Transition From Juvenile Dermatomyositis to Spondyloarthritis: A Novel Overlapping Inflammatory Phenotype

Abstract

Objective

Juvenile dermatomyositis (JDM) is a rare childhood inflammatory myopathy, whereas spondyloarthritis (SpA) is an inflammatory arthropathy characterized by enthesitis and peripheral or axial involvement. We describe a series of patients diagnosed with JDM in childhood who later fulfilled classification criteria for SpA, a sequential phenotype that has not been well characterized.

Methods

This institutional review board–exempt retrospective study (2024‐180) reviewed demographics, clinical features, laboratory tests, imaging, treatments, and outcomes in a convenience sample of patients who met criteria for JDM before age 18 and subsequently fulfilled the Assessment of Spondyloarthritis International Society (ASAS) or Classification for Psoriatic Arthritis (CASPAR) criteria. Patients with SpA preceding JDM were excluded. Descriptive statistics were used to explore potential predictors of SpA development.

Results

Seven patients met the inclusion criteria; 57% were female and 86% were White. The median age at JDM diagnosis was 8 years (interquartile range [IQR] 5–10 years). The median interval time to SpA diagnosis was 7 years (IQR 5–14 years). All tested patients had elevated neopterin and magnetic resonance imaging evidence of myositis; 71% experienced complications before SpA onset. All met ASAS criteria for peripheral SpA, and 29% met CASPAR criteria. HLA‐B27 was positive in 20% of those tested. Most (83%) responded to standard SpA therapies; one required rituximab and azathioprine for calcinosis.

Conclusion

This preliminary small study highlights a possible evolution from JDM, an adaptive immune disease, to SpA, an innate‐driven illness. These findings suggest potential predictors of disease overlap and warrant future studies to clarify mechanisms and inform predictive models.

INTRODUCTION

Juvenile dermatomyositis (JDM), within the spectrum of juvenile idiopathic inflammatory myopathies (JIIMs), is a rare systemic autoimmune disease of childhood characterized by symmetric myositis and characteristic cutaneous findings, with potential pulmonary, gastrointestinal, cardiovascular, and central nervous system involvement. ¹ , ² , ³ , ⁴ It has an estimated incidence of approximately 2–4 per million, with a median age at diagnosis near seven years. ⁵ , ⁶ Overall, JIIM pathogenesis reflects an interplay between innate and adaptive immunity affecting muscle, skin, and vascular tissues. ¹ , ² , ³ Environmental exposures, genetic risk, and type I interferon (IFN) signaling have all been implicated in initiating and sustaining inflammation in JDM. ⁷ , ⁸ , ⁹ , ¹⁰ , ¹¹ , ¹²

A key mechanistic framework in JDM centers on a type I IFN–linked inflammatory cascade, with up‐regulation of major histocompatibility complex class I expression, dendritic cell activation, lymphocyte trafficking into muscle, and reinforcement loops that promote ongoing inflammation and autoantibody production. ¹² , ¹³ Consistent with this biology, approximately 60% of patients demonstrate myositis‐specific autoantibodies (eg, anti‐Mi2, anti‐NXP2, anti‐TIF1γ, and anti‐MDA5), supporting clinically meaningful immunologic heterogeneity within JDM. ¹⁴ , ¹⁵ , ¹⁶ Although mortality has declined in modern cohorts, medium‐ and long‐term morbidity remains substantial: functional impairment and cumulative damage are common, emphasizing the importance of understanding long‐term trajectories beyond initial myositis control. ¹⁷ , ¹⁸ , ¹⁹ , ²⁰

Spondyloarthritis (SpA) comprises a family of inflammatory disorders, including axial SpA, peripheral SpA, and psoriatic arthritis (PsA), unified by enthesitis, synovitis, and characteristic imaging features. ²¹ , ²² , ²³ In the United States, SpA affects roughly 0.9% to 1.4% of adults, and diagnosis is frequently delayed. ²⁴ , ²⁵ , ²⁶ Contemporary classification integrates imaging (magnetic resonance imaging [MRI] or radiographs) and/or an HLA‐B27–based clinical arm in the Assessment of Spondyloarthritis International Society (ASAS) framework. ²⁴ , ²⁵ , ²⁶ , ²⁷ Notably, nonradiographic axial SpA can exhibit clinically significant disease activity despite lower C‐reactive protein (CRP) and less active MRI inflammation than radiographic disease. ²⁸ PsA further expands phenotypic breadth, and peripheral or axial manifestations may coexist. ²³ The Classification for Psoriatic Arthritis (CASPAR) criteria provide highly specific classification anchors that permit PsA classification even in the absence of current psoriasis when other characteristic features are present. ²⁹

