Inflammatory arthritis in systemic sclerosis: What to do?

Abstract

Musculoskeletal involvement, including arthritis and tendinopathy, is a common and important determinant of disability and impaired quality of life in systemic sclerosis. However, the treatment of arthritis in systemic sclerosis has not been studied as a primary outcome in randomized controlled trials, and arthritis-specific outcome measures for systemic sclerosis have not been sufficiently validated to date. Rheumatologists caring for patients with systemic sclerosis must address these complaints regularly despite the fact that the level of evidence for the treatment of systemic sclerosis–related inflammatory arthritis is limited. Consensus statements, based on treatments for related musculoskeletal aspects of rheumatoid arthritis, systemic lupus erythematosus, and other autoimmune diseases, support the use of methotrexate and hydroxychloroquine. Newer biologics, which have efficacy in the treatment of other autoimmune conditions, may show promise in the treatment of arthritis in systemic sclerosis. In this article, we review the current literature on the assessment and treatment of systemic sclerosis arthritis in order to address management considerations.

Introduction

Musculoskeletal (MSK) involvement, including arthritis and tendinopathy, is a common feature of systemic sclerosis (SSc). Joint symptoms have been described in 24%–97% of patients with SSc during the course of their illness and in 12%–66% of patients at the time of diagnosis.^1–3 Joint involvement may be the presenting feature of SSc or may predate the typical diagnostic features of the disease. It contributes to disability and impaired quality of life.^4–6 Unfortunately, the treatment of arthritis in SSc has not yet been studied as a primary outcome in randomized controlled trials (RCTs), and arthritis-specific outcome measures for SSc have not been well validated. ⁷ Although the best strategies for managing arthritis in SSc have not yet been borne out by arthritis-specific RCTs, rheumatologists caring for patients with SSc must address this complaint regularly. In this review, we describe the spectrum of arthritis and discuss management considerations.

Arthritis in SSc

MSK manifestations of SSc are diverse, and there is no formal classification system. ⁸ Patients may present with complaints of stiffness and/or pain and this may be explained by a combination of joint, tendon, skin, and/or muscle involvement. Patients with SSc report joint pain, stiffness, reduced dexterity, and reduced grip strength in up to 80% of cases. ⁹ MSK pain syndromes in SSc ¹⁰ include polyarthralgia/polyarthritis, tendonitis, bursitis, and overlapping rheumatoid arthritis (RA). Arthritis in SSc may appear similar to RA with anti-cyclic citrullinated peptide (CCP) positivity and RA-like erosions in some patients. In others, arthritis may be more similar to that of systemic lupus erythematosus (SLE) with non-erosive arthropathy. In some, arthritis mutilans is observed.

Arthralgia, synovitis, contracture, and tendon friction rubs (TFRs) may be present.^3,11 Synovitis can be seen in both early and late stage disease. ¹¹ Synovitis, joint contracture, and TFR are also associated with disease severity and progression in SSc^11–13 and are more frequent in patients with diffuse cutaneous systemic sclerosis (dcSSc). Pain may also occur secondarily, due to ischemic or traumatic digital ulcerations, calcinosis, or concomitant fibromyalgia. A sense of stiffness and development of contractures can occur from an inter-play of skin and MSK pathology. (Figure 1) Small joint contractures, an aspect of irreparable damage, occur in up to 30% of patients within 4 years of disease onset ¹⁴ and are associated with an increased mortality risk. ¹⁵ Pain in a single joint may relate to multiple factors, and the rheumatologist may need a multi-faceted approach to improve pain and function. (Figure 2)

Figure 1.

Contractures developing from primarily cutaneous and subcutaneous disease: (a) photograph of the hands of a male patient with early dcSSc with flexion deformities at the MCP joints. (Photo credit: John Indalecio, OTR/L, CHT), (b) longitudinal ultrasound image at the right third finger MCP demonstrates a normal appearance, with no synovial thickening or hyperemia, and (c) ultrasound measurement of the skin over the right third finger proximal phalanx reveals abnormal thickening (calipers).

Figure 2.

Calcinosis and contractures: (a) photograph of the hands of a female patient with long-standing lcSSc and CCP-positive RA overlap with inflammatory arthritis and calcinosis-demonstrating deformities and joint swelling in the hands and wrists and (b) frontal radiographs of the left and right hands in this patient demonstrate flexion-extension deformities in the fingers and multiple foci of bulky mineralization in the wrists and hands. Acro-osteolysis is also noted at both index fingers.

