosporine A in a number of studies, with respect to baseline subjective and objective measurements of dry eyes (Table 1).24,25,26,27,28,29,30,31,32,33,34 The largest RCT,32 which included 877 patients with dry eye syndrome (31% with SS) who were evaluated after 6 months, tested two different cyclosporine A doses (0.05% and 0.1%). A significant improvement in Schirmer test scores was observed for both groups (P<0.007) compared with placebo, but corneal staining scores only significantly improved after treatment with 0.05% cyclosporine A (P = 0.008, P = 0.06 in the 0.1% group);32 the 0.05% group also showed significantly reduced blurred vision—one of seven symptoms of ocular discomfort—(P <0.01) and artificial tear use (P = 0.006).32 Therefore, the recommended dosing regimen is 0.05% cyclosporine A drops twice daily. In a publication reporting the extension of two RCTs in patients initially treated for 6–12 months, no additional improvement in subjective and objective measurements was observed during the extended study period (mean of 20 months).34 With respect to adverse events, in the largest RCT of cyclosporine A eye drops,32 the only marked difference between the cyclosporine A and placebo groups was in the percentage of patients who experienced burning eye (15% versus 6%, respectively); most events were mild-to-moderate and transient, with only 2% of patients discontinuing treatment because of burning or stinging.32 Only two patients, both in the placebo group, experienced ocular infection during the study period.32 Taken together, these studies have led to ocular cyclosporine A being licensed for use in the USA and 27 other countries worldwide, but not in Europe, Canada or Australia.

Diquafosol

Tauber et al.35 performed a RCT using an ophthalmic formulation of 2% diquafosol tetrasodium (INS365), an agonist of the P2Y purinoceptor 2 protein that is involved in ocular surface hydration. The study, which enrolled 527 patients with keratoconjunctivitis sicca (76 secondary to SS), revealed no marked difference in therapeutic effect between topical diquafosol and placebo, except for a greater improvement in corneal staining score with the former.

Systemic medications—level of evidence

Secretagogues

For patients with SS who have residual salivary gland function, stimulation of saliva flow with a secretagogue is the treatment of choice and is, at present, the most efficacious means of preventing long-term oral complications (Table 2). Nonpharmacological secretagogues might be useful in patients with reasonable levels of salivary flow, such as mechanical and/or gustatory stimulation with sugar-free gums and candies. With respect to pharmacological salivary stimulation, some choleretic (anetholtrithione) and mucolytic (bromhexine, N-acetylcysteine) agents have been used as secretagogues in primary SS since the 1980s, although without solid scientific evidence of efficacy. Uncontrolled studies found only marginal therapeutic benefits for anetholtrithione and bromhexine,36,37 whereas a small trial, in which 26 patients were treated with 200 mg N-acetylcysteine three times daily, demonstrated substantial improvements in sicca symptoms and ocular test results.38

Table 2 Studies of oral secretagogues in patients with SS
Full size table

Two muscarinic acetylcholine receptor agonists (pilocarpine and cevimeline) are licensed for the treatment of sicca symptoms in SS. These agents stimulate the muscarinic acetylcholine receptors M1 and M3 present on salivary glands, leading to increased secretory function.

Pilocarpine

11 studies have analysed the use of the muscarinic acetylcholine receptor agonist pilocarpine in patients with SS.39,40,41,42,43,44,45,46,47,48,49 In the two largest RCTs,42,48 pilocarpine doses of 5 mg and 7.5 mg every 6 h resulted in markedly improved oral, ocular, nasal, vaginal and skin dryness, and salivary flow rates. Uncontrolled studies have also reported a substantial therapeutic benefit of this drug, based on the findings of ocular tests and reduction of candidiasis.39,40,41,43,44,45,46 Data from RCTs have revealed a high frequency of adverse events associated with pilocarpine use, including sweating, increased urinary frequency and flushing (observed in 43%, 10% and 10% of patients, respectively).42,47,48,49 In a dose-adjustment RCT,42 23% of patients were switched from the 7.5 mg to the 5 mg regimen after 6 weeks of treatment, suggesting the lower dose is better tolerated.

