IFNAR2 (interferon alpha/beta receptor subunit 2)
The high-affinity signal-transducing subunit of the heterodimeric type I interferon receptor. IFNAR2 binds IFN-α and IFN-β with high affinity (K_D ~10–100 pM for IFN-α2; IFNAR1 binds with ~1000× lower affinity alone) and constitutively associates with jak1. When type I IFN ligand induces pairing with IFNAR1 (the TYK2-associated low-affinity subunit), the activated IFNAR1/IFNAR2 heterodimer initiates JAK-STAT signaling that drives expression of several hundred interferon-stimulated genes (ISGs). A soluble circulating isoform (sIFNAR2) lacking the transmembrane domain acts as a natural decoy receptor. Aging-context relevance: IFNAR2 is the gatekeeper of all type I IFN downstream signaling; its chronic tonic activation by self-derived nucleic acids (leaked mtDNA, LINE-1 cDNA, cytosolic chromatin) drives inflammaging, while paradoxically ~4–6% of individuals aged >70 years harbor neutralizing autoantibodies against IFN-α/ω that functionally antagonize IFNAR2-mediated signaling at the ligand level.
Identity
- UniProt: P48551 (INAR2_HUMAN); Swiss-Prot curated entry
- NCBI Gene: 3455
- HGNC: 5433 (symbol IFNAR2; synonyms IFNABR, IFNARB)
- Ensembl: ENSG00000159110
- Chromosomal location: 21q22.11 (same locus cluster as IFNAR1, IL10RB, and IFNGR2 — type I/II cytokine receptor gene cluster)
- Mouse ortholog: Ifnar2 (Mus musculus; one-to-one ortholog; close functional conservation across mammals)
- Protein length: 515 amino acids (canonical long isoform, IFNAR2c)
Naming note. IFNAR2 has no separate pathway page; [[type-i-interferon-signaling]] is the canonical pathway wikilink. The protein file ifnar2.md and any future ifnar1.md file are the atomic protein pages.
Protein structure
Isoforms
Three isoforms arise from alternative splicing of the same primary transcript 1:
| Isoform | Also called | Length | Localization | Role |
|---|---|---|---|---|
| Isoform 1 (IFNAR2c) | Long form | 515 aa | Cell membrane (single-pass type I) | Canonical signaling subunit; constitutively associates with JAK1 via intracellular tail |
| Isoform 2 (IFNAR2b) | Short form | 249 aa | Cell membrane | Contains ectodomain + short intracellular stub; unable to activate JAK1; dominant-negative function proposed |
| Isoform 3 (IFNAR2a / P40) | Soluble form (sIFNAR2) | 243 aa | Secreted | Ectodomain only; circulates in plasma and urine; functions as decoy receptor |
Domain topology (canonical long isoform)
- Signal peptide: residues 1–26 (cleaved in mature protein)
- Extracellular domain (ECD): residues 27–243
- Sub-domain D1 (FN-III repeat 1): residues 27–130; primary ligand-binding surface; contacts IFN-α/β
- Sub-domain D2 (FN-III repeat 2): residues 131–243; involved in receptor dimerization with IFNAR1
- Three disulfide bonds (Cys39–Cys122; Cys85–Cys93; Cys207–Cys227) stabilize the FN-III folds
- Five N-glycosylation sites: Asn58, Asn87, Asn116, Asn188, Asn192
- Transmembrane helix: residues 244–264
- Intracellular tail: residues 265–515; constitutively binds JAK1 via the Box1/Box2 JAK-binding motif
Function — IFN-α/β signal transduction
High-affinity ligand binding and receptor dimerization
IFNAR2 binds IFN-α2 (and most other IFN-α subtypes) with high affinity as a monomeric receptor, while IFNAR1 alone binds with approximately 1000-fold lower affinity 2. The binary IFN-α·IFNAR2 complex then recruits IFNAR1, which stabilizes the ternary signaling-competent heterotrimer. This asymmetric binding architecture (IFNAR2 = primary affinity determinant; IFNAR1 = signal-amplifying co-receptor) explains why IFNAR2 density is the dominant factor controlling cellular IFN sensitivity in dose-response experiments. IFN-β binds both subunits with higher affinity than IFN-α and forms a more stable complex, contributing to IFN-β’s greater potency per molecule 2. needs-replication — quantitative K_D values in primary cells under physiological conditions are sparse; most binding kinetics data derive from expressed recombinant ectodomains.
