Understanding the Role of Type I Interferons in Cutaneous Lupus and Dermatomyositis: Toward Better Therapeutics

Abstract

A 29‐year‐old female presented to a rheumatology‐dermatology clinic with a pruritic rash that began 6 months prior, after a viral illness. She had previously been diagnosed with eczema and treated with antihistamines and topical steroids without improvement. She also noted fatigue, hair loss, and severe scalp pruritus. Physical examination was notable for violaceous periorbital edema, scaly erythematous papules on the metacarpophalangeal joints of bilateral hands, dilated capillaries of the proximal nail folds, scaly plaques on bilateral elbows, and excoriated erythematous plaques on upper chest, back and hips. The patient reported no muscle weakness, and strength testing and creatinine phosphokinase were normal. Magnetic resonance imaging of the thigh showed no evidence of inflammation or edema. Antibody testing was negative. A diagnosis of clinically amyopathic dermatomyositis was made. Computed tomography scans of the chest, abdomen and pelvis, colonoscopy, and mammogram showed no evidence of cancer. The patient was initiated on methotrexate. Her cutaneous manifestations persisted with debilitating intractable pruritus, and thus, she was transitioned to mycophenolate mofetil, again with minimal improvement. Intravenous immunoglobulin was not approved by insurance given the lack of muscle involvement in her disease. This patient's case highlights a common clinical scenario in rheumatology and dermatology and raises several important issues related to the immunologic underpinnings of cutaneous lupus erythematosus (CLE) and dermatomyositis (DM): What is the role of type I interferon (IFN) in triggering skin disease in CLE and DM? What is the role of IFN in the pathogenesis of skin inflammation in CLE and DM? Can we apply what we know about IFN‐targeted therapeutics in CLE and DM to develop better treatments for skin disease?

Introduction

Dermatomyositis (DM) is a chronic autoimmune disease with characteristic skin findings and often affects the muscles, heart, gastrointestinal, joints, and lungs but can also occur only or predominantly on the skin. Patients with amyopathic or hypomyopathic DM demonstrate absent or clinically insignificant muscle findings. In the absence of myopathy, the diagnosis of DM is often delayed. Studies suggest that the clinically amyopathic DM subphenotype is more common than previously thought, and such patients comprise up to 60% of patients with DM in dermatology clinics. ¹ Skin disease in DM is often relapsing and recalcitrant to therapy, even when systemic disease is well controlled. Importantly, DM skin activity correlates with a poorer quality of life, even compared with other chronic diseases. ²

Lack of knowledge regarding the pathogenesis of DM and drivers that instigate skin disease has delayed effective therapy development. To date, there are no US Food and Drug Administration (FDA)–approved therapies for the treatment of cutaneous lupus erythematosus (CLE) or skin disease in DM. As in our case, many patients have refractory skin lesions and pruritus. Additionally, a lack of inclusion criteria for skin disease activity and damage in ongoing clinical trials limit therapeutic options. Current therapies are broadly immunosuppressive, often ineffective, and cause significant morbidity.

DM rashes appear similar clinically and histologically to those seen in CLE and can be challenging to differentiate, particularly if disease is limited to the skin (Figure 1). Like in CLE, type I interferons (IFNs) are highly up‐regulated in DM blood, muscle, and skin and correlate with disease activity. ³ Additionally, both CLE and DM skin are thought to be induced or “triggered” by heightened IFN exposure, such as after sunlight, medications such as hydroxyurea and immune checkpoint inhibitors, and immune‐stimulating supplements or infections such as COVID‐19 in genetically predisposed individuals (Figure 2A). Although DM and CLE share common triggers, the type I IFN subtype and efficacy of current therapeutics differ and may indicate key differences in disease pathogenesis, such as CLE being IFNα driven, whereas IFNβ may play a dominant role in DM (Figure 2B). A number of novel therapeutics are emerging that act on the IFN system and the type I IFN receptor–associated JAK/STAT signaling pathway (Figure 2A). In this review, we will discuss our current understanding of the role of type I IFN in the immunopathogenesis of CLE and DM skin inflammation. Furthermore, we will discuss emerging IFN‐targeted therapies used in ongoing trials in systemic lupus erythematosus (SLE) and how they may be applicable and efficacious in the treatment of DM skin disease.

Figure 1.

