Idiopathic Inflammatory Myopathies-Associated Interstitial Lung Disease in Adults
Article information
Abstract
Idiopathic inflammatory myopathies (IIM) are a heterogeneous group of autoimmune diseases characterized by muscle involvement and various extramuscular manifestations. Interstitial lung disease (ILD) is one of the most common extramuscular manifestations of IIM and is associated with significant mortality and morbidity. The clinical phenotypes, treatment responses, and prognosis of IIM-ILD are significantly related to myositis-specific antibody (MSA) profiles, with some racial differences. The features associated with MSA in IIM-ILD could also be relevant to cases of ILD where MSA is present but does not meet the criteria for IIM. The anti-melanoma differentiation-associated gene 5 antibody is highly associated with rapidly progressive ILD (RP-ILD), especially in Asian populations, and with characteristic cutaneous manifestations, such as skin ulcers. Radiologically, ground-glass opacities, consolidations, and nonsegmental linear opacities were more predominant than reticular opacities and honeycombing. While the mortality rate is still around 30%, the prognosis can be improved with early intensive therapy with corticosteroids and multiple immunosuppressants. In contrast, anti-aminoacyl-tRNA synthetase (ARS) antibodies are associated with chronic ILD, although RP-ILD is also common. Patients with anti-ARS antibodies often show lung-predominant presentations, with subtle muscle and skin involvement. Radiologically, reticular opacities, with or without consolidation, are predominant and may progress to honeycombing over time. Combination therapy with corticosteroids and a single immunosuppressant is recommended to prevent relapses, which often lead to a decline in lung function and fatal long-term outcomes. Significant advances in immunology and genetics holds promise for fostering more personalized approaches to managing IIM-ILD.
Introduction
Idiopathic inflammatory myopathies (IIM) are a heterogeneous group of autoimmune diseases characterized by muscle involvement and various extramuscular manifestations [1,2]. Interstitial lung disease (ILD) is one of the most common extramuscular manifestations, with various phenotypes, and contributes to significant mortality and morbidity [3]. Recent advances have uncovered several myositis-specific antibodies (MSA), revealing a significant correlation between the MSA profile and ILD phenotype. In addition, ILD with MSA that does not meet the criteria for IIM often has clinical characteristics similar to those of IIM-ILD and may be included in IIM-associated ILD (IIM-ILD) [4,5]. Given the increasing availability of comprehensive MSA detection assays, it is beneficial to understand IIM-ILD according to the MSA profile when determining appropriate management [3,6,7]. This narrative review discusses the current diagnostic approach and management strategies for IIM-ILD in adults, including ILD with MSA without evident IIM, emphasizing on the distinct clinical phenotype per MSA profile from a pulmonologist’s perspective. Finally, we summarize the current knowledge regarding the pathogenesis of IIM and IIM-ILD, which will improve our understanding of this complex disease.
Current Classifications of IIM
IIM was initially subclassified as polymyositis (PM) or dermatomyositis (DM) based on skin manifestations, muscle weakness, the elevation of serum muscle enzymes, myopathy on electromyography, and muscle biopsy findings, according to the Bohan and Peter criteria [8]. Subsequently, some DM patients with typical skin manifestations with minimal or no muscle involvement have been termed clinically amyopathic dermatomyositis (CADM) [9]. In contrast, some patients diagnosed with PM have been subclassified as inclusion body myositis or immune-mediated necrotizing myopathy (IMNM) based on characteristic clinical features and pathological findings [10,11]. In addition, IIM has been reported to occur in association with other connective tissue diseases (CTDs) and has been termed overlap myositis (OM) [1]. Concurrent with developing these subclassifications, several autoantibodies have been identified in patients with IIM. These autoantibodies have traditionally been divided into two subsets: MSAs, which are specific for IIM, and myositis-associated antibodies (MAAs), which are also found in other CTDs [12,13]. As MSA is associated with the distinct clinical features of IIM and is generally mutually exclusive, several authors have proposed subclassifications and management strategies for IIM based on the MSA profile, along with classical classifications [14]. On the other hand, MAAs are also associated with clinical features, some of which present with OM, although beyond the scope of this review [1,13].