JAK-STAT cascade activation
Upon ligand-induced heterodimerization 2:
- Constitutive IFNAR2–JAK1 association — JAK1 is bound to the intracellular tail of IFNAR2 (long isoform) in the resting state; this association is required for receptor surface stability and IFN-I response competence.
- TYK2 is the cognate kinase for IFNAR1 — ligand-induced proximity of the two receptor subunits brings JAK1 (via IFNAR2) and TYK2 (via IFNAR1) into juxtaposition.
- Trans-phosphorylation of JAK1 and TYK2 — each kinase phosphorylates the other in a reciprocal activation step.
- STAT1 (Tyr701) and STAT2 (Tyr689) phosphorylation — activated JAK1/TYK2 phosphorylate recruited STAT1 and STAT2.
- ISGF3 complex assembly — p-STAT1/p-STAT2 heterodimer recruits IRF9 (interferon regulatory factor 9) to form the trimeric ISGF3 transcription factor.
- Nuclear import and ISG induction — ISGF3 binds ISRE elements (consensus 5′-AGTTTCNNTTTCC-3′) and drives transcription of interferon-stimulated genes (ISGs), including antiviral effectors (OAS1, MX1, IFIT1), immune-modulatory genes, and feedback regulators.
Negative regulation: USP18
A critical negative-feedback node is USP18 (ubiquitin-specific protease 18), which binds the intracellular tail of IFNAR2 and displaces JAK1 — preventing sustained STAT phosphorylation and attenuating the IFN-I response 1. USP18 is itself an ISG (induced by IFN-I signaling), creating a delayed-negative-feedback loop that limits chronic signaling. Loss-of-function mutations in USP18 cause severe autoinflammatory disease in humans, underscoring that unrestrained IFNAR2-JAK1 signaling is intrinsically pathogenic.
Soluble IFNAR2 (sIFNAR2): natural IFN-I antagonist
The secreted P40 isoform (sIFNAR2, also IFNAR2a) is detectable in human plasma and urine 3. As the ectodomain alone — without transmembrane or intracellular components — sIFNAR2 can bind IFN-α and IFN-β and sequester them in the extracellular space, preventing engagement of membrane-bound IFNAR1/IFNAR2 heteroreceptor. This decoy-receptor mechanism is analogous to soluble TNF receptor or IL-1Ra in attenuating inflammatory cytokine signaling.
Clinical correlate: In multiple sclerosis patients untreated with interferon therapy, serum sIFNAR2 levels are significantly lower than in healthy controls. IFN-β therapy increases sIFNAR2 levels substantially within one year; other MS therapies do not 4. In severe COVID-19, higher sIFNAR2 levels in plasma correlated with survival relative to non-survivors (Fricke-Galindo 2022) 5, suggesting that greater decoy-receptor expression may buffer against excessive systemic IFN-I signaling in the context of acute viral infection. no-mechanism — the directionality of sIFNAR2 as protective vs. IFN-I–dampening (which could impair antiviral defense) is context-dependent and not fully characterized in aging.
| Dimension | Status | Notes |
|---|---|---|
| Pathway conserved in humans? | yes | JAK1/TYK2/STAT1/STAT2 assignments identical in human and mouse; high sequence conservation |
| Phenotype conserved in humans? | yes | IFNAR2-null mice are susceptible to viral infection; human IFNAR2-deficiency (immunodeficiency-45) causes similar vulnerability |
| Replicated in humans? | yes (mechanistic) | Core JAK-STAT signaling mechanism well established; sIFNAR2 physiology needs further human characterization |
Role in aging
IFNAR2 as gatekeeper of inflammaging
Inflammaging — the chronic low-grade sterile inflammation of aging — is mechanistically driven in part by tonically elevated type I IFN production. Upstream inducers include:
- cGAS-STING activation by leaked mitochondrial DNA, cytosolic chromatin from senescent cells, and nuclear lamina-disruption fragments (see cgas-sting and sting)
- Retroelement reactivation — LINE-1 and endogenous retrovirus (ERV) elements that become transcriptionally de-repressed in aged epigenomes generate cytosolic cDNA sensed by cgas
- Persistent DNA damage (unrepaired DSBs in senescent cells) activates the dna-damage-response → IRF3/IRF7 → IFN production
IFNAR2 is the obligate transducer of all of these upstream IFN-I signals: without IFNAR2, no ISG induction occurs regardless of how much IFN-I is produced. Therefore, IFNAR2-level modulation (whether by anifrolumab, sIFNAR2 decoy expression, or USP18 induction) is among the most proximal pharmacological handles on the inflammaging phenotype 6. needs-human-replication — direct demonstration that reducing IFNAR2 signaling amplitude reduces inflammaging biomarkers in aged humans is lacking; evidence is primarily mechanistic and from autoimmune-indication trials.
