Antisynthetase antibodies

Aminoacyl transfer RNA (tRNA) synthetases (ARSs) are a group of cytoplasmic enzymes, each of which catalyse the binding of a specific amino acid to the cognate tRNA during protein synthesis. Antibodies directed to these enzymes (anti-ARS) interact with conformational epitopes in conserved regions of the enzyme to inhibit enzyme activity . The anti-ARS are highly specific, can be detected preceding disease onset and are thus considered to be critical in disease pathogenesis .

Antibodies directed to eight different ARSs are currently recognised ( Table 1 ); it is plausible that the remaining twelve tRNA synthetases are also antigenic and that antibody specificities directed to them will be identified. The antisynthetases are the most common MSAs and are identified in 35–40% of patients with IIM .

The antisynthetases are associated with a unique constellation of clinical features including myositis, interstitial lung disease (ILD), inflammatory arthritis, fever, mechanics’ hands and Raynaud’s phenomenon . Antibodies to Jo-1 (the most common antisynthetase) are associated with greater risk of myositis than the less common non-Jo-1 antisynthetases, in which ILD is often predominant . Individual differences are apparent, however, for each of the antisynthetases. For instance, anti-PL7 is associated with milder muscle disease and lower CK levels than anti-Jo-1 antibody , and myositis is a predominant clinical feature of anti-Jo-1, anti-EJ and anti-PL7 .

Aggarwal et al. evaluated the disease manifestations and long-term outcomes of patients with antisynthetase syndrome (120 anti-Jo-1 and 80 non-Jo-1) . Patients with non-Jo-1 antisynthetase syndrome initially presented with symptoms of non-myositis-type CTD together with ILD, often had a delayed diagnosis and had increased pulmonary morbidity and worse survival than those with anti-Jo-1 antisynthetase syndrome . Another study revealed that 12/45 (27%) patients with antisynthetase syndrome presented with polyarthritis .

DM-specific antibodies (Mi-2, TIF1γ, MDA5, NXP-2 and SAE)

In addition to anti-Mi-2 antibodies, a number of other autoantibody specificities are found in DM , namely antibodies directed to transcriptional intermediary factor-1ϒ (TIF1γ), melanoma differentiation-associated gene-5 (MDA5), nuclear matrix protein-2 (NXP-2) and SUMO-1-activating enzyme heterodimer (SAE). These antibodies are associated with distinct clinical syndromes, including differential risk of cutaneous manifestations, ILD and malignancy. As such, DM is now considered a heterogeneous condition of characteristic syndromes associated with specific antibodies. It is likely that serological categorisation of DM for clinical and prognostic purposes will become routine as testing for antibodies becomes more widely available.

IMNM-specific antibodies

Among patients with necrotising myopathy (which may be associated with infections or malignancy), a subset has a clearly identified autoantibody association, hence the term IMNM or necrotising autoimmune myositis (NAM) . At present, two autoantibodies associated with IMNM have been identified, namely antibodies to signal recognition particle (SRP) and 3-hydroxy-3-methylglutaryl-CoA reductase (HMGCR).

SRP

SRP is a ribonucleoprotein particle involved in targeting secretory proteins to the rough endoplasmic reticulum, and antibodies to SRP are found in 4–6% of patients with IIM . Myopathy associated with anti-SRP tends to be rapidly progressive (in contrast to PM, which has an indolent presentation), accompanied by marked CK elevation and significant dysphagia . Anti-SRP antibodies were previously considered as markers for cardiac involvement , although subsequent studies have not confirmed this .

Previously reported to be associated with an aggressive form of PM, the dominant histological finding is now recognised as a necrotising myopathy . Variation in clinical presentation and muscle histopathology associated with anti-SRP antibody has been demonstrated in a large cohort of 100 Japanese patients with anti-SRP myositis .

HMGCR

IMNM has been found in association with an antibody directed to 100-kDa/200-kDa proteins , since identified as HMGCR , the target of statins. These antibodies are rarely detected in patients administered statins with self-limited musculoskeletal symptoms . However, these antibodies were found in 45/750 (6%) patients with suspected IIM at the Johns Hopkins Center and in 19/207 (9%) South Australian patients with histologically confirmed IIM .