Arthritis in SSc plays an important role in the morbidity associated with this condition. Joint involvement contributes to disability and pain. These aspects are independent risk factors for depression in patients with SSc which occurs in 36%–65% of patients and is associated with a decrease in quality of life.^16,17

Assessing arthritis in SSc

As with other forms of arthritis, the rheumatologist uses a combination of physical examination, laboratory assessment, and radiographic studies to evaluate arthritis in SSc. Assessment of function can be performed by using Hand Mobility in SSc (HAMIS), Cochin Hand Function Scale (CHFS), and Scleroderma HAQ (sHAQ), although these are used more in clinical trials than in clinical practice. Tender and swollen joint counts (TJC and SJC, respectively) are used commonly in the assessment of RA patients. These assessments have been studied in SSc with the TJC providing high interobserver and intraobserver reliability, whereas the SJC provides only moderate interobserver and intraobserver reliability. However, neither of these assessments corresponded with the abnormalities seen on ultrasound (US). ¹⁸ The Disease Activity Score 28 using erythrocyte sedimentation rate (DAS28-ESR) and C-reactive protein (DAS28-CRP), the Simplified Disease Activity Index (SDAI) and the Clinical Disease Activity Index CDAI (all used in RA) have also been shown to be valid for SSc patients, with the DAS28-ESR showing the best performance regarding reliability and construct validity. ⁷ Joint involvement, specifically synovitis, is associated with elevated levels of acute phase reactants. ¹¹ The responsiveness of erythrocyte sedimentation rate (ESR) and C-reactive protein (CRP) to change with respect to arthritis symptoms has not been specifically recognized in SSc.

Auto-antibodies can be helpful in the evaluation of the SSc patient with arthritis. Anti-CCP antibodies are seen in 2.6%–8.3% of patients^19,20 with SSc and may identify patients with SSc-RA overlap who are more likely to develop erosive arthropathy. However, rheumatoid factor (RF) is identified in approximately 30%–50% of patients with SSc but is nonspecific. ²¹ Other antibodies that have been associated with diverse joint involvement include U1RNP, U3RNP, and RNA polymerase 3 antibodies. ²²

Imaging is also useful in assessing MSK involvement in SSc. Plain radiographs may demonstrate changes related to inflammatory, degenerative, and fibrotic processes ²³ such as periarticular osteopenia, erosions, joint-space narrowing, contractures, and calcinosis (Figure 3). US can detect SSc-related articular pathology including joint effusion, synovial thickening, and osseous erosion (Figure 4). US can also evaluate for extra-articular abnormalities such as tenosynovitis, tendon pulley thickening, soft tissue mineralization, and skin thickening. (Figure 5) Joint synovitis ± effusion and tenosynovitis are the most frequently reported findings in US series of SSc patients.^24,25 Flexor tendon pulley thickening as detected on US has been shown to correlate with hand mobility and disease duration. ²⁶ Elhai et al. ²⁷ showed that articular involvement in SSc may be underestimated by clinical examination and that sclerosing tenosynovitis and calcification are specific SSc features seen on US. Magnetic resonance imaging (MRI) is highly sensitive for inflammatory arthropathy due to SSc, including synovitis and osseous erosion ²⁸ (Figure 6). Magnetic resonance angiography confers an additional benefit of diagnosing vasculopathy, ²⁹ commonly in the hands and can detect arterial attenuation, beading, and occlusion. ³⁰ These imaging modalities can provide a window into understanding the anatomy and pathology in SSc arthropathy and further investigation will allow us to maximize their clinical utility.

Figure 3.

Calcinosis and arthropathy on radiographs: (a) frontal and (b) lateral radiographs of the right hand and wrist in a female patient with lcSSc and inflammatory arthritis demonstrate flexion deformities at the IP joints, as well as prominent mineralization at the CMC and radiocarpal joints and in the dorsal soft tissues at the wrist. The distal ulna is obscured by the calcinosis. Acro-osteolysis is noted at the thumb and index and middle fingers. There is also multifocal MCP and IP arthrosis. IP: interphalangeal; CMC: carpometacarpal; MCP: metacarpophalangeal.

Figure 4.

Ultrasound features of arthropathy in SSc: (a) ultrasound image of the second MCP joint in a 38-year-old woman demonstrating a normal dorsal joint capsule (thin arrows) without synovial thickening and (b) ultrasound image of a 45-year-old woman with SSc with marked synovial thickening at the second MCP dorsal capsule (thin arrows). MCP: metacarpophalangeal.

Figure 5.