Clinical Significance of MSA
1. Association between clinical features of IIM and MSA profile
MSA is detected in 45%–85% of adult IIM patients, most commonly anti-aminoacyl-tRNA synthetase (ARS) antibody (20%–40%), followed by anti-melanoma differentiation-associated gene 5 (MDA5) antibody (8%–20%), and others [12]. As shown in Table 1, the MSA profile correlates significantly with the phenotype of IIM: anti-MDA5, Mi2, transcriptional intermediary factor 1-γ (TIF1-γ), nuclear matrix protein-2 (NXP2), and small ubiquitin-like modifier 1 activating enzyme (SAE) antibodies are associated with the DM phenotype [1,12,13]. In particular, anti-MDA5 antibody are highly associated with the CADM phenotype, while anti-TIF1-γ and NXP2 antibodies are associated with malignancy [1,12,13]. Anti-ARS antibodies, including histidyl (Jo-1), threonyl (PL-7), alanyl (PL-12), glycyl (EJ), isoleucyl (OJ), asparaginyl (KS), phenylalanyl (Zo), tyrosyl (Ha), valyl (VRS), and cysteinyl (CRS) antibodies, are associated with an anti-synthetase syndrome (ASyS) characterized by ILD, myositis, inflammatory polyarthritis, fever, Raynaud’s phenomenon, and mechanic’s hands [1,12,13]. Anti-signal recognition particle (SRP) and 3-hydroxy-3-methylglutaryl-coenzyme A reductase (HMGCR) antibodies are associated with IMNM, which is characterized by predominant muscle fiber necrosis with little or no lymphocytic infiltrate [1,11-13]. Additionally, each antibody was correlated with distinct features of extramuscular manifestations, including ILD [1,12,13,15]. It is worth noting that some relationships between the MSA profiles and clinical features were significantly affected by racial differences, as discussed in the subsequent sections.
2. MSA detection assays
Despite the clinical significance of MSA in IIM, there is no standardized universal MSA detection assay [13]. While immunoprecipitation (IP) assays using radioisotope-labeled cell lysates have traditionally served as the gold standard for MSA detection, the complexity of this procedure has significantly restricted the number of facilities capable of performing it [7]. Several commercially available comprehensive autoantibody detection assays have been developed to overcome these disadvantages. However, no assay has been validated for its sensitivity and specificity. For example, line blotting, which can simultaneously measure multiple autoantibodies, is extensively used in clinical practice. However, its sensitivity and specificity for detecting MSA vary widely depending on the type of MSA, which often leads to false-positive or false-negative results [16]. Therefore, clinicians should consider the accuracy of these tests in various ways. Anti-nucleolar antibody (ANA) indirect immunofluorescence (IIF), which is widely used to screen for autoantibodies in the CTD, often helps to increase the reliability of MSA results when concordance between the ANA IIF staining pattern and detected MSA is observed [17]. As a caution, when analyzing ANA IIF, it is essential to examine not only the nuclear staining pattern but also the cytoplasmic staining pattern. This is because certain MSA, particularly those associated with ILD, such as anti-MDA5 and ARS antibodies, are anti-cytoplasmic antibodies [1,17]. Regarding monospecific immunoassays, commercial enzyme-linked immunosorbent assays are available for anti-TIF1-γ, Mi2, MDA5, and ARS-antibodies (only measuring Jo-1, PL-7, PL-12, EJ, and KS) with almost 100% sensitivity and specificity compared with their gold standard assay, IP [18,19].
Spectrum of IIM-ILD
IIM-ILD typically denotes ILD occurring in patients diagnosed with established IIM, but it may also include ILD with MSA that doesn’t meet the criteria for IIM [3]. IIM is usually diagnosed based on the muscle and skin manifestations, according to the Bohan and Peter Sontheimer criteria or, more recently, the European Alliance of Associations for Rheumatology/American College of Rheumatology (ACR) 2017 classifications [2,8,9]. However, due to the diverse range of manifestations associated with IIM, there are currently no classification criteria that exhibit perfect sensitivity and specificity [2,20]. Thus, patients with features of IIM may not be diagnosed with IIM according to the existing classification criteria. These instances are notably pertinent for patients with MSA exhibiting lung-predominant presentations with subtle muscular or cutaneous manifestations. They are often classified under idiopathic interstitial pneumonias (IIPs) or interstitial pneumonia with autoimmune features (IPAF) as a research category [21-27]. However, accumulating evidence indicates that ILD patients with MSA who do not meet the criteria for IIM exhibit similar clinical features and outcomes to those with established IIM [4,5]. In addition, among patients meeting the IPAF criteria, those with MSA showed different characteristics and prognoses compared to those with other autoantibodies [4]. Although not yet to be formally accepted, these results suggest the rationale for including ILD patients with MSA who do not meet the criteria for IIM in the category of IIM-ILD rather than IIPs or IPAF [4,5,27,28]. However, it should be noted that most of these findings are derived from ILD patients with anti-ARS antibodies. Hence, further investigations are necessary to ascertain whether ILD patients with other MSAs demonstrate comparable clinical features and outcomes to those with the corresponding MSAs in the context of IIM. The schematic spectrum and classification of IIM and IIM-ILD based on MSA profiles are shown in Figure 1.