Cevimeline

The efficacy of another muscarinic acetylcholine receptor agonist, cevimeline, has been assessed in eight studies that enrolled patients with SS.50,51,52,53,54,55,56,57 Three RCTs have compared dosages ranging between 15 mg and 60 mg taken three times daily, and the best results—including considerable improvements in dry mouth and dry eyes, salivary flow rates and ocular test results—were achieved with a dose of 30 mg.50,52,54 A crossover trial also found a marked reduction in candidiasis, dental plaque burden and gingival bleeding.52 In the two largest trials of cevimeline in SS,50,54 both the 30 mg and 60 mg regimens were associated with higher frequencies of nausea (relative risk [RR] = 1.68 and 2.77, respectively) and sweating (RR = 2.16 and 3.00, respectively) in comparison with placebo, and rigors were more common with the 60 mg dose than with placebo (RR = 1.92).50

Glucocorticoids

The frequent use of glucocorticoids in clinical practice for the treatment of primary SS is not supported by reliable scientific evidence, as no study has specifically evaluated these agents as therapies for the extraglandular features of this disease. Five studies—all but one uncontrolled—have analysed the use of corticosteroids in SS (Table 3),58,59,60,61,62 but have focused principally on their effects on sicca symptoms. In 1993, Fox et al.60 performed a RCT in a small series of patients with SS (eight patients in each arm) comparing placebo to 30 mg daily prednisone combined with 20 mg daily piroxicam, and found appreciable differences in subjective symptoms but not in objective tests. Increased salivary flow with prednisolone, administered either orally59 or by parotid irrigation,58 together with reduced levels of IgG, rheumatoid factor (RF) and anti-Ro/SSA and anti-La/SSB antibodies has been demonstrated in uncontrolled studies in SS; however, limited information can be drawn from these studies owing to substantial methodological flaws in their design. By contrast, a prospective study in 60 patients with primary SS, followed for a mean of nearly 4 years,61 found that the use of corticosteroids did not influence the progressive disease-associated decrease in salivary flow rates. The use of systemic corticosteroids in patients with SS has been associated with adverse events, including increased appetite and weight gain in comparison with placebo,60 and a twofold higher frequency of diabetes mellitus compared with age-matched and gender-matched individuals without systemic autoimmune diseases.62

Table 3 Studies of oral glucocorticoids, immunosuppressive and immunomodulatory agents use in patients with SS
Full size table

Antimalarial agents

The use of antimalarial agents in primary SS is based on their effectiveness in similar diseases, such as systemic lupus erythematosus. Only five studies have analysed the use of hydroxychloroquine (HCQ) in patients with SS (Table 3).63,64,65,66,67,68,69 A retrospective study of 50 patients described considerable improvements in sicca features, parotid gland enlargement, oral infection, myalgia, arthralgia, fatigue and joint swelling after 12 months of HCQ therapy.67 All studies of HCQ use in patients with SS found substantial improvements in baseline analytical and immunological parameters, including ESR and levels of gamma globulins, IgG, IgM, C-reactive protein (CRP), RF, anti-La/SSB antibodies and haemoglobin.63,64,65,66,67,68,69 No cases of retinal toxicity or severe adverse events were reported in any of these studies.63,64,65,66,67,68,69 Furthermore, a retrospective case–control study noted a lower prevalence of diabetes mellitus in patients with SS treated with antimalarial agents than in age-matched and gender-matched individuals without systemic autoimmune diseases.62

Immunosuppressive agents

The use of immunosuppressive agents in primary SS is based on the same level of evidence as that of glucocorticoids (Table 3). The use of cyclosporine A, azathioprine, methotrexate, leflunomide or mycophenolic acid has been studied in five small series of patients with SS.70,71,72,73,74 One of these studies, a RCT using azathioprine, observed no improvements in disease.74 Although improvements in sicca symptoms were found in the studies using cyclosporine A,70 methotrexate71 and mycophenolic acid,73 no study found statistically significant improvements in objective tests. Treatment with leflunomide72 or mycophenolic acid73 was associated with substantial improvements in some SF-36® (36-item Short Form Health Survey) and Multidimensional Fatigue Inventory domains, and with a reduction in analytical abnormalities, including levels of gamma globulin, IgM, IgG, RF and C3 and/or C4. The results of these studies provided three key messages regarding the use of immunosuppressive agents: these agents offer limited benefits for sicca features; specific analysis of their effects on extraglandular features is inadequate; and they are associated with unacceptable rates of adverse events (frequency of 41–100%).70,71,72,73,74