Extra-articular musculoskeletal manifestations of SSc on ultrasound: (a) ultrasound image of dorsal skin measurement at the left second finger depicts abnormally thick skin (arrowheads, calipers) in a 48-year-old man with scleroderma. Normal skin thickness in a 42-year-old woman is depicted in (b). (c) Diffuse tendon sheath thickening (block arrows) consistent with tenosynovitis along a finger flexor tendon in a 34-year-old male patient. (d) Focal mineralization (block arrows) consistent with calcinosis along the flexor tendon at the DIP joint in a 38-year-old female patient. DIP: distal interphalangeal.

Figure 6.

Calcinosis-related arthropathy and tendinopathy on MRI: (a) sagittal and (b) axial proton density MR images of the wrist in a 54-year-old female patient with lcSSc and inflammatory arthritis (CCP and RF negative) demonstrate intra-articular calcinosis at the CMC joint (arrowheads). There is osseous destruction at the distal ulna. Prominent soft tissue calcinosis is within the dorsal extensor tendons (asterisks), and a superimposed tendon tear is present that likely contributes to the patient’s finger flexion deformities. The overlying dorsal skin is diffusely thickened, and there are multiple regions of focal skin breakdown (block arrows). IP: interphalangeal; CMC: carpometacarpal; MCP: metacarpophalangeal.

Approach to treatment in SSc-related inflammatory arthritis

To date, there have been no definitive RCTs assessing the treatment of arthritis in SSc. Currently, outcome measures are not standardized with respect to MSK disease except for HAQ-DI, the CHFS or the HAMIS. Points to consider with respect to trial design in SSc arthritis have been described. ³¹ It is reasonable to consider therapies used for arthritis in other rheumatological conditions for patients with SSc despite the limitations of the current evidence. Below, we review the literature for various therapeutic options for arthritis in SSc. Table 1 lists studies that specifically looked at MSK involvement as one of their secondary outcome measures.

Table 1.

Selected studies addressing arthritis in systemic sclerosis.

Treatment	Study type	Effects	Scores	Reference
Benoxaprofen	Open label trial	10 patients received 600 mg benoxaprofen daily for 6 months. Four out of 10 patients felt subjective improvement in pain and stiffness at 6 months. Overall, there were no statistically significant changes in joint motion, PFTs, skin thickness and elasticity.	“joint motion disorder” and patient subjective awareness of pain and stiffness	Halkier-Sørensen et al. 32
Leflunomide	Case series	Three female SSc patients with arthritic symptoms unresponsive to other treatments were treated with 20 mg/d of leflunomide and had either complete resolution (two patients) or significant improvement (one patient) of articular involvement after a few weeks of treatment. Ritchie index decreased from 6 to 2, 6 to 0, and 12 to 2 in the three patients. Likewise, VAS decreased from 80 to 10, 90 to 10, and 100 to 40 after treatment. Symptom resolution remained stable after 1 year of follow-up on leflunomide.	Ritchie index, Visual Analog Scale, ESR, and CRP	Sebastiani et al. 33
Cyclophosphamide	RCT	Analysis of the data from Scleroderma Lung Studies I and II found no significant difference in SJC and TJCs between cyclophosphamide (79 patients) and placebo group (79 patients) at baseline, 12, and 24 months. The median SJC and TJC score in both groups was 0 at baseline which could limit this analysis. In addition, there was a significant decrease in SJC over the 2 years of the study regardless of treatment (p = 0.002), and there was a trend toward decreased TJC (p = 0.07).	SJC and TJC	Au et al. 34
TNF inhibitors	Retrospective chart review	10 SSc patients with inflammatory arthritis received TNF-α inhibitor ⩾12 months. After 3 months, all patients had improvement in TJC and SJC and, significantly, by 12 months median SJC and TJC decreased from 10 to 0 (p < 0.01) and 15 to 3 (p = 0.02) respectively. Median pain score decreased from 6 to 3.5 (p = 0.10). 30% of patients developed malignancy during 30 month observation.	SJC and TJC and self reported pain score	Omair et al. 35
Abatacept	Prospective observational trial	11 patients with articular symptoms were treated with 10 mg/kg/month of abatacept for 11 months resulting in significant improvement in TJC and SJC, morning stiffness, DAS28, VAS scores. Median TJC declined from 6 to 1 (p = 0.002); median SJC declined from 3 to 0 (p = 0.008); median morning stiffness (min) declined from 60 to 15 (p = 0.004); median DAS declined from 4.5 to 2.3 (p = 0.001); median VAS declined from 43 to 20 (p = 0.014) .Six patients were good EULAR responders. Concomitant steroid use was discontinued in two patients and reduced in five patients.	SJC and TJC, morning stiffness, DAS28, and VAS articular pain score	Elhai et al. 36
TCZ	Prospective observational trial	15 patients with articular symptoms were treated with 8 mg/kg/month of TCZ for 5 months resulting in significant improvement in TJC and SJC, morning stiffness, DAS28, VAS scores. Median TJC declined from 9 to 1.5 (p = 0.001); median SJC declined from 4 to 0 (p < 0.001); median morning stiffness (min) declined from 60 to 25 (p = 0.008); median DAS declined from 5.2 to 2.8 (p < 0.001); median VAS declined from 80 to 25 (p = 0.031). 10 patients were good EULAR responders. Concomitant steroid use was discontinued in two patients and reduced in three patients.	SJC and TJC, morning stiffness, DAS28, and VAS articular pain score	Elhai et al. 36
TCZ	RCT	43 patients assigned to weekly 162 mg subcutaneous TCZ and 44 patients assigned to placebo were compared at 24 and 48 weeks. Treatment with TCZ showed better scores in the treatment group with respect to mRSS, clinican global VAS, patient global VAS, and HAQ-DI at 48 weeks that did not achieve statistical significance. 47% of patients in the TCZ group and 49% in the placebo group were described as having joint involvement. Among these patients, mean TJC changed from baseline by −4.3 (SD, 7.3) at 24 weeks and −5.1 (SD, 7.3) at 48 weeks with TCZ compared to −2.1 (SD, 6.3) at 24 weeks and −2.9 (SD, 7.1) at 48 weeks in the placebo group, but that was not considered significant.	TJC and VAS	Khanna et al. 37

