Matrix metalloproteinase–inhibitor imbalance in Kawasaki disease and multisystem inflammatory syndrome in children

Matrix metalloproteinase–inhibitor imbalance in Kawasaki disease and multisystem inflammatory syndrome in children | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Article Matrix metalloproteinase–inhibitor imbalance in Kawasaki disease and multisystem inflammatory syndrome in children Kacper Toczylowski, Dawid Lewandowski, Diana Martonik, Artur Sulik This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-7754190/v1 This work is licensed under a CC BY 4.0 License Status: Under Review Version 1 posted 7 You are reading this latest preprint version Abstract Multisystem inflammatory syndrome in children (MIS-C) and Kawasaki disease (KD) are pediatric hyperinflammatory conditions with overlapping features, but their molecular basis remains poorly defined. Matrix metalloproteinases (MMPs) and their endogenous tissue inhibitors of metalloproteinases (TIMPs), and inducers like EMMPRIN, are a tightly regulated proteolytic system that controls extracellular matrix ECM homeostasis and modulate immune responses. We profiled serum concentrations of MMPs, TIMPs, and related mediators (EMMPRIN and TNF-α) in 42 children with MIS-C, 5 with KD, 41 febrile controls, and 25 healthy controls. At presentation, MIS-C/KD showed a distinct protease–inhibitor signature characterized by markedly elevated TIMP-1, MMP-8, MMP-3, and TNF-α, with ratios indicating a net proteolytic bias despite compensatory TIMP upregulation. TIMP-1 and MMP-8 demonstrated excellent diagnostic performance (AUC >0.94 vs controls), and TIMP-1 correlated most strongly with clinical severity (ρ=0.55). Treatment with IVIG and corticosteroids reduced EMMPRIN but was paradoxically associated with rebound increases in MMP-1 activity. High-dose steroids blunted the post-treatment decline of neutrophil-derived proteases (MMP-8 and MMP-13). Subgroup analyses linked renal and cardiovascular involvement, shock, and gastrointestinal disease to the most pronounced biomarker alterations, driven by TIMP-1 and EMMPRIN elevations. Direct comparisons between MIS-C and KD revealed only modest, non-significant differences after correction, while SARS-CoV-2 serology status did not affect the main biomarker signals. These findings identify dysregulated MMP/TIMP balance as a central feature of pediatric hyperinflammatory conditions and highlight TIMP-1 as a marker of disease severity. Health sciences/Biomarkers Health sciences/Diseases Biological sciences/Immunology Health sciences/Medical research Multisystem Inflammatory Syndrome in Children Kawasaki Disease Matrix Metalloproteinases Tissue Inhibitor of Metalloproteinases Vasculitis Biomarkers Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 1. Introduction Multisystem Inflammatory Syndrome in Children (MIS-C) is a rare but serious post-infectious complication following SARS-CoV-2 infection, typically occurring 2–6 weeks after the acute phase.[ 1 ] MIS-C presents with fever, systemic inflammation, and multi-organ involvement, often including cardiovascular, gastrointestinal, and neurological symptoms.[ 2 , 3 ] While its clinical features overlap substantially with Kawasaki Disease (KD) and other pediatric inflammatory syndromes, MIS-C appears to involve more pronounced cardiovascular dysfunction, including higher rates of myocardial depression and shock requiring intensive care support.[ 4 – 6 ] Both MIS-C and KD are characterized by systemic vasculitis and endothelial dysfunction,[ 7 , 8 ] yet the underlying molecular mechanisms driving tissue injury and organ damage remain incompletely understood. A key pathological feature common to both conditions is inflammation affecting the vascular endothelium and extracellular matrix (ECM), which may lead to coronary artery abnormalities, myocardial dysfunction, and other serious complications.[ 9 ] Understanding the proteolytic processes underlying ECM degradation and vascular remodeling in these conditions could provide critical insights into disease pathogenesis and identify novel therapeutic targets.[ 10 ] Matrix metalloproteinases (MMPs) and their endogenous inhibitors, tissue inhibitors of metalloproteinases (TIMPs), constitute a tightly regulated proteolytic system that controls ECM homeostasis and remodeling.[ 11 , 12 ] Under physiological conditions, this system maintains tissue architecture and facilitates normal wound healing and angiogenesis.[ 13 , 14 ] Nevertheless, dysregulation of the MMP/TIMP balance can lead to pathological ECM degradation, endothelial dysfunction, and vascular damage.[ 15 ] MMPs not only remodel tissue structure but also modulate immune responses by processing cytokines, chemokines, and cell surface receptors, thereby amplifying inflammatory cascades. Recent evidence has highlighted the critical role of MMPs in cardiovascular pathology. MMP-8 (neutrophil collagenase) is predominantly released by activated neutrophils and is considered a key mediator of acute inflammatory responses.[ 16 , 17 ] Studies in Kawasaki disease have demonstrated that MMP-8, MMP-9, and MMP-12 are significantly elevated and correlate with disease severity and coronary artery involvement.[ 18 – 20 ] Similarly, TIMP-1, the primary inhibitor of most MMPs, has been shown to be markedly upregulated in inflammatory conditions and may serve as both a protective response and a biomarker of disease activity.[ 21 , 22 ] In MIS-C, emerging evidence suggests that matrix remodeling plays a central role in pathogenesis. Pavan Kumar et al. demonstrated that children with MIS-C exhibit elevated levels of multiple MMPs, particularly MMP-8, MMP-12, and MMP-13, which could distinguish MIS-C from acute COVID-19 and other tropical diseases with high sensitivity (84–100%) and specificity (80–100%).[ 23 ] These findings suggest that neutrophil-derived proteases may potentially be particularly important in MIS-C pathogenesis, potentially reflecting the intense neutrophilic inflammation characteristic of this syndrome. The balance between MMPs and TIMPs may potentially be particularly crucial in determining clinical outcomes. While absolute concentrations of individual proteases provide valuable information, MMP/TIMP ratios may better reflect the net proteolytic activity and tissue remodeling potential. Elevated MMP/TIMP ratios have been associated with increased ECM degradation, endothelial permeability, and vascular dysfunction in various inflammatory conditions. In the context of pediatric inflammatory syndromes, these ratios may serve as more sensitive indicators of ongoing tissue damage and provide insights into the balance between destructive and protective responses.[ 24 ] Despite the clinical overlap between MIS-C and KD, few studies have systematically compared their MMP/TIMP profiles or examined how these biomarkers change with immunomodulatory treatment. Furthermore, the relationship between MMP/TIMP profiles and clinical severity, organ involvement, and treatment response remains poorly characterized. Previous studies of COVID-19 have demonstrated altered MMP profiles in both adults and children, with higher MMP-3, MMP-8, and MMP-9 levels in adults than in pediatric patients, and partial normalization after immunomodulatory therapy. [ 25 ] Whether similar MMP/TIMP dysregulation underlies hyperinflammatory conditions such as MIS-C and Kawasaki disease has not been systematically assessed. 2. Results 2.1 Demographic characteristics of the study groups We analyzed serum samples from children with MIS-C (n = 42) and Kawasaki disease (n = 5), which were pooled for the primary analysis as hyperinflammatory vasculopathy (pediatric syndromes with hyperinflammation that cause vascular inflammation), together with febrile controls (FC, n = 41) and healthy controls (HC, n = 25). Several children initially labeled as KD during clinical care were reclassified as MIS-C after review according to CDC criteria, on the basis of SARS-CoV-2 serology and multisystem involvement. Two cases fulfilled partial criteria and were categorized as KD after multidisciplinary discussion. One case lacked serologic confirmation but occurred during high community SARS-CoV-2 circulation and fulfilled full clinical criteria; this was classified as MIS-C. The febrile controls group comprised children with acute COVID-19 (n = 9), viral meningitis (n = 25), viral respiratory infections (n = 5), and bacterial infections (n = 2). Age distributions did not differ significantly between study groups (p = 0.196), and the male-to-female ratio was comparable across cohorts (p = 0.331). Demographic characteristics including age and sex distribution are shown in Supplementary Figure S1 . Detailed information about control groups are provided in Supplementary Table S1 . 2.2 Treatment All MIS-C and KD children were treated with intravenous infusion of immunoglobulins of 2g/kg body weight. A subset of children received additional glucocorticoids. Steroid regimens were harmonized by conversion to prednisone-equivalent doses (PEQ) using standard equivalences (5 mg prednisone ≈ 4 mg methylprednisolone ≈ 0.75 mg dexamethasone). For each patient, daily doses were converted to mg/kg/day PEQ. Patients were then categorized according to the highest dose received before T2 sampling: Low dose: ≤ 2 mg/kg/day PEQ, Moderate dose: > 2 to < 12.5 mg/kg/day PEQ, High dose (pulse): ≥ 12.5 mg/kg/day PEQ (corresponding to ≥ 10 mg/kg/day methylprednisolone or ≥ 1.9 mg/kg/day dexamethasone). Patients were assigned to the high-dose group if any pulse-dose day (≥ 12.5 mg/kg/day PEQ) occurred before T2 sampling. 