Epidemiology of IIM-ILD
The prevalence of ILD in IIM varies widely by subclass of IIM, type of MSA, and race. A recent meta-analysis reported that the prevalence rates of ILD in patients with PM, DM, and CADM were 35%, 42%, and 53%, respectively [29]. Anti-ARS and MDA5 antibodies are most closely associated with ILD and can be found in up to 80%–100% of these antibody-positive patients [15]. Other MSAs associated with ILD are anti-SAE (20%–70%) and SRP (26%–53%) antibodies [15,29-31]. Patients with other MSAs, including anti-Mi2, TIF1-γ, NXP2, or HMGCR, rarely develop ILD [15,29]. Racial differences were also observed between groups. The prevalence of ILD in patients with PM/DM was higher in Asia (50%) than in the US (23%) and Europe (26%) [29].
As mentioned above, a significant number of patients with ILD have MSA with no or subtle muscle or skin manifestations that do not meet the classification criteria for IIM [3,23-28,32]. A Japanese study reported that anti-ARS antibodies were detected in 6.6% of the patients with IIPs [26]. Conversely, 17%–23% of patients with anti-ARS antibodies have isolated ILD without evidence of IIM [23,32]. Although the exact prevalence is unknown and probably rare, other MSAs have also been reported in ILD patients without evidence of IIM [4,24,27].
Clinical Features of IIM-ILD
IIM-ILD has various presentations, ranging from subclinical to rapidly progressive, with characteristic extrapulmonary manifestations. As mentioned above, the features of ILD and extrapulmonary manifestations are strongly associated with MSA profile [1,3,12,15,33-35]. Additionally, treatment response and prognosis are related to MSA profile [34,35]. Therefore, understanding the clinical features of IIM-ILD per MSA profile is crucial for early diagnosis and appropriate management of IIM-ILD. In this section, we focus on ILD associated with anti-ARS antibody (ARS-ILD) and ILD with MDA5 antibody (MDA5-ILD), and briefly discuss ILD with anti-SAE antibody (SAE-ILD) and ILD with anti-SRP antibody (SRP-ILD).
1. ARS-ILD
Between 53% and 66% of individuals with ARS-ILD experience a chronic onset lasting over 3 months (>3 months), with some being asymptomatic (15%). Additionally, 18%–47% of individuals with ARS-ILD exhibit significant progression within 3 months (<3 months), termed as rapidly progressive ILD (RP-ILD) [15,25]. Regardless of presentation, they generally respond well to initial treatment with 64%–95% response rates and have a comparatively favorable prognosis, with a 5-year survival rate of approximately 90% [25,28,34-37]. However, ARS-ILD often relapses during glucocorticoid tapering, leading to a subsequent decline in lung function and fatal long-term outcome [36-38]. In two studies with a median follow-up of more than 5 years, the recurrence rate of ARS-ILD was as high as 38%–56%, with a median time from treatment initiation to the first recurrence of 2 to 3 years [36,38].
ASyS frequently manifests with distinct extrapulmonary symptoms, serving as significant indicators for suspecting ARS-ILD. These symptoms include myositis, polyarthritis, fever, Raynaud’s phenomenon, and mechanic’s hands [33]. These findings may be present inconsistently or absent altogether (33%) during the initial work-up but could develop throughout the disease [23,25,32]. Additionally, the prevalence of extrapulmonary manifestations is associated with individual anti-ARS antibodies. Myositis is one of the most common extrapulmonary manifestations, occurring in 40%–55% of cases at disease onset and rising to 50%–80% during the course of the disease. However, it is uncommon in cases with anti-PL-12 antibodies (17%–38%) and anti-KS antibodies (0%) [23,25]. Similar to that in IIM without anti-ARS antibodies, myositis in ASyS predominantly affects the symmetric proximal muscles, causing muscle weakness, muscle pain, or fatigue, with elevated serum muscle enzyme levels [1]. Polyarthritis is also a common manifestation of ASyS, with a higher prevalence in patients with Jo-1 antibody (58%–61.1%) than in those with other anti-ARS antibodies (20%–35%) [23,25]. Only 32%–40% of patients with ASyS have typical skin manifestations of DM, such as heliotrope rash, Gottron’s papules, or Gottron’s sign at diagnosis [23,33]. Thus, a significant number of ASyS, especially anti-non-Jo1 ARS-antibodies, do not meet the criteria for IIM [2,3,23]. Another characteristic cutaneous manifestation of ASyS is the mechanic’s hands, a fissure with hyperkeratotic papules and scales along the sides of the fingers. Although once considered a pathognomonic feature of ASyS, mechanic’s hands occur in only 28%–40% of patients with ASyS and can be seen in patients with other MSAs [23,33,39]. Fever and Raynaud’s phenomenon occur in 20%–45% of patients with ASyS [23,33].