IFN‐driven cutaneous manifestations of dermatomyositis (A–D) and cutaneous lupus erythematosus (E–H) are pictured: (A) widespread erythematous to violaceous papules and plaques on extensor surfaces; (B) heliotrope sign; (C) poikilodermatous erythema on anterior neck in “V” distribution; (D) Gottron's papules overlying joints; (E) psoriasiform SCLE; (F) malar erythema of acute LE; (G) annular erythematous plaques of SCLE; (H) dorsal hand erythema sparing joints of LE. IFN, interferon; LE, lupus erythematosus; SCLE, subacute cutaneous lupus erythematosus.

Figure 2.

(A) Summary of the type I IFN pathway, potential disease triggers, and mechanisms of new therapeutics is given. (B) Similarities and differences between CLE and DM are shown. BDCA2, blood DC antigen 2; cGAS, Cyclic GMP‐AMP synthase; CLE, cutaneous lupus erythematosus; DM, dermatomyositis; dsRNA, double‐stranded RNA; HCQ, hydroxychloroquine; IFN, interferon; IFNAR, IFNα receptor; IMO, Isomalto‐oligosaccharide; IRF, IFN response factor; IRG, IFN response gene; ISGF3, IFN‐stimulated gene factor 3; ISRE, IFN‐stimulated response element; LPS, Lipopolysaccharide; mAb, monoclonal antibody; MDA5, melanoma differentation‐associated protein 5; pDC, plasmacytoid dendritic cell; RIG‐I, retinoic acid–iducible gene I; ST2825, small molecule inhibitor of MyD88; STING, stimulator of IFN genes; TLR, Toll‐like receptor; TNF, tumor necrosis factor; TYK2, Tyrosine Kinase 2.

Type I IFNs

Type I IFNs are an immunomodulatory class of cytokines that are involved in the innate and adaptive immune system and serve to protect against viral infection and provide cancer immunosurveillance. ⁴ Mounting evidence suggests that type I IFNs are important to the initiation and maintenance of autoimmune and inflammatory skin diseases, such as CLE and DM (Figure 2A). IFNs promote antigen presentation and natural killer cell function, regulate inflammatory pathways, and activate the adaptive immune system to form immunologic memory. ⁵ The type I IFN family of cytokines is composed of 17 members: 13 subtypes of IFNα, IFNβ, IFNε, IFNκ, and IFNω. ⁶ , ⁷ All type I IFNs signal through the same heterodimeric transmembrane receptor, the IFNα receptor (IFNAR), which is composed of two subunits: IFNAR1 and IFNAR2. The IFNAR is expressed in nearly all cell types. Activation of JAK1 and Tyrosine Kinase 2 (TYK2) result in the tyrosine phosphorylation and activation of several STAT family members. Activation of STAT1 and STAT2 lead to the recruitment of IFN response factor 9 (IRF9) and the formation of the STAT1–STAT2–IRF9 complex, which is known as the IFN‐stimulated gene factor (ISGF3) complex. This complex then migrates to the nucleus and binds to IFN‐stimulated response elements in the promoters of ISGs to initiate gene transcription. Activated genes include up‐regulation of IFN signaling components, IFN genes themselves, and effector genes for antiviral responses. A number of therapies that target type I IFN and downstream signaling are emerging and are exciting avenues of interest in the treatment of autoimmune skin disease, as discussed below.

Most cell types can produce type I IFNs (leukocytes, fibroblasts, endothelial cells, keratinocytes), but the primary producers of IFNs may be context and disease dependent. The dominant cytokine profile and inflammatory cell populations of CLE lesions are heterogeneous and have been shown to reflect different responsiveness to antimalarials. ⁸ Hydroxychloroquine (HCQ)–responsive patients had significantly higher type I IFN–regulated transcripts and lower tumor necrosis factor α (TNFα) levels than HCQ‐refractory patients. ⁸ In contrast, patients who were refractory to HCQ, thus needing addition of quinacrine, had significantly up‐regulated TNFα expression produced by myeloid dendritic cells. ⁸

Triggers of skin disease

UV light

Photosensitivity, a sensitivity to UV light whereby even ambient light can result in inflammatory skin lesions, is a hallmark of CLE and SLE and occurs in 60% to 80% of patients. ⁹ The exact mechanism of photosensitivity is still unfolding, but type I IFNs likely play an important role (Figure 2A and B). SLE keratinocytes (KCs) exhibit a more robust type I IFN response after UV exposure when compared with healthy controls. ¹⁰ This proinflammatory effect of ultraviolet B (UVB) in patients with SLE with photosensitivity likely stems from the IFN‐rich environment. ¹¹

