Janus kinase inhibitor, tofacitinib, in refractory juvenile dermatomyositis: a retrospective multi-central study in China

This study was a retrospective study from May 1 to December 31 in the year of 2021. All patients were diagnosed based on Bohan and Peter’s criteria in 1975 [2]. Beijing Children’s Hospital, Children’s Hospital of Fudan University, Henan Children’s Hospital, Xi’an Children’s Hospital, and Children’s Hospital of Zhejiang University School of Medicine participated in this study. Patients were included if they were younger than 18, had ever received over 3 months of tofacitinib treatment, and regularly followed up in the respective center. Patients were excluded if receiving less than 3 months of tofacitinib treatment. Indications of starting tofacitinib therapy were active skin and muscle involvement.

This study gained approval from the Ethics Committee of a hospital on 1 Aug 2022 ([2022]-E-156-R). Written informed consent was obtained from the patient and their legal guardians.

All clinical information including clinical course and laboratory results were collected. Laboratory tests like liver and renal function, coagulation profile, and routine blood and urine tests were done during each follow-up. Myositis-specific antibodies (MSAs) and myositis-associated antibodies (MAAs) of five centers were evaluated by enzyme-linked immunosorbent assay (ELISA). Physical examination and essential images were done to assess disease activity. Cutaneous Assessment Tool-binary method (CAT-BM, skin disease activity score range 0–17, corresponding to “inactive,” “mild,” “moderate,” “severe,” and “very severe” were 0.7 ± 0.6, 5.0 ± 2.4, 8.0 ± 2.0, 9.8 ± 1.8, 10.5 ± 0.7; skin disease damage score range 0–11, corresponding to “mild,” “moderate,” and “severe” were 2.3 ± 1.6, 4.2 ± 2.0, 5.7 ± 1.7) and Childhood Myositis Assessment Scale (CMAS, score range 0–52, with high score indicate minimal disease) were used to evaluate skin involvement and muscle strength. Aspartate aminotransferase (AST, normal range 14–44 U/L), alanine aminotransferase (ALT, normal range 7–30 U/L), creatine kinase (CK, normal range 25–200 U/L), lactate dehydrogenase (LDH, normal range 110–295 U/L) were covered. Pulmonary function was investigated in accordance with respiratory symptoms and a high-resolution CT scan. International Myositis Assessment and Clinical Studies Group (IMACS) defined complete clinical response as more than 6 months of inactive disease (e.g., normal muscle strength, no skin rashes, and normal kinase under myositis treatment). Complete clinical remission was defined as over 6 months of inactive disease without treatment. Medical escalation was defined as an increase of over 25% of the original dose and as an increase of therapy due to increased disease activity [10]. Opportunistic infections (OI), gastrointestinal perforation, thromboembolism, malignancy, and thrombocytopenia were common adverse effects of tofacitinib and monitored carefully during the respective follow-up. Baseline (month = 0) was the time that tofacitinib was initiated.

All data were studied using SPSS 20.0. The follow-up data were compared with the baseline data through the Wilcoxon single rank test and Mann–Whitney U test. The treatment response of MSAs and MAAs was compared using the chi-square test. P value < 0.05 indicated a difference with statistical significance.

Of all cases assessed, 49 (55.6%) were female and all Han Chinese. Median age of disease onset was 82.0 ± 4.1 months (ranging from 17 to 176 months). Table 1 lists the general information. At the time of the first prescription, 86 (97.7%) of patients had skin involvement. Their CAT activity score ranged from 0 to 6, with 33 inactive (38.3%), 51 (59.3%) mild and 2 (2.3%) moderate. CAT damage score ranged from 0 to 7, with 38 (43.2%) mild, 39 (44.3%) moderate, and 9 (10.2%) severe. The muscular system was the second most affected system with 65(73.9%) patients suffering from decreased muscle strength. Six (6.8%) patients had dysphagia. Medium CMAS at first prescription was 34.0 with 5 (5.9%) patients in severe states (less than 15 score). At first prescription, 17 (19.3%) patients suffered from calcinosis. Seven (8.0%) patients had lipoatrophy. A total of 50 (56.8%) patients were reported with interstitial lung disease (ILD) at first prescription, as indicated by the assessment result of high-resolution pulmonary CT scan and respiratory symptoms. The joint, digestive, and central nervous systems are also affected. All patients had finished myositis-specific antibodies (MSAs) and myositis-associated antibodies (MAAs) tests. Forty-seven patients (53.4%) had positive MSAs and MAAs. To be specific, anti-NXP2 and anti-MDA5 were most abundant. Seventeen out of 19 (89.4%) patients with positive anti-MDA antibodies presented with ILD at first prescription. Six patients (50.0%) with positive anti-NXP2 antibodies presented with ILD.