2.3 Protease–inhibitor profiles at presentation At presentation (T1), children with MIS-C/KD displayed a distinct protease–inhibitor profile compared with febrile and healthy controls (Fig. 1 and Fig. 2 ). The most prominent differences were seen in TIMP-1, MMP-8, MMP-3, and TNF-α. TIMP-1 was markedly elevated in MIS-C/KD (median 347 ng/mL, IQR 278–466) compared with febrile (218 ng/mL, IQR 121–233) and healthy controls (151 ng/mL, IQR 141–161; KW p = 9.1×10⁻¹², η²=0.49). MMP-8 concentrations were similarly increased (median 40 ng/mL, IQR 26–88) relative to febrile (8 ng/mL) and healthy controls (5 ng/mL; KW p = 4.4×10⁻¹², η²=0.47). MMP-3 followed the same pattern (9 ng/mL vs 2 and 2; KW p = 1.3×10⁻¹¹, η²=0.45), as did TNF-α (40 pg/mL vs 7 and 4; KW p = 1.2×10⁻¹⁰, η²=0.44), and EMMPRIN (9 ng/mL vs 6 and 7; KW p = 1.3×10⁻⁴, η²=0.15). Other notable differences included higher total TIMPs, calculated as the sum of TIMP-1, TIMP-2, TIMP-3, and TIMP-4 concentrations (623 vs 410 and 454 ng/mL; KW p = 2.3×10⁻¹⁰, η²=0.42), elevated MMP-7 (4 vs 3 and 2 ng/mL; KW p = 1.9×10⁻⁷, η²=0.27) and MMP-12 (60 vs 36 and 58 pg/mL; KW p = 7.3×10⁻⁸, η²=0.34), and a complex pattern of TIMP-3, which was reduced compared with healthy controls (63 vs 87 ng/mL) but remained higher than in febrile illness (39 ng/mL; KW p = 4.7×10⁻⁶, η²=0.23). In contrast, MMP-1 was not consistently elevated, and its activity appeared relatively suppressed when normalized to inhibitors. 2.4 Relative MMP/TIMP profiles The MMP-8/TIMP-2 ratio was increased almost six-fold in MIS-C/KD (median 0.395) compared with febrile (0.069) and healthy controls (0.059; KW p = 3.8×10⁻⁹, η²=0.37), and MMP-8/TIMP-4 was similarly increased (28.3 vs 5.41 and 8.00; KW p = 5.7×10⁻⁸, η²=0.32). The MMP-8/TIMP-1 ratio was also higher (0.113 vs 0.046 and 0.043; KW p = 7.1×10⁻⁶, η²=0.22), as was MMP-3/TIMP-2 (0.063 vs 0.018 and 0.015; KW p = 4.2×10⁻⁹, η²=0.37). In contrast, the MMP-2/TIMP-1 ratio was reduced (0.93 vs 1.59 and 2.00; KW p = 1.5×10⁻⁷, η²=0.33). All relative concentrations are presented in Supplementary Table S2 . 2.5 Correlation matrix Correlation analysis revealed coordinated regulation of proteases and their inhibitors (Fig. 3 , Supplementary Table S3 ). TIMPs were strongly correlated with each other and with TNF-α, with TIMP-1 showing positive associations with TIMP-2 (ρ = 0.61), TIMP-4 (ρ = 0.40), total TIMPs (ρ = 0.80), EMMPRIN (ρ = 0.66), MMP-2 (ρ = 0.47), MMP-12 (ρ = 0.57), and TNF-α itself (ρ = 0.56). Additionally, TIMP-1 demonstrated a significant negative correlation with serum albumin (ρ=-0.48), suggesting that higher tissue inhibitor levels associate with hypoalbuminemia and increased vascular permeability. TIMP-2 was most closely linked to MMP-2. A neutrophil-related cluster of MMP-8, MMP-12, and MMP-13 was evident (ρ up to 0.67), with MMP-8 showing strong correlation with white blood cell count (ρ = 0.51), reflecting neutrophilic inflammation. MMP-3 correlated with both MMP-2 (ρ = 0.34) and TNF-α (ρ = 0.56), and demonstrated significant associations with C-reactive protein (ρ = 0.48), as did the MMP-3/TIMP-3 ratio (ρ = 0.49), indicating that stromelysin activity relates to systemic inflammatory burden. EMMPRIN (also called CD147), a known inducer, was associated with multiple TIMPs and MMPs, including MMP-12 (ρ = 0.83), MMP-10 (ρ = 0.69), TIMP-2 (ρ = 0.66), and TIMP-1 (ρ = 0.66). 2.6 Discriminatory performance of biomarkers Receiver operating characteristic analyses are presented in Supplementary Table S4 and in, Fig. 4 . TIMP-1 demonstrated the strongest discriminatory performance against healthy controls with an AUC of 0.962, achieving optimal sensitivity of 91.5% (95%CI: 80.1–96.6%) and perfect specificity of 100.0% (95% CI: 79.6–100%) at a threshold of 204 ng/mL. MMP-3 and MMP-8 showed similarly robust performance (AUC 0.943 and 0.941, respectively), with MMP-3 achieving 91.5% sensitivity and 91.3% specificity, while MMP-8 reached 87.2% sensitivity and 91.7% specificity at their respective optimal thresholds. The MMP-8/TIMP-2 ratio also demonstrated strong accuracy (AUC 0.907) with 72.3% sensitivity and perfect specificity against healthy controls. Against febrile controls, discrimination remained clinically meaningful though somewhat reduced. TIMP-1 maintained good performance (AUC 0.873) with 74.5% sensitivity and 95.1% specificity, while MMP-3 (AUC 0.859, sensitivity 83.0%, specificity 87.5%) and MMP-8 (AUC 0.847, sensitivity 74.5%, specificity 85.4%) showed comparable discriminatory capacity. The MMP-8/TIMP-2 ratio achieved an AUC of 0.840 with 70.2% sensitivity and 85.4% specificity against febrile controls. Importantly, these biomarkers failed to distinguish between healthy and febrile control groups (mean AUC 0.385, range 0.293–0.448). MMP-7 shows moderate discriminatory performance compared to the other biomarkers (against HC AUC = 0.893, against FC AUC = 0.709). 2.7 Treatment-associated changes Paired sera were available for 31 patients. Treatment with IVIG and/or corticosteroids induced marked shifts in the protease–inhibitor profile between T1 and T2 (Supplementary Table S5 , Fig. 5 ). EMMPRIN decreased significantly (mean reduction − 36.8%, p < 1×10⁻⁵), consistent with suppression of upstream protease induction. In contrast, MMP-1 increased sharply (median 2 to 8 ng/mL, + 218%, p = 2.0×10⁻⁸), with parallel rises in its ratios to TIMPs, indicating a relative rebound of collagenase activity. Several other ratios, including MMP-1/TIMP-3 and MMP-1/TIMP-4, also shifted upward (p < 1×10⁻⁵), while MMP-10 declined modestly (Wilcoxon p = 0.024). Most other proteases did not change significantly in absolute terms, although selective ratio changes indicated nuanced shifts in balance rather than uniform suppression. Overall, treatment was associated with downregulation of the inducer EMMPRIN and most TIMP-dominant profiles, but with a rebound of MMP-1 activity relative to inhibitors, suggesting complex re-equilibration rather than uniform dampening of the MMP/TIMP axis. 2.8 Impact of high-dose steroids on biomarkers When stratified by steroid exposure, high-dose regimens were associated with less pronounced suppression of neutrophil-associated proteases. MMP-8 decreased more strongly in children receiving no or low-dose steroids (median fold-change 0.12) than in those treated with high-dose regimens (0.32; p = 0.008, r = 0.50). A similar pattern was seen for MMP-13 (0.98 vs 1.58; p = 0.012, r = 0.51), with parallel but non-significant trends for MMP-9, MMP-2, and TIMPs. Conversely, some analytes including MMP-10, MMP-12, and TIMP-4 showed slightly greater reductions with high-dose therapy, but effect sizes were small. To assess the independent effect of steroid treatment on biomarker expression while controlling for disease severity, we performed PSM analysis. Propensity score matching successfully created balanced treatment groups of 10 high steroid and 10 none/low steroid patients. To avoid overfitting with limited sample size, we matched on the composite severity score (pre-treatment) as the principal confounder; age and sex were balanced post-match (SMDs < 0.05). Among 16 biomarkers analyzed, two showed significant differences after multiple comparison correction. MMP-8 fold changes were significantly higher in the high steroid group compared to none/low steroid group (FC = 0.477 vs 0.122, p = 0.003, Cohen's d = 1.17), indicating that steroid treatment attenuated the decrease in MMP-8 levels: MMP-8 was reduced on average by 52% in high dose group vs. 88% in low/none steroids group. MMP-13 showed significant upregulation in the high steroid group (2.029 vs 0.822, p = 0.008, Cohen's d = 1.71). Several other biomarkers including TIMP-1, EMMPRIN, and MMP-9 showed trends towards greater reduction in none/low steroids group (p < 0.10), while MMP-10 decreased more strongly in the high-dose group (Fig. 6 ). 2.9 Biomarker differences by clinical severity TIMP-1 concentrations rose stepwise across mild, moderate, and severe cases (median 292 → 360 → 567 ng/mL; H = 18.7, p = 8.5×10⁻⁵, η²=0.38), paralleled by total TIMPs (524 → 627 → 829 ng/mL; H = 10.8, p = 0.0045). TNF-α also increased with severity (32 → 36 → 75 pg/mL; H = 8.8, p = 0.012), and MMP-7 was higher in severe disease (2.9 → 3.4 → 5.9 ng/mL; H = 7.6, p = 0.023). TIMP-4 showed a similar gradient (1.6 → 2.1 → 2.5 ng/mL; H = 8.1, p = 0.017). Several ratios, including MMP-9/TIMP-1, MMP-1/TIMP-1, and MMP-13/TIMP-1, decreased progressively with severity, suggesting stronger dominance of inhibitors relative to proteases in the sickest children. To link biomarker profiles with clinical disease burden, we analyzed correlations with a continuous severity score. TIMP-1 at presentation showed the strongest association (ρ = 0.55, p = 6.1×10⁻⁵), followed by TIMP-4 (ρ = 0.42, p = 0.0036) and total TIMPs (ρ = 0.41, p = 0.0044). Among MMPs, MMP-3 (ρ = 0.30, p = 0.043) and MMP-8 (ρ = 0.30, p = 0.044) were significant, as was TNF-α (ρ = 0.37, p = 0.011). No associations were observed for MMP-1, MMP-9, MMP-10, or MMP-13. Analyses of clinical course showed only weak relationships with length of hospital stay. At follow-up (T2), moderate correlations were observed for MMP-9/TIMP-2 (r = 0.47, p = 0.006), MMP-8/TIMP-2 (r = 0.44, p = 0.010), MMP-9 (r = 0.42, p = 0.014), MMP-8 (r = 0.42, p = 0.014), and TIMP-3 (r = 0.41, p = 0.017). None remained significant after correction for multiple comparisons, and no consistent associations were seen across timepoints, suggesting that biomarker profiles do not robustly predict hospital stay in this cohort. 