UV stimulation likely up‐regulates IFN production through endogenous nucleic acid triggering cyclic GMP‐AMP synthase (cGAS) and the stimulator of IFN genes (STING) pathway. This is supported by a study showing that UV stimulation of murine skin results in rapid up‐regulation of Ifnb that is blocked by absence of cGAS. ¹² UVB has also been noted to induce activation of human endogenous retroviruses, which leads to production of double‐stranded RNA and activation of retinoic acid–inducible gene I (RIG‐1)/melanoma differentiation‐associated protein 5 (MDA5)‐mediated type I IFN induction. ¹² Interestingly, the presence of enhanced IFN genes in the skin has been translated to systemic increases in IFN‐regulated genes, suggesting that systemic IFN responses can result from skin exposure. ¹² UVB can also increase the IFN‐stimulating capacity of nucleic acids because cells exposed to UV‐treated DNA have increased IFN production compared with cells not exposed to UVB‐treated DNA. ¹² , ¹³ Injection of DNA treated with UV light into lupus‐prone mice can induce CLE‐like lesions, and deletion of the enzyme that removes degradation of UV‐treated DNA leads to CLE‐like lesions with enhanced IFN production in a STING‐dependent manner. ¹⁴ Together, these data suggest that UV modulation of nucleic acids and nucleic acid sensing is critical for UV‐mediated IFN responses.

Similar to CLE, cutaneous signs of DM are frequently present on sun‐exposed areas such as the upper neck and back and/or shoulders. Photosensitivity may be an initial presenting sign of DM and should prompt consideration of the diagnosis. Photoprovocation testing in patients with DM resulted in increased photosensitivity and inducement of new lesions in ~25% of patients. ¹⁵ Formal UVB testing of these patients demonstrated a reduced minimal erythema dose. Certain myositis‐specific antibodies have been correlated with increased incidence of photosensitivity in DM, including anti‐Mi2 and anti–transcription intermediary factor 1γ. ¹⁶ The role of type I IFNs in the photosensitive response of DM has not been studied.

Medications

Drug‐induced lupus erythematosus is a lupus‐like autoimmune disorder that occurs with exposure to certain medications. Terbinafine (an allylamine derivative) is one of the most widely prescribed medications associated with drug‐induced subacute CLE, with all studied patients exhibiting resolution after drug cessation. Although the mechanism of the terbinafine‐induced autoimmune reaction is not entirely clear, it may be related to the lipophilic and keratinophilic properties that contribute to altered nuclear antigen configuration, antinuclear and antihistone antibodies, and enhanced IFN production. ¹⁷

Several medications have been linked to DM onset or flare, including hydroxyurea, TNFα inhibitors, immune checkpoint inhibitors (ICIs), penicillamine, and statins. ¹⁸ Hydroxyurea is a cytotoxic agent used to treat myeloproliferative disorders and has been reported as the most common culprit for drug‐induced DM, accounting for 51% of patients (n = 36/70). ¹⁹ The mechanism of hydroxyurea‐induced DM is not entirely clear but may be due to the synergistic effects of hydroxyurea blocking DNA site repair and UV irradiation, causing DNA strand breaks and likely up‐regulating KC production of type I IFNs. ²⁰ TNFα inhibitors have also been implicated in DM onset or flare. ²¹ This may be explained by TNF inhibition paradoxically increasing the production of type I IFN, which activates antigen‐presenting cells and stimulates autoantigen production. ²² Supporting this, in an open pilot study of infliximab in patients with refractory inflammatory myopathies, several patients experienced acute worsening, and these flares were associated with increased type I IFN activity in both blood and muscle. ²³

ICIs, which target programmed cell death 1 (PD‐1), programmed cell death ligand (PDL‐1) mDC and CTLA‐4, are increasingly being used to treat both solid and hematologic malignancies by harnessing the immune system's ability to recognize and destroy malignant cells. ICIs result in a wide range of autoimmune skin manifestations, including LE and DM, as a consequence of unleashing immune system regulation. ²⁴ The mechanism underlying ICI induction of autoimmune skin disease is complex but likely in part due to enhanced CD4+ and CD8+ T cell activation with subsequent release of cytokines such as type I IFNs. ²⁵