Immunomodulatory agents

Several immunomodulatory agents have been tested in SS (Table 3), with marginal benefits (D-penicillamine)75 or with an unacceptable rate of adverse events (thalidomide).76 Mizoribine, an imidazole nucleoside, was evaluated in two uncontrolled studies performed by the same group,77,78 with improvements in sicca visual analogue scale scores and salivary flow being reported. Somewhat surprisingly, however, 29 adverse drug reactions were reported in 18 of 59 patients included in the first study,77 compared with none in the second study.78

Other therapies

A wide variety of other drugs have been tested in patients with primary SS, including steroid hormones, antimicrobial agents, antiulcer drugs and fatty acids. Some of the drugs tested showed no change in objective and subjective sicca features in comparison with placebo, including rebamipide79 and doxycycline.80 Other agents were found to be associated with an unacceptable rate of adverse events (zidovudine),81 or marginal therapeutic benefits (dehydroepiandrosterone, linolenic and γ-linolenic acids and nizatidine).82,83,84,85,86,87,88

Biologic agents—level of evidence

The emergence of biologic therapies has increased the therapeutic armamentarium available for the treatment of SS, but use of these agents is limited by the lack of licensing. 17 studies have analysed the therapeutic potential of five biologic agents in primary SS (Table 4): IFN-α;89,90,91,92 two anti-TNF agents—infliximab93,94 and etanercept95,96 (a monoclonal antibody against TNF and a recombinant soluble TNF receptor, respectively); and two B-cell-targeted therapies—rituximab97,98,99,100,101,102,103,104,105 and epratuzumab106 (monoclonal antibodies against CD20 and CD22, respectively).

Table 4 Studies of biologic agents in patients with SS
Full size table

IFN-α

Controlled studies (two double-blind, placebo-controlled and one single-blind, controlled with sucralfate) comprising a total of 569 patients with SS have evaluated the use of 150 IU (international units) of oral IFN-α three times daily.89,90,91 Two of these studies reported that sicca symptoms, salivary flow and lymphocytic infiltration all improved with IFN-α treatment.90,91 However, the largest of the three controlled trials—a RCT that included 497 of the 569 patients treated with IFN-α in controlled trials to date—observed that IFN-α use only improved unstimulated salivary flow and was associated with a higher percentage of gastrointestinal adverse events in comparison with placebo.89 A small uncontrolled comparative study found increased tear and saliva production in patients treated with 3.1 MU (million units) of oral IFN-α2 three times per week in comparison with patients receiving 6 mg per kg of HCQ daily.92

TNF blockade

Infliximab

A prospective open-label study in 16 patients with SS demonstrated efficacy of infliximab in this disease, based on subjective and objective measures.93 By contrast, the subsequent TRIPSS study,94 a RCT that enrolled 103 patients with SS, found no statistically significant changes in symptoms, salivary flow rates, quality of life or in the results of ocular tests or salivary biopsy in comparison with placebo. The only therapeutic benefits of infliximab treatment reported in the TRIPSS study were improvements in some analytical parameters.94 With respect to adverse events, infusion-related reactions were reported in 10% of patients and infections in 4% (one patient developed pneumococcal septicaemia).

Etanercept

Two studies using etanercept (one RCT, one prospective) revealed no difference in the main sicca characteristics and symptoms associated with SS after treatment.95,96 Nevertheless, substantial improvements in some analytical parameters (ESR and CRP levels) were observed. Three of 29 patients treated with etanercept in these trials experienced adverse events.