Because early recognition of SpA is probabilistic, additive approaches using likelihood ratios have been proposed to estimate axial SpA probability in clinical practice; additionally, MRI of the sacroiliac joints can carry substantial diagnostic weight in early disease. ³⁰ Mechanistically, SpA is increasingly conceptualized along an autoimmunity–autoinflammation continuum involving innate and innate‐like IL‐17 sources within interconnected gut and enthesis ecosystems. ³¹ In this context, the sequential emergence of an SpA phenotype after childhood‐onset JDM is biologically plausible but poorly described. In this retrospective case series, we characterize patients meeting stringent JDM criteria in childhood who later fulfilled ASAS criteria (with a subset meeting CASPAR), aiming to (i) describe clinical, serologic, imaging, and treatment features across both disease phases and (ii) generate testable hypotheses regarding potential predictors and immunologic intermediates underlying a JDM to SpA transition. ³² , ³³ , ³⁴ , ³⁵ , ³⁶

METHODS

Objectives

The primary objective was to compile a case series of patients diagnosed with JDM who subsequently developed clinical features of SpA, and to comprehensively describe their clinical presentations, biologic markers, imaging findings, and responses to therapy.

Criteria

This institutional review board–exempt (2024‐180) retrospective study collected demographic information, clinical symptoms, laboratory findings, imaging, treatments, and outcomes from a convenience sample of patients with JDM who subsequently developed SpA. Patients selected for the study received medical treatment at Corewell Health William Beaumont University Hospital or Corewell Health Grand Rapids Hospital Butterworth.

Inclusion criteria included patients who met diagnostic criteria for JDM before age 18 and who later developed SpA according to ASAS and CASPAR classification criteria. Exclusion criteria were patients who developed SpA before JDM or who did not meet ASAS or CASPAR criteria.

Data collection

Data collection was completed using a standardized form created by two authors (AG and IK). Patients were identified using their medical record number (MRN) for data collection purposes. A linking key document listing patients’ MRNs to patient numbers was used.

Data collected upon JDM manifestation included sex, date of birth, date of disease onset, duration of follow‐up, family history of myositis or other rheumatic diseases, site of muscle weakness, cutaneous skin findings, other unique systemic symptoms present at time of onset, markers of inflammation at time of onset, biologic markers of disease, muscle biopsy results, radiographic imaging, treatments received, and therapeutic response (complete, partial, or failure).

Data collected for the same patients during SpA manifestations included the date of disease onset, duration of JDM up to SpA onset, clinical examination findings of joints with arthritis, clinical examination findings of tendons with enthesitis, psoriatic skin findings, markers of inflammation at time of onset, any biomarkers positive on disease onset, HLA‐B27 status, radiologic imaging, fulfillment of axial or peripheral ASAS, fulfillment of CASPAR criteria, treatments received and therapeutic response (complete, partial, or failure).

Markers of inflammation included elevated serum levels of creatine kinase (CK), erythrocyte sedimentation rate (ESR), CRP, lactate dehydrogenase (LDH), aldolase, neopterin, aspartate aminotransferase (AST), and alanine aminotransferase. Biomarkers included anti‐Mi2 antibody, NXP2 antibody, TIF1 antibody, MDA5 antibody, SAE antibody, immune‐mediated necrotizing myositis antibody, signal recognition particle, 3‐hydroxy‐3‐methylglutaryl‐coenzyme A reductase, and HLA‐B27. Complete remission under treatment was defined as the absence of any clinical and biologic signs of the disease (ie, the disappearance of muscle weakness, resolution of skin findings, and normalization of markers of inflammation). Partial remission was defined by improvement in symptoms and markers of inflammation, but with some physical features persisting after treatment started. Treatment failure was considered if symptoms remained unchanged during treatment. Complete remission, partial remission, and failure to treatment were assessed by a dual‐certified American Board of Internal Medicine & American Board of Pediatrics certified rheumatologist (JB) who oversaw treatment of each patient. A large language model, ChatGPT, was used for checking spelling and grammar.

Statistical analysis

Statistical analysis was performed by biostatisticians within the Corewell Health Research Institute. Descriptive statistics were used to examine demographic, clinical, and laboratory data from patients upon JDM onset and SpA onset. Descriptive statistics were used due to the small sample size (n = 7). Chi‐square and logistic regression were not used because the study was statistically underpowered. Instead, proportions, medians, and interquartile ranges (IQR) were reported to highlight patterns.

RESULTS

Demographics

A total of seven patients met the inclusion criteria (Table 1); 57% (4/7) were female and 86% (6/7) were White. Four patients (57%) had a family history of autoimmune disease (thyroiditis, sarcoidosis, rheumatoid arthritis, and Raynaud syndrome). Age at JDM onset had a median of 8 years (IQR 5–10 years). Age at SpA onset had a median of 20 years (IQR 16–22 years). The interval from JDM onset to SpA diagnosis had a median of 7 years (IQR 5–14 years). Physician follow‐up had a median of 14 years (IQR 13–15 years).

Table 1.

Patient demographics^*

	Case 1	Case 2	Case 3	Case 4	Case 5	Case 6	Case 7
Sex	Male	Male	Female	Female	Female	Female	Male
Ethnicity	White	White	White	White	White	White	Hispanic
Current age, y	26	38	20	26	9	21	34
Age at onset of JDM, y	10	8	5	13	2	6	8
Age at onset of SpA, y	16	22	18	20	7	20	33
Follow‐up, y	16	a	15	13	7	14	a
Time from JDM to SpA, y	6	17	13	7	5	14	2
Family history	None	None	None	Hashimoto thyroiditis, RA, Crohn disease	Raynaud syndrome, sarcoidosis, RA	Graves disease	None

^*JDM, juvenile dermatomyositis; RA, rheumatoid arthritis; SpA, spondyloarthritis.

^a Lost to follow‐up until onset of SpA symptoms.

Juvenile dermatomyositis features

All patients met diagnostic criteria for JDM (Table 2). At presentation, heliotrope rash was present in 43% (3/7) and Gottron papules in 86% (6/7). Symmetric proximal weakness occurred in 100% (7/7), MRI findings consistent with myositis were present in 100% (7/7), muscle enzyme elevation was documented in 57% (4/7), and muscle biopsy supported JDM in 57% (4/7).

Table 2.

Features at onset of JDM diagnosis^*

	Case 1	Case 2	Case 3	Case 4	Case 5	Case 6	Case 7
Site of muscle weakness	Shoulders, hips	Shoulders, hips	Cervical, shoulders, hips	Cervical, shoulders, hips, ankles	Shoulders, hips	Shoulders, hips	Hips
Cutaneous findings	Gottron sign	Gottron sign, Raynaud syndrome, telangiectasia	Gottron sign	Heliotrope rash, Shawl sign, Gottron sign, Raynaud syndrome	Heliotrope rash, Shawl sign	Gottron sign, heliotrope rash, oral ulcers	Gottron sign
Other systematic symptoms	Abdominal pain, fever	Photosensitivity	Arthritis of knees, abdominal pain, fever	Arthritis of wrists and ankles, abdominal pain	Abdominal pain	Fever, pancytopenia	None
Muscle biopsy positive for JDM	Yes	NA	Yes	Yes	Yes	Insufficient sample	NA
MRI consistent for myositis	Yes	Yes	Yes	Yes	Yes	Yes	Yes

^*JDM, juvenile dermatomyositis; MRI, magnetic resonance imaging; NA, not available.

Two patients (cases 4 and 5) had pelvis images on their MRI images. Case 4 had subchondral edema and narrowing bilaterally of their sacroiliac joints. Case 5 did not have abnormal findings of their pelvis or sacroiliac joints. Two patients (cases 2 and 7) had incomplete documentation from the time of initial JDM diagnosis due to institutional record‐retention limits; therefore, laboratory values and biopsy data at JDM onset were not available for those patients. Muscle biopsy was not confirmatory in three patients: one biopsy was nondiagnostic due to insufficient tissue sampling, and two patients did not have biopsy results available in the record. Additional systemic features at onset were documented in 85% (6/7), including abdominal pain (4/7), arthritis (2/7), fever >38°C (3/7), photosensitivity (1/7), and pancytopenia (1/7).