Supplements

Treatment with a number of herbal supplements, including Spirulina plantensis, Aphanizomenon flos‐aquae, Chlorella, Echinacea, and alfalfa, stimulates the immune system, and treatment with them has been associated with autoimmune skin disease onset and exacerbations. ²⁶ , ²⁷ , ²⁸ , ²⁹ , ³⁰ Patients with DM have been reported to have a higher rate of herbal supplement treatment compared with patients with CLE or autoimmune blistering diseases and healthy controls. In vitro testing showed that IsaLean weight loss powder dose‐dependently stimulates peripheral blood mononuclear cell secretion of inflammatory cytokines, including TNFα, IFNα, and IFNβ, primarily via Toll‐like receptor 4 activation. ²⁹ Practitioners should be aware of the potential immunostimulatory effects of herbal supplements on DM and specifically ask their patients about possible consumption.

Viral illness

In genetically susceptible individuals, viral infections have been suspected to induce or exacerbate autoimmune skin disease, including LE and DM, through up‐regulation of type I IFN. Recent reports have shown an increased incidence of disease flares in patients with autoimmune skin disease after COVID‐19 vaccination, with an increased proportion of patients with DM compared with patients with lupus erythematosus. ³¹ Of the patients who had autoimmune exacerbations after vaccination, 20% had symptoms after the first dose, 82% after the second dose, and 4% after the third dose. ³¹ The underlying mechanisms of immune stimulation are believed to be in part be driven by type I IFN production, which the COVID‐19 vaccines have been shown to up‐regulate. ³² It is unclear why DM appears to be disproportionally affected; however, differences in cells and type of IFN or an increased propensity to flare at any given time may be contributing factors. ³¹

Role of IFN in the immunopathogenesis of skin disease

CLE

It has been well‐documented that CLE has a strong type I IFN signature. ³³ , ³⁴ , ³⁵ , ³⁶ Histologically normal skin in patients with SLE produces type I IFN ¹⁰ , ³⁷ with a marked concentration of type I IFN response in the absence of infiltrating leukocytes. ³⁷ In a recent study using single‐cell RNA sequencing of paired lesional and nonlesional skin samples from patients with SLE, an IFN‐rich signature was seen in both lesional and nonlesional KCs as well as other cell populations, such as fibroblasts and T cells. ³⁸ In addition, a proinflammatory cellular communication network between stromal and inflammatory cells was seen in lesional and nonlesional skin that supports the role of KCs priming myeloid recruitment and inflammatory responses. ³⁸ Further investigation into the transcriptional IFN dominant phenotypes and communication networks may provide targeted therapeutic strategies.

Other cell populations besides KCs likely contribute to enhanced IFN production in the skin. Most research on type I IFN production has focused on plasmacytoid dendritic cells (pDCs). In otherwise healthy individuals, pDCs produce large amounts of IFN upon recognition of viral antigens via endosomal Toll‐like receptors TLR7 and TLR9. ³⁹ , ⁴⁰ However, the role of pDCs in type I IFN production in autoimmune diseases such as SLE is less clear. Psarras et al show that, in preclinical autoimmunity and in SLE, pDCs are not effector cells, but rather have lost their capacity for TLR‐mediated IFNα and TNFα production and fail to induce T cell activation, independently of disease activity and blood IFN signature. ³⁷ pDCs in lesional CLE were not major type I IFN producers. ⁴¹ This may be in part due to inflammatory environments with high TNFα production, ⁴² such as can be found in the skin of patients with SLE. ⁴³ Additionally, monocytes exhibit STING‐dependent type I IFN production after UVB stimulation. ⁴⁴ How the inflammatory environment modulates the functional status of pDCs and other IFN‐producing cells in lupus is still being uncovered.

DM

Overexpression of type I IFN also has been linked to the immunopathogenesis of DM. IFN‐inducible transcripts and activated type I IFN signatures are observed in patients with DM. ³ Increased expression of IFNα/β‐inducible genes and proteins induced by type I IFNs such as myxovirus resistance protein are seen in peripheral blood mononuclear cells and affected muscle and skin of patients with DM. ⁴⁵ , ⁴⁶ , ⁴⁷ Elevated blood IFNβ protein is seen in DM muscle and blood and correlates with high blood type I IFN–inducible genes. ⁴⁸ In the skin, IFNβ protein is highly up‐regulated in the T cell, macrophage, dendritic cell, and endothelial cell populations. ⁴¹ Additionally, IFN signatures, in particular IFNβ, closely correlate with cutaneous disease activity. ⁴⁹ IFNβ expression is abundant in all cells present in the epidermis and dermis of DM skin. ⁴¹ TLR3 has been shown to be elevated in vascular endothelial cells, myeloid dendritic cells, and regenerating myofibers in patients with juvenile DM (JDM). Type I IFN signatures are increased in B cells of patients with JDM, and the activation of TLR7 and IFNα may lead to expansion of immature B cell populations. ⁵⁰ In vivo and in vitro studies show that myogenic precursor cells are a source of type I IFNs. ⁵⁰