2.10 Biomarker differences by clinical phenotype Finally, subgroup analyses revealed phenotype-specific biomarker associations, with 157 significant differences identified across 1,646 tested. Renal and cardiovascular involvement showed the densest signal clusters. Children with renal involvement demonstrated marked elevations of EMMPRIN (median 14 vs 8 ng/mL, p = 1.7 × 10⁻⁵) and TIMP-1 (median 680 vs 329 ng/mL, p = 0.0001) at baseline, while shock on admission was similarly associated with high TIMP-1 and total TIMP concentrations. Cardiac complications also exhibited distinct protease–inhibitor profiles: pericardial effusion was linked to differential EMMPRIN dynamics post-treatment (median fold change 0.37 vs 0.81, p = 0.0008), and reduced left ventricular contractility showed 14 significant biomarker associations. Gastrointestinal disease and hypotension each displayed around 15 associations, again enriched for TIMP and EMMPRIN changes. In contrast, neurological, pulmonary, and hematological features were associated with fewer or less consistent biomarker differences. Together, these findings highlight TIMP-1 and EMMPRIN as recurring signals across multiple severe phenotypes, suggesting that dysregulation of the protease–inhibitor axis underlies multi-organ involvement in MIS-C/KD. Detailed results are provided in Supplementary Table S6 . 2.11 Comparison of MIS-C and Kawasaki disease Although MIS-C patients generally presented with more severe cardiovascular involvement and higher systemic inflammation (Supplementary Table S7 ), direct biomarker comparisons with Kawasaki disease revealed only modest differences. At the uncorrected level, several analytes showed higher concentrations in KD, including MMP-7 (notably at follow-up), MMP-10 at baseline, and their ratios to TIMPs (MMP-7/TIMP-1, MMP-7/TIMP-2, MMP-10/TIMP-1, MMP-10/TIMP-4). In contrast, MMP-1 tended to be lower in KD than MIS-C. However, none of these associations remained significant after correction for multiple comparisons, reflecting both the small KD sample size and the substantial overlap between groups. Because MIS-C and KD overlap clinically, particularly after 2020, we also analyzed biomarker profiles according to the evidence of of SARS-CoV-2 recent infection. Several markers, including MMP-13 (0.98 vs 1.30 ng/mL) and related ratios (MMP-13/TIMP-1, MMP-13/TIMP-2, MMP-13/TIMP-4), as well as MMP-10/TIMP-1, were nominally lower in virus-exposed children, but none remained significant after correction. Importantly, the major discriminatory signals identified in our primary analysis — TIMP-1, MMP-8, MMP-3, TNF-α, and MMP-8/TIMP ratios — did not differ between MIS-C and KD, underscoring the shared protease–inhibitor dysregulation that defines hyperinflammatory state in both conditions. 3. Discussion This study shows that children with MIS-C and Kawasaki disease (KD) share a common protease–inhibitor signature that clearly distinguishes them from febrile and healthy controls. The profile is dominated by elevated TIMP-1, MMP-8, MMP-3, and TNF-α, together with shifts in MMP/TIMP ratios indicating a net proteolytic bias. These findings place dysregulation of the protease–inhibitor system at the center of pediatric hyperinflammatory conditions and extend prior observations from KD into MIS-C. Recent reviews underscore endothelial injury and microvascular dysfunction as key elements of MIS-C pathobiology, with variable benefits of IVIG and corticosteroids across studies [ 6 ]. Our results also align with prior pediatric COVID-19 work showing elevations of MMP-3, MMP-8, and MMP-9 compared with healthy children and lower levels than in adults [ 25 ]. Independent cohorts document endothelial dysfunction and arterial stiffness in MIS-C, and endothelial biomarkers such as endocan (ESM-1) are elevated—consistent with our observation that the upstream inducer EMMPRIN decreased after therapy [ 26 , 27 ]. The strong signal from neutrophil-associated proteases, particularly MMP-8 with its correlated cluster (MMP-12, and MMP-13 which can be activated by neutrophil elastase), underscores innate immune activation in MIS-C and KD; many of these signals declined with treatment [ 28 , 29 ]. In KD, elevated MMP-8 and MMP-9 have long been linked to coronary artery damage and aneurysm formation [ 18 , 30 ], while in MIS-C, MMPs are among the most discriminatory serum proteins differentiating cases from acute COVID-19 or dengue [ 23 ]. Our ROC analyses reinforce these observations: TIMP-1, MMP-3, and MMP-8 each achieved AUCs > 0.9, suggesting promising diagnostic utility that warrants validation in larger and more diverse cohorts, particularly where overlapping infections complicate clinical diagnosis. TIMP-1 emerged as the biomarker most consistently associated with severity, showing the strongest correlations with a composite severity score and a clear stepwise increase across mild, moderate, and severe disease. Across clinical phenotypes, TIMP-1 and EMMPRIN were the most recurrent signals, suggesting a shared upstream axis tied to multi-organ involvement. This is compatible with the multifunctional role of TIMP-1 as both an MMP inhibitor and an acute-phase, cytokine-like protein [ 31 ]. Notably, the most severe cases exhibited declining MMP/TIMP ratios, indicating relative inhibitor dominance despite escalating inflammation—potentially reflecting compensatory TIMP upregulation to restrain proteolysis. Similar patterns have been reported in critically ill MIS-C patients in India [ 23 ], supporting the idea that the balance between protease activity and TIMP response, rather than absolute MMP abundance alone, may track clinical severity. Conversely, lower post-treatment TIMP-1 has been associated with persistent coronary artery lesions in KD, raising the possibility that TIMP-1 contributes to vascular healing [ 28 ]. Classical inflammatory markers (CRP, ESR, ferritin, procalcitonin) remain central to MIS-C and KD assessment, but they primarily reflect systemic inflammation. By contrast, MMPs and TIMPs capture extracellular matrix remodeling, neutrophil and macrophage activation, and vascular injury. The robust elevation of TIMP-1 and MMP-8 in our cohort highlights how matrix-directed biomarkers may complement conventional laboratory tests by reporting a distinct dimension of the hyperinflammatory response [ 32 ]. Treatment dynamics provide further insight into regulation of this system. Immunomodulation was associated with a marked reduction in EMMPRIN, a known inducer of MMP expression [ 33 ], indicating upstream suppression of protease activation. In parallel, MMP-1 concentrations rose with corresponding increases in MMP-1/TIMP ratios. This rebound may reflect activation of collagenase-dependent remodeling during vascular repair, consistent with reported roles of MMP-1 in angiogenesis and endothelial permeability [ 15 ]. Overall, rather than uniform suppression, therapy appears to selectively rebalance individual MMP/TIMP axes with potential implications for vascular recovery [ 34 ]. Within this context, our exploratory dose–response analysis—situated among heterogeneous real-world immunomodulation data [ 35 ] —suggested that high-dose pulse steroids blunted the post-treatment decline of neutrophil-related proteases (MMP-8 and MMP-13). Although limited by sample size, these results raise the hypothesis that very high steroid exposure may incompletely suppress neutrophil protease activity, potentially prolonging vascular injury. Prior KD studies have similarly reported variable effects of IVIG and steroids on MMP expression [ 36 ], underscoring the need to evaluate treatment effects on protease biology systematically. Phenotype-stratified analyses indicated that biomarker changes mirror organ involvement. Liver disease associated with broad upregulation of TIMPs and neutrophil proteases, whereas gastrointestinal presentations showed selective reductions in MMP-1/TIMP ratios. These patterns are consistent with literature framing extracellular matrix remodeling as a driver of organ injury in systemic inflammation [ 13 ]. Recent proteomic work likewise separates MIS-C and KD from other pediatric hyperinflammatory states and highlights cytokine/immune-activation axes that co-vary with our MMP/TIMP signature [ 37 , 38 ]. Direct MIS-C vs KD comparisons in our cohort revealed only modest differences that did not survive correction for multiple testing, supporting the view that the protease–inhibitor signature reflects a shared hyperinflammatory vasculopathy phenotype rather than the infectious trigger per se and that MIS-C and KD occupy overlapping immunopathologic space [ 39 – 41 ]. This study has limitations. It is single-center, the KD group was relatively small (limiting power for disease-specific contrasts), and we measured protein concentrations rather than enzymatic activity (which may diverge due to post-translational regulation and endogenous inhibitors). Treatment heterogeneity and the observational design also constrain causal inference about therapy effects. Nevertheless, strengths include a comparatively large MIS-C cohort, inclusion of both febrile and healthy controls, broad multiplex profiling of MMPs, TIMPs, and inducers, and paired sampling before and after immunomodulation—features that together provide an integrated view of biomarkers at presentation, their diagnostic performance, their relation to clinical severity and organ involvement, and their evolution with treatment. In summary, dysregulation of the MMP/TIMP system is a defining feature of pediatric hyperinflammatory vasculopathy. TIMP-1 and MMP-8 were the most informative markers, capturing diagnostic separation and severity associations, while treatment produced selective rebalancing of the protease–inhibitor network. Pooling MIS-C and KD for primary analyses, with disease-specific and serology-based contrasts treated as exploratory, allowed us to interrogate shared and divergent biology across this spectrum. These data support further evaluation of the protease–inhibitor axis as a mechanistic and biomarker framework for MIS-C and KD, with prospective validation needed to define clinical utility. 