B-cell-targeted therapies

The largest number of clinical trials investigating the efficacy of biologic agents in patients with SS used the B-cell-depleting antibody rituximab;97,98,99,100,101,102,103,104,105 Table 5 summarizes data reported for the 125 patients with SS treated with this agent. Two placebo-controlled RCTs comprising 47 patients demonstrated considerable improvements in sicca features, salivary flow, ocular tests results, fatigue and quality of life scores after treatment with rituximab.97,98

Table 5 Rituximab use in controlled trials and observational studies in SS97,98,99,100,101,102,103,104,105
Full size table

With respect to systemic involvement, one of these trials also reported a reduction in extraglandular features in patients treated with rituximab, compared with placebo.97 Similar results of rituximab therapy were also observed in uncontrolled studies, especially for articular, vasculitic, pulmonary and neurological involvement. Seror et al.101 reported a decrease in the daily dose of corticosteroids used by patients with systemic involvement of primary SS after treatment with rituximab, which might have implications for reduction of the risk of steroid-associated adverse events. All studies that have investigated the effects of rituximab in patients with SS97,98,99,100,101,102,103,104,105 have reported statistically significant reductions in analytical parameters, such as ESR, CRP levels, cryoglobulinemia and/or RF titres.

In one study,100 antichimeric-antibodies were found in five of 23 (22%) patients with SS, four of whom had early disease. An overall analysis of rituximab-related adverse events in patients with SS revealed early and late infusion-related reactions in 19% of patients—particularly in individuals with early disease who were not pretreated with corticosteroids—and infections in 13% (Table 5).

The anti-CD22 antibody epratuzumab has only been tested in one study in patients with SS. This open-label prospective study, which included 16 patients treated with epratuzumab, demonstrated marked improvements in fatigue and subjective patient-reported and physician-assessed outcomes.106

Therapy tailored to clinical scenarios

Management of sicca features

The primary therapeutic approach for sicca manifestations should be centred on symptomatic treatment, using artificial tears and saliva substitutes (Figure 1). This recommendation is based on evidence from a number of studies demonstrating that daily use of these treatments offers relief from sicca symptoms and can improve quality of life, without the risk of adverse events.

Figure 1: Proposed therapeutic algorithm for the treatment of sicca features in SS.
figure 1

Level of evidence (1–4; recommendations based on evidence from studies that predominantly included patients with sicca syndrome or patients with SS are denoted by + and ++, respectively) and strength of recommendation (A–D) are shown in parentheses for each therapy according to the grading recommendations of Harbour and Miller.123 The first step in the management of sicca symptoms always consists of topical therapies; systemic therapies represent the second step, if topical interventions are not effective. Patients with acceptable salivary flow outputs might have poor tolerance of saliva substitutes owing to the 'sticky' feeling they can cause; in these patients, mechanical and/or gustatory stimulation (sugar-free candies and chewing gum) might be a useful first-line therapy before intervention with muscarinic receptor agonists (pilocarpine or cevimeline) is considered. Although a high level of evidence supports the use of certain topical ocular therapies, these agents have a grade B recommendation because studies were overwhelmingly performed in patients with keratoconjunctivitis sicca, with only a variable proportion of patients with SS included. Limited evidence is available for the treatment of sicca features other than those affecting the eyes and mouth in primary SS.1 Lacrimal punctal occlusion using plugs or thermal cautery can be useful in refractory and/or severe cases. Abbreviations: ENT, ear, nose and throat; SS, Sjögren's syndrome.

Full size image

No controlled studies have directly compared saliva substitutes and secretagogues in individuals with moderate-to-severe sicca symptoms; therefore, it seems reasonable that, in clinical practice, patients with mild symptoms should first be treated with agents not associated with adverse events (such as saliva substitutes) before drugs with a higher rate of adverse effects (including muscarinic acetylcholine receptor agonists) are introduced.107 The available data on management of oral dryness do not show conclusively that one specific saliva substitute is superior to another in patients with SS; thus, more than one substitute should always be tried before turning to alternative approaches, especially when the first agent prescribed is poorly tolerated. In addition, patients with acceptable salivary flow outputs might show poor tolerance of saliva substitutes owing to the 'sticky feeling' caused by these agents, which can provoke a sensation of oral dryness. In these patients, mechanical and/or gustatory stimulation of saliva production using sugar-free chewing gums and sweets might be useful. The use of anticholinergic medications, intake of alcohol (including mouthwashes and fluoride rinses that contain alcohol) and smoking should be discouraged, whereas nonpharmacological approaches such as water intake, mechanical and gustatory stimulation, and fluoride toothpaste can be useful and should be promoted.