2. MDA5-ILD
MDA5-ILD often presents as a RP-ILD (69%) with a poor prognosis, whereas chronic forms of ILD are less common (7%) [15,34,35]. However, there are racial differences in MDA5-ILD, with RP-ILD as low as 18%–21% in non-Asian populations [24,40]. MDA5-ILD, especially RP-ILD, shows a poor treatment response, and approximately one-third of patients die of respiratory failure within 1 year despite intensive treatment [34,35].
Between 80% and 90% of patients with MDA5-ILD have typical cutaneous manifestations at initial presentation, allowing the diagnosis of DM, especially CADM [24,40]. In particular, patients with MDA5-ILD present unique mucocutaneous manifestations reflecting the underlying vasculopathy, which is an important clue for the early diagnosis of MDA5-ILD. Skin ulcers are the most distinctive cutaneous manifestation in patients with MDA5-antibody with a 40%–82% prevalence and are less common in IIM with other MSAs [39]. Skin ulcers are often associated with severe pain and occur at specific sites, such as the proximal interphalangeal and metacarpophalangeal joints of the fingers, elbows, shoulders, knees, ankles, digital pulp, and nail folds [39]. More importantly, the presence of skin ulcers can be a prognostic factor of MDA5-ILD [41]. Therefore, clinicians should look for skin ulcers at these predilection sites when evaluating patients with ILD, especially RP-ILD. Palmar papules, also known as the inverted Gottron’s sign, are another unique cutaneous manifestation in patients with anti-MDA5 antibodies, with a prevalence of 20%–60% [39]. They occur on the palms and palmar aspects of the fingers, with the predicted involvement of the metacarpophalangeal or interphalangeal joint creases [39]. Additionally, they are often associated with pain and ulceration. Other characteristic skin manifestations in patients with anti-MDA5 antibodies include nonscarring alopecia (78%), oral ulcers (50%), and panniculitis (20%) [39]. The mechanic’s hands were also often seen [39]. In addition to skin lesions, arthritis/arthralgia was common (72.2%) [40]. Although rare, patients with anti-MDA5 antibodies may present with ILD without skin or muscle manifestations [27].
3. SAE-ILD and SRP-ILD
SAE-ILD is usually a mild ILD with no or mild respiratory symptoms and responds well to treatment without relapse and progression [42]. Anti-SAE antibodies correlate well with typical DM skin manifestations, often preceding muscle involvement [31,42].
SRP-ILD is usually a mild ILD with no or mild respiratory symptoms and does not show RP-ILD [30]. SRP-ILD responds well to treatment and rarely develops into a progressive course [30]. Patients with anti-SRP antibodies present severe myopathy and elevated serum creatine kinase [11]. Approximately 40% of the patients with anti-SRP antibodies develop dysphagia, which may lead to aspiration pneumonia or may even be associated with the development of ILD [30,43].
Radiological Features of IIM-ILD
IIM-ILD shows various ILD patterns on computed tomography (CT) with predominant inflammation, with or without fibrosis, depending on the disease activity and time course. Similar to clinical features, the radiological features of IIM-ILD are highly associated with the MSA profile [44]. Therefore, it is desirable to understand the chest CT features per MSA to predict the MSA profile based on CT findings.
1. ARS-ILD
The most common chest CT findings in ARS-ILD are ground-glass opacities (GGO) (98%–100%), followed by reticular opacities (67%–87%) and consolidations (43%–48%) [45-47]. Subpleural curved linear shadows are seen in 26%–38% [46,47]. Honeycombing is rare (3%–9%) or, if present, limited at presentation but increases or occurs in approximately one-third of patients over time [45-47]. The axial distribution of the abnormality is frequently peripherally predominant (70%–95%), often along with a peribronchovascular distribution (73%) [47]. The craniocaudal distribution is almost always lower lobe predominant, and reticular opacities with traction bronchiectasis in the lower lobes are suggestive of ARS-ILD [45,47]. According to the radiological IIPs classification [22], nonspecific interstitial pneumonia (NSIP) pattern is the most common (44%–55%), followed by NSIP+organizing pneumonia (OP) pattern (19%–34%) and OP pattern (6%–28%) (Figure 2) [21,45-47]. The usual interstitial pneumonia (UIP) pattern has also been reported (1%–4%), but this probably represents an advanced stage of NSIP or NSIP+OP pattern [44-47].
