Increased histological severity and chronicity of interface inflammation in anti‐MDA5 dermatomyositis—A comparative dermatopathological analysis

Anti‐melanoma differentiation‐associated gene 5 (MDA5) dermatomyositis is characterized by serological detection of anti‐MDA5 antibody and rapidly progressive interstitial lung disease. In this study, the largest cohort of skin biopsies to date of anti‐MDA5 dermatomyositis was reviewed and compared with cases of dermatomyositis with negative serology. Findings contribute to the histological diagnosis and evaluation of the severity of cutaneous inflammation in anti‐MDA5 dermatomyositis. Skin biopsies collected over a 7‐year period from individuals with clinically and histologically confirmed dermatomyositis with anti‐MDA5 serology were reviewed. A total of 46 cases with 17 anti‐MDA5 positive cases were retrieved. Patients with positive antibody were younger (53.7 vs. 60.6 years, p = .013). No differences in epidermal changes (p > .05) were observed. Pertaining to interface changes, anti‐MDA5 dermatomyositis showed a higher degree of pigmentary incontinence (p = .014), suggesting increased and sustained cutaneous inflammation. Periodic acid–Schiff (PAS) stain demonstrated a greater degree of basement membrane thickening (p = .045). Other parameters, including dermal inflammation, dermal mucin deposition and vasculitic/vasculopathic features did not show statistical difference between anti‐MDA5 positive and negative dermatomyositis (p > .05). Findings suggest increased cutaneous inflammation for anti‐MDA5 dermatomyositis. In skin biopsies, marked pigmentary incontinence or basement membrane thickening should raise suspicion of anti‐MDA5 dermatomyositis.

evaluation of the severity of cutaneous inflammation in anti-MDA5 dermatomyositis.Skin biopsies collected over a 7-year period from individuals with clinically and histologically confirmed dermatomyositis with anti-MDA5 serology were reviewed.
A total of 46 cases with 17 anti-MDA5 positive cases were retrieved.Patients with positive antibody were younger (53.7 vs. 60.6 years, p = .013).No differences in epidermal changes (p > .05)were observed.Pertaining to interface changes, anti-MDA5 dermatomyositis showed a higher degree of pigmentary incontinence (p = .014),suggesting increased and sustained cutaneous inflammation.Periodic acid-Schiff (PAS) stain demonstrated a greater degree of basement membrane thickening (p = .045).
Other parameters, including dermal inflammation, dermal mucin deposition and vasculitic/vasculopathic features did not show statistical difference between anti-MDA5 positive and negative dermatomyositis (p > .05).Findings suggest increased cutaneous inflammation for anti-MDA5 dermatomyositis.In skin biopsies, marked pigmentary incontinence or basement membrane thickening should raise suspicion of anti-MDA5 dermatomyositis.

K E Y W O R D S
anti-MDA5 antibody, anti-MDA5 dermatomyositis, dermatomyositis, dermatopathology, histology

| INTRODUC TI ON
Anti-melanoma differentiation-associated gene 5 (MDA5) dermatomyositis is an emerging entity characterized by serological detection of anti-MDA5 antibody.Patients present with classical cutaneous manifestation of dermatomyositis along with cutaneous ulcerations, and a significant proportion do not have musculoskeletal symptoms and are amyopathic. 1 Recognition and timely diagnosis of anti-MDA5 dermatomyositis are important, as a significant proportion of these patients develop rapidly progressive interstitial lung disease, leading to an increased mortality. 2In this study, the largest cohort of skin biopsies to date of anti-MDA5 dermatomyositis was reviewed and compared with cases of dermatomyositis with negative serology to anti-MDA5 antibody.Findings will contribute to the histological diagnosis of anti-MDA5 dermatomyositis, as well as characterization of the pattern and severity of cutaneous inflammation in anti-MDA5 dermatomyositis.