Emerging therapies targeting type I IFNs

With mounting evidence suggesting a role of type I IFN signaling in the pathogenesis of autoimmune diseases, an increasing number of monoclonal antibodies (mAb) and small molecule inhibitors are being developed to target the type I IFN signaling pathway (summarized in Figure 2A and Table 1). Therapies targeting IFNα or IFNβ, the IFNAR chain, pDCs, TLRs, and their downstream IFN signaling pathways, including JAK inhibitors and TYK2 inhibitors, will be discussed below. We also will discuss cannabinoid receptor type 2 agonist lenabasum.

Table 1.

Emerging therapies in CLE and DM^*

Therapy	Mechanism of action	CLE				DM
		Level of evidence	Clinical endpoints for skin	Comments	Ref	Level of evidence	Clinical endpoints for skin	Comments	Ref
Targeting IFN
Sifalimumab	Anti‐IFNα mAb	Phase IIb, randomized, double‐masked, placebo‐controlled study	≥4‐point reduction in CLASI	Slight reduction in CLASI subset of patients with CLASI activity score ≥10, after week 28	82	Phase Ib, randomized, double‐masked placebo‐ controlled trial	15% reduction in MMT8, no skin‐specific endpoints	Inhibited ISG in skin (and muscle)	52
F‐06823859 (dazukibart)	Anti‐IFNβ IgG1 neutralizing Ab					Phase II, double‐masked, placebo‐controlled trial	>5‐point reduction in CDASI score or >40% CDASI improvement from baseline at 12 weeks	>5‐point reduction in 100% and 95% of 150 mg and 600 mg doses vs. placebo of 35.7% R>40% reduction in CDASI shown in 80%, 82.1% and 7.1% for the 150 mg, 600 mg and placebo groups	83
Anifrolumab	Anti‐IFNAR1 mAb	Phase IIb (moderate to severe SLE); phase III trial (TULIP‐I and TULIP‐II)	≥50% improvement in CLASI; reduction of 50% or more in CLASI	Phase IIb and TULIP‐II met primary endpoint; 50% reduction in CLASI was noted at week 12 in patients with baseline CLASI‐A>10	53, 55, 56
Inhibition of pDCs
Litifilimab		Randomized, double‐blind, placebo‐controlled trial	CLASI‐A	The difference from placebo in the change from baseline in CLASI‐A score at week 16 was –33.4 percentage points in the 150‐mg group, and –28.0 percentage points in the 450‐mg group.	59
Downstream IFN signaling
Deucravacitinib	TYK2 inhibitor	Phase II, Randomized, Double‐masked, placebo‐controlled trial	CLASI‐50 at 52 weeks	CLASI‐50 was significant in patients with baseline CLASI‐A score ≥10	78
Cannabinoid receptor type 2 agonist
Lenabasum	Selective CB2 agonist					Phase 2 randomized, placebo‐controlled	CDASI	Reduced IFNβ gene expression and IL‐31 protein expression	81

^*Ab, antibody; CDASI, Cutaneous Dermatomyositis Disease Area and Severity Index; CLASI; Cutaneous lupus erythematosus Disease Area and Severity Index; CLASI‐50, decrease of ≥50% from baseline CLASI activity score; CLE, cutaneous lupus erythematosus; DM; dermatomyositis; HCQ, hydroxychloroquine; IFN, interferon; IFNAR, IFNα receptor; IL, interleukin; ISG, interferon‐stimulated gene; mAb, monoclonal antibody; MMT8l, Manual Muscle Testing; pDC, plasmacytoid dendritic cell; SLE, systemic lupus erythematosus; TYK2, Tyrosine Kinase 2.