4. Methods 4.1 Study Design and Participants This was a prospective observational study with both cross-sectional and longitudinal components, conducted at the University Children’s Clinical Hospital of the Medical University of Bialystok, Poland. Four cohorts were included: children with multisystem inflammatory syndrome in children (MIS-C), Kawasaki disease (KD), febrile pediatric controls (FC), and healthy pediatric controls (HC). Patients with MIS-C and KD were diagnosed according to the Centers for Disease Control and Prevention (CDC) and American Heart Association (AHA) criteria, respectively. Because MIS-C and KD share overlapping clinical phenotypes and host responses, and differentiation can be challenging in children with positive SARS-CoV-2 IgG antibodies, we prespecified pooled primary analyses (MIS-C/KD vs. FC/HC), with MIS-C versus KD comparisons treated as exploratory. Blood samples for MIS-C and KD were collected prospectively at baseline (T1, before treatment) and 4–5 days after defervescence (T2, following IVIG and/or corticosteroid therapy). Controls were matched for sex and age when feasible. Clinical laboratory information was abstracted retrospectively from electronic medical records on a standardized case report form without access to the outcome of the biomarkers. 4.2 Sample Collection and Storage Peripheral blood samples were collected into serum tubes, centrifuged within 2 hours, and stored at − 80°C until analysis. All samples were processed using standardized protocols to minimize pre-analytical variability. 4.3 Protein Quantification Serum concentrations of 9 matrix metalloproteinases (MMP-1, -2, -3, -7, -8, -9, -10, -12, -13), 4 tissue inhibitors of metalloproteinases (TIMP-1, -2, -3, -4), TNF-α, and EMMPRIN were measured using custom and commercially available Luminex® multiplex bead-based immunoassays (R&D Systems/Bio-Techne, Minneapolis, MN). Samples were assayed in three batches over two years; batch-to-batch variation was addressed by normalization using internal control samples repeated across batches and by median ratio correction for each marker. 4.4 Clinical severity score To capture the extent of clinical involvement in MIS-C and Kawasaki disease, we developed a composite severity score integrating organ dysfunction and standard laboratory indices, adapted from commonly used severity markers in pediatric inflammatory syndromes.[ 42 – 45 ] Twelve domains were included (Supplementary table S8 ). Each domain was assigned a score reflecting the degree of abnormality, and the total score ranged from 0 to 20, with higher values indicating greater severity. The domains were: (i) hemodynamic instability/shock, (ii) cardiac involvement, (iii) respiratory support, (iv) neurological involvement, (v) gastrointestinal symptoms, (vi) inflammatory markers (CRP, PCT), (vii) coagulopathy (D-dimer, APTT/INR), (viii) liver function and albumin, (ix) hyponatremia, (x) lymphopenia, (xi) prolonged fever, and (xii) ICU admission. This score was used both categorically (mild, moderate, severe) and continuously. For categorical analysis, children were initially stratified into mild (≤ 7 points), moderate (8–11 points), and severe (≥ 12 points). To minimize information loss, correlations between continuous severity scores and biomarker concentrations were also performed. 4.5 Statistical Analysis Because biomarker distributions were non-normal, non-parametric tests were used throughout. Overall comparisons between MIS-C/KD, febrile controls, and healthy controls were performed with the Kruskal–Wallis test, followed by post hoc pairwise Mann–Whitney U tests. Effect sizes were reported as η² for Kruskal–Wallis tests and rank-biserial correlation (r) for Mann–Whitney and Wilcoxon signed-rank tests. Where parametric comparisons or propensity score–matched analyses were appropriate, effect sizes were expressed as Cohen’s d. To control for multiple comparisons, Bonferroni correction was applied; unadjusted p-values are presented for exploratory analyses. Diagnostic performance of candidate biomarkers was assessed using receiver operating characteristic (ROC) curves. Area under the curve (AUC) values were reported with 95% confidence intervals, with thresholds of ≥ 0.7, ≥ 0.8, and ≥ 0.9 interpreted as acceptable, good, and excellent discrimination, respectively. Optimal cut-off values were defined by the Youden index, and corresponding sensitivity and specificity values were calculated. To evaluate the impact of steroid treatment, fold changes (FC) were calculated as the ratio of follow-up (T2) to baseline (T1) concentrations. Propensity score matching (PSM) was used to minimize confounding by disease severity, with propensity scores estimated from logistic regression models including the calculated clinical severity score. One-to-one nearest-neighbor matching without replacement was performed, and covariate balance was verified using standardized mean differences (SMD), with SMD < 0.1 indicating adequate balance. Biomarker fold changes were compared between matched groups using Mann–Whitney U tests. All analyses were two-sided with a significance threshold of 0.05. Missing data were handled by pairwise deletion without imputation. Statistical analyses were performed in Python (SciPy, scikit-learn), GraphPad Prism (GraphPad Software, San Diego, CA), and Statistica (TIBCO Software Inc., v14). 4.6 Ethics Statement This study was conducted according to the declaration of Helsinki and approved by the Ethics Committee of Medical University of Bialystok (approval no. APK.002.259.2020 and APK.002.428.2022). A written informed consent was obtained from all studied patients and/or patient parents/caregivers. Declarations Author contributions Conceptualization: KT, AS; Data curation: DL, DM; Formal Analysis: KT, DM; Funding acquisition: KT; Investigation: KT, DM; Project administration: AS; Resources: KT, DL, AS;; Supervision: AS; Visualization: KT; Writing – original draft: KT; Writing – review & editing: KT, DL, DM, AS Acknowledgements We thank Prof. Robert Flisiak, Head of the Department of Infectious Diseases and Hepatology at the Medical University of Białystok, Poland, for providing institutional support and maintaining the laboratory environment that enabled this work. Artificial intelligence–assisted tools (ChatGPT, OpenAI) were used to improve the clarity and grammar of the manuscript text. The authors reviewed and take full responsibility for all content. Data availability statement The datasets generated and analyzed during the current study are available in the Harvard Dataverse repository, https://doi.org/10.7910/DVN/XAHZHY. All other data supporting the findings of this study are provided within the article and its Supplementary Information files. Competing interests KT, DL, and AS were employed by Pfizer in a clinical trial program evaluating a Lyme disease vaccine. KT has received honoraria from GSK for lectures on meningococcal disease and vaccines. The authors declare no other competing interests. Funding This work was supported by the Medical University of Bialystok, Poland (Grants No. B.SUB.25.440 and B.SUB.24.448). References Multisystem Inflammatory Syndrome in Children (MIS-C) Associated with SARS-CoV-2 Infection 2023 Case Definition | CDC. (2024). at Consiglio, C. R. et al. The Immunology of Multisystem Inflammatory Syndrome in Children with COVID-19. Cell 183, 968-981.e7 (2020). Hoste, L., Van Paemel, R. & Haerynck, F. Multisystem inflammatory syndrome in children related to COVID-19: a systematic review. Eur J Pediatr 180, 2019–2034 (2021). Jiang, L. et al. COVID-19 and multisystem inflammatory syndrome in children and adolescents. Lancet Infect Dis 20, e276–e288 (2020). Ramcharan, T. et al. Paediatric Inflammatory Multisystem Syndrome: Temporally Associated with SARS-CoV-2 (PIMS-TS): Cardiac Features, Management and Short-Term Outcomes at a UK Tertiary Paediatric Hospital. Pediatr Cardiol 41, 1391–1401 (2020). Lampidi, S. et al. Multisystem inflammatory syndrome in children (MIS-C): A nationwide collaborative study in the Greek population. Eur J Pediatr 183, 1693–1702 (2024). Kawasaki, T. [Acute febrile mucocutaneous syndrome with lymphoid involvement with specific desquamation of the fingers and toes in children]. Arerugi 16, 178–222 (1967). Burns, J. C. & Glodé, M. P. Kawasaki syndrome. The Lancet 364, 533–544 (2004). Onouchi, Y. The genetics of Kawasaki disease. Int J Rheum Dis 21, 26–30 (2018). Olejarz, W., Łacheta, D. & Kubiak-Tomaszewska, G. Matrix Metalloproteinases as Biomarkers of Atherosclerotic Plaque Instability. International Journal of Molecular Sciences 21, 3946 (2020). Visse, R. & Nagase, H. Matrix metalloproteinases and tissue inhibitors of metalloproteinases: structure, function, and biochemistry. Circ Res 92, 827–839 (2003). Parks, W. C., Wilson, C. L. & López-Boado, Y. S. Matrix metalloproteinases as modulators of inflammation and innate immunity. Nat Rev Immunol 4, 617–629 (2004). Bonnans, C., Chou, J. & Werb, Z. Remodelling the extracellular matrix in development and disease. Nat Rev Mol Cell Biol 15, 786–801 (2014). Lu, P., Takai, K., Weaver, V. M. & Werb, Z. Extracellular Matrix Degradation and Remodeling in Development and Disease. Cold Spring Harb Perspect Biol 3, a005058 (2011). Galis, Z. S. & Khatri, J. J. Matrix metalloproteinases in vascular remodeling and atherogenesis: the good, the bad, and the ugly. Circ Res 90, 251–262 (2002). Lenglet, S., Mach, F. & Montecucco, F. Role of Matrix Metalloproteinase-8 in Atherosclerosis. Mediators Inflamm 2013, 659282 (2013). Lee, E.