In the South Australian cohort, no preferential occurrence of anti-HMGCR with IMNM was found. Rather, the antibody was equally distributed among disease subgroups (including DM, PM and IBM), suggesting that anti-HMGCR antibody may be an MSA triggered by statin exposure and unrelated to the myositis type . Certainly, an inflammatory (rather than purely necrotising) component to anti-HMGCR myopathy is supported by the finding of scattered CD4+ and CD8+ T lymphocytic cells together with plasmacytoid dendritic cells in muscle specimens .

Development of anti-HMGCR is associated with DRB1*11 and is strongly associated with statin exposure (OR = 39, p = 0.0001) . Among patients with myositis, statin-exposed HLA-DR11 carriers have the highest risk of developing anti-HMGCR antibodies (90% positive predictive value), suggesting a two-hit phenomenon . Recommendations for genetic screening prior to statin therapy in the general population are premature at present, and the risk of developing IIM among anti-HMGCR-positive individuals is unclear.

Testing for anti-HMGCR antibodies by ELISA has high sensitivity (94%) and specificity (99%) using immunoprecipitation as the gold standard . The levels of anti-HMGCR antibodies have been shown to correlate with CK levels and muscle weakness, and the antibodies persist despite clinical improvement following immunosuppressive therapy . Testing for anti-HMGCR antibodies is considered a useful diagnostic tool in patients suspected of statin-mediated IMNM .

Antisynthetase antibodies

Aminoacyl transfer RNA (tRNA) synthetases (ARSs) are a group of cytoplasmic enzymes, each of which catalyse the binding of a specific amino acid to the cognate tRNA during protein synthesis. Antibodies directed to these enzymes (anti-ARS) interact with conformational epitopes in conserved regions of the enzyme to inhibit enzyme activity . The anti-ARS are highly specific, can be detected preceding disease onset and are thus considered to be critical in disease pathogenesis .

Antibodies directed to eight different ARSs are currently recognised ( Table 1 ); it is plausible that the remaining twelve tRNA synthetases are also antigenic and that antibody specificities directed to them will be identified. The antisynthetases are the most common MSAs and are identified in 35–40% of patients with IIM .

The antisynthetases are associated with a unique constellation of clinical features including myositis, interstitial lung disease (ILD), inflammatory arthritis, fever, mechanics’ hands and Raynaud’s phenomenon . Antibodies to Jo-1 (the most common antisynthetase) are associated with greater risk of myositis than the less common non-Jo-1 antisynthetases, in which ILD is often predominant . Individual differences are apparent, however, for each of the antisynthetases. For instance, anti-PL7 is associated with milder muscle disease and lower CK levels than anti-Jo-1 antibody , and myositis is a predominant clinical feature of anti-Jo-1, anti-EJ and anti-PL7 .

Aggarwal et al. evaluated the disease manifestations and long-term outcomes of patients with antisynthetase syndrome (120 anti-Jo-1 and 80 non-Jo-1) . Patients with non-Jo-1 antisynthetase syndrome initially presented with symptoms of non-myositis-type CTD together with ILD, often had a delayed diagnosis and had increased pulmonary morbidity and worse survival than those with anti-Jo-1 antisynthetase syndrome . Another study revealed that 12/45 (27%) patients with antisynthetase syndrome presented with polyarthritis .

DM-specific antibodies (Mi-2, TIF1γ, MDA5, NXP-2 and SAE)

In addition to anti-Mi-2 antibodies, a number of other autoantibody specificities are found in DM , namely antibodies directed to transcriptional intermediary factor-1ϒ (TIF1γ), melanoma differentiation-associated gene-5 (MDA5), nuclear matrix protein-2 (NXP-2) and SUMO-1-activating enzyme heterodimer (SAE). These antibodies are associated with distinct clinical syndromes, including differential risk of cutaneous manifestations, ILD and malignancy. As such, DM is now considered a heterogeneous condition of characteristic syndromes associated with specific antibodies. It is likely that serological categorisation of DM for clinical and prognostic purposes will become routine as testing for antibodies becomes more widely available.