2. MDA5-ILD
The chest CT findings of MDA5-ILD are mainly made up of GGO (83%–100%) and consolidations (58.3%–75%), while reticular opacities are less common (0%–15%) and honeycombing is extremely rare [48,49]. Nonseptal linear opacities, such as subpleural curvilinear shadows or perilobular opacities, were frequently observed (83.3%–90%) (Figure 3) [48-50]. The axial distribution of the abnormality was almost always peripheral, occasionally with peribronchovascular distribution (25%) [49]. The craniocaudal distribution is almost always lower lobe predominance [49]. Particularly, consolidation with volume loss in the lower lobes suggests RP-ILD with a poor prognosis [51,52]. However, even subtle subpleural linear opacities or GGOs at initial evaluation may rapidly progress to widespread GGOs, indicating diffuse alveolar damage (DAD) (Figure 4). According to the radiological IIPs classification [22], most cases are classified as OP or OP+NSIP patterns, which are sometimes considered unclassifiable. In contrast, minor cases are classified as NSIP patterns, and UIP patterns are extremely rare in the Japanese cohort [49,50]. In a recent cluster analysis that divided IIM-ILD into three clusters based on CT findings, more subpleural consolidation, fewer reticular opacities, and no honeycombing were considered characteristic features in a cluster composed mainly of MDA5-ILD [44]. Thus, these CT features help to differentiate MDA5-ILD from other IIM-ILD subtypes, although there is considerable overlap. Of note, a recent study of non-Asian patients with MDA5-ILD showed that the NSIP pattern was the most common (51%) and the UIP pattern was present in 13%, suggesting that, as in clinical features, there may be substantial racial differences in the CT findings of MDA5-ILD [24].
3. SAE-ILD and SRP-ILD
SAE-ILD shows limited subpleural GGO predominantly in the bilateral basal lungs, which is compatible with NSIP pattern [53]. SRP-ILD showed GGO (37%), reticulation (63%), and consolidation (33.3%), with lower lobe and peribrochovascular predominance [30]. Traction bronchiectasis (7.4%) and cysts (3.7%) were less common [30]. According to the radiological IIPs classification [22], the NSIP pattern was the most common (37%), followed by the OP pattern (33%) [30].
Pathological Features of IIM-ILD
Few studies have investigated the histopathologic findings in IIM-ILD because surgical lung biopsy (SLB) is rarely performed in ILD patients with well-established IIM because of a lack of evidence of its utility in predicting prognosis and guiding treatment strategies, along with concerns about the morbidity and mortality associated with SLB. Among IIM-ILD, ARS-ILD accounts for most pathological studies because ARS-ILD often masquerades as IIPs owing to a lower prevalence of muscle or skin lesions than IIM-ILD with other MSAs [23,25,32]. Historically, the NSIP pattern has accounted for a majority of histopathologic patterns of ARS-ILD (Figure 5). A recent comprehensive review revealed that NSIP is the most predominant pathologic pattern in ARS-ILD but occurs in less than half of all cases (41%), followed by UIP (24%) and OP (19%), with varying frequencies of individual anti-ARS antibodies [54]. However, the relationship between histopathological patterns and prognosis or treatment response in ARS-ILD has not yet been fully established. Notably, some studies have suggested that ARS-ILD with UIP pattern has a better prognosis than idiopathic pulmonary fibrosis and responds to immunosuppressive drugs [55]. Thus, the significance of histologic acquisition in ARS-ILD remains unknown, and usually, only atypical cases are indicative of SLB. Recently, transbronchial lung cryobiopsy (TBLC) has been widely used and performed for various ILDs. TBLC can obtain the characteristic pathologic patterns of ARS-ILD, such as NSIP, NSIP with OP, or UIP with an acceptable safety profile (Figure 6) [56]. Given its less invasive nature compared with SLB, TBLC could be utilized in a broader spectrum of patients with ARS-ILD, and its clinical significance should be discussed in the future studies.
Although MDA5-ILD has the second highest prevalence among IIM-ILD, SLB is usually not indicated because of its fulminant course. According to several case studies, almost all autopsy cases exhibit a DAD pattern that can be detected by biopsy at an early stage [57]. A non-DAD pattern has also been reported in SLB with a potentially better prognosis [58,59]. Notably, several biopsy case studies have shown perilobular-predominant poorly aerated alveoli and intra-alveolar organization, which may correspond to perilobular opacities on CT [50,58,59].
Treatment Strategy of IIM-ILD
Given the heterogeneity of IIM-ILD, an individualized treatment strategy that considers ILD phenotype is preferable. Although an evidence-based treatment approach has not yet been established, several recommendations and, most recently, a guideline from the ACR/American College of Chest Physicians have been published [6,60-63]. These suggest an individualized treatment strategy stratified by ILD onset, MSA profile, and treatment response. Therefore, we discuss the treatment strategy for IIM-ILD, considering these factors. Finally, we address the current evidence of antifibrotic therapy for IIM-ILD. The proposed treatment algorithm, which is a modification of the Japanese Respiratory Society and Japan College of Rheumatology, is shown in Figure 7. The typical initial doses of each drug are listed in Table 2.