EULAR: European League Against Rheumatism; RCT: randomized controlled trial; VAS: Visual Analog Scale; CRP: C-reactive protein; TNF: tumor necrosis factor: SJC: swollen joint count; TJC: tender joint count; TCZ: tocilizumab; SSc: systemic sclerosis; ESR: erythrocyte sedimentation rate; DAS28: Disease Activity Score 28; mRSS: Rodnan Skin Score.

Physical and occupational therapy

Treatment of MSK manifestations of SSc, including arthritis, with physical and occupational therapy is crucial. Connective tissue massage, joint manipulation, and institution of a home exercise program have been shown to lead to improvements in HAMIS, CHFS, and HAQ. ³⁸ It is essential for occupational therapy to start early before contractures become intractable. Due to the fact that SSc is a chronic illness, these rehabilitative programs need to extend over the long term, and different approaches to motivating continued compliance with an exercise program must be considered, ³⁹ as a short or limited course of occupational therapy will likely not suffice. Home exercise plans are important because the stretch and massage that is required must occur regularly and is not adequately carried out only at the time in the therapist’s office. When possible, caregivers can be involved in occupational therapies as well.

Nonsteroidal anti-inflammatory drugs

There are over 20 different nonsteroidal anti-inflammatory drugs (NSAIDs) available, and this class of medication is one of the most commonly used worldwide. NSAIDs act primarily by inhibiting cyclooxygenase, have important analgesic effects, and are indicated for mild to moderate pain from somatic origin. NSAIDs are used commonly to treat arthralgia and arthritis in SSc. One double-blind RCT studying the effects of aspirin (975 mg/day) plus dipyridamole (225 mg/day) versus placebo in SSc found no significant effects on joints or any other clinical outcome measures after 1 year of treatment. ⁴⁰ There were similarly no significant results in an SSc-specific open-label study of benoxaprofen, which is currently off the market. ³² In the latter study, 4 of 10 patients noted an improvement in pain and stiffness. Both of these studies were performed in the 1980s and did not use standard outcome measures that would be used today. Despite the lack of dedicated studies in SSc, it is reasonable to utilize NSAIDs to address these vital aspects of pain and stiffness. It is important to monitor for adverse events and be mindful of the important gastrointestinal, renal, cardiovascular, and hematologic side effects that can occur in patients with co-morbidities, such as SSc patients with multisystem involvement.

Corticosteroids

Corticosteroids are a mainstay in reducing MSK pain and inflammation in rheumatic diseases. In a retrospective case-control study, Steen and Medsger found a significant association between antecedent corticosteroid therapy, particularly at high doses and the development of scleroderma renal crisis (SRC). In this article, low-dose prednisone was defined as <15 mg and high-dose prednisone as ⩾15 mg/day. ⁴¹ This is especially important, as the patients most likely to receive corticosteroids are those with early, inflammatory disease, who are also those at greatest risk for SRC. An RCT predating this work compared dexamethasone 100 mg IV monthly pulses versus placebo and showed modest improvement in total skin score (TSS) and functional disability index in the treatment group versus worsening in controls. ⁴² This dosing regimen would not be used presently, given the findings of Steen and Medsger. Low-dose prednisone (⩽10 mg) is commonly used in SSc-related arthritis and is relatively safe in most instances. Based on the association of high-dose steroids and SRC, we avoid discourage higher steroid doses in the treatment of arthritis. Patients requiring corticosteroids at any dosage are counseled regarding the risk of SRC and are instructed in home blood pressure monitoring.