IMNM-specific antibodies

Among patients with necrotising myopathy (which may be associated with infections or malignancy), a subset has a clearly identified autoantibody association, hence the term IMNM or necrotising autoimmune myositis (NAM) . At present, two autoantibodies associated with IMNM have been identified, namely antibodies to signal recognition particle (SRP) and 3-hydroxy-3-methylglutaryl-CoA reductase (HMGCR).

SRP

SRP is a ribonucleoprotein particle involved in targeting secretory proteins to the rough endoplasmic reticulum, and antibodies to SRP are found in 4–6% of patients with IIM . Myopathy associated with anti-SRP tends to be rapidly progressive (in contrast to PM, which has an indolent presentation), accompanied by marked CK elevation and significant dysphagia . Anti-SRP antibodies were previously considered as markers for cardiac involvement , although subsequent studies have not confirmed this .

Previously reported to be associated with an aggressive form of PM, the dominant histological finding is now recognised as a necrotising myopathy . Variation in clinical presentation and muscle histopathology associated with anti-SRP antibody has been demonstrated in a large cohort of 100 Japanese patients with anti-SRP myositis .

HMGCR

IMNM has been found in association with an antibody directed to 100-kDa/200-kDa proteins , since identified as HMGCR , the target of statins. These antibodies are rarely detected in patients administered statins with self-limited musculoskeletal symptoms . However, these antibodies were found in 45/750 (6%) patients with suspected IIM at the Johns Hopkins Center and in 19/207 (9%) South Australian patients with histologically confirmed IIM .

In the South Australian cohort, no preferential occurrence of anti-HMGCR with IMNM was found. Rather, the antibody was equally distributed among disease subgroups (including DM, PM and IBM), suggesting that anti-HMGCR antibody may be an MSA triggered by statin exposure and unrelated to the myositis type . Certainly, an inflammatory (rather than purely necrotising) component to anti-HMGCR myopathy is supported by the finding of scattered CD4+ and CD8+ T lymphocytic cells together with plasmacytoid dendritic cells in muscle specimens .

Development of anti-HMGCR is associated with DRB1*11 and is strongly associated with statin exposure (OR = 39, p = 0.0001) . Among patients with myositis, statin-exposed HLA-DR11 carriers have the highest risk of developing anti-HMGCR antibodies (90% positive predictive value), suggesting a two-hit phenomenon . Recommendations for genetic screening prior to statin therapy in the general population are premature at present, and the risk of developing IIM among anti-HMGCR-positive individuals is unclear.

Testing for anti-HMGCR antibodies by ELISA has high sensitivity (94%) and specificity (99%) using immunoprecipitation as the gold standard . The levels of anti-HMGCR antibodies have been shown to correlate with CK levels and muscle weakness, and the antibodies persist despite clinical improvement following immunosuppressive therapy . Testing for anti-HMGCR antibodies is considered a useful diagnostic tool in patients suspected of statin-mediated IMNM .

cN1A antibody

Antibodies to cytoplasmic 5′-nucleotidase 1A (cN1A), a 43-kDa muscle autoantigen, have recently been identified in patients with IBM, supporting a role for B-cell-mediated humoral immunity in its pathogenesis . Anti-cN1A antibodies have high specificity for IBM, but they may also be found in patients with other autoimmune diseases, for example, systemic lupus erythematosus (SLE) or Sjögren’s syndrome . In a South Australian cohort of patients with histologically proven IBM, 24/69 (34.8%) patients exhibited antibodies to cN1A, although sensitivity has been reported to be as high as 70% .

Novel autoantibodies in IIM

Concurrence of MSA/MAA

Koenig et al. investigated the relationships and outcomes of MSA/MAA in 100 patients with DM or PM classified by the Bohan and Peter criteria. A number of antibodies were found to occur in combination. Antibodies to SRP, Mi-2 and antisynthetases were mutually exclusive. Notably, antibodies to Ku, PmScl, U1RNP and antisynthetases were found in association with another antibody in a large proportion of patients and with two other antibodies in one-third of sera . Antibodies to Ro52 most frequently occurred in combination with antisynthetase antibodies (14/31, 45%), especially Jo-1 (11/31, 35%) , although antisynthetases themselves are largely mutually exclusive .