1. RP-ILD
When treating IIM-ILD, it is essential to initially consider the manner of onset of ILD. As mentioned above, IIM-ILD can present with RP-ILD, a majority of which corresponds to MDA5-RP-ILD with a poor prognosis. Several recommendations have suggested triple combination therapy, including high-dose corticosteroids, calcineurin inhibitors, and intravenous cyclophosphamide for MDA5-RP-ILD [6,60-62]. Given the potential superiority of initial triple combination therapy over a stepwise approach in treating MDA5-RP-ILD, early diagnosis and initiation of triple combination therapy are desirable to improve the prognosis of MDA5-RP-ILD [64]. In contrast, triple combination therapy may result in overtreatment for certain patients, particularly those with MDA-5-negative RP-ILD, as it raises the risk of opportunistic infections [64]. Therefore, early detection of anti-MDA5 antibodies in patients with RP-ILD is critical. However, as the measurement of anti-MDA5 antibodies typically takes several days or weeks, in the case of fulminant ILD, we should speculate on the presence of anti-MDA5 antibodies from the aforementioned characteristic skin manifestations, such as skin ulcers, CADM presentation, and/or CT findings. However, given that minor patients with MDA5-ILD show a lung-dominant presentation, the possibility of MDA5-ILD should not be ruled out even in the absence of overt skin or muscle manifestations [24,27]. Despite the intensive treatments, a significant number of patients with MDA5-RP-ILD show progression with a fatal outcome. In refractory cases, the potential use of tofacitinib, rituximab, and plasma exchange as rescue therapies has been suggested [65-67]. Previous studies have identified predictive factors for poor prognosis in patients with MDA5-ILD. They include older age, higher C-reactive protein and ferritin levels, low oxygenation, higher serum titer of the anti-MDA5 antibody, and the coexistence of anti-Ro52 antibody [68-70]. Higher serum levels of Krebs von den Lungen-6, a useful biomarker reflecting disease activity in various ILDs, have also been reported to be associated with a poor prognosis in MDA5-ILD [68]. Therefore, clinicians should consider early initiation of rescue therapy in addition to triple combination therapy in patients with these risk factors. Conversely, patients with MDA5-ILD without these poor prognostic factors might be treated with corticosteroids and single immunosuppressive agent [71]. MDA5-negative RP-ILD, predominantly comprising ARS-ILD cases, typically show favorable responses to corticosteroids combined with immunosuppressants [34,35]. Nonetheless, due to the potential for fatal outcomes associated with MDA5-negative RP-ILD, additional immunosuppressive medications should be considered for cases of progressive ILD despite initial treatment, regardless of the presence or absence of anti-MDA5 antibodies [6,35]. Following the acute phase, the maintenance therapy for RP-ILD mirrors that of the chronic form of ILD, as elaborated below.
2. Chronic form of ILD
For patients with the chronic IIM-ILD, clinicians should aim to improve prognosis not only in the short term, but also in the long term. Although there is no consensus regarding the initiation of therapy, treatment initiation is determined based on a comprehensive assessment of the severity and progression of the patient’s symptoms, radiographic findings, and/or lung function impairment, as well as the patient’s age and comorbidities. Some patients with chronic ILD have a stable course without progression and can be managed with caution, adopting a watch-and-wait approach [6].
Similar to RP-ILD, the MSA profile is a useful guide for an appropriate treatment strategy in the chronic form of IIM-ILD. In ARS-ILD, initial combination therapy with corticosteroids and a single immunosuppressant is recommended to prevent relapse and subsequent decline in lung function [6,36,38]. If there is no relapse, a dose reduction of corticosteroids and immunosuppressants should be considered to minimize side effects. Notably, a recent study suggested an association between obesity and relapse in patients with ARS-ILD, likely due to the influence of adipokines and cytokines released from the adipose tissue [72]. As corticosteroids may induce obesity, weight control may be important in ARS-ILD to prevent recurrence. Despite careful management, many patients experience recurrence of ILD. If recurrence occurs, clinicians should consider redosing the current treatment, switching, or adding other immunosuppressants (tacrolimus, cyclosporine A, azathioprine, cyclophosphamide, mycophenolate mofetil, and rituximab) or intravenous immunoglobulin. However, the current evidence of their usefulness is limited [73-75]. In addition to pharmacotherapy, supplemental oxygen therapy and pulmonary rehabilitation are recommended. Lung transplantation should be considered in patients who develop severe lung impairment despite treatment.