Corticosteroid injections

Corticosteroid injections may be helpful for localized symptomatic relief. In one small study of patients with painful scleroderma of the hand, hydrodissection of the carpal tunnel with 1% lidocaine followed by 80 mg triamcinolone injection found that pain scores were significantly improved by 67%. ⁴³ One case reported pain relief for 3 months after 20 mg methylprednisolone injection to the sternoclavicular joint in an lcSSc patient. ⁴⁴ At our center, we find corticosteroid injections to be helpful in the treatment of well-localized syndromes, for example, mono-arthritis or localized tenosynovitis, and they can provide temporary relief.

Hydroxychloroquine

Hydroxychloroquine (HCQ) and other anti-malarials are widely used in the treatment of rheumatic diseases. ⁴⁵ There have been over 90 RCTs and observational studies providing evidence for HCQ treatment in SLE. ⁴⁶ Placebo-controlled randomized trials have been performed in RA showing moderate efficacy and excellent tolerability.^47–51 SSc, like SLE, is within the spectrum of interferon-mediated diseases, ⁵² and there is good rationale for the use of HCQ in SSc. Consensus statements are supportive of the use of HCQ in SSc-related inflammatory arthritis. ⁵³ A small retrospective study using the EUSTAR database compared 10 SSc patients with joint involvement treated with HCQ for at least 6 months with 10 sex-, age-, and disease duration–matched SSc patients also with joint involvement but not treated with HCQ. Significant improvement was seen in global health (Likert-type scale) with a trend in improvement of synovitis on US. However, differences in the presence of tender or swollen joints were not observed. ⁵⁴ To our knowledge, prospective studies of HCQ in SSc have not been performed. HCQ is a relatively safe option for SSc patients with MSK complaints. A trial of 3–6 months is prudent to assess effect. A maximum dose of 5 mg/kg and ophthalmology screening according to risk factors is recommended as with other forms of arthritis. ⁵⁵

Methotrexate

Methotrexate (MTX) is a common initial disease-modifying anti-rheumatic drug (DMARD) for RA,^56,57 and it shows modest benefit in treating the cutaneous manifestations of SSc.^58,59 It is frequently chosen to ameliorate SSc arthritis although there is a lack of published evidence to support its use for this indication. One of the earliest double-blind RCTs to study MTX in SSc looked at TSS and found that patients on MTX 15 mg subcutaneously fared better than those on placebo at 24 weeks with respect to TSS and VAS general health, approaching statistical significance. By the end of the trial, of all the patients treated with MTX for at least 24 weeks, 68% were considered responders to treatment but arthritis-specific measurements were not evaluated. ⁶⁰ Another study that noted that 8 of 33 patients had joint complaints at baseline studied patients treated with oral MTX for 6 months in which the majority of patients failed to have improvement in modified Rodnan Skin Score (mRSS). Those few patients who did have improvement extended their oral MTX treatment to a year and at were found to have significant mRSS improvement. ⁶¹ This study included patients with relatively long disease duration (up to 15 years), and arthritic outcomes were not specifically measured. Another double-blind RCT of early dcSSc looked at MTX treatment over 12 months and found significant improvement in the physician global scores in the treated group but only a trend toward superiority in the mRSS and no significant difference in the patient global assessment. ⁵⁸ This study was underpowered, used a low dose of MTX, did not look at arthritis specifically, and had more people in the placebo group who were also taking prednisone which may have confounded results. A re-analysis of the above studies using Bayesian methods found that MTX had a high probability of out-performing placebo in each of the three primary outcome measurements: 90% on mRSS, 92% on University of California, Los Angeles (UCLA) skin score, and 98% on physician global assessment. ⁵⁹

Given the efficacy of MTX in RA and in SSc skin disease, it is an appropriate agent to try in patients with predominant MSK and cutaneous complaints. In a survey to determine consensus for management of SSc, MTX was the initial management choice favored by SSc experts. ⁵³ As with the treatment of patients with RA, if response is inadequate, therapy should be advanced or changed. In the SSc expert consensus paper, for SSc-related inflammatory arthritis, the treatment approach favored was: MTX (60%) followed by adding corticosteroids (37%) or HCQ (31%) with 62% in agreement. The decision for DMARD therapy in SSc-related arthritis will, of course, depend on other organ system manifestation, and MTX as the first-line therapy for arthritis may or may not be the best choice in a patient with other severe internal organ manifestations.