The association of anti-Ro52 with anti-Jo-1 has long been recognised . More recently, anti-Ro52 antibodies were identified in 57% of anti-Jo-11 sera but also in 67% of sera with anti-PL7 antibodies . The concurrence of antibody specificities is considered to reflect a heterogeneous B-cell response in IIM .

MHC I and II

MHC class I antigens are not expressed by normal muscle fibres but are up-regulated in IIM in response to interferons and other cytokines . MHC class I up-regulation on muscle fibres may make them targets for CD8+ cytotoxic T cells and contribute to muscle fibre necrosis. The antigen target in muscle fibres and mechanisms for muscle fibre destruction have not been identified. In addition to enabling adaptive immune responses, MHC I molecules may themselves be pathogenic as MHC I transgenic mice develop weakness before lymphocytic infiltration has occurred , supporting the role of MHC I in non-immune pathways. Expression of MHC I is an early feature in IIM, preceding lymphocytic infiltrates , and persisting late into the disease, without being attenuated by immunosuppressive therapy . These attributes make this an attractive immunohistochemical marker for the diagnosis of IIM. The role of MHC class II is less well defined. This marker is absent in normal muscle, but it is present on antigen-presenting cells such as myofibres .

In a review of 120 muscle biopsies from patients with IIM (61 PM, 14 DM and 45 IBM), Das et al. showed that the sensitivity for MHC I and II in IIM was 100% and 93%, respectively . The proportion of patients with >25% of MHC II-positive myofibres was higher in IBM than DM or PM. All 45 biopsies from IBM were immunopositive for MHC I and II. Interestingly, in 30/45 biopsies, 100% of fibres expressed these markers . Rodriguez Cruz et al. confirmed the high sensitivity (and low specificity) for MHC I in diagnosis of IIM, and reported a much higher specificity of MHC II, especially in IBM. Hence, they proposed that combined immunostaining for MHC I and II increases specificity for the diagnosis of IIM, and that positive immunostaining for MHC II in biopsies with inflammatory features supports a diagnosis of an immune-mediated myopathy . The proportion of internal fibres positive for MHC I is useful for distinguishing IIM from non-inflammatory myopathies; a threshold of 50% is optimal, with 100% positive predictive value and 94% negative predictive value . Importantly, the validity of quantitative assessment of MHC I for diagnosis of IIM has been confirmed according to Standards for Reporting of Diagnostic Accuracy recommendations. The marker showed almost perfect inter-observer agreement and was readily reproducible .

The distribution of MHC I expression may also serve to distinguish subsets of disease: in DM, expression of MHC I may typically show perifascicular accentuation . This can also be seen in patients without cutaneous manifestations of DM, and in patients with antisynthetase antibodies.

Membrane attack complex

The role of complement activation in DM with deposition of MAC in endomysial vessels increases the possibility of using MAC as a diagnostic tool to differentiate DM from other subsets of disease. Immunostaining for MAC showed high sensitivity (80.9%) and specificity (85%) for a diagnosis of DM when comparing 21 cases of DM with 42 cases of PM/IBM, and an even higher specificity (98.4%) for distinguishing IIM from non-inflammatory myopathies . A retrospective analysis of 33 cases of IIM and 59 biopsies with non-inflammatory myopathies did not confirm the ability of MAC positivity to distinguish subsets of disease. However, a high negative predictive value (95%) was noted for a diagnosis of IIM if both MAC and MHC I were negative . Similarly, another study showed that MAC deposition is not exclusive to DM, and the complex was detected in necrotic fibres and vessels from patients with DM and PM .

MAC staining in non-necrotic fibres has been used to distinguish IIM from dysferlinopathies (in which inflammatory infiltrates may lead to confusion with IIM) . Brunn et al. found the cellular infiltrates in dysferlinopathies to be mainly CD4-positive, CD25-negative T cells and macrophages, and that MAC was detected on intact and necrotic myofibres . Dysferlinopathies have been reported to have a more dominant infiltrate of macrophages than T cells, with absence of T-cell-mediated cytotoxicity (in contrast to PM), but together with positive immunostaining for MHC I .