Chronic IIM-ILD without anti-ARS antibodies is less likely to recur than ARS-ILD [38]. Therefore, corticosteroid monotherapy may be permitted in patients with chronic IIM-ILD without anti-ARS antibodies [6]. Notably, a significant number of patients with MDA5-ILD, especially non-Asian patients, have a chronic form of ILD [24,40]. In French cohorts, clusters composed mainly of chronic MDA5-ILD showed good long-term prognosis [40]. In addition, another large study that included 76 patients with chronic MDA5-ILD showed that only one patient died from ILD progression, although most patients received immunosuppressants during the follow-up [24]. While no study has specifically explored the recurrence rate and detailed clinical course of the chronic form of MDA5-ILD, it’s plausible that recurrence and progression of ILD may be less frequent in patients with chronic MDA5-ILD. The management approach for recurrence and progression of ILD in non-ARS chronic IIM-ILD remains consistent with that for ARS-ILD.
3. Current evidence of antifibrotic therapy for IIM-ILD
Given the current evidence supporting the efficacy of antifibrotic agents across a broad spectrum of ILDs, their use is appealing for patients with chronic progressive ILD [76]. However, the INBUILD trial, which evaluated the efficacy and safety of nintedanib in various ILDs with progressive phenotypes, included only two patients with ARS-ILD and one with PM-related ILD, resulting in weaker evidence for antifibrotic therapy in IIM-ILD than in other ILDs [77]. Although other studies demonstrating the efficacy in chronic progressive IIM-ILD are also lacking, interestingly, recent retrospective studies have demonstrated the potential efficacy of antifibrotic therapy for RP-ILD in patients with IIM along with its tolerability [78,79]. As mentioned above, although the efficacy of antifibrotic agents in IIM-ILD is under investigation, a recent guideline recommends antifibrotic agents for patients with progression of ILD on initial therapy, considering the pace of progression and the amount or pattern of fibrotic change on CT [63].
Current Knowledge on the Pathogenesis of IIM and IIM-ILD
IIM is a complex disease thought to be caused by aberrant immune activation that occurs in a particular environment in genetically predisposed individuals as depicted in Figure 8 [80]. The possible triggers of IIM include smoking, ultraviolet (UV) exposure, and viral infections [81-83]. Molecular mimicry or the similarity between self-antigens and pathogen proteins may be a cause of autoimmunity. For instance, myosin, tropomyosin, keratin, and alanyl-tRNA synthetase (the target proteins of anti-PL-12 autoantibodies) share homologous sequences with proteins from adenoviruses, influenza viruses, and Epstein-Barr viruses [84,85]. Therefore, the muscle or skin becomes a secondary initiator of the autoimmune response if autoantigens are upregulated due to injury, such as trauma or UV radiation. Considerable advancements have been made in the study of IIM genetics, which may offer insights into the etiology of these illnesses. The most significant genetic risk factor for disease susceptibility is the major histocompatibility complex, specifically with alleles of the specific human leukocyte antigen (HLA). The HLA 8.1 ancestral haplotype (HLA-A*01:01, HLA-C*07:01, HLA-B*08:01, HLA-DRB1*03:01, HLA-DQA1*05:01, and HLA-DQB*02:01) is associated with IIM in Caucasians [86]. In contrast, Asian populations have a different immunogenetic background (HLA-DRB1*08:03 or HLA-DRB1*04:05 with ARS-antibodies; DRB1*12:02, HLA-DRB1*01:01, or HLA-DRB1*04:05 with MDA5-antibodies) [87,88]. Genome-wide association studies (GWAS) have identified myositis-related loci in addition to HLA connections under various conditions [89]. This study identified 22 significant genome-wide loci related to IIM [89]. These loci mainly function as expression quantitative trait loci (eQTLs) that influence the mRNA expression levels of immune-related genes [89,90]. Disease-associated eQTLs, along with their protein interaction networks, can elucidate potential pharmacological targets. For example, the tyrosine kinase 2 (TYK2) gene product, which is implicated in myositis, is a target of the Janus kinase inhibitor tofacitinib [89]. Similarly, the gene product of family with sequence similarity 167, member A (FAM167A)-B lymphocyte kinase (BLK) is modulated by nintedanib through the inhibition of platelet-derived growth factor receptor beta [89]. This genetic information will pave the way for novel treatment strategies and drug repositioning for IIM.