Other DMARDs

Azathioprine

Azathioprine (AZA) is used commonly in the management of RA and arthritis related to SLE. In SSc, AZA has been studied in the context of interstitial lung disease (ILD) and has not been particularly helpful in that context. ⁶² It has also been studied in remission maintenance in SSc following 1 year of IV pulse cyclophosphamide treatment. ⁶³ Arthritis-specific outcome measures were not reported in either study although in the context of its use for remission maintenance, stability was seen in the HAQ-DI. Although there is a lack of evidence supporting the use of AZA for the treatment of arthritis in SSc, it may be considered in this context, given its utility in RA and SLE-related arthritis as well as its cost and availability. It may be helpful in the treatment of myositis related to SSc as well but is not generally favored in the treatment of SSc-related ILD, unless other medications are not available.

Leflunomide

Leflunomide is a pyrimidine synthesis inhibitor used in the treatment of RA. In a letter, Sebastiani et al. ³³ reported a case series of three SSc patients with variable articular involvement (one erosive and two non-erosive) who had failed treatment with steroids, MTX, cyclosporine and D-penicillamine. Leflunomide was well tolerated and successfully treated arthritis in these patients. ³³ Although the literature supporting use of leflunomide in SSc is sparse, this is a reasonable consideration in patients who have failed, not tolerated, or are contraindicated to use other oral DMARDs.

Mycophenolate mofetil

Mycophenolate mofetil (MMF) is an inhibitor of lymphocyte proliferation used originally to prevent allograft rejection, and now used commonly as a steroid-sparing agent in a variety of rheumatic diseases. In SSc, there is evidence to support its use to treat skin and lung disease.^64–67 However, MSK outcomes have not been presented from prospective trials to our knowledge. Future studies assessing MMF efficacy in SSc should include MSK outcomes as secondary or tertiary measures.

Cyclophosphamide

Cyclophosphamide is considered in the management of SSc-ILD and has been shown to have beneficial effects on skin disease as well. Due to its significant side effects, it would not be considered for use in SSc solely for MSK manifestations, but if it were considered for lung manifestations, we asked whether the literature could guide us on whether it might secondarily help MSK disease. A number of studies combining cyclophosphamide and corticosteroids in treating SSc have focused on ameliorating symptoms of ILD and found encouraging results in skin scores and joint involvement although skin and joint involvement were not the primary outcome measures, the studies lacked a control group, and it is not clear whether improved scores were related to cyclophosphamide alone, steroid alone, or a combination of both.^62,68–71 In a secondary analysis using the data from the Scleroderma Lung Study, Au et al. ³⁴ analyzed the following MSK outcomes: joint swelling or tenderness noted at the bilateral wrists, elbows, knees, and metacarpophalangeal joints; presence of contractures; muscle tenderness; hand extension; and fist closure. There was a significant decrease in SJC over the 2 years of the study regardless of treatment (p = 0.002), and there was a trend toward decreased TJC (p = 0.07). There was no significant difference in the rate of decline in SJC or TJC between the CYC and placebo groups. ³⁴ However, there were very few patients in either treatment or placebo group that complained of swollen and tender joints at baseline thereby, limiting generalizability to populations with prominent joint complaints. At this point in time, it is not clear if cyclophosphamide would be of benefit in arthritis and based on its significant side effects, other modalities would be favored.

Biologics

TNF inhibitors

Tumor necrosis factor (TNF) alpha inhibitors are approved for the treatment of RA and psoriatic arthritis as well as other autoimmune diseases. These medications have been considered in the treatment of SSc-related arthritis, but response and safety in SSc patients is not clear. Some case reports of SSc patients describe improved joint symptoms and overall clinical stability when treated with infliximab.^72,73 Bosello et al. ⁷⁴ described TNF-alpha inhibition with infliximab followed by etanercept that led to a decrease in joint swelling and tenderness in a case series of four patients with SSc and erosive polyarthritis. A retrospective chart review described that 83% of patients who were treated with etanercept for active joint disease (edema, erythema, warmth, pain, and synovitis) had a positive response including decreased signs of inflammation or synovitis and/or complete resolution of joint symptoms. The mean mRSS improved but did not reach statistical significance, and most patients received concomitant treatment with NSAIDs, MTX, HCQ, and/or prednisone. ⁷⁵ Omair et al. ³⁵ reported retrospectively on 10 SSc patients treated with TNF inhibitors, noting improvement in SJC and TJC from 10 to 0 (p < 0.01) and 15 to 3 (p = 0.02), respectively. However, median pain score did not significantly decrease, and 3 of 10 patients in that group developed malignancy within 30 months. A 26-week prospective, open-label, pilot study of infliximab for patients with dcSSc did not show clear benefit in mRSS, HAQ-DI, functional status, or physician global assessment but was associated with clinical stability and a decrease in two markers of collagen synthesis. Around 44% of the treated patients prematurely discontinued the study due to infusion reactions considered to be related to the formation of anti-chimeric antibodies. ⁷⁶