With respect to eye drops for relief of ocular symptoms, evidence supports the use of preservative-free formulations of artificial tears (containing hyaluronate or methylcellulose) 3–4 times daily, reducing the interval between dosing to as little as 1 h when necessary.1,2,3 Alternatives include topical 0.05% retinyl palmitate and artificial tears containing polyethylene or propylene glycols.

Individuals with severe sicca features can require more intensive therapy than those with mild or moderate sicca symptoms, owing to the substantial implications for the patient's general health status. In patients with moderate or severe oral dryness and residual salivary gland function, oral muscarinic receptor agonists (pilocarpine or cevimeline) are the treatment of choice, as long as contraindications are taken into account (Figure 1). Evidence suggests that the doses with the best ratios of efficacy to adverse events are 5 mg every 6 h for pilocarpine and 30 mg every 8 h for cevimeline, although no studies have directly compared the efficacy of the two drugs. Tolerance must be considered, as the frequency of reported adverse events for these agents was as high as 40%, with some events being more commonly associated with pilocarpine (increased urinary frequency and flushing) and others with cevimeline (gastrointestinal effects). In patients intolerant to muscarinic receptor agonists, N-acetylcysteine might be an alternative. Although RCTs have demonstrated a considerable improvement in sicca features after rituximab therapy,97,100 we consider that its off-label use to treat only sicca involvement—even when severe—is not warranted.

Patients with SS-associated severe or refractory keratoconjunctivitis sicca might require the addition of topical anti-inflammatory agents. Ocular NSAIDs or corticosteroids should only be prescribed by ophthalmologists and for the minimum time necessary, as adverse events seem to be more frequently reported after use of these agents for >2 weeks. By contrast, controlled trials support the use of topical 0.05% cyclosporine A twice daily, as this therapy has an acceptable safety profile, although further benefits are not seen beyond 6 months of treatment.

The role of autologous serum has only been tested in two small studies at present,108,109 and its efficacy in the treatment of sicca symptoms should be confirmed in larger studies. For the most refractory cases of occular dryness, temporary occlusion of the puncta through the insertion of plugs is recommended.1,2,3

Management of general symptoms

Patients with primary SS often present with nonspecific general symptoms, including noninflammatory muscle and joint involvement, fatigue and weakness, which can have a greater impact on the quality of life than sicca features. In patients with these manifestations, the first step should be a differential diagnosis with associated conditions such as hypothyroidism, neoplasia, depression and especially fibromyalgia, which is reported in 22–33% of patients with primary SS and can heavily influence both patient and physician evaluations of health status.110,111 After excluding these disorders, HCQ should be the cornerstone of therapy, with clinical benefits of this agent being reported beyond fatigue and musculoskeletal pain; uncontrolled studies found additional improvements in subjective and objective sicca features as well as analytical and immunological parameters, and reductions in parotid enlargement and oral infections.63,64,65,66,67,68,69

The four biologic agents investigated in trials that enrolled patients with primary SS have been associated with improvements in fatigue (including one small RCT using rituximab).93,95,97,98,106 However, we consider that the off-label use of these new drugs to treat only general symptoms of SS (even when severe) is not justified.

Management of systemic involvement

As a general rule, the management of extraglandular features in primary SS should be organ-specific, using mainly corticosteroids and immunosuppressive agents (Figure 2).1,2,3 Unfortunately, studies analysing the effects of immunosuppressive agents in patients with primary SS, which are overwhelmingly uncontrolled, are designed to evaluate sicca rather than systemic outcomes. These trials have shown poor results of immunosuppressive therapies, with an excess of adverse events.