| RE SULTS
A total of 46 cases were retrieved, consisting of 26 female patients and 20 male patients.All biopsies were obtained from clinically nonulcerated skin lesions, mostly over the upper limbs (n = 15), trunk (n = 11), and neck (n = 5).There were 17 cases with positive anti-MDA5 antibody and 29 cases with negative serology.In anti-MDA5 antibody positive cases, a younger average age (53.7 vs. 60.6 years, p = .013)was observed, but no difference was seen in the patient sex distribution (Table 1).
Comparison regarding epidermal changes, including presence of a basketweave cornifying layer, parakeratosis, hyperkeratosis, epidermal hyperplasia/thinning and spongiosis, demonstrated no significant differences between anti-MDA5 antibody positive and negative cases (p > .05)(Table 2, Figure 1A-C).As for histological parameters related to interface changes, there were no differences for the presence of necrotic keratinocytes, degree of basal vacuolar alteration, or interface smudging (p > .05),but anti-MDA5 antibody positive cases showed a higher degree of pigmentary incontinence (p = .014)(Table 2, Figure 1D).Dermal histological parameters, including presence of dermal edema, telangiectasia, and extravasated red cells, or the amount/composition of dermal inflammatory infiltrates (lymphocytes, plasma cells, and eosinophils) did not demonstrate statistical differences (p > .05)(Table 2, Figure 2A,B).
For histochemical stains, the degree of basement membrane thickening assessed on PAS stain was greater (thicker) for anti-MDA5 antibody positive cases (p = .045).There was no difference for the degree of dermal mucin deposition on AB stain (p = 1) (Table 3, Figure 3A-D).

| DISCUSS ION
Anti-MDA5 dermatomyositis is a subgroup of dermatomyositis characterized by anti-MDA5 antibody serology. 4,5Anti-MDA5 dermatomyositis presents with the characteristic cutaneous manifestation of dermatomyositis with periorbital heliotrope rash, poikiloderma over chest (V sign) and shoulder (shawl sign), papules over dorsal phalangeal joints (Gottron's papules), and digital flexoral papules (inverse Gottron's sign). 1,5,6Additionally, palmar papules and auricular involvement, 7,8 and clinical signs of vasculopathy, including Raynaud's phenomenon, ulceration, splinter hemorrhage, and digital necrosis, are associated with anti-MDA5 dermatomyositis. 2 Musculoskeletal involvement of anti-MDA5 dermatomyositis is uncommon or can even be absent clinically (i.e., amyopathic). 1,9tramuscular/ cutaneous manifestations, in particular pulmonary and hepatic involvement, are not uncommon.Severe cardiomyopathy has also been reported. 10 for muscle biopsy, compatible with clinical findings of lack of muscle involvement, histological changes are less prominent in anti-MDA5 antibody positive dermatomyositis patients.Inflammation is focal and to a lesser degree, whereas other features such as fiber atrophy, capillary loss, and tubuloreticular formations are less frequent or absent. 2 Literature on the histology of skin biopsy for anti-MDA5 dermatomyositis is limited.There are isolated case reports describing interface inflammation, the hallmark dermatopathological pattern of dermatomyositis, in skin biopsies of anti-MDA5 dermatomyositis. 11,12Vasculitic and vasculopathic changes are also observed, from individual reports 10,13 and a series of five cases by Shakshouk et al. 14 However, those biopsies were taken from clinically livedoid or ulcerated lesions. 10,13e current cohort reviewed, to date, represents the largest number of skin biopsies from anti-MDA5 antibody positive dermatomyositis with comparison to anti-MDA5 antibody negative cases.
In as ulceration precludes histological assessment of interface changes, but is known to be common in anti-MDA5 dermatomyositis. 15travasated red blood cells, a histological sign associated with with basal vacuolar alteration, interface smudging, and the presence of necrotic keratinocytes, 16 among which there were no statistical differences between the two groups (p > .05).In contrast to the other histological features, the development of pigmentary incontinence requires sustained interface inflammation, 17 involving the release (damage) of melanosomes from junctional melanocytes, and transfer of the melanosomes to the dermis with or without phagocytosis by dermal melanophages. 18Similarly, as demonstrated by PAS stain, 19 increased basement membrane thickening is observed in anti-MDA5 dermatomyositis.Basement membrane thickening is seen in many forms of interface dermatitis and is associated with chronicity. 20,21Both findings signify intense and persistent inflammation at the dermo-epidermal junction in anti-MDA5 dermatomyositis.Conversely, differences were not observed for basal vacuolar alteration, interface smudging and necrotic keratinocytes, which can be acute changes that indicate interface inflammation, but not chronicity or persistency.