-J. et al. Regulation of neuroinflammation by matrix metalloproteinase-8 inhibitor derivatives in activated microglia and astrocytes. Oncotarget 8, 78677–78690 (2017). Chua, P. K. et al. Elevated Levels of Matrix Metalloproteinase 9 and Tissue Inhibitor of Metalloproteinase 1 during the Acute Phase of Kawasaki Disease. Clin Vaccine Immunol 10, 308–314 (2003). Gavin, P. J., Crawford, S. E., Shulman, S. T., Garcia, F. L. & Rowley, A. H. Systemic arterial expression of matrix metalloproteinases 2 and 9 in acute Kawasaki disease. Arterioscler Thromb Vasc Biol 23, 576–581 (2003). Sekiguchi, K. et al. The Imbalance between Matrix Metalloproteinase-9, -2 and Tissue Inhibi tor of Metalloproteinases-1 in Acute Phase Kawasaki Disease. Pediatr Res 53, 170–170 (2003). Ries, C. Cytokine functions of TIMP-1. Cell Mol Life Sci 71, 659–672 (2014). Moore, C. S. & Crocker, S. J. An alternate perspective on the roles of TIMPs and MMPs in pathology. Am J Pathol 180, 12–16 (2012). Pavan Kumar, N. et al. Role of matrix metalloproteinases in multi-system inflammatory syndrome and acute COVID-19 in children. Front Med (Lausanne) 9, 1050804 (2022). Newby, A. C. Dual role of matrix metalloproteinases (matrixins) in intimal thickening and atherosclerotic plaque rupture. Physiol Rev 85, 1–31 (2005). Toczyłowski, K. et al. Differential Inflammatory Responses in Adult and Pediatric COVID-19 Patients: Implications for Long-Term Consequences and Anti-Inflammatory Treatment. Med Sci Monit 30, e944052 (2024). Cannavo, A. et al. Serum endocan (ESM-1) as diagnostic and prognostic biomarker in Multisystem inflammatory syndrome in children (MIS-C). Cytokine 184, 156797 (2024). Ahmadi, A. et al. Vascular function and arterial stiffness in multisystem inflammatory syndrome in children with Covid-19. ARYA Atheroscler 21, 54–62 (2025). Li, Y. et al. Dynamic changes of serum matrix metalloproteinases and tissue inhibitors in Kawasaki disease: implications for coronary artery lesions persistence. Ital J Pediatr 51, 169 (2025). Vandooren, J., Van den Steen, P. E. & Opdenakker, G. Biochemistry and molecular biology of gelatinase B or matrix metalloproteinase-9 (MMP-9): the next decade. Crit Rev Biochem Mol Biol 48, 222–272 (2013). Senzaki, H. The pathophysiology of coronary artery aneurysms in Kawasaki disease: role of matrix metalloproteinases. Arch Dis Child 91, 847–851 (2006). Jackson, H. W., Defamie, V., Waterhouse, P. & Khokha, R. TIMPs: versatile extracellular regulators in cancer. Nat Rev Cancer 17, 38–53 (2017). Henderson, L. A. et al. American College of Rheumatology Clinical Guidance for Multisystem Inflammatory Syndrome in Children Associated With SARS–CoV-2 and Hyperinflammation in Pediatric COVID-19: Version 2. Arthritis & Rheumatology 73, e13–e29 (2021). von Ungern-Sternberg, S. N. I., Zernecke, A. & Seizer, P. Extracellular Matrix Metalloproteinase Inducer EMMPRIN (CD147) in Cardiovascular Disease. Int J Mol Sci 19, 507 (2018). Tian, F. et al. Correlation Between Matrix Metalloproteinases With Coronary Artery Lesion Caused by Kawasaki Disease. Front Pediatr 10, 802217 (2022). McArdle, A. J. et al. Immunomodulatory Treatment of Multisystem Inflammatory Syndrome in Children. The New England journal of medicine 385, 11 (2021). Sakata, K. et al. Matrix Metalloproteinase-9 in Vascular Lesions and Endothelial Regulat ion in Kawasaki Disease. Circ J 74, 1670–1675 (2010). Reiter, A. et al. Proteomic mapping identifies serum marker signatures associated with MIS-C specific hyperinflammation and cardiovascular manifestation. Clinical Immunology 264, 110237 (2024). Dourdouna, M.-M., Tatsi, E.-B., Syriopoulou, V. & Michos, A. Proteomic Signatures of Multisystem Inflammatory Syndrome in Children (MIS-C) Associated with COVID-19: A Narrative Review. Children (Basel) 11, 1174 (2024). Tsoukas, P. & Yeung, R. S. M. Kawasaki disease and MIS-C share a host immune response. Nat Rev Rheumatol 18, 555–556 (2022). Ghosh, P. et al. An Artificial Intelligence-guided signature reveals the shared host immune response in MIS-C and Kawasaki disease. Nat Commun 13, 2687 (2022). Noval Rivas, M. & Arditi, M. Kawasaki Disease and Multisystem Inflammatory Syndrome in Children. Rheum Dis Clin North Am 49, 647–659 (2023). Molloy, M. J. et al. Epidemiology and Severity of Illness of MIS-C and Kawasaki Disease During the COVID-19 Pandemic. Pediatrics 152, e2023062101 (2023). Savorgnan, F. et al. Physiologic profile associated with severe multisystem inflammatory syndrome in children: a retrospective study. Pediatr Res 93, 102–109 (2023). Tran, D. M. et al. Severity predictors for multisystemic inflammatory syndrome in children after SARS-CoV-2 infection in Vietnam. Sci Rep 14, 15810 (2024). Martin, B., Rao, S. & Bennett, T. D. Disparities in Multisystem Inflammatory Syndrome in Children and COVID-19 Across the Organ Dysfunction Continuum. JAMA Netw Open 6, e2249552 (2023). Additional Declarations Competing interest reported. KT, DL, and AS were employed by Pfizer in a clinical trial program evaluating a Lyme disease vaccine. KT has received honoraria from GSK for lectures on meningococcal disease and vaccines. The authors declare no other competing interests. Supplementary Files TableS1Controls.xlsx TableS2CompleteGroupComparisonResultsT1.xlsx TableS3biomarkercorrelationmatrix.xlsx TableS4ROCAnalysisResultswith95CI.xlsx TableS5T1vsT2AnalysisKDMISCPatients.xlsx TableS6biomarkerclinicalanalysisresults.xlsx TableS7MISCvsKD.xlsx TableS8clinicalseverity.xlsx FigureS1.png Cite Share Download PDF Status: Under Review Version 1 posted Reviews received at journal 15 Apr, 2026 Reviewers agreed at journal 06 Apr, 2026 Reviewers invited by journal 06 Apr, 2026 Editor assigned by journal 03 Feb, 2026 Editor invited by journal 17 Oct, 2025 Submission checks completed at journal 09 Oct, 2025 First submitted to journal 06 Oct, 2025 You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-7754190","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":618487779,"identity":"358599d5-c811-4a9b-bf76-10cd95eeb595","order_by":0,"name":"Kacper Toczylowski","email":"data:image/png;base64,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","orcid":"","institution":"Medical University of Białystok","correspondingAuthor":true,"prefix":"","firstName":"Kacper","middleName":"","lastName":"Toczylowski","suffix":""},{"id":618487780,"identity":"8fcb3cb5-33cc-4a5c-829f-035e4947b179","order_by":1,"name":"Dawid Lewandowski","email":"","orcid":"","institution":"Medical University of Białystok","correspondingAuthor":false,"prefix":"","firstName":"Dawid","middleName":"","lastName":"Lewandowski","suffix":""},{"id":618487781,"identity":"ed1d0350-12c9-47a3-b87b-df220faa89f0","order_by":2,"name":"Diana Martonik","email":"","orcid":"","institution":"Medical University of Białystok","correspondingAuthor":false,"prefix":"","firstName":"Diana","middleName":"","lastName":"Martonik","suffix":""},{"id":618487782,"identity":"7e8f57cd-abeb-4aac-aecf-05a9025891b3","order_by":3,"name":"Artur Sulik","email":"","orcid":"","institution":"Medical University of Białystok","correspondingAuthor":false,"prefix":"","firstName":"Artur","middleName":"","lastName":"Sulik","suffix":""}],"badges":[],"createdAt":"2025-09-30 18:23:16","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-7754190/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-7754190/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":106635941,"identity":"174d2228-891e-4104-b531-7ccf11fcca6b","added_by":"auto","created_at":"2026-04-10 16:51:41","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":1151815,"visible":true,"origin":"","legend":"\u003cp\u003eSerum MMP concentrations across study groups.\u003c/p\u003e\n\u003cp\u003eBoxplots show baseline serum concentrations of matrix metalloproteinases (MMPs) in children with MIS-C/KD, febrile controls, and healthy controls. Each panel displays median, interquartile range, and outliers for the indicated MMP.\u003c/p\u003e","description":"","filename":"Figure1.png","url":"https://assets-eu.researchsquare.com/files/rs-7754190/v1/6d3719a82523b872a07ec44a.png"},{"id":106726827,"identity":"a1e19d61-c90b-4cc3-b608-3cf036d665a7","added_by":"auto","created_at":"2026-04-12 18:37:23","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":482749,"visible":true,"origin":"","legend":"\u003cp\u003eSerum concentrations of TIMPs, EMMPRIN, and TNF-α across study groups.\u003c/p\u003e\n\u003cp\u003eBoxplots show baseline concentrations of tissue inhibitors of metalloproteinases (TIMPs), extracellular matrix metalloproteinase inducer (EMMPRIN), and tumor necrosis factor alpha (TNF-α) in MIS-C/KD, febrile controls, and healthy controls.\u003c/p\u003e","description":"","filename":"Figure2.png","url":"https://assets-eu.researchsquare.com/files/rs-7754190/v1/616f1d2511c1b13f7f276225.png"},{"id":106726384,"identity":"e4668af6-19be-4019-a857-9e631e435ea7","added_by":"auto","created_at":"2026-04-12 18:35:58","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":450973,"visible":true,"origin":"","legend":"\u003cp\u003eCorrelation matrix of protease–inhibitor system at admission.\u003c/p\u003e\n\u003cp\u003eSpearman correlation coefficients are shown for MMPs, TIMPs, EMMPRIN, and TNF-α in the entire study cohort at baseline (T1). Positive correlations are indicated in red, negative correlations in blue, with correlation values annotated in each cell.\u003c/p\u003e","description":"","filename":"Figure3.png","url":"https://assets-eu.researchsquare.com/files/rs-7754190/v1/3c66ff8e8055c0f5d9dda515.png"},{"id":106635946,"identity":"49b7528d-560e-43dd-aae9-e20cb149db53","added_by":"auto","created_at":"2026-04-10 16:51:41","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":300962,"visible":true,"origin":"","legend":"\u003cp\u003eDiagnostic performance of key biomarkers.