Anti‐IFNα

A phase Ib clinical trial study (NCT00533091) observed that the anti‐IFNα mAb sifalimumab inhibited type I IFN signatures in patients with DM. ⁵¹ Sifalimumab has specificity only for IFNα, leaving other type I IFNs unaffected and able to bind to the IFNAR. Given that current data suggest that DM seems to be predominantly driven by IFNβ compared with IFNα, it is unclear whether further investigations into IFNα inhibition will yield efficacy in treating DM skin disease.

Anti‐IFNβ

A phase II trial (NCT03181893, Ddzukibart) tested anti‐IFNβ, an immunoglobin G1 neutralizing antibody, in adults with DM and showed efficacy in skin. ⁵² Clinical end point Cutaneous Dermatomyositis Disease Area and Severity Index (CDASI) reduction of <5 points was achieved in 100% and 96% for the 150‐mg and 600‐mg doses, respectively. The placebo response was 35.7%. For the Cutaneous Dermatomyositis Disease Area and Severity Index CDASI reduction of >40%, the rates were 80%, 82.1%, and 7.1% for the 150 mg, 600 mg, and placebo arms, respectively. ⁵²

Anti‐IFNAR1 mAb

An anti‐IFNAR1 mAb, anifrolumab, showed significant effects in decreasing the disease activity of SLE in phase II trials, and another phase IIb study in patients with moderate to severe SLE was considered very successful. ⁵³ , ⁵⁴ Two phase III trials (known as TULIP‐1 and TULIP‐2) in patients with SLE were completed in 2019 and showed promise in treatment of skin disease in lupus erythematosus with a >50% reduction in CLASI was noted at week 12 in patients with a baseline CLASI‐A >10. ⁵⁵ , ⁵⁶ Several recent case series have shown promise of anifrolumab for the treatment of refractory CLE ⁵⁷ , ⁵⁸ ; further studies are needed for skin limited disease.

IFNAR blockade blunts the effects of both IFNα and IFNβ and has been shown to reverse many of the proteomic and cellular changes of enhanced type I IFN seen in SLE. Given similar IFN‐driven changes in DM, further studies are warranted to determine if IFNAR blockade may provide clinical benefit.

Inhibition of pDCs

Litifilimab is an mAb that binds to blood DC antigen 2, an inhibitory receptor on pDCs, and induces rapid internalization to inhibit the production of type I IFNs. ⁵⁶ Litifilimab was beneficial to patients with CLE. ⁵⁹

pDCs have also been implicated in DM, with increased numbers noted in DM skin, ⁴¹ , ⁶⁰ , ⁶¹ although there were more pDCs in CLE than DM. ⁶² Single‐cell analysis using mass cytometry of DM skin supports this, showing increased pDC numbers in DM skin. However, this study suggests that pDCs may not be the main drivers of type I IFN production. Monocyte and/or macrophages, monocytoid dendritic cells (mDCs), and T cells were shown to be main contributors, with mDCs being the main drivers of IFNβ production. In DM, pDCs were shown to produce high IFNγ, which activates a receptor complex distinct from that of the type I interferons. Whether inhibition of pDCs through down‐regulation of IFNγ will result in clinical improvement in DM is not known.

TLR inhibition

HCQ is a TLR7, TLR8, and TLR9 antagonist and a conventional antimalarial drug for autoimmune skin diseases, particularly CLE. ⁶³ A significant proportion (20%–30%) of patients with DM develop a cutaneous reaction, most commonly a morbilliform rash, ⁶⁴ limiting treatment with HCQ in DM. Additionally, studies suggest that only 25% of patients with DM respond to HCQ. ⁶⁵ This may be explained by elevated numbers of CD11c+ mDCs, which are important producers of IFNβ, in HCQ nonresponders compared with HCQ responders in DM. ⁶² Interestingly, IFNα was lower in HCQ nonresponders, and there was no difference in IFNβ expression between the two groups. ⁶² This suggests that HCQ may have effects on high IFNα–expressing responders, whereas HCQ nonresponders may not respond due to differences in cell sources such as mDCs that alter the type I IFN composition. ⁶² It is unknown whether additional mechanisms of TLR inhibition will be more efficacious in DM or result in cutaneous adverse events such as occur with HCQ.