STAT1 modifications impair IFNAR2 interaction with aging
A 2025 study found that aging promotes β-hydroxybutyrylation of STAT1 at lysine 592 (Kbhb-STAT1), which specifically impairs STAT1’s interaction with the IFNAR2 intracellular domain, dampening downstream ISG induction in response to IFN-I stimulation 7. This represents an aging-specific post-translational mechanism that functionally decouples IFNAR2 from STAT1, weakening the antiviral IFN-I response in aged tissues even when surface receptor levels are unchanged. O-GlcNAcylation of STAT1 competes with Kbhb modification and preserves receptor–STAT1 coupling; fructose-mediated enhancement of O-GlcNAc on STAT1 partially restored antiviral competence in aged mice 7. needs-replication — single study; human evidence absent. no-fulltext-access — Zuo 2025 full text is closed-access (not_oa per a local paper archive); claims sourced from abstract only.
Bastard 2021: anti-IFN-α/ω autoantibodies as a natural IFNAR2-pathway antagonist
The most striking aging-context finding for IFNAR2 comes not from genetics or pharmacology but from a large population-level study of spontaneously occurring autoantibodies. Bastard et al. (2021) demonstrated that neutralizing autoantibodies against IFN-α2 and/or IFN-ω are present in ~0.18% of individuals aged 18–69 years but rise to ~1.1% at 70–79 years and ~3.4% at >80 years (high-concentration assay; 10 ng/mL; n=34,159 uninfected general-population individuals). By a more sensitive lower-concentration assay (~100 pg/mL, closer to physiological IFN concentrations), overall prevalence in those >80 years reaches ~6.3% 8.
These autoantibodies function as pre-formed soluble IFNAR pathway antagonists: by binding and neutralizing secreted IFN-α/ω in the extracellular space, they prevent IFNAR2 engagement before ligand-receptor contact can occur. The result is a functional IFN-I signaling deficiency in otherwise immunocompetent elderly individuals — invisible to standard clinical immune workups. The same 2021 study showed that these autoantibodies account for approximately 20% of COVID-19 deaths (18.5% of 1,124 deceased patients at the sensitive assay; 13.3% at the stringent assay) 8. Autoantibodies predated SARS-CoV-2 infection in six individuals with pre-pandemic samples available.
Mechanistic paradox of aging. Two opposing states coexist in the aged type I IFN system:
- Chronic tonic IFN-I production (from cGAS-STING, mtDNA leakage, LINE-1 reactivation) → drives inflammaging via IFNAR2 → STAT1/2 → ISG induction
- Autoantibody-mediated extracellular IFN-α/ω neutralization (affects ~4–6% of >70-year-olds) → impairs antiviral IFNAR2 signaling at the ligand-availability level despite normal receptor expression
These two states may overlap in the same individual. The functional consequence of combined chronic intracellular IFN-I signaling with extracellular IFN-I neutralization is not known. no-mechanism
See bastard-2021-anti-ifn-autoantibody-age-prevalence for full methodological details and verified quantitative figures.
Pharmacology
Anifrolumab (IFNAR1-targeting mAb; FDA-approved 2021)
Anifrolumab (Saphnelo; AstraZeneca) is a fully human IgG1κ monoclonal antibody that binds IFNAR1 (not IFNAR2 directly) and blocks the assembly of the functional IFN-α/β receptor heterodimer 9. By occupying IFNAR1 and preventing its association with the IFN-α/β·IFNAR2 binary complex, anifrolumab effectively antagonizes signaling through both receptor subunits — including IFNAR2-dependent JAK1 activation and downstream STAT phosphorylation.