Pathology of IBM

Autophagy, the process by which intracellular proteins are cleared, is impaired in IBM. This process, together with misfolding of proteins, results in protein aggregates and rimmed vacuoles . Overexpression of a number of markers of autophagy including LC3a, LC3b and Beclin 1 has been noted in IBM and is considered critical to vacuolar degeneration . P62 is a protein that transports ubiquitinated proteins for degradation in either the proteasome or lysosome , and increased expression of p62 protein and mRNA has been found in IBM . Moreover, immunoreactivity to p62 was found to be a specific marker for IBM and therefore useful in distinguishing this condition from PM . The autophagic markers p62 and LC3 were confirmed to be sensitive markers for IBM, and the protein aggregation marker TDP-43 was shown to be highly specific for IBM . This study by Hiniker et al. suggested that quantitative assessment of the proportion of fibres positive for LC3 and TDP-43 helps diagnose IBM; <14% LC3-positive fibres makes a diagnosis of IBM unlikely, whereas a finding of >7% of fibres positive for TDP-43 supports a diagnosis of IBM .

Antibodies and histopathology

The question as to how an autoantibody response to myositis autoantigens, known to be ubiquitously expressed , leads to localised disease remains unanswered. Target tissues may regulate the response , and to better understand the effects of myositis-specific autoantibodies on skeletal muscle, concerted efforts to determine whether different antibodies are associated with distinctive histopathology are under way.

Aouizerate J et al. compared muscle MHC I and II expression in DM patients with and without antisynthetase syndrome. Although MHC I expression was ubiquitously observed in ASS and DM without antisynthetase antibodies, MHCII expression was found more frequently in ASS (27/33 patients) than in DM patients without antibodies (4/17; p < 0.001) . Furthermore, in patients with antisynthetase myositis, a distinctive perifascicular expression of MHCII was observed.

An investigation into the autoantibody–histopathological correlations among patients with DM (probable or definite by the Bohan and Peter criteria) at Johns Hopkins Centre showed, for the first time, that individual myositis (DM)-specific antibodies are associated with unique combinations of pathological features. Interestingly, antibodies to NXP-2 and TIF1γ showed similar histological features of prominent perifascicular atrophy and perivascular inflammation, with a paucity of inflammatory infiltrates. The distinguishing pathological feature between these antibodies was the frequency of mitochondrial dysfunction, seen commonly in association with anti-TIF1γ. Antibodies to Mi-2 were associated with the highest prevalence of perifascicular atrophy and perivascular inflammation and showed prominent inflammation. Consistent with the association of anti-MDA5 antibodies with CADM , the muscle histological results in these patients showed a paucity of primary and perivascular inflammation and perifascicular atrophy. The concurrence of anti-Ro52 with anti-Jo-1 antibodies was associated with a higher prevalence of perivascular inflammation, suggesting that antibody specificities may act synergistically or additively.

Among patients with anti-Jo-1 antibodies, there were no differences between histopathological features when comparing patients with a Bohan and Peter diagnosis of DM versus PM. Further, Pinal-Fernandez et al. suggested these patients not be classified as DM or PM, but rather as anti-Jo-1 syndrome . Mescam-Mancini et al. also investigated the histopathology associated with anti-Jo-1 antibodies and found perifascicular necrosis (rather than perifascicular atrophy), perimysial inflammation and fragmentation and diffusely positive MHC I expression with perifascicular accentuation, together with sarcolemmal complement deposition .

Collective study of biospecimens from large national and international repositories in the future is expected to further our understanding of autoantibody–muscle pathology correlations, which may enable the development of targeted therapies.

CTD/IIM

The coexistence of IIM and other CTDs is widely recognised. There may be a true overlap of two disease entities where both classification criteria are met, or myositis complicating another CTD, as is seen in mixed CTD, SSc, Sjögren’s syndrome (SS) and vasculitis. This remains an area of interest, with implications for treatment approaches.