Among patients with available enzyme data (n = 5), aldolase was elevated in 80% (4/5), CK in 40% (2/5), LDH in 80% (4/5), and AST in 80% (4/5) (Table 3). Neopterin was elevated in 100% (4/4) of those tested. Autoantibody testing was performed in 86% (6/7). Antinuclear antibody (ANA) positivity was observed in 50% (3/6), and a myositis‐specific antibody was identified in one patient (NXP2, 1/6); the remaining tested patients did not have a documented myositis‐specific antibody identified in their records (Table 3).

Table 3.

Laboratory findings at onset of JDM^*

	Case 1	Case 2	Case 3	Case 4	Case 5	Case 6	Case 7
Aldolase (0–14.4 U/L)	12.2	NA	18.3	341	7.8	20.8	NA
Neopterin (2.0–10.0 nmol/L)	33.7	NA	25	50	NA	26.6	NA
ESR (0–15 mm/h)	8	NA	17	12	2	19	NA
CRP (0.0–9.9 mg/L)	0.3	NA	0.2	0.9	0.3	1	NA
AST (10–40 U/L)	61	NA	88	363	38	346	NA
ALT (10–40 U/L)	56	NA	86	264	18	156	NA
Elevated CK (age dependent, U/L)	No (82)	NA	Yes (324)	Yes (17,194)	No (172)	No (71)	NA
Elevated LDH (age dependent, U/L)	Yes (383)	NA	Yes (409)	Yes (1,130)	No (305)	Yes (436)	NA
Autoantibodies	ANA, anti‐Jo‐1, RF, SCL‐70 (−)	NA	ANA (+)	ANA, anti‐Jo‐1, centromere, chromatin (+), SSA/SSB (−)	ANA (+)	ANA (−)	ANA (−), NXP2 (+)

^*CK range: case 1 (30–225 U/L), case 3 (20–175 U/L), case 4 (20–175 U/L), case 5 (20–175 U/L), and case 6 (20–175 U/L); LDH range: case 1 (120–300 IU/L), case 3 (165–340 U/L), case 4 (120–300 U/L), case 5 (180–380 U/L), and case 6 (165–340 U/L). ALT, alanine aminotransferase; ANA, antinuclear antibody; AST, aspartate aminotransferase; CK, creatine kinase; CRP, C‐reactive protein; ESR, erythrocyte sedimentation rate; JDM, juvenile dermatomyositis; LDH, lactate dehydrogenase; NA, not available; RF, rheumatoid factor.

Initial treatment and responses are summarized in Table 4. All patients received initial combination therapy including hydroxychloroquine, methotrexate, and prednisone. Most patients achieved complete remission on standard therapy or with treatment escalation when needed (eg, intravenous immunoglobulin (IVIG) for persistent rash in one case; rituximab for JDM‐associated complications in one case), with additional patient‐level treatment details provided in Table 4.

Table 4.

Treatment course for JDM and SpA cases^*

	Case 1	Case 2	Case 3	Case 4	Case 5	Case 6	Case 7
JDM Dx (response)	HCQ, MTX, PDN (CR)	HCQ, MTX, PDN a	HCQ, MTX, PDN (CR)	HCQ, MTX, PDN (PR)	HCQ, MTX, PDN (CR)	HCQ, MTX, PDN (PR)	HCQ, MTX, PDN a
JDM subsequent Tx (response)	HCQ, meloxicam (CR)	AZA (PR)	HCQ, meloxicam (CR)	RTX for ILD (CR); HCQ, MTX, PDN (CR)	HCQ, MTX, meloxicam (CR)	IVIG (CR); HCQ, MTX, meloxicam (CR)	a
SpA at Dx	HCQ, meloxicam	AZA	HCQ, meloxicam	HCQ, MTX, PDN	HCQ, MTX, meloxicam	HCQ, meloxicam	HCQ, MTX, PDN
Initial Tx (response)	ETN, HCQ, meloxicam (PR)	RTX, AZA (CR)	ETN, HCQ, meloxicam (CR)	GOL, HCQ, meloxicam, MTX (PR)	TOF, HCQ, MTX, meloxicam (CR)	HCQ, MTX, meloxicam (FR)	ETN, HCQ, MTX, PDN (CR)
Subsequent Tx (response)	ADA, HCQ, meloxicam (CR)	–	–	ABA, HCQ, MTX, meloxicam (PR); UPA, HCQ, MTX, meloxicam (CR)	–	ENT, HCQ, meloxicam (CR)	–