IFNAR1 vs IFNAR2 targeting distinction. Anifrolumab does not bind IFNAR2. However, because IFNAR2 requires IFNAR1 as a co-receptor for signaling (the heterodimer is the functional unit), anifrolumab achieves downstream functional antagonism of IFNAR2-JAK1 signaling by disrupting complex formation. A therapeutic antibody targeting IFNAR2 directly would have the same functional consequence but different binding epitope. No IFNAR2-specific clinical-stage antibody is currently FDA-approved.
Clinical evidence (SLE). The TULIP-2 phase 3 RCT (n=362; 180 anifrolumab 300 mg IV every 4 weeks vs 182 placebo) achieved 47.8% BICLA (British Isles Lupus Assessment Group-based Composite Lupus Assessment) response rate (86/180) vs 31.5% (57/182) for placebo (adjusted difference 16.3 percentage points; 95% CI 6.3–26.3; P=0.001); significant glucocorticoid dose reduction and skin disease improvement 9. FDA approved July 2021 for moderate-to-severe systemic lupus erythematosus (SLE). A 3-year blinded extension study (TULIP LTE; n from TULIP extension cohort) showed no new safety signals, sustained glucocorticoid-sparing effects, and comparable serious adverse event rates between anifrolumab and placebo arms 10. Herpes zoster occurred in 7.2% of treated patients in TULIP-2, consistent with expected IFN-I suppression effects on viral surveillance.
Active trials. As of 2026-05-13, ClinicalTrials.gov lists 20 active or recruiting studies with anifrolumab (RECRUITING or ACTIVE_NOT_RECRUITING). These include Phase 3 interventional trials (subcutaneous formulation SLE — NCT04877691; cutaneous lupus — NCT06015737; proliferative lupus nephritis — NCT05138133; subcutaneous SLE — NCT06455449, NCT05835310, NCT07430306), a Phase 2 cardiovascular-risk biomarker study in SLE (NCT05440422), a Phase 4 post-marketing pregnancy registry (NCT06594068), an exploratory Phase 3 systemic sclerosis study (DAISY; NCT05925803), and multiple non-interventional real-world effectiveness registries. No registered trial has an explicit aging or inflammaging primary endpoint.
Aging-translational candidate status. Anifrolumab’s mechanism — blunting tonic IFNAR-mediated ISG induction — directly targets the molecular driver of type I IFN-driven inflammaging. Off-label use in elderly patients with autoimmune overlap syndromes driven by IFN-I signatures (e.g., dermatomyositis, Aicardi-Goutières-spectrum conditions in elderly) is an emerging investigational area. Dermatomyositis serum activates type I IFN signaling in skeletal muscle via IFNAR1/2; IFNAR1 blockade or JAK1 inhibition with ruxolitinib prevented serum-induced muscle weakness in an ex vivo model (Kaewin et al. 2026, ARD) 11. However, direct evidence that anifrolumab or equivalent IFNAR-targeting agents reduce inflammaging biomarkers (e.g., ISG score, CRP, IL-6) or improve healthspan endpoints in aging humans is absent. needs-human-replication
druggability-tier rationale. Tier 1 is assigned under the aging-context convention (CLAUDE.md): a clinical drug (anifrolumab) exists that directly engages the IFNAR1/IFNAR2 signaling complex and is FDA-approved. Anifrolumab binds IFNAR1 rather than IFNAR2 directly, but blocks IFNAR2-dependent downstream signaling by heterodimer disruption — making the receptor complex effectively tier-1 pharmacologically addressable. A strict per-subunit reading (anifrolumab targets IFNAR1, not IFNAR2) could argue for tier-2 for IFNAR2 specifically; however, the functional consequence for IFNAR2 signaling is identical whether IFNAR1 or IFNAR2 is targeted, and the complex-level pharmacology is the relevant unit for this receptor pair. This choice is documented here per the aging-context convention.
Immunodeficiency-45 (loss-of-function context)
Homozygous or compound-heterozygous loss-of-function variants in IFNAR2 cause Immunodeficiency-45 (OMIM 614172) — autosomal recessive susceptibility to severe viral infections (particularly herpes viruses, viral encephalitis) with relative protection from autoimmune disease. This Mendelian phenotype confirms that IFNAR2 is non-redundant for antiviral IFN-I signaling in humans and establishes the directionality: reduced IFNAR2 function → more severe viral infections, not simply IFN-I pathway attenuation.