Figure 2: Proposed therapeutic algorithm for treatment of the main extraglandular manifestations of SS.
figure 2

Available data for treatment of extraglandular SS symptoms come from nonanalytical studies, such as retrospective series or case reports (evidence level 3, on a scale of 1–4) representing the lowest strength of recommendation (grade D, rated from A–D) according to the grading recommendations of Harbour and Miller.123 The sole exception is a RCT using rituximab, which showed a reduction of the number of reported extraglandular manifestations compared with placebo (though the number of patients was too small to provide a specific recommendation). Corticosteroids should be used at the minimum dose and length of time necessary. Classification of drugs as first-line, second-line and third-line therapies is based on the number of case reports that have shown them to be effective and the authors' experience. No data are available on immunosuppressive maintenance therapy; thus, regimens similar to those used in other autoimmune diseases (systemic lupus erythematosus, vasculitis) are recommended. Abbreviations: CNS, central nervous system; IVIg, intravenous immunoglobulin; SS, Sjögren's syndrome.

Full size image

Similarly, studies of biologic agents in SS have primarily focused on evaluating their effects on sicca involvement; nevertheless, several studies have reported promising results regarding the influence of these therapies on the extraglandular features of SS. These studies include a small RCT97 and uncontrolled trials of rituximab,98,99,100,101 which have reported a clinical response in >80% of patients with systemic involvement, improvements in several immunological parameters and a reduction in the mean daily corticosteroid dose. Therefore, current scientific evidence suggests that rituximab can be considered for the treatment of patients with systemic involvement refractory to standard therapies (limited response or intolerance to corticosteroids and immunosuppressive agents).112 The amount and quality of evidence on the off-label use of rituximab in the management of SS-related extraglandular features is higher than that available for corticosteroids and immunosuppressive drugs; however, a reasonable assessment of costs and the risk of serious adverse events versus the benefits of treatment should always be made on an individual basis.

Some retrospective studies have specifically analysed the use of corticosteroids and immunosuppressive agents for the treatment of organ-specific manifestations in patients with primary SS. These studies support the use of corticosteroids and cyclophosphamide in myelitis, azathioprine in interstitial lung disease and methotrexate in joint involvement (Figure 2).113 Evidence for the therapeutic management of other extraglandular SS features comes from isolated case reports or small case series.1,2,3 The most-frequently used immunosuppressive agent for the treatment of glomerulonephritis, vasculitis, multiple neuritis and central nervous system (CNS) involvement is cyclophosphamide.1,2,3 By contrast, some extraglandular features, such as interstitial nephritis or ataxic neuronopathy, seem to have a poor or no response to corticosteroids and immunosuppressive agents.113 Given the low level of evidence from studies in patients with SS, the choice of drugs for organ-by-organ management is, therefore, usually heavily influenced by therapeutic strategies accepted in clinically similar, but aetiopathogenically different, diseases such as systemic lupus erythematosus or systemic vasculitis.

Management of life-threatening situations

Severe, life-threatening disease manifestations have rarely been reported in patients with primary SS. In nine studies comprising 2,241 patients with primary SS, in which mortality rates and causes of death were detailed, only 17 deaths were attributable to SS-related systemic involvement, representing less than 8% of the 221 deaths reported.114 Vasculitis—overwhelmingly cryoglobulinemic and involving vital organs such as the kidneys, lungs and gastrointestinal tract—was the major life-threatening presentation of primary SS.114 Other severe complications of SS include CNS involvement, progressive ataxic neuronopathy, pulmonary arterial hypertension and severe cytopenia.115

No controlled studies have evaluated the therapeutic management of patients with SS-associated life-threatening conditions; at present, data is only available from small (<10 patients) retrospective studies and isolated case reports. However, this small body of evidence, together with expert opinion, suggests that methylprednisolone and cyclophosphamide pulses should be used in patients with severe systemic vasculitis or CNS involvement, combined with plasma exchange in the most severe cases.115 Rituximab is increasingly reported as a promising therapy, not only in patients with life-threatening SS but also in individuals with associated B-cell lymphoma (Figure 2).115,116