\u003c/p\u003e\n\u003cp\u003eReceiver operating characteristic (ROC) curves demonstrate the ability of TIMP-1, MMP-8, MMP-3, and the MMP-8/TIMP-2 ratio to distinguish MIS-C/KD from healthy and febrile controls. Curves are shown separately for each biomarker, with AUC values summarized in the figure.\u003c/p\u003e","description":"","filename":"Figure4.png","url":"https://assets-eu.researchsquare.com/files/rs-7754190/v1/398c116b67d8ce7d0dafd845.png"},{"id":106635948,"identity":"2f1ffdf4-3e3d-4697-a08c-3d042d3eac5d","added_by":"auto","created_at":"2026-04-10 16:51:41","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":109265,"visible":true,"origin":"","legend":"\u003cp\u003eTreatment-associated changes in serum protein concentrations.\u003c/p\u003e\n\u003cp\u003eHeatmap displays log2 fold change (T2/T1) for MMPs, TIMPs, TNF-α, and EMMPRIN in MIS-C/KD patients (n=31). Each row represents a patient and each column a protein. Red denotes increased concentrations at T2, blue denotes decreased concentrations. Patients are grouped according to corticosteroid dose.\u003c/p\u003e","description":"","filename":"Figure5.png","url":"https://assets-eu.researchsquare.com/files/rs-7754190/v1/a65150f3abd8b104153ed8c5.png"},{"id":106726359,"identity":"3854cc94-bfa7-4891-9063-0a6921dbcd42","added_by":"auto","created_at":"2026-04-12 18:35:56","extension":"png","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":112501,"visible":true,"origin":"","legend":"\u003cp\u003eImpact of high-dose steroids on biomarker trajectories after matching.\u003c/p\u003e\n\u003cp\u003ePropensity score matching created balanced groups of high-dose (n=10) and none/low-dose (n=10) steroid-treated patients. Left panel: statistical significance of biomarker fold-change (T2/T1) differences, shown as –log10(p-value) with thresholds for p=0.05 (red) and p=0.10 (orange). Right panel: effect sizes (Cohen’s d) for the same comparisons. Positive values indicate relatively higher fold change in the high-dose group (attenuated reduction or greater increase), while negative values indicate greater suppression with high-dose steroids.\u003c/p\u003e","description":"","filename":"Figure6.png","url":"https://assets-eu.researchsquare.com/files/rs-7754190/v1/d1e02e55e2668ed9fb310063.png"},{"id":106728128,"identity":"741579e0-dee4-4f96-bb93-5a267ac6be79","added_by":"auto","created_at":"2026-04-12 18:41:54","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":3379607,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7754190/v1/f822601f-f758-4af3-9de9-f8fd39b4ec4d.pdf"},{"id":106635942,"identity":"583ea362-f00f-4214-b62e-4b8914e542f8","added_by":"auto","created_at":"2026-04-10 16:51:41","extension":"xlsx","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":13001,"visible":true,"origin":"","legend":"","description":"","filename":"TableS1Controls.xlsx","url":"https://assets-eu.researchsquare.com/files/rs-7754190/v1/d34af3f96e3c28c74144e502.xlsx"},{"id":106635944,"identity":"2b25a117-fb64-4c7d-8584-6a65ebbc4dbc","added_by":"auto","created_at":"2026-04-10 16:51:41","extension":"xlsx","order_by":2,"title":"","display":"","copyAsset":false,"role":"supplement","size":30712,"visible":true,"origin":"","legend":"","description":"","filename":"TableS2CompleteGroupComparisonResultsT1.xlsx","url":"https://assets-eu.researchsquare.com/files/rs-7754190/v1/e2cf96f043382ce36719e635.xlsx"},{"id":106726393,"identity":"5fad2e39-ed43-4355-9625-73103037d584","added_by":"auto","created_at":"2026-04-12 18:35:59","extension":"xlsx","order_by":3,"title":"","display":"","copyAsset":false,"role":"supplement","size":106590,"visible":true,"origin":"","legend":"","description":"","filename":"TableS3biomarkercorrelationmatrix.xlsx","url":"https://assets-eu.researchsquare.com/files/rs-7754190/v1/de5c687b8f147f86b178baf2.xlsx"},{"id":106726399,"identity":"9d82d843-4cab-4945-95c4-5f8e003c33e4","added_by":"auto","created_at":"2026-04-12 18:36:02","extension":"xlsx","order_by":4,"title":"","display":"","copyAsset":false,"role":"supplement","size":10551,"visible":true,"origin":"","legend":"","description":"","filename":"TableS4ROCAnalysisResultswith95CI.xlsx","url":"https://assets-eu.researchsquare.com/files/rs-7754190/v1/6a958fc80276fc59337368ff.xlsx"},{"id":106635949,"identity":"296b42ea-c2c0-4a89-867d-5a262908ab78","added_by":"auto","created_at":"2026-04-10 16:51:42","extension":"xlsx","order_by":5,"title":"","display":"","copyAsset":false,"role":"supplement","size":38770,"visible":true,"origin":"","legend":"","description":"","filename":"TableS5T1vsT2AnalysisKDMISCPatients.xlsx","url":"https://assets-eu.researchsquare.com/files/rs-7754190/v1/b02dc0fc91c5bd348080d164.xlsx"},{"id":106726846,"identity":"74753a33-4df1-4b8f-a067-6de8cb594784","added_by":"auto","created_at":"2026-04-12 18:37:24","extension":"xlsx","order_by":6,"title":"","display":"","copyAsset":false,"role":"supplement","size":124364,"visible":true,"origin":"","legend":"","description":"","filename":"TableS6biomarkerclinicalanalysisresults.xlsx","url":"https://assets-eu.researchsquare.com/files/rs-7754190/v1/4ff1f9e4d5038831cffd4695.xlsx"},{"id":106635951,"identity":"34952ac3-5f96-4be9-b9bb-038bd3faa80e","added_by":"auto","created_at":"2026-04-10 16:51:42","extension":"xlsx","order_by":7,"title":"","display":"","copyAsset":false,"role":"supplement","size":20234,"visible":true,"origin":"","legend":"","description":"","filename":"TableS7MISCvsKD.xlsx","url":"https://assets-eu.researchsquare.com/files/rs-7754190/v1/edee5a19c8d3ad4493b26a9d.xlsx"},{"id":106635954,"identity":"17b30799-a34e-432a-9279-934c2a704ebe","added_by":"auto","created_at":"2026-04-10 16:51:42","extension":"xlsx","order_by":8,"title":"","display":"","copyAsset":false,"role":"supplement","size":11252,"visible":true,"origin":"","legend":"","description":"","filename":"TableS8clinicalseverity.xlsx","url":"https://assets-eu.researchsquare.com/files/rs-7754190/v1/a3fec0d818b4082552832399.xlsx"},{"id":106727212,"identity":"4103e185-d924-46e2-af6e-2078ac6b2a0b","added_by":"auto","created_at":"2026-04-12 18:38:19","extension":"png","order_by":9,"title":"","display":"","copyAsset":false,"role":"supplement","size":338143,"visible":true,"origin":"","legend":"","description":"","filename":"FigureS1.png","url":"https://assets-eu.researchsquare.com/files/rs-7754190/v1/7d863fc9383b192b6ace206f.png"}],"financialInterests":"Competing interest reported. KT, DL, and AS were employed by Pfizer in a clinical trial program evaluating a Lyme disease vaccine. KT has received honoraria from GSK for lectures on meningococcal disease and vaccines. The authors declare no other competing interests.","formattedTitle":"Matrix metalloproteinase–inhibitor imbalance in Kawasaki disease and multisystem inflammatory syndrome in children","fulltext":[{"header":"1. Introduction","content":"\u003cp\u003eMultisystem Inflammatory Syndrome in Children (MIS-C) is a rare but serious post-infectious complication following SARS-CoV-2 infection, typically occurring 2\u0026ndash;6 weeks after the acute phase.[\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e] MIS-C presents with fever, systemic inflammation, and multi-organ involvement, often including cardiovascular, gastrointestinal, and neurological symptoms.[\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e, \u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e] While its clinical features overlap substantially with Kawasaki Disease (KD) and other pediatric inflammatory syndromes, MIS-C appears to involve more pronounced cardiovascular dysfunction, including higher rates of myocardial depression and shock requiring intensive care support.[\u003cspan additionalcitationids=\"CR5\" citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e]\u003c/p\u003e \u003cp\u003eBoth MIS-C and KD are characterized by systemic vasculitis and endothelial dysfunction,[\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e, \u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e] yet the underlying molecular mechanisms driving tissue injury and organ damage remain incompletely understood. A key pathological feature common to both conditions is inflammation affecting the vascular endothelium and extracellular matrix (ECM), which may lead to coronary artery abnormalities, myocardial dysfunction, and other serious complications.[\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e] Understanding the proteolytic processes underlying ECM degradation and vascular remodeling in these conditions could provide critical insights into disease pathogenesis and identify novel therapeutic targets.[\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e]\u003c/p\u003e \u003cp\u003eMatrix metalloproteinases (MMPs) and their endogenous inhibitors, tissue inhibitors of metalloproteinases (TIMPs), constitute a tightly regulated proteolytic system that controls ECM homeostasis and remodeling.[\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e, \u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e] Under physiological conditions, this system maintains tissue architecture and facilitates normal wound healing and angiogenesis.[\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e, \u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e] Nevertheless, dysregulation of the MMP/TIMP balance can lead to pathological ECM degradation, endothelial dysfunction, and vascular damage.[\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e] MMPs not only remodel tissue structure but also modulate immune responses by processing cytokines, chemokines, and cell surface receptors, thereby amplifying inflammatory cascades.\u003c/p\u003e \u003cp\u003eRecent evidence has highlighted the critical role of MMPs in cardiovascular pathology. MMP-8 (neutrophil collagenase) is predominantly released by activated neutrophils and is considered a key mediator of acute inflammatory responses.