Pathway membership
- type-i-interferon-signaling — IFNAR2 is a core node (key-nodes list); the obligate signal-transducing subunit of the IFN-I receptor heterodimer
- jak-stat-pathway — IFNAR2 → JAK1 → STAT1/STAT2 → ISGF3 is the canonical JAK-STAT1 branch for IFN-I
Key interactors
- ifnar1 (implicit stub) — the TYK2-associated low-affinity co-receptor subunit; pairs with IFNAR2 to form the functional heterodimer; anifrolumab’s direct binding target
- jak1 — constitutively bound to IFNAR2 intracellular tail; required for signal competence
- tyk2 — constitutively bound to IFNAR1; cross-phosphorylated with JAK1 upon receptor activation
- stat1 — primary downstream STAT recruited; phosphorylated Tyr701
- stat2 (implicit stub) — STAT2 Tyr689 phosphorylation; heterodimerizes with STAT1 in ISGF3
- irf9 — associates with p-STAT1/p-STAT2 to complete ISGF3 trimeric TF
- usp18 — negative regulator; binds IFNAR2 intracellular tail to displace JAK1; limits sustained signaling; USP18 is an ISG (delayed feedback)
- cgas / sting — upstream inducers of IFN-I ligand production (not direct IFNAR2 interactors, but key upstream nodes in the aging-dysregulation chain)
Limitations and gaps
- sIFNAR2 physiology in aging is poorly characterized. Whether sIFNAR2 plasma levels change with aging, and whether higher sIFNAR2 is net-protective (buffers inflammaging) or net-deleterious (limits antiviral defense), is not established in aged cohorts. needs-human-replication
- STAT1–IFNAR2 uncoupling in aging (Zuo 2025) requires independent replication. The β-hydroxybutyrylation mechanism is a single study in mice; human confirmation is absent. needs-replication
- No aging-indication clinical data for anifrolumab. All clinical evidence is in autoimmune disease (SLE, dermatomyositis). Whether IFNAR-blockade reduces inflammaging biomarkers or extends healthspan in elderly individuals without diagnosed autoimmune disease is untested. needs-human-replication
- IFNAR1 vs IFNAR2 separate protein pages. The functional unit is the heterodimer; mechanistic claims about subunit-specific contributions are partially inferred. ifnar1 lacks its own protein page as of 2026-05-13.
- GTEx aging expression data not populated.
gtex-aging-correlationfield is null; requires SOPsops/finding-tissue-expression.mdto extract tissue-by-age correlation data. not-populated - Mendelian randomization evidence absent. No MR instrument has been published for IFNAR2 germline variation;
mr-causal-evidence: not-tested.
Cross-references
- type-i-interferon-signaling — canonical pathway page; IFNAR2 is listed as a key node; mechanistic details in receptor-signaling section
- jak-stat-pathway — downstream signaling relay
- chronic-inflammation — primary aging-relevant hallmark modulated by IFNAR2
- cellular-senescence — senescent cells produce IFN-I via cGAS-STING; IFNAR2 mediates ISG induction and paracrine senescence spread
- disabled-adaptive-immunity — Bastard 2021 anti-IFN autoantibodies represent acquired IFN-I immunodeficiency in the aged; cross-link via IFNAR2 pathway
- bastard-2021-anti-ifn-autoantibody-age-prevalence — population evidence for anti-IFN-α/ω autoantibody prevalence with aging; verified study page with full quantitative figures
- ifnar1 (implicit stub) — IFNAR1 subunit page; anifrolumab’s direct target
- jak1 — constitutive IFNAR2 intracellular partner
- stat1 — primary signaling effector downstream of JAK1
- usp18 (implicit stub) — critical feedback suppressor of IFNAR2 signaling
- anifrolumab (implicit stub) — anti-IFNAR1 mAb; FDA-approved for SLE 2021; potential inflammaging modulator
Footnotes
Footnotes
-
UniProt P48551 (INAR2_HUMAN), accessed 2026-05-13 via REST endpoint
https://rest.uniprot.org/uniprotkb/P48551· Swiss-Prot (manually curated) entry; isoforms, domain topology, PTMs, and interaction data confirmed. ↩ ↩2 -