^*ABA, abatacept; ADA, adalimumab; AZA, azathioprine; CR, complete response; Dx, diagnosis; ETN, etanercept; FR, failed response; GOL, golimumab; HCQ, hydroxychloroquine; ILD, interstitial lung disease; IVIG, intravenous Ig; JDM, juvenile dermatomyositis; MTX, methotrexate; PDN, prednisone; PR, partial response; RTX, rituximab; SpA, spondyloarthritis; TOF, tofacitinib; Tx, treatment; UPA, upadacitinib.

^a Lost to follow‐up.

Before the onset of SpA symptoms, 71% (5/7) developed additional complications (Table 5), including calcinosis and squamous cell carcinoma, Hashimoto thyroiditis, interstitial lung disease, septic shock with adrenal insufficiency, and esophagitis.

Table 5.

Features at onset of SpA diagnosis^*

	Case 1	Case 2	Case 3	Case 4	Case 5	Case 6	Case 7
Complications before SpA	None	Calcinosis, squamous cell carcinoma	Hashimoto thyroiditis	ILD	None	Septic shock, adrenal insufficiency	Esophagitis
Arthritis (swollen joint count)	Yes (6)	Yes (6)	Yes (18)	Yes (6)	Yes (4)	Yes (12)	Yes (2)
Enthesitis (enthesitis count)	Yes (5)	Yes (2)	Yes (6)	Yes (4)	Yes (4)	Yes (11)	Yes (4)
Inflammatory back pain	Yes	No	Yes	Yes	No	Yes	Yes
Current skin psoriasis	No	No	No	Yes	Yes	No	No
Sacroiliitis on imaging	Yes	NA	NA	Yes	NA	No	NA
Meets ASAS axial criteria	Yes	No	No	Yes	No	No	No
Meets ASAS peripheral criteria	Yes	Yes	Yes	Yes	Yes	Yes	Yes
Meets CASPAR PsA criteria	No	No	No	Yes	Yes	No	No

^*ASAS, Assessment of Spondyloarthritis International Society Criteria; CASPAR, Classification for Psoriatic Arthritis; ILD, interstitial lung disease; PsA, psoriatic arthritis; SpA, spondyloarthritis.

Spondyloarthropathy features

SpA features at onset for each case are summarized in Table 5. Arthritis and enthesitis were present in 100% (7/7) at SpA onset. The mean arthritis count was 7.7, and the mean enthesitis count was 5.1. Inflammatory back pain was present in 71% (5/7), and psoriasis in 29% (2/7). Among those who underwent sacroiliac imaging at onset (n = 3), sacroiliitis was present in 67% (2/3). Case 1 had newly diagnosed sacroiliitis, and case 4 had slightly improved subchondral edema from prior imaging. All patients met ASAS criteria for peripheral SpA (7/7), whereas 29% (2/7) met ASAS axial criteria and 29% (2/7) met CASPAR criteria.

Laboratory data at SpA onset were available for six patients (Table 6); data for case 2 was unavailable due to record‐retention limits. At SpA onset, aldolase was elevated in 17% (1/6), CK in 33% (2/5), LDH in 50% (3/6), and AST in 17% (1/5), whereas ESR and CRP were not elevated among those tested (ESR 0/5; CRP 0/5). HLA‐B27 was positive in 20% (1/5) of those tested.

Table 6.