[\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e, \u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e] Studies in Kawasaki disease have demonstrated that MMP-8, MMP-9, and MMP-12 are significantly elevated and correlate with disease severity and coronary artery involvement.[\u003cspan additionalcitationids=\"CR19\" citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e] Similarly, TIMP-1, the primary inhibitor of most MMPs, has been shown to be markedly upregulated in inflammatory conditions and may serve as both a protective response and a biomarker of disease activity.[\u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e, \u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e]\u003c/p\u003e \u003cp\u003eIn MIS-C, emerging evidence suggests that matrix remodeling plays a central role in pathogenesis. Pavan Kumar et al. demonstrated that children with MIS-C exhibit elevated levels of multiple MMPs, particularly MMP-8, MMP-12, and MMP-13, which could distinguish MIS-C from acute COVID-19 and other tropical diseases with high sensitivity (84\u0026ndash;100%) and specificity (80\u0026ndash;100%).[\u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e] These findings suggest that neutrophil-derived proteases may potentially be particularly important in MIS-C pathogenesis, potentially reflecting the intense neutrophilic inflammation characteristic of this syndrome.\u003c/p\u003e \u003cp\u003eThe balance between MMPs and TIMPs may potentially be particularly crucial in determining clinical outcomes. While absolute concentrations of individual proteases provide valuable information, MMP/TIMP ratios may better reflect the net proteolytic activity and tissue remodeling potential. Elevated MMP/TIMP ratios have been associated with increased ECM degradation, endothelial permeability, and vascular dysfunction in various inflammatory conditions. In the context of pediatric inflammatory syndromes, these ratios may serve as more sensitive indicators of ongoing tissue damage and provide insights into the balance between destructive and protective responses.[\u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e]\u003c/p\u003e \u003cp\u003eDespite the clinical overlap between MIS-C and KD, few studies have systematically compared their MMP/TIMP profiles or examined how these biomarkers change with immunomodulatory treatment. Furthermore, the relationship between MMP/TIMP profiles and clinical severity, organ involvement, and treatment response remains poorly characterized. Previous studies of COVID-19 have demonstrated altered MMP profiles in both adults and children, with higher MMP-3, MMP-8, and MMP-9 levels in adults than in pediatric patients, and partial normalization after immunomodulatory therapy. [\u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e] Whether similar MMP/TIMP dysregulation underlies hyperinflammatory conditions such as MIS-C and Kawasaki disease has not been systematically assessed.\u003c/p\u003e"},{"header":"2. Results","content":"\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e \u003ch2\u003e2.1 Demographic characteristics of the study groups\u003c/h2\u003e \u003cp\u003eWe analyzed serum samples from children with MIS-C (n\u0026thinsp;=\u0026thinsp;42) and Kawasaki disease (n\u0026thinsp;=\u0026thinsp;5), which were pooled for the primary analysis as hyperinflammatory vasculopathy (pediatric syndromes with hyperinflammation that cause vascular inflammation), together with febrile controls (FC, n\u0026thinsp;=\u0026thinsp;41) and healthy controls (HC, n\u0026thinsp;=\u0026thinsp;25). Several children initially labeled as KD during clinical care were reclassified as MIS-C after review according to CDC criteria, on the basis of SARS-CoV-2 serology and multisystem involvement. Two cases fulfilled partial criteria and were categorized as KD after multidisciplinary discussion. One case lacked serologic confirmation but occurred during high community SARS-CoV-2 circulation and fulfilled full clinical criteria; this was classified as MIS-C. The febrile controls group comprised children with acute COVID-19 (n\u0026thinsp;=\u0026thinsp;9), viral meningitis (n\u0026thinsp;=\u0026thinsp;25), viral respiratory infections (n\u0026thinsp;=\u0026thinsp;5), and bacterial infections (n\u0026thinsp;=\u0026thinsp;2). Age distributions did not differ significantly between study groups (p\u0026thinsp;=\u0026thinsp;0.196), and the male-to-female ratio was comparable across cohorts (p\u0026thinsp;=\u0026thinsp;0.331). Demographic characteristics including age and sex distribution are shown in Supplementary Figure \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003e. Detailed information about control groups are provided in Supplementary Table \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003e.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec4\" class=\"Section2\"\u003e \u003ch2\u003e2.2 Treatment\u003c/h2\u003e \u003cp\u003eAll MIS-C and KD children were treated with intravenous infusion of immunoglobulins of 2g/kg body weight. A subset of children received additional glucocorticoids. Steroid regimens were harmonized by conversion to prednisone-equivalent doses (PEQ) using standard equivalences (5 mg prednisone\u0026thinsp;\u0026asymp;\u0026thinsp;4 mg methylprednisolone\u0026thinsp;\u0026asymp;\u0026thinsp;0.75 mg dexamethasone). For each patient, daily doses were converted to mg/kg/day PEQ. Patients were then categorized according to the highest dose received before T2 sampling: Low dose: \u0026le; 2 mg/kg/day PEQ, Moderate dose: \u0026gt; 2 to \u0026lt;\u0026thinsp;12.5 mg/kg/day PEQ, High dose (pulse): \u0026ge; 12.5 mg/kg/day PEQ (corresponding to \u0026ge;\u0026thinsp;10 mg/kg/day methylprednisolone or \u0026ge;\u0026thinsp;1.9 mg/kg/day dexamethasone). Patients were assigned to the high-dose group if any pulse-dose day (\u0026ge;\u0026thinsp;12.5 mg/kg/day PEQ) occurred before T2 sampling.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec5\" class=\"Section2\"\u003e \u003ch2\u003e2.3 Protease\u0026ndash;inhibitor profiles at presentation\u003c/h2\u003e \u003cp\u003eAt presentation (T1), children with MIS-C/KD displayed a distinct protease\u0026ndash;inhibitor profile compared with febrile and healthy controls (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e and Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e). The most prominent differences were seen in TIMP-1, MMP-8, MMP-3, and TNF-α. TIMP-1 was markedly elevated in MIS-C/KD (median 347 ng/mL, IQR 278\u0026ndash;466) compared with febrile (218 ng/mL, IQR 121\u0026ndash;233) and healthy controls (151 ng/mL, IQR 141\u0026ndash;161; KW p\u0026thinsp;=\u0026thinsp;9.1\u0026times;10⁻\u0026sup1;\u0026sup2;, η\u0026sup2;=0.49). MMP-8 concentrations were similarly increased (median 40 ng/mL, IQR 26\u0026ndash;88) relative to febrile (8 ng/mL) and healthy controls (5 ng/mL; KW p\u0026thinsp;=\u0026thinsp;4.4\u0026times;10⁻\u0026sup1;\u0026sup2;, η\u0026sup2;=0.47). MMP-3 followed the same pattern (9 ng/mL vs 2 and 2; KW p\u0026thinsp;=\u0026thinsp;1.3\u0026times;10⁻\u0026sup1;\u0026sup1;, η\u0026sup2;=0.45), as did TNF-α (40 pg/mL vs 7 and 4; KW p\u0026thinsp;=\u0026thinsp;1.2\u0026times;10⁻\u0026sup1;⁰, η\u0026sup2;=0.44), and EMMPRIN (9 ng/mL vs 6 and 7; KW p\u0026thinsp;=\u0026thinsp;1.3\u0026times;10⁻⁴, η\u0026sup2;=0.15). Other notable differences included higher total TIMPs, calculated as the sum of TIMP-1, TIMP-2, TIMP-3, and TIMP-4 concentrations (623 vs 410 and 454 ng/mL; KW p\u0026thinsp;=\u0026thinsp;2.3\u0026times;10⁻\u0026sup1;⁰, η\u0026sup2;=0.42), elevated MMP-7 (4 vs 3 and 2 ng/mL; KW p\u0026thinsp;=\u0026thinsp;1.9\u0026times;10⁻⁷, η\u0026sup2;=0.27) and MMP-12 (60 vs 36 and 58 pg/mL; KW p\u0026thinsp;=\u0026thinsp;7.3\u0026times;10⁻⁸, η\u0026sup2;=0.34), and a complex pattern of TIMP-3, which was reduced compared with healthy controls (63 vs 87 ng/mL) but remained higher than in febrile illness (39 ng/mL; KW p\u0026thinsp;=\u0026thinsp;4.7\u0026times;10⁻⁶, η\u0026sup2;=0.23). In contrast, MMP-1 was not consistently elevated, and its activity appeared relatively suppressed when normalized to inhibitors.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec6\" class=\"Section2\"\u003e \u003ch2\u003e2.4 Relative MMP/TIMP profiles\u003c/h2\u003e \u003cp\u003eThe MMP-8/TIMP-2 ratio was increased almost six-fold in MIS-C/KD (median 0.395) compared with febrile (0.069) and healthy controls (0.059; KW p\u0026thinsp;=\u0026thinsp;3.8\u0026times;10⁻⁹, η\u0026sup2;=0.37), and MMP-8/TIMP-4 was similarly increased (28.3 vs 5.41 and 8.00; KW p\u0026thinsp;=\u0026thinsp;5.7\u0026times;10⁻⁸, η\u0026sup2;=0.32). The MMP-8/TIMP-1 ratio was also higher (0.113 vs 0.046 and 0.043; KW p\u0026thinsp;=\u0026thinsp;7.1\u0026times;10⁻⁶, η\u0026sup2;=0.22), as was MMP-3/TIMP-2 (0.063 vs 0.018 and 0.015; KW p\u0026thinsp;=\u0026thinsp;4.2\u0026times;10⁻⁹, η\u0026sup2;=0.37). In contrast, the MMP-2/TIMP-1 ratio was reduced (0.93 vs 1.59 and 2.00; KW p\u0026thinsp;=\u0026thinsp;1.5\u0026times;10⁻⁷, η\u0026sup2;=0.33). All relative concentrations are presented in Supplementary Table \u003cspan refid=\"MOESM2\" class=\"InternalRef\"\u003eS2\u003c/span\u003e.