Future directions

Advances in our understanding of the molecular mechanisms underlying the aetiopathogenesis of primary SS could lead to the development of novel therapies. After the demonstration of the effectiveness of TNF inhibitors in rheumatoid arthritis, these agents were studied in patients with SS, but RCTs revealed a lack of efficacy. At present, B-cell-targeted biologic agents seem to be the most promising therapies in patients with SS, especially rituximab, which has been used in more than 100 reported cases. Other B-cell-targeting agents with therapeutic potential include those against CD22+ cells (epratuzumab) and, in particular, inhibitors of B lymphocyte stimulator (also known as TNF ligand superfamily member 13B or B-cell-activating factor)-signalling (belimumab). On the basis of data from preliminary studies,117,118,119 the suggestion of a role for B-cell-depleting therapies in the modification of the aetiopathogenesis of primary SS—a disease characterized by B-cell hyperactivity—seems reasonable. Ongoing trials in patients with SS are concentrated on the evaluation of rituximab and belimumab, but the efficacy of HCQ, anakinra and oromucosal surfactants is also being investigated. In the future, the therapeutic management of sicca features will probably be based on muscarinic receptor agents with increased selectivity and, possibly, on salivary gland electrostimulation.120 New immunosuppressive agents, especially B-cell-targeted therapies, will probably have key roles in the treatment of complicated cases of primary SS, especially those with extraglandular involvement.

Conclusions

During the past three decades, the use of substitute agents for sicca features and corticosteroids and immunosuppressive agents for systemic involvement has been the foundation of SS therapy. Unfortunately, a major problem in reviewing therapies and offering solid therapeutic recommendations in primary SS is the limited level of reported evidence (Box 1).3 The overall low level of evidence from therapeutic studies in primary SS highlights the requirement for substantially larger trials of the most promising therapies. Large RCTs, which are considered the gold standard in clinical research for the assessment of treatment efficacy and safety, are scarce in primary SS. This limitation might be explained by the heterogeneous clinical presentation (both sicca-related and systemic), the lack of consensus end points for evaluating outcomes and the often poor correlation between the intensity of self-reported symptoms and the results of objective tests. Consequently, therapeutic decisions are based on a combination of personal experience and reported scientific evidence, principally from uncontrolled studies. The complexity of the therapeutic approach in primary SS is increased by the fact that not all patients respond to first-line therapies, and even less scientific evidence is available to guide treatment interventions in this situation; thus, allocation of second-line drugs is frequently based on the opinion of the treating clinician. Objective, standardized evaluation of disease burden has been facilitated by two proposed international indexes.121,122 However, international efforts are also required to enrol and characterize large, multicentre cohorts of patients with SS, and to develop consensus end points for clinical trials.

Management of patients with SS requires a multidisciplinary approach; ophthalmologists, odontologists and gynaecologists should all be routinely involved in the follow-up monitoring of sicca manifestations. Likewise, extraglandular involvement requires organ-specific specialist care, and patients with lymphoproliferative neoplasia should be referred to a haematologist.

The emergence of new immunosuppressive and biologic agents has increased the therapeutic armamentarium available for the treatment of severe disease, but their use in primary SS is limited by the absence of specific regulatory approval. Nevertheless, considerable progress has been made in the targeting of specific aetiopathogenic pathways in primary SS. Ultimately, this approach could result in the development of new effective and highly selective therapies, without the adverse effects often associated with the less selective immunosuppressive agents that represent the current standard-of-care therapies.

Review criteria

We searched MEDLINE for English-language articles published between 1st January 1986 and 30th November 2011 that reported studies in adult humans using the MeSH term “Sjögren's syndrome” and the MeSH subheading “therapy”. We read the titles and abstracts (if available) looking for articles on currently available drugs for Sjögren's syndrome therapy (inclusion criterion). Duplicate publications, case reports, reviews, experimental studies and uncontrolled series with fewer than 10 patients were excluded. We also manually searched the reference list of relevant articles retrieved for further publications.