| CON CLUS ION
Skin biopsies of anti-MDA5 positive and negative dermatomyositis were characterized by interface inflammation.In cases of anti-MDA5 dermatomyositis, more pronounced pigmentary incontinence (p = .014)and basement membrane thickening (p = .045)were seen, supporting the theory that there is intense and persistent inflammation at the dermo-epidermal junction in anti-MDA5 antibody positive dermatomyositis patients.These findings suggest that there is increased cutaneous inflammation for anti-MDA5 dermatomyositis compared with its anti-MDA5 antibody negative counterpart.As for histological assessment of skin biopsies, marked pigmentary incontinence or basement membrane thickening should raise suspicion of anti-MDA5 dermatomyositis.

A
computerized search of the hospital pathology archives from the year 2015 to 2022 for skin biopsies with a clinical hospital diagnosis code of dermatomyositis was performed.Case notes of the respective patients were reviewed for demographic data and anti-MDA5 serology results.The clinical diagnosis of dermatomyositis was based on the European Alliance of Associations for Rheumatology (EULAR) diagnostic criteria. 3Myositis autoantibody panel was performed in a laboratory using a commercial line immunoblot assay (EUROLINE autoimmune inflammatory myopathies 16 Ag; EUROIMMUN).Histology slides, including hematoxylin & eosin (H&E), periodic acid-Schiff (PAS), and Alcian blue (AB) were retrieved and assessed by two independent pathologists for histological parameters including epidermal and dermal changes, and histochemical (PAS and AB) staining.Discrepancies were resolved by viewing the slides on a multiheaded microscope until a consensus was reached.Cases were excluded if (1) anti-MDA5 serology was not performed, (2) histological features incompatible with dermatomyositis were present, or (3) biopsy was performed on clinically ulcerated or vasculitic lesions.Chi-squared test, Fisher exact test and t test were used for continuous and categorical variables against anti-MDA5 serology results.Descriptive and comparative statistical analyses were performed using Microsoft Excel 2021 and SPSS version 23.0 (IBM).A p value of <.05 was considered significant.The study was approved by the Joint Chinese University of Hong Kong-New Territories East Cluster Clinical Research Ethics Committee with the waiver of requirement for written consent (Reference number: 2020.065).

2 3
cutaneous vasculitis, was not seen at different rates in anti-MDA5 antibody positive and negative cases.Other frank vasculitic or vascuolopathic changes, including fibrinoid necrosis of vessel walls, intravascular fibrin thrombi, and neutrophilic vascular infiltration were not identified.Of all histological parameters compared, only pigmentary incontinence (p = .014)and basement membrane thickening (p = .045)showed a significant difference between anti-MDA5 antibody F I G U R E 1 Epidermal and interface changes in dermatomyositis.(A) Thinning of epidermis, hematoxylin & eosin (H&E), magnification.(B) Necrotic keratinocytes, as indicated by deeply eosinophilic bodies in the epidermis, H&E, 200× magnification.(C) Marked basal vacuolar alteration with apparent vacuoles at the dermo-epidermal junction, H&E, 400× magnification.(D) Marked pigmentary incontinence with melanin deposits in the superficial dermis, H&E, 200× magnification.Dermal changes in dermatomyositis.(A) Dermal edema, hematoxylin & eosin (H&E), 100× magnification.(B) Perivascular lymphoplasmacytic infiltrates, H&E, 200× magnification.Comparison of histochemical staining between anti-MDA5 antibody positive and negative dermatomyositis.Positive Negative p value Periodic acid-Schiff (basement membrane thickening) dermatomyositis cases.Pigmentary incontinence is one of the hallmark features of interface dermatitis, along

TA B L E 1
Demographics of the cohort.
Comparison of histological features anti-MDA5 antibody positive and negative dermatomyositis.
cerated or vasculitic/vasculopathic lesions itself is indicative of anti-MDA5 dermatomyositis.Identification of key differentiating features between anti-MDA5 positive and negative biopsies on histology is of importance and has diagnostic value, especially in anti-MDA5 dermatomyositis patients who present without clinical vasculopathic manifestation.Ulcerated cases were excluded for this study TA B L E 2 CHAN et al.