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec7\" class=\"Section2\"\u003e \u003ch2\u003e2.5 Correlation matrix\u003c/h2\u003e \u003cp\u003eCorrelation analysis revealed coordinated regulation of proteases and their inhibitors (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e, Supplementary Table \u003cspan refid=\"MOESM3\" class=\"InternalRef\"\u003eS3\u003c/span\u003e). TIMPs were strongly correlated with each other and with TNF-α, with TIMP-1 showing positive associations with TIMP-2 (ρ\u0026thinsp;=\u0026thinsp;0.61), TIMP-4 (ρ\u0026thinsp;=\u0026thinsp;0.40), total TIMPs (ρ\u0026thinsp;=\u0026thinsp;0.80), EMMPRIN (ρ\u0026thinsp;=\u0026thinsp;0.66), MMP-2 (ρ\u0026thinsp;=\u0026thinsp;0.47), MMP-12 (ρ\u0026thinsp;=\u0026thinsp;0.57), and TNF-α itself (ρ\u0026thinsp;=\u0026thinsp;0.56). Additionally, TIMP-1 demonstrated a significant negative correlation with serum albumin (ρ=-0.48), suggesting that higher tissue inhibitor levels associate with hypoalbuminemia and increased vascular permeability. TIMP-2 was most closely linked to MMP-2. A neutrophil-related cluster of MMP-8, MMP-12, and MMP-13 was evident (ρ up to 0.67), with MMP-8 showing strong correlation with white blood cell count (ρ\u0026thinsp;=\u0026thinsp;0.51), reflecting neutrophilic inflammation. MMP-3 correlated with both MMP-2 (ρ\u0026thinsp;=\u0026thinsp;0.34) and TNF-α (ρ\u0026thinsp;=\u0026thinsp;0.56), and demonstrated significant associations with C-reactive protein (ρ\u0026thinsp;=\u0026thinsp;0.48), as did the MMP-3/TIMP-3 ratio (ρ\u0026thinsp;=\u0026thinsp;0.49), indicating that stromelysin activity relates to systemic inflammatory burden. EMMPRIN (also called CD147), a known inducer, was associated with multiple TIMPs and MMPs, including MMP-12 (ρ\u0026thinsp;=\u0026thinsp;0.83), MMP-10 (ρ\u0026thinsp;=\u0026thinsp;0.69), TIMP-2 (ρ\u0026thinsp;=\u0026thinsp;0.66), and TIMP-1 (ρ\u0026thinsp;=\u0026thinsp;0.66).\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec8\" class=\"Section2\"\u003e \u003ch2\u003e2.6 Discriminatory performance of biomarkers\u003c/h2\u003e \u003cp\u003eReceiver operating characteristic analyses are presented in Supplementary Table \u003cspan refid=\"MOESM4\" class=\"InternalRef\"\u003eS4\u003c/span\u003e and in, Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003e. TIMP-1 demonstrated the strongest discriminatory performance against healthy controls with an AUC of 0.962, achieving optimal sensitivity of 91.5% (95%CI: 80.1\u0026ndash;96.6%) and perfect specificity of 100.0% (95% CI: 79.6\u0026ndash;100%) at a threshold of 204 ng/mL. MMP-3 and MMP-8 showed similarly robust performance (AUC 0.943 and 0.941, respectively), with MMP-3 achieving 91.5% sensitivity and 91.3% specificity, while MMP-8 reached 87.2% sensitivity and 91.7% specificity at their respective optimal thresholds. The MMP-8/TIMP-2 ratio also demonstrated strong accuracy (AUC 0.907) with 72.3% sensitivity and perfect specificity against healthy controls. Against febrile controls, discrimination remained clinically meaningful though somewhat reduced. TIMP-1 maintained good performance (AUC 0.873) with 74.5% sensitivity and 95.1% specificity, while MMP-3 (AUC 0.859, sensitivity 83.0%, specificity 87.5%) and MMP-8 (AUC 0.847, sensitivity 74.5%, specificity 85.4%) showed comparable discriminatory capacity. The MMP-8/TIMP-2 ratio achieved an AUC of 0.840 with 70.2% sensitivity and 85.4% specificity against febrile controls. Importantly, these biomarkers failed to distinguish between healthy and febrile control groups (mean AUC 0.385, range 0.293\u0026ndash;0.448). MMP-7 shows moderate discriminatory performance compared to the other biomarkers (against HC AUC\u0026thinsp;=\u0026thinsp;0.893, against FC AUC\u0026thinsp;=\u0026thinsp;0.709).\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec9\" class=\"Section2\"\u003e \u003ch2\u003e2.7 Treatment-associated changes\u003c/h2\u003e \u003cp\u003ePaired sera were available for 31 patients. Treatment with IVIG and/or corticosteroids induced marked shifts in the protease\u0026ndash;inhibitor profile between T1 and T2 (Supplementary Table \u003cspan refid=\"MOESM5\" class=\"InternalRef\"\u003eS5\u003c/span\u003e, Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003e). EMMPRIN decreased significantly (mean reduction \u0026minus;\u0026thinsp;36.8%, p\u0026thinsp;\u0026lt;\u0026thinsp;1\u0026times;10⁻⁵), consistent with suppression of upstream protease induction. In contrast, MMP-1 increased sharply (median 2 to 8 ng/mL, +\u0026thinsp;218%, p\u0026thinsp;=\u0026thinsp;2.0\u0026times;10⁻⁸), with parallel rises in its ratios to TIMPs, indicating a relative rebound of collagenase activity. Several other ratios, including MMP-1/TIMP-3 and MMP-1/TIMP-4, also shifted upward (p\u0026thinsp;\u0026lt;\u0026thinsp;1\u0026times;10⁻⁵), while MMP-10 declined modestly (Wilcoxon p\u0026thinsp;=\u0026thinsp;0.024). Most other proteases did not change significantly in absolute terms, although selective ratio changes indicated nuanced shifts in balance rather than uniform suppression. Overall, treatment was associated with downregulation of the inducer EMMPRIN and most TIMP-dominant profiles, but with a rebound of MMP-1 activity relative to inhibitors, suggesting complex re-equilibration rather than uniform dampening of the MMP/TIMP axis.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec10\" class=\"Section2\"\u003e \u003ch2\u003e2.8 Impact of high-dose steroids on biomarkers\u003c/h2\u003e \u003cp\u003eWhen stratified by steroid exposure, high-dose regimens were associated with less pronounced suppression of neutrophil-associated proteases. MMP-8 decreased more strongly in children receiving no or low-dose steroids (median fold-change 0.12) than in those treated with high-dose regimens (0.32; p\u0026thinsp;=\u0026thinsp;0.008, r\u0026thinsp;=\u0026thinsp;0.50). A similar pattern was seen for MMP-13 (0.98 vs 1.58; p\u0026thinsp;=\u0026thinsp;0.012, r\u0026thinsp;=\u0026thinsp;0.51), with parallel but non-significant trends for MMP-9, MMP-2, and TIMPs. Conversely, some analytes including MMP-10, MMP-12, and TIMP-4 showed slightly greater reductions with high-dose therapy, but effect sizes were small.\u003c/p\u003e \u003cp\u003eTo assess the independent effect of steroid treatment on biomarker expression while controlling for disease severity, we performed PSM analysis. Propensity score matching successfully created balanced treatment groups of 10 high steroid and 10 none/low steroid patients. To avoid overfitting with limited sample size, we matched on the composite severity score (pre-treatment) as the principal confounder; age and sex were balanced post-match (SMDs\u0026thinsp;\u0026lt;\u0026thinsp;0.05). Among 16 biomarkers analyzed, two showed significant differences after multiple comparison correction. MMP-8 fold changes were significantly higher in the high steroid group compared to none/low steroid group (FC\u0026thinsp;=\u0026thinsp;0.477 vs 0.122, p\u0026thinsp;=\u0026thinsp;0.003, Cohen's d\u0026thinsp;=\u0026thinsp;1.17), indicating that steroid treatment attenuated the decrease in MMP-8 levels: MMP-8 was reduced on average by 52% in high dose group vs. 88% in low/none steroids group. MMP-13 showed significant upregulation in the high steroid group (2.029 vs 0.822, p\u0026thinsp;=\u0026thinsp;0.008, Cohen's d\u0026thinsp;=\u0026thinsp;1.71). Several other biomarkers including TIMP-1, EMMPRIN, and MMP-9 showed trends towards greater reduction in none/low steroids group (p\u0026thinsp;\u0026lt;\u0026thinsp;0.10), while MMP-10 decreased more strongly in the high-dose group (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003e).\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec11\" class=\"Section2\"\u003e \u003ch2\u003e2.9 Biomarker differences by clinical severity\u003c/h2\u003e \u003cp\u003eTIMP-1 concentrations rose stepwise across mild, moderate, and severe cases (median 292 \u0026rarr; 360 \u0026rarr; 567 ng/mL; H\u0026thinsp;=\u0026thinsp;18.7, p\u0026thinsp;=\u0026thinsp;8.5\u0026times;10⁻⁵, η\u0026sup2;=0.38), paralleled by total TIMPs (524 \u0026rarr; 627 \u0026rarr; 829 ng/mL; H\u0026thinsp;=\u0026thinsp;10.8, p\u0026thinsp;=\u0026thinsp;0.0045). TNF-α also increased with severity (32 \u0026rarr; 36 \u0026rarr; 75 pg/mL; H\u0026thinsp;=\u0026thinsp;8.8, p\u0026thinsp;=\u0026thinsp;0.012), and MMP-7 was higher in severe disease (2.9 \u0026rarr; 3.4 \u0026rarr; 5.9 ng/mL; H\u0026thinsp;=\u0026thinsp;7.6, p\u0026thinsp;=\u0026thinsp;0.023). TIMP-4 showed a similar gradient (1.6 \u0026rarr; 2.1 \u0026rarr; 2.5 ng/mL; H\u0026thinsp;=\u0026thinsp;8.1, p\u0026thinsp;=\u0026thinsp;0.017). Several ratios, including MMP-9/TIMP-1, MMP-1/TIMP-1, and MMP-13/TIMP-1, decreased progressively with severity, suggesting stronger dominance of inhibitors relative to proteases in the sickest children.\u003c/p\u003e \u003cp\u003eTo link biomarker profiles with clinical disease burden, we analyzed correlations with a continuous severity score. TIMP-1 at presentation showed the strongest association (ρ\u0026thinsp;=\u0026thinsp;0.55, p\u0026thinsp;=\u0026thinsp;6.1\u0026times;10⁻⁵), followed by TIMP-4 (ρ\u0026thinsp;=\u0026thinsp;0.42, p\u0026thinsp;=\u0026thinsp;0.0036) and total TIMPs (ρ\u0026thinsp;=\u0026thinsp;0.41, p\u0026thinsp;=\u0026thinsp;0.0044). Among MMPs, MMP-3 (ρ\u0026thinsp;=\u0026thinsp;0.30, p\u0026thinsp;=\u0026thinsp;0.043) and MMP-8 (ρ\u0026thinsp;=\u0026thinsp;0.30, p\u0026thinsp;=\u0026thinsp;0.044) were significant, as was TNF-α (ρ\u0026thinsp;=\u0026thinsp;0.37, p\u0026thinsp;=\u0026thinsp;0.011). No associations were observed for MMP-1, MMP-9, MMP-10, or MMP-13.\u003c/p\u003e \u003cp\u003eAnalyses of clinical course showed only weak relationships with length of hospital stay. At follow-up (T2), moderate correlations were observed for MMP-9/TIMP-2 (r\u0026thinsp;=\u0026thinsp;0.47, p\u0026thinsp;=\u0026thinsp;0.006), MMP-8/TIMP-2 (r\u0026thinsp;=\u0026thinsp;0.44, p\u0026thinsp;=\u0026thinsp;0.010), MMP-9 (r\u0026thinsp;=\u0026thinsp;0.42, p\u0026thinsp;=\u0026thinsp;0.014), MMP-8 (r\u0026thinsp;=\u0026thinsp;0.42, p\u0026thinsp;=\u0026thinsp;0.014), and TIMP-3 (r\u0026thinsp;=\u