Initial laboratory results at SpA diagnosis^*

	Case 1	Case 2	Case 3	Case 4	Case 5	Case 6	Case 7
Aldolase (0–14.4 U/L)	8	NA	2	5.1	5	4.4	15.8
ESR (0–15 mm/h)	2	NA	7	7	1	NA	8
CRP (0.0–9.9 mg/L)	NA	NA	<3.0	1.4	<0.3	<0.3	4.5
AST (10–40 U/L)	14	NA	17	21	34	25	64
ALT (10–40 U/L)	22	NA	14	16	20	23	40
Elevated CK (age dependent, U/L)	Yes (347)	NA	No (68)	No (68)	No (156)	No (97)	Yes (573)
Elevated LDH (age dependent, U/L)	No (176)	NA	No (121)	Yes (245)	Yes (286)	NA	Yes (334)
HLA‐B27	Negative	Positive	Negative	NA	Negative	Negative	NA

^*CK range: case 1 (30–225 U/L), case 3 (20–175 U/L), case 4 (20–175 U/L), case 5 (26–180 U/L), case 6 (29–143 U/L), and case 7 (30–225 U/L); LDH range: case 1 (100–220 U/L), case 3 (100–220 U/L), case 4 (100–220 U/L), case 5 (140–280 U/L), and case7 (100–220 U/L). ALT, alanine aminotransferase; AST, aspartate aminotransferase; CK, creatine kinase; CRP, C‐reactive protein; ESR, erythrocyte sedimentation rate; LDH, lactate dehydrogenase; NA, not available; SpA, spondyloarthritis.

Treatment and response

Treatments and responses are summarized in Table 4. In brief, most patients improved with standard SpA‐directed therapy, most commonly tumor necrosis factor (TNF) inhibitors; selected patients required therapy changes due to incomplete response or adverse effects; and two patients improved with JAK inhibition. Patient‐level treatment sequences and outcomes are provided in Table 4.

DISCUSSION

We describe a sequential phenotype in which children who met stringent criteria for JDM later developed SpA after a median of 7 years (IQR 5–14 years). At JDM onset, MRI‐confirmed myositis was universal, and neopterin was elevated in all tested patients, consistent with type I IFN–skewed dermatomyositis biology. ¹ , ² , ³ , ⁴ At JDM onset, patients were evaluated for arthritis and enthesitis; two patients (cases 3 and 4) had peripheral arthritis, whereas none had evidence of enthesitis. It is not uncommon for patients with myositis to have arthritis at the onset of disease. ³⁷ Later emergence of arthritis, enthesitis, inflammatory back pain, and imaging resulting in meeting ASAS/CASPAR criteria, however, is suggestive of a change in disease phenotype. Among the two patients whose MRIs included pelvic imaging, one demonstrated bilateral sacroiliac subchondral edema and joint space narrowing, whereas the other showed no evidence of sacroiliitis. Although incidental early sacroiliac inflammation cannot be excluded in patients without pelvic imaging, the absence of inflammatory back pain or enthesitis at JDM onset in the cohort suggests clinically that SpA manifestations generally emerged later.

At SpA onset, every patient had enthesitis with peripheral arthritis on clinical examination and met ASAS peripheral SpA criteria; 28.5% met CASPAR and 28.5% met ASAS axial criteria. Psoriasis was present in 28.5%, and sacroiliitis was observed by imaging in two of three cases. Most patients improved with TNF inhibitors and, in selected cases, JAK inhibition; one patient received rituximab for calcinosis. HLA‐B27 positivity (1/5 tested) was lower than expected for axial SpA but is compatible with predominantly peripheral/enthesis‐predominant disease, and reinforces that HLA‐B27 is neither necessary nor sufficient for SpA classification. ²¹ , ²⁵ , ²⁶ , ³¹ Overall, these descriptive findings support a clinically observed JDM to SpA sequence in a subset of patients; however, mechanistic conclusions should remain cautious in a small retrospective series. ¹ , ² , ³ , ⁴ , ³¹ , ³² , ³³ , ³⁴

Long‐term follow‐up is essential in JDM given the frequency of relapsing/chronic disease and cumulative damage. In the retrospective inception cohort study by Feldman et al, 63% experienced flares over 13.9 years, and calcinosis/chronic continuous disease predicted worse function. ¹⁸ In our cohort, 71% developed additional complications before SpA (calcinosis, interstitial lung disease, Hashimoto thyroiditis, severe infection with adrenal insufficiency, esophagitis). Universal MRI‐proven myositis and frequent neopterin elevation are consistent with a strong IFN‐1 signature and vasculopathy in dermatomyositis, and may justify long‐term surveillance beyond the initial myositis phase. ⁸ , ¹³