In 2018, a GWAS of a Japanese cohort identified an intronic variant of WD repeat and FYVE domain-containing 4 (WDFY4) (rs7919656) association with CADM [91], subsequently confirmed in Chinese patients with MDA5-RP-ILD [92]. This risk allele was linked to higher expression of a truncated WDFY4 isoform (tr-WDFY4) with a trans-eQTL effect that enhanced nuclear factor κB (NF-κB)-associated genes [91]. Notably, tr-WDFY4 markedly enhanced the MDA5-mediated NF-κB activation in response to poly (I:C), a synthetic analog of double-stranded RNA mimicking viral RNA [91]. NF-κB activation induces the production of type I interferon, which is the hallmark of the inflammatory response in the affected organs and the peripheral blood [93,94]. These findings underscore the importance of the WDFY4-MDA5-type I interferon axis as a pivotal element in CADM pathogenesis.
Numerous activated CD8+ T lymphocytes were present in affected and comparatively preserved lung tissues in lung biopsy specimens from patients with IIM-ILD [95]. Mass cytometry analysis of immune cells from the bronchoalveolar lavage fluid revealed distinct characteristics of increased T-cell immunoglobulin domain and mucin domain 3 (TIM-3+) CD8+ T-cells in myositis-associated ILDs [96]. Single-cell RNA-sequencing (scRNA-seq) has further highlighted robust antibody-producing capability and CD8+ T-cell responses as cellular immunological signatures in affected lungs and peripheral blood [97]. A high frequency of circulating CD8+ T-cells expressing interferon-stimulated gene 15 (ISG15) at baseline is correlated with poor 1-year survival in patients with MDA5-antibody [97]. Overactivation of type I interferon and fibrosis signaling pathways was also found in the affected lungs of patients with MDA5-antibody [97], aligning with the identified dysregulation of the tr-WDFY4-MDA5-type I interferon pathway [91,93,94]. Additionally, the clonal expansion of muscle-infiltrating CD8+ T-cells, characterized by the expression of homolog of blimp-1 In T-cell (HOBIT), X-C motif chemokine ligand 1 (XCL1), and CXC motif chemokine receptor 6 (CXCR6), was observed [98].
A plausible scenario involves viral infections and cellular damage leading to an immune response in CD8+ T-cells. Granzyme B released from CD8+ T-cells cleaves several ARS and PM/Scl or alters MDA5 and Ro52 [99], rendering them immunogenic. The modified proteins are then transported from the lungs through antigen-presenting cells to the lymph nodes, eliciting a response in T- and B-cells, particularly in individuals with certain HLA haplotypes. CD4+ helper T-cells facilitate B-cell activation and differentiation into autoantibody-producing plasma cells. Tissue damage results from the systemic circulation of activated lymphocytes and autoantibodies that come into contact with autoantigens in the lungs and react to them. Aberrant type I interferon signaling is the hallmark of the inflammatory response in affected organs, as illustrated in Figure 8 [100].
Conclusion
ILD represents one of the most prevalent extramuscular manifestations of IIM, contributing significantly to mortality and morbidity. The clinical features of IIM-ILD are notably associated with the MSA profile, with observed racial distinctions. This correlation could extend to ILD cases accompanied by MSA that fall short of meeting the IIM criteria. A treatment strategy that takes into account ILD onset, MSA profile, and treatment response has been proposed. Considerable progress has been made in understanding the pathogenesis of IIM and IIM-ILD.
Notes
Authors’ Contributions
Conceptualization: Moda M, Inoue Y. Writing - original draft preparation: Moda M, Yanagihara T. Writing - review and editing: all authors. Approval of final manuscript: all authors.
Conflicts of Interest
Mitsuhiro Moda has received lecture fees from Boehringer Ingelheim, and Shionogi & Co. outside of this work. Toyoshi Yanagihara has received grants from Boehringer Ingelheim outside of this work. Ran Nakashima has received grants from Medical & Biological Laboratories, and lecture fees from Bristol-Myers Squibb, Astellas Pharma, Boehringer Ingelheim, Actelion Pharmaceuticals, Mitsubishi Tanabe Pharma, Asahi Kasei Pharma Corp., Eli Lilly Japan, and Daiichi Sankyo, Inc. outside of this work. Toru Arai has received lecture fees from Boehringer Ingelheim, and Shionogi & Co. outside of this work. Yoshikazu Inoue has received grants from Japanese Ministry of Health, Labour, and Welfare, and lecture fees from Boehringer Ingelheim, Shionogi& Co., Kyorin pharmaceutical, GSK, and Astra Zeneca outside of this work. Yoshikazu Inoue is a consultant and steering/advisory committee member for Boehringer Ingelheim, Roche, Galapagos, Taiho, Kyorin pharmaceutical, Tanabe Mitsubishi, CSL Behring, and Vicore Pharma AB. Other authors have no conflicts of interest.
Funding
No funding to declare.
Acknowledgements
The authors would like to thank Editage (www.editage.jp) for English language editing.