Some case reports of the use of anti-TNF agents in SSc describe potential side effects related to immunological effects. A 74-year-old SSc patient with pulmonary and joint involvement treated with adalimumab initially had arthritic symptom improvement but ultimately died from respiratory failure secondary to an exacerbation of pulmonary fibrosis thought to be due to adalimumab treatment. ⁷⁷ A separate case report links etanercept treatment with development of macrophage activation syndrome. ⁷⁸ Another describes a patient diagnosed with anti-CCP-positive RA who was treated with prednisone and adalimumab then developing an overlap syndrome of RA, polymyositis, and scleroderma possibly relating to adalimumab. ⁷⁹ Infliximab-induced scleroderma in an RA patient has also been noted. ⁸⁰ No papers were found describing the use of certolizumab or golimumab in this group.

There has been concern with the use of anti-TNF agents in other autoimmune diseases such as RA, particularly in inducing lung fibrosis. There have been a number of case reports of anti-TNF-induced ILD in RA. ⁸¹ A retrospective cohort study using diagnostic codes and prescription records did not find an association of increased risk of ILD with anti-TNF therapy among 8417 RA patients. ⁸²

Given that the safety and efficacy of using anti-TNF in SSc is not clearly established, in 2011, a EUSTAR study group performed a 3-part Delphi exercise to obtain expert opinion on this question. ⁸³ Experts did not recommend the routine use of TNF antagonists in SSc although they agreed that arthritis was a potential indication for their use in SSc. Controlled clinical trials would be needed before general recommendations should be given.

Overall, anti-TNF treatment in SSc is not strictly contraindicated in an individual with severe and erosive arthritis as long as there is careful monitoring and counseling regarding potential risks. Other biologic agents, if available, may be preferred given the experience described below and a likely more favorable patient safety profile.

Rituximab and B-cell depleting therapies

Rituximab is a chimeric, monoclonal antibody targeted to CD20 that results in B-cell depletion. It is approved in the United States for use in combination with MTX for patients with moderately or severely active RA with incomplete response to anti-TNF therapy. Rituximab has been studied in SSc primarily in observational studies and small trials in the context of skin and lung disease. DcSSc patients in the EUSTAR cohort treated with rituximab had a greater decrease in mRSS and a smaller decline in forced vital capacity (FVC) versus matched controls when examined retrospectively. ⁸⁴ In a prospective open-label study, Lafyatis et al. ⁸⁵ observed no significant change in mRSS after 6 months of treatment in very early dcSSc. However, in a separate study, Daoussis et al. ⁸⁶ observed improved mRSS and FVC in later stage patients with ILD treated with rituximab, while such improvements were not observed in a control group not treated with rituximab in this open-label study compared to those treated with standard therapies in a prospective, open-label study without randomization. Keir et al. ⁸⁷ have described its use as a potential treatment for refractory progressive ILD cases. In each of the studies that showed significant improvement in mRSS and disease activity scores, all patients were treated with both rituximab and some other form of immunosuppression (such as methylprednisolone, MTX, MMF, or cyclophosphamide).^88–90 None were specifically focused on arthritis measures but one study did report significant improvement in one patient’s arthritis symptoms throughout the duration of her treatment. ⁸⁸ There were no unexpected adverse events in the studies reported, with rates of infection in SSc patients similar to those in other patients with rheumatic disease.

Belimumab is a recombinant human monoclonal antibody approved for the treatment of certain SLE patients. Our group reported a randomized, double-blind, placebo-controlled, pilot study with patients treated with MMF therapy as background therapy. ⁹¹ Although TJC and SJC were not assessed, and the patients were selected primarily for active skin disease, there was no difference noted between groups with respect to VAS for pain. There were no differences in number of adverse effects or serious adverse effects between patients treated with belimumab versus placebo in this study. The efficacy of B-cell depleting agents for SSc-associated arthritis is not clear at this point in time based on the literature, but safety in the reported literature to date is reassuring that SSc patients do not experience unexpected adverse events. This is an area of great interest for future studies.