Overlap across inflammatory myopathy, psoriasis, and SpA spectrum disease has precedent, although the pathogenesis is unclear and may reflect shared genetic risk, environmental triggers, and/or immune abnormalities rather than a single unifying pathway. ¹ Ge and He reported axSpA patients who later developed idiopathic inflammatory myopathy roughly a decade after SpA onset, mirroring our reverse sequence and supporting the possibility of time‐dependent overlap across the muscle–enthesis–skin axis. ³⁶ Co‐occurrence of dermatomyositis and psoriasis is uncommon but biologically plausible, and management can be challenging because some therapies for one domain may worsen another in select contexts. ³⁸ , ³⁹ , ⁴⁰ , ⁴¹ In our series, psoriasis occurred in 28.5% of patients at SpA onset, and most responded to TNF inhibitors or JAK inhibitors without obvious myositis destabilization, but careful skin and muscle monitoring remains prudent. Population‐level data also suggest nonrandom comorbidity patterns in ankylosing spondylitis (eg, uveitis, psoriasis, Behçet disease, and sarcoidosis), supporting targeted screening as SpA biology emerges. ⁴²

From a practical standpoint, SpA can be viewed along an autoimmunity–autoinflammation continuum with important roles for IL‐17 biology, and the relative contributions of pathways are likely to vary across axial, peripheral, and psoriatic phenotypes. ³¹ , ³² , ³³ , ³⁴ , ³⁵ This framework aligns with the favorable responses observed here to TNF inhibition (a cornerstone of axSpA care) ³⁶ , ³⁸ and the clinical benefit seen with JAK inhibition in some patients, which is biologically plausible given JAK‐STAT involvement in IFN‐driven inflammation and downstream inflammatory circuits. ¹ , ²¹ At the same time, the failure of IL‐23p19 blockade in ankylosing spondylitis despite efficacy in PsA underscores heterogeneity within the SpA spectrum and cautions against overgeneralized mechanistic conclusions. ³² Accordingly, when patients transition from JDM to SpA, therapy should prioritize agents with established SpA efficacy (eg, NSAIDs and TNF/IL‐17/JAK inhibitors, as appropriate) whereas remaining alert to dermatologic and myopathic trade‐offs described in overlap settings. ¹ , ²¹ , ³⁸

All cases developed peripheral SpA features in late adolescence/young adulthood—an interval when SpA recognition is often delayed. ²⁶ We suggest routine screening for enthesitis and inflammatory back pain in JDM survivors and a low threshold for sacroiliac MRI when features accumulate, given its diagnostic weight early in disease. ²⁶ , ³⁰ For patients classified as nonradiographic axial SpA, counseling should emphasize that radiographic progression is not inevitable, and management should be guided by symptoms and objective inflammation rather than radiographic change alone. ²⁸

Strengths of this study include a clearly defined sequential phenotype, dual‐phase characterization using standardized classification anchors (ASAS and CASPAR), and treatment outcomes over long follow‐up. Limitations include small sample size, retrospective convenience sampling, descriptions of arthritis and enthesitis from clinical examination findings without image confirmation studies, incomplete serology/HLA‐B27 testing, limited sacroiliac imaging on JDM and SpA onset, and missing historical data due to record‐retention policies; causality cannot be inferred. Alternative explanations (eg, shared predisposition producing two independent conditions) remain possible. Still, the recurrent pattern in our cohort—enthesitis at SpA onset in all cases, sacroiliitis in two of three patients who underwent sacroiliac imaging, and elevated neopterin among those tested during the JDM phase (4/4)—supports prospective studies to define predictors and the frequency of a potential JDM to SpA phenotype.

AUTHOR CONTRIBUTIONS

All authors contributed to at least one of the following manuscript preparation roles: conceptualization AND/OR methodology, software, investigation, formal analysis, data curation, visualization, and validation AND drafting or reviewing/editing the final draft. As corresponding author, Dr Grooms confirms that all authors have provided the final approval of the version to be published and takes responsibility for the affirmations regarding article submission (eg, not under consideration by another journal), the integrity of the data presented, and the statements regarding compliance with institutional review board/Declaration of Helsinki requirements.

Supporting information

Disclosure Form: