Patients with muscle biopsy confirmed IIM presenting to Hannover Medical School between 2015 and 2024 were screened for coexisting SjD. Screening included Saxon- und Schirmer-testing as well as detection of SS-A antibodies in the first place and a glandular biopsy in antibody-negative cases with xerophthalmia and/or xerostomia. Salivary gland biopsy was also offered to antibody-positive cases, but not all of these patients consented to perform the biopsy since the diagnosis of SjD was already established applying the screening procedure only. Regardless of any prior SjD diagnosis established in outpatient or inpatient settings elsewhere, all patients underwent a systematic assessment for SjD at Hannover Medical School. SjD was diagnosed according to the 2016 ACR/EULAR classification criteria for primary SjD (9). If SjD was confirmed, the patient was included in the further analyses of this study. Clinical and paraclinical data were retrospectively analyzed with a special focus on diagnosis, involvement of IIM besides striated skeletal musculature, disease course and treatment. In addition to demographic data, results of electromyography and electroneurography, laboratory analyses, treatment details, patient acceptable symptom state (PASS), and clinical improvement were assessed (3, 31).

To assess disease activity in patients with SjD, the ESSDAI (EULAR Primary Sjögren’s Syndrome Disease Activity Index) and ESSPRI (EULAR Primary Sjögren’s Syndrome Patient-Reported Index) scores were calculated (32, 33). To evaluate clinical improvement in patients with IIM, the ACR/EULAR criteria for minimal, moderate, and major clinical response in adult dermatomyositis and polymyositis were applied (3, 34–36). Due to the retrospective nature of the study, only the minimal dataset comprising Manual Muscle Testing, Extra-muscular Disease Activity, and Muscle Enzymes was used (3, 34–36).

The findings were compared to sex- and age-matched controls with biopsy confirmed IIM in whom SjD was ruled out applying the 2016 ACR/EULAR classification criteria for primary SjD applying a frequency matching approach (9). To achieve a balanced matching for sex and age while ensuring exclusion of SjD, 13 patients with IBM and 11 patients with dermato-/polymyositis were included as controls. Patients and controls were excluded from the present study if the diagnosis of IIM was not confirmed by muscle biopsy. Histopathological results of muscle biopsy of the included patients was defining the clinical diagnosis of IBM, dermatomyositis and polymyositis, thus matching also accounted for histopathological findings. Figure 1 depicts the flowchart of the study.

All statistical analyses were performed using standard software (IBM SPSS Statistics, version 29.0, IBM Corp., Armonk, NY, USA). Continuous variables were described as median and interquartile range (IQR), and categorical variables as absolute and relative frequencies. A two-sided significance level of 5% (p < 0.05) was considered statistically significant for all comparisons. For variables with missing data, a complete-case analysis was performed; no imputation methods were applied. Prior to analysis, data were tested for normal distribution using the Shapiro-Wilk test. For normally distributed continuous variables, group comparisons were conducted using the unpaired (independent) t-test. In cases where the assumption of normality was violated, the non-parametric Mann–Whitney U test was applied instead. For categorical variables, group comparisons were conducted using the chi-square test or Fisher’s exact test where appropriate. For within-subject comparisons between baseline and follow-up, paired t-tests or Wilcoxon signed-rank tests were used depending on data distribution. To explore predictors of disease activity and treatment intensity, multivariate linear regression analyses were performed. Independent variables included demographic factors (SjD diagnosis, age at SjD diagnosis, age at IIM diagnosis, sex), clinical variables (IBM subtype, baseline ESSDAI, ESSPRI), immunological markers (anti-SSA/Ro and anti-SSB/La antibodies), and treatment-related variables (number of immunosuppressive agents, use of rituximab or cyclophosphamide). The dependent variables were ESSDAI score at last follow-up, delta-ESSDAI, manual muscle testing score at follow-up, delta muscle strength, and total improvement score. All models were tested for standard regression assumptions, including linearity, homoscedasticity, and normal distribution of residuals. Multicollinearity was assessed using variance inflation factors (VIFs), which did not indicate critical collinearity among predictors. In addition, an exploratory binary logistic regression model was applied to assess predictors of high-efficacy therapy use (rituximab, cyclophosphamide). Due to the small sample size and convergence issues, the results of this model are reported in the results section for exploratory purposes only but should be interpreted with caution.

A total of 23 patients with SjD and concomitant IIM were included in the study. Of these, 39% (9 out of 23) were diagnosed with IBM, and 61% (14 out of 23) with either polymyositis (11 patients) or dermatomyositis (3 patients). At the time of SjD diagnosis, patients with IBM were significantly older than those with polymyositis or dermatomyositis (63 years versus 55 years, p=0.0089). In most cases, SjD was diagnosed concurrently with or after the diagnosis of IIM. Further details on SjD characteristics are provided in Table 1 .

When comparing baseline characteristics of patients with IIM and concomitant SjD to those without SjD, asymmetrical muscle weakness was observed significantly more often in patients with SjD ( Table 2 ). This was particularly evident in patients with polymyositis and dermatomyositis, where asymmetrical muscle weakness occurred in 11 out of 14 patients with SjD, compared to only 1 out of 11 patients without SjD (p=0.0172). Manifestations of IIM besides of the striated skeletal musculature (involving heart, lungs, skin, or joints) were also significantly more frequent in patients with concomitant SjD ( Table 2 ).

Further details on the comparison of IIM patients with and without SjD are presented in Table 2 and show no other clear differences between the two groups. Non-neurological comorbidities including cardiovascular (e.g. arterial hypertension, heart failure), metabolical (e.g. obesity, glucose metabolism disorders) and hormonal diseases (thyroid gland insufficiency) were frequently observed in both cohorts. Electromyography and electroneurography revealed significant differences only in the prevalence of reduced amplitudes in electromyography ( Table 2 ). Regarding laboratory analyses, myositis-specific antibodies were detected with similar frequencies in patients with concomitant SjD (PM-Scl-100 + SAE1 + PM-Scl75, PM-Scl100, SRP, MDA5 + SRP, Mi-2alpha + Mi2ß, each n=1) and those without coexisting SjD (SRP (n=2), NXP2 (n=1)). SSA-antibodies were found in a total of 17/47 patients (IIM with and without SjD), thus being prevalent in 36% in the present cohort. Testing for anti-cN1A antibodies was not performed in patients with IBM.

As shown in Table 2 and Supplementary Table 1 , IIM-SjD patients received significantly more different therapeutics than IIM patients without SjD during the disease course (median follow-up duration with SjD: 68 months (IQR: 32-102); median follow-up duration without SjD: 92 months (IQR: 49-147); p=0.2638). High-efficacy therapies (rituximab, cyclophosphamide) were significantly more often used in IIM-SjD patients than controls ( Table 2 ).

Median ESSPRI scores (IBM: 6.3, interquartile range (IQR): 3.7-6.5; polymyositis/dermatomyositis: 6, IQR: 4.3-6.5) as well as ESSDAI (IBM: 35, IQR: 25-41; polymyositis/dermatomyositis: 21, IQR: 17-36) at diagnosis were not significantly different comparing patients with IBM or poly-/dermatomyositis and comorbid SjD ( Figure 2 ). At last follow-up (median of 68 months after diagnosis, IQR 32–102 months; IBM and SjD: 68 months (IQR: 36–144 months); poly-/dermatomyositis and SjD: 67 months (IQR: 24-93)), ESSDAI scores were significantly lower in poly-/dermatomyositis and comorbid SjD than IBM (IBM: 33, IQR: 28-42; polymyositis/dermatomyositis: 12, IQR: 11-23) as ESSDAI significantly improved in patients with polymyositis/dermatomyositis and SjD (p=0.0038), but did not change in patients with IBM (p=0.7518).

In terms of treatment response, manual muscle testing was not significantly different comparing median scores (and IQR) at diagnosis and last follow-up in IIM-SjD patients ( Table 3 ; all patients: p=0.8559; IBM: 67 (58-73) vs. 65 (52-73), p=0.0749; polymyositis/dermatomyositis: 74 (70-77) vs. 78 (70-80), p=0.1226). In contrast, in IIM patients without SjD, manual muscle testing worsened significantly, mainly due to the worsening of the IBM patient group (all patients: p=0.0171; IBM: 69 (62-73) vs. 63 (35-71), p=0.0283; poly-/dermatomyositis: 72 (65-78) vs. 70 (67-75), p=0.3342).

Although only a small proportion of patients fulfilled the ACR/EULAR criteria for minimal and moderate clinical improvement, total improvement score was significantly higher in patients with SjD ( Table 3 ), which was due to the SjD patients with polymyositis/dermatomyositis (median total improvement score with SjD: 30 (IQR: 15-39); median total improvement score without SjD: 6 (IQR: 2-21); p=0.0114).

PASS was achieved equally at last follow-up in patients with IIM with and without SjD ( Table 3 ).

At last follow-up, serum creatine kinase (CK) concentration was significantly lower in SjD patients than patients without SjD, mainly due to IBM patients (median serum CK concentration (U/l) in patients with SjD: 43 (IQR: 28-178); median serum CK concentration (U/l) in patients without SjD: 218 (IQR: 130-495); p=0.0122). Further information on treatment and outcome parameters is shown in Table 3 .

To assess for correlations of baseline characteristics with outcome parameters in IIM-SjD patients, multivariate linear regression analyses were employed with the dependent variables being ESSDAI score at last follow-up, delta-ESSDAI, manual muscle testing score at follow-up, delta muscle strength, and total improvement score. Of these analyses, only prediction of ESSDAI-score at last follow-up was statistically significant (delta-ESSDAI: p=0.092, manual muscle testing score: p=0.391, delta muscle strength: p=0.956, total improvement score: p=0.213, ESSDAI-score at last follow-up: p=0.009, adjusted R²: 0.778), with evidence of IBM as type of IIM being the most relevant predictor (p=0.014) as shown in Supplementary Table 2 .

To assess for predictors of usage of a higher number of immunosuppressants indicating a high disease activity, a multivariate linear regression analysis with number of immunosuppressants being the dependent variable was employed. The model was statistically significant (p=0.002, adjusted R²=0.831). As shown in Supplementary Table 3 , higher baseline ESSDAI scores (p=0.021) and the presence of anti-SSB/La antibodies (p=0.025) were positively associated with the number of immunosuppressive agents used, whereas higher age at SjD diagnosis was negatively associated (p=0.011). In addition, an exploratory binary logistic regression was performed to assess predictors of high-efficacy therapy use. Although the overall model was statistically significant (χ²=21.270, p=0.006), convergence issues and extremely large standard errors indicated numerical instability, thus individual predictor estimates were not reported in detail.