doi:10.1016/j.immuni.2012.03.013 · Stark GR, Darnell JE Jr · Immunity 2012 · n=not applicable · review · model: human; canonical review of JAK-STAT signaling mechanism; IFNAR2–JAK1 and IFNAR1–TYK2 constitutive-association assignments confirmed; IFN-β higher-affinity binding than IFN-α confirmed; ISGF3 assembly mechanism (p-STAT1/p-STAT2/IRF9; ISRE motif); PDF locally available at DOI lookup
10.1016/j.immuni.2012.03.013↩ ↩2 ↩3 -
PMID 9208871 · Pestka S · Seminars in Oncology 1997 · review · model: human/general; foundational review documenting three IFNAR2 isoforms (IFNAR2a soluble/P40; IFNAR2b short membrane; IFNAR2c long membrane); DOI not confirmed via archive ↩
-
doi:10.1177/1352458516667564 · Órpez-Zafra T et al. · Multiple Sclerosis Journal 2017 · n=MS patients + controls · observational · model: humans; sIFNAR2 significantly lower in naïve MS and CIS vs healthy controls; IFN-β therapy increased sIFNAR2 within 1 year; archive status pending download ↩
-
doi:10.3389/fimmu.2022.949413 · Fricke-Galindo I et al. · Frontiers in Immunology 2022 · n=severe COVID-19 cohort · observational · model: humans; higher sIFNAR2 plasma levels in COVID-19 survivors vs non-survivors; archive status pending download ↩
-
doi:10.1016/j.cytogfr.2020.04.005 · Bonafè M, Prattichizzo F et al. · Cytokine & Growth Factor Reviews 2020 · review · model: human aging context; proposes that aging-related impaired type I IFN response from chronic inflammation interacts with COVID-19 severity; archive status pending download ↩
-
doi:10.1038/s41423-025-01266-x · Zuo Y, Wang Q et al. · Cellular & Molecular Immunology 2025 · experimental (molecular; aged mouse) · model: aging mice and cell lines; aging-induced STAT1 Kbhb at Lys592 impairs STAT1–IFNAR2 interaction; O-GlcNAc modification on STAT1 competes with Kbhb and preserves receptor–STAT1 coupling; fructose partially restores antiviral IFN-I immunity in aged mice; NOT VERIFIED — paper is closed-access (not_oa per a local paper archive); claims sourced from abstract only no-fulltext-access ↩ ↩2
-
bastard-2021-anti-ifn-autoantibody-age-prevalence · Bastard P, Casanova JL et al. · Science Immunology 2021 · n=40,016 (34,159 uninfected + 3,595 critical COVID-19 + 623 severe COVID-19 + 1,639 mild/asymptomatic) · observational · model: humans; prevalence of neutralizing anti-IFN-α/ω autoantibodies 0.18% in <70y, 1.1% at 70–79y, 3.4% at >80y (high-concentration assay; n=34,159); ~6.3% at >80y by sensitive assay; ~20% of COVID-19 deaths attributed to these autoantibodies; autoantibodies predate infection; PDF locally available at DOI lookup
10.1126/sciimmunol.abl4340↩ ↩2 -
doi:10.1056/NEJMoa1912196 · Morand EF et al. · New England Journal of Medicine 2020 (TULIP-2 trial) · n=362 (180 anifrolumab, 182 placebo; modified ITT) · rct · model: human SLE patients; BICLA response 47.8% (86/180) vs 31.5% (57/182) placebo (adjusted difference 16.3 pp; 95% CI 6.3–26.3; P=0.001); herpes zoster 7.2% treated; bronchitis 12.2% treated; one death from pneumonia in anifrolumab arm; FDA approved July 2021 for SLE; PDF locally available via a local paper archive ↩ ↩2
-
doi:10.1002/art.42392 · Kalunian KC et al. · Arthritis & Rheumatology 2022 (published 2023 in print) · rct (3-year extension) · model: human SLE (TULIP extension cohort); no new safety signals; sustained glucocorticoid-sparing effect; comparable SAE rates between arms; archive status pending download ↩
-
doi:10.1016/j.ard.2026.02.018 · Kaewin S et al. · Annals of the Rheumatic Diseases 2026 · experimental (ex vivo) · model: human dermatomyositis serum + muscle tissue; IFNAR1 blockade or JAK1/2 inhibition (ruxolitinib) prevented dermatomyositis-serum-induced muscle weakness; DOI not yet indexed in archive (2026 publication) ↩