Indication of starting JAK inhibitor treatment comprised active skin and muscle involvement and failure with previous DMARDs and biologics. The median disease duration of starting of JAK inhibitor treatment and median time of JAK inhibitor treatment were 22.0 ± 2.7 months and 10.1 ± 0.6 months (ranging from 3 to 21 months). Fifteen (17.0%) patients were administrated with the combination of tofacitinib, steroids, and immunosuppressive agents at first prescription due to long medical history in external hospitals and high disease activity. The longest time from the first prescription to starting tofacitinib reached 141 months.

As for previous medication, all cases used either prednisolone, immunosuppressant, or biologics as therapy. A combination of prednisolone (2 mg/kg) and methotrexate was used as first-line treatment in 50 (56.8%) patients. Fifty-one (57.9%) patients received methylprednisolone pulse, 70 (79.5%) received IVIG, 37 (42.0%) received cyclosporine, nine (10.2%) received cyclophosphamide, four (4.5%) received mycophenolate mofetil, 22 (25.0%) received hydroxychloroquine, 18 (20.4%) received thalidomide, 9 (10.2%) received tocilizumab, and 1 (1.1%) received infliximab (Table 2). Cyclosporine and thalidomide have been generally used in combination with other medications. At the last follow-up visit, the median dosage of steroids was taped to 0.05 mg/kg. Ten (11.4%) patients reached steroid discontinuation. For patients reaching glucocorticoid discontinuation, six (60.0%) were on tofacitinib monotherapy, one (10.0%) was on methotrexate monotherapy and three (30.0%) were on hydroxychloroquine (HCQ) monotherapy.

Disease activity was improved in patients after at least 3 months of tofacitinib treatment. CAT-BM, CMAS (Fig. 1), and pulmonary function progressed significantly. At the time of tofacitinib introduction (baseline), the median CAT-BM activity score was 1.53 ± 0.15 and decreased significantly during each follow-up visit (3, 6, 12, 18, and 21 months). CAT-BM damage score also dropped significantly (3, 6, 12, 18, and 21 months). CMAS in baseline was 42.94 ± 1.00 and it increased to 51.00 ± 0.40 at 21 months. Significant improvements of CMAS and CAT-BM during each follow-up visit indicated the efficacy of tofacitinib in skin and muscle manifestations. Fifty-three (60.2%) refractory JDM patients presented with pulmonary involvement recovered with no respiratory symptoms and had a reduction in the size or number of lesions under high-resolution computerized tomography (HRCT).

There was no significant change in the level of white blood cells, neutrophils, hemoglobin, platelet, and erythrocyte sedimentation rate before and after tofacitinib treatment (Fig. 2A–D). In terms ofe kinases, most patients had normal kinases during baseline and follow-up (Fig. 2E). The AST level was 45.4 ± 5.8 U/L at baseline and it decreased significantly at 3, 6, 9, and 12 months. The serum level of ALT decreased significantly during follow-up (3,6,9 and 12 months). CK level at baseline was 121.2 ± 16.2 U/L which was in the normal range. The average LDH level at the tofacitinib baseline was 330.2 ± 16.1 U/L and decreased significantly at 3, 6, 9, and 12 months. Ferritin level at baseline was 245.2 ± 63.2 μg/L, significant decrease in ferritin was found after 3 months of therapy (Fig. 2F). At baseline, 3 patients should be considered the diagnosis as macrophage activation syndrome in accordance with sJIA-MAS criteria [11]. However, they can’t be diagnosed as macrophage activation syndrome since these patients’ temperature fell into a normal range and was not at hyper-inflammatory state.

During disease progression, calcinosis emerged in 3 (3.4%) patients after tofacitinib started. At the last follow-up visit, 15 (75%) patients presented with calcinosis showed a reduction in the size or number of calcinosis, and calcinosis completely resolved in 5 (25%) patients.

Sixteen (18.2%) patients reached IMACS-defined complete clinical response after 6 months of therapy. Thirty-seven (42.0%) patients reached complete clinical response in 9 months, 42 (47.7%) patients reached complete clinical response in 12 months and 44 patients reached complete clinical response in 18 months. In general, 44 (50.0%) refractory patients reached complete clinical response (Fig. 3). Of patients who reached clinical response, 2 (2.3%) of them had medicine escalation. One got medicine escalation one month after reaching a complete clinical response and reached a response again 8 months later. One got medical escalation 3 months after reaching a complete clinical response and now remains in an active state. Most patients reached complete clinical response at 9 months since active skin rashes and muscle weakness served as the indicator of tofacitinib start. No patient in this study reached complete clinical remission.

Common adverse effects (e.g., tuberculosis, perforation of the gastrointestinal system, thrombosis, and allergic reactions) were not reported in our study. Only one patient had herpes zoster infection in 9 months after initiation. After drug withdrawal, herpes recovered and tofacitinib was given again.