In a well-characterised cohort of 648 patients in the South Australian myositis registry, 118 (11.1%) patients with IIM had a concurrent CTD identified. Maundrell and colleagues identified PM to be the predominant histological subtype on muscle biopsy where another CTD was present (personal communication). In a cohort of patients with SSc from the Johns Hopkins Scleroderma Center database, 42 patients underwent muscle biopsy for muscle weakness . Interestingly, in this cohort, PM was uncommon and non-specific myositis and IMNM were the predominant histological subtypes. Importantly, a significant proportion also showed evidence of acute motor denervation, highlighting the role of other factors in muscle weakness in these patients. This was reiterated in a small cohort of 15 patients with SLE who underwent muscle biopsy for weakness, fatigue or myalgia . Seven demonstrated evidence of non-specific myositis, and 13 of the 15 showed evidence of type 2 atrophy, seen in glucocorticoid use and disuse.

Disease activity measures

With the rapid development of definitions and understanding of the pathophysiology of IIM, there has been growing understanding of clinically meaningful outcomes relevant to IIM. Robust clinical trials require reliable, validated outcome measures to adequately assess the efficacy of therapy. As outcome measures and definitions of improvement (DOIs) have changed to reflect the evolving understanding of disease, this has led to difficulty interpreting the evidence base for therapies in IIM.

As IIM is a complex, multisystem condition, there is no single gold-standard measure for assessing disease activity. Multiple disease activity measures have been developed and used both in practice and in clinical trials. A single tool cannot cover all facets of relevant disease outcome; thus, combinations have been used to reflect disease activity. In recent years, this has been most clearly defined by the International Myositis Assessment and Clinical Studies (IMACS) group .

One of the challenges in IIM is to distinguish disease activity (considered amenable to treatment and potentially reversible) from damage (considered largely irreversible, and often the result of cumulative disease activity or therapy). To this end, IMACS has established separate tools to assess disease activity and cumulative damage.

Assessment of disease activity

Over several years, the IMACS group comprising both adult and paediatric experts in myositis have agreed on a core set of disease activity measures, after reviewing the reliability, validity and feasibility.

The IMACS core set measures include

• –
  

  physician’s global disease activity assessment by visual analogue and five-point Likert scale

  
  
• –
  

  patient’s global disease activity assessment on visual analogue scale

  
  
• –
  

  manual muscle testing (MMT)

  
  
• –
  

  physical function measured by Health Assessment Questionnaire (HAQ)

  
  
• –
  

  serum levels of muscle enzymes

  
  
• –
  

  Myositis Disease Activity Assessment Tool (MDAAT) capturing extra-muscular disease activity

These measures have good content validity and have been partially validated on construct and criterion validity .

Preliminary DOIs were proposed in 2004 . The DOI that ranked highest was improvement by ≥20% in any three of the six core set measures with no more than two worsening by ≥25% (which cannot include MMT). This dichotomous outcome measure does not allow differential weighting for individual components. As such, these measures are currently being re-evaluated, and efforts to develop new consensus criteria for improvement and to allow for weighting of the six core set measures are ongoing.

Assessment of damage

IMACS has also developed tools to evaluate cumulative damage. Global damage can be scored by physicians and patients using visual analogue scales. The myositis damage index (MDI) captures the presence or absence of damage in nine domains (muscle, skeletal, cutaneous, gastrointestinal, pulmonary, cardiovascular, peripheral vascular, endocrine and ocular), in addition to infections, malignancy and death. Each domain is further evaluated by visual analogue scales for severity of damage.

While an indication of active disease on the MDAAT may prompt escalation of therapy, the use of the MDI in clinical practice allows for longitudinal assessment of the effects of disease and therapy not only on skeletal muscle but also on other organ systems. Such a tool in routine clinical practice documents cumulative muscle and extra-muscular damage, and can potentially be used to compare patient cohorts treated with different therapies. The MDI has been validated for use in IIM and is sensitive to change. Damage is defined as ‘persistent or permanent change in anatomy, physiology, and function that develops from previously active disease, complications of therapy, or other events’ despite treatment, after 6 months .