Abatacept

Abatacept is a soluble fusion protein of cytotoxic T-lymphocyte-associated protein 4 (CTLA-4) and the Fc portion of immunoglobulin G1 (CTLA4-Ig) that blocks T-cell co-stimulation. It is used in RA and has been studied in other rheumatologic disorders as well. In one case series describing four dcSSc patients who had failed conventional therapies and had ongoing skin involvement, addition of abatacept to their current treatment regimen improved subjective symptoms, mRSS, and myositis in one patient ⁹² In a double-blind, placebo-controlled RCT, Chakravarty et al. ⁹³ described a trend toward improvement in mRSS in patients treated with abatacept. Significant differences were not seen in pain VAS or in ESR, but the patients were not selected for arthritis. ³⁶ There were no unusual safety signals in the study and no differences in safety between groups. Elhai et al. retrospectively assessed patients in the EUSTAR network with refractory polyarthritis or refractory myopathy treated with abatacept or tocilizumab (TCZ). A group of 11 patients with SSc-polyarthritis was treated with abatacept. These patients had statistically significant improvement in TJC, SJC, morning stiffness (minutes), pain VAS, and DAS28. ³⁶ The authors noted that treatment with abatacept was associated with a trend toward improvement of refractory myopathy, but statistical significance was not achieved. During treatment, steroids were stopped in two patients and the dose was reduced in five patients. ³⁶ At the present time the ASSET trial, which is a larger, placebo-controlled RCT, is ongoing, and we eagerly await the results. ⁹⁴

Tocilizumab

TCZ is a humanized anti-IL-6 receptor antibody that inhibits the effects of this cytokine. It is used for the treatment of RA and giant-cell arteritis. Elhai et al. reported retrospectively on 15 patients in the EUSTAR cohort with refractory SSc-polyarthritis treated with TCZ. These patients had significant improvement in TJC, SJC, morning stiffness (minutes), VAS pain, and DAS28. ³⁶ Khanna et al. ³⁷ reported a phase 2, double-blind, placebo-controlled RCT of subcutaneous TCZ in 87 patients with early dcSSc. In this trial, patients were selected to have worsening skin disease and an elevated acute phase reactant. Treatment with TCZ showed better scores in the treatment group with respect to mRSS, clinician global VAS, patient global VAS, and HAQ-DI at 48 weeks but which did not achieve statistical significance. More detailed information was presented with respect to joint involvement in this study than in most previous SSc studies. In the TCZ group and placebo group, 47% of 43 patients and 49% of 43 patients were described as having joint involvement, respectively. Among these patients, mean TJC changed from baseline by −4.3 (standard deviation (SD), 7.3) at 24 weeks and −5.1 (SD, 7.3) at 48 weeks with TCZ compared to −2.1 (SD, 6.3) at 24 weeks and −2.9 (SD, 7.1) at 48 weeks in the placebo group. Adverse events were similar between groups, but serious infections were more common in the TCZ group.

Other

Aside from biologics, NSAIDS and DMARDs, a few trials have studied other treatments with mixed results. In one study, intravenous immunoglobulin (IVIG) treatment of seven patients with joint involvement who were treated with IVIG found improvements in both mRSS and joint pain, tenderness, and hand function. ⁹⁵ Another RCT, utilizing hyperimmune caprine serum (AIMSPRO), confirmed tolerability and a trend toward improvement in mRSS and SF-36 scores and a significant improvement in neuropathic pain score.^96,97

Conclusion

MSK complaints are an important source of morbidity in SSc. At this time the evidence for the treatment of SSc-related inflammatory arthritis is limited. Consensus statements support the use of MTX and HCQ. Low-dose prednisone (⩽10 mg) can be used for symptomatic relief with counseling regarding the risk of scleroderma renal crisis at higher dose and vigilant blood pressure monitoring. Preliminary work with abatacept and TCZ show significant promise for these particular biologics for SSc-related MSK disease. Further study of both of these biologics is ongoing, and B-cell-depleting therapies have the potential to be helpful as well. Anti-TNF agents may be helpful in some patients, but there are some safety concerns in SSc patients which have been observed. It is clear that more work is needed on the validation of SSc-specific MSK outcome measures as well as the use of US and other radiologic modalities. Outcome measures like the TJC and the SJC can be included as exploratory outcome measures in SSc clinical trials, especially those investigating medications indicated for other forms of arthritis. RCTs specifically addressing MSK issues in SSc are needed.