Downstream interferon signaling

JAK inhibitors target inhibition of cytokine‐mediated signaling via the JAK/STAT pathway, which is important for signaling of inflammatory cytokines, immunoregulation, and different cell growth factors. ⁶⁶ , ⁶⁷ Baricitinib is a selective inhibitor of Janus kinase inhibitor JAK1/JAK2 that has shown significant efficacy in patients with SLE in a phase II trial but failed in two phase II trials. ⁶⁸ Tofacitinib, an inhibitor of JAK1/3, is approved by the FDA for the treatment of rheumatoid arthritis, psoriatic arthritis, and ulcerative colitis. ⁶⁹ Limited reports have shown that tofacitinib improves recalcitrant cutaneous lupus. ⁷⁰ In a recent phase I double‐masked randomized safety trial of tofacitinib, authors suggest that inhibition of aberrant IFN signaling results in improvement of cardiometabolic and immunologic parameters associated with the premature atherosclerosis in SLE, ⁷¹ the results of which will need to be followed up with long‐term studies. Clinical trials of tofacitinib for SLE and CLE are still in progress. ⁶⁹ Filgotinib, a highly selective JAK1 inhibitor, failed in a phase II clinical trial. ⁶⁹ , ⁷²

A few studies have reported a positive response to JAK inhibitors in DM ⁷³ , ⁷⁴ as well as in a subset of patients with refractory JDM. ⁷⁴ In a recent systematic review, 38 female patients and 15 male patients with DM had been treated with JAK inhibition without serious side effects. ⁷⁵ Tofacitinib was the most frequently prescribed, followed by ruxolitinib (inhibitor of JAK1/2). ⁷⁵ Improvement of CDASI was reported in most of the studies; however, the duration of follow‐up was limited to 1 to 15 months. ⁷⁵ The first prospective, open‐label clinical trial of tofacitinib in 10 patients with DM was recently published and showed promising efficacy and decreased STAT1 signaling and IFN gene expression in the skin. ⁷³ Further studies are warranted to determine the long‐term safety and efficacy of JAK inhibitors to treat DM skin disease.

TYK2 inhibition

Ducravacitinib is an oral, selective, allosteric inhibitor of TYK2 that binds the regulatory domain and locks the enzyme in an inactive state, distinguishing it from inhibitors of JAK1, JAK2, and/or JAK3 that bind the highly conserved active domains. ⁷⁶ Ducravacitinib is FDA approved for the treatment of adults with moderate to severe plaque psoriasis. ⁷⁷ A recent phase II randomized, double‐masked, placebo‐controlled trial suggests higher response, including for CLASI‐50, in patients with SLE compared with placebo, with acceptable safety profile. ⁷⁸

Cannabinoid receptor type 2 agonist

The endocannabinoid system can modulate inflammatory and fibrotic responses. ⁷⁹ Cannabinoid receptor type 2 (CB2) is mainly expressed on immune cells and plays a role in returning an activated immune response to homeostasis through modulation of inflammation, ⁸⁰ leading to reductions in type I IFNs. Lenabasum is a selective CB2 agonist. A recent phase II randomized, placebo‐controlled clinical trial study showed that lenabasum treatment was well tolerated and associated with improvement of CDASI and multiple efficacy outcomes in patients with moderate to severe skin involvement. ⁸¹ Additionally, lenabasum had a significant effect on IFNβ gene expression in the skin and reduced interleukin‐31 protein, a cytokine associated with itching of the skin. ⁸¹ The degree and consistency of clinical benefit, combined with a favorable safety profile, warrant further evaluation of lenabasum for skin predominant DM.

Conclusion

There is a critical need for better treatments for dermatomyositis, particularly skin disease. Understanding the pathogenesis of skin diseases and drivers that instigate disease, in particular the role of type I IFNs, is key to propel effective therapy development. Skin only, skin predominant, and postmyopathic skin disease are more common than previously reported. As for our patient, skin disease is often refractory to current systemic directed therapies and results in debilitating pruritus and increased morbidity. Current and ongoing trial data mostly require active myositis, excluding patients with predominantly skin disease, and this often limits clinicians’ ability to get medications covered by insurance companies for skin disease. Novel therapeutics are emerging that act on the IFN system and associated signaling and are promising new avenues for treatment (Table 1). To expand use of this armamentarium for refractory skin disease in DM, validated disease severity scores need to be accepted by the FDA and used as primary outcomes. Given the effectiveness of intravenous immunoglobulin (IVIg) and the likelihood of success with IVIg in this case, it is essential that DM trials include patients with amyopathic DM so that they can access appropriate therapies.

AUTHOR CONTRIBUTIONS

All authors were involved in drafting the article or revising it critically for important intellectual content, and all authors approved the final version to be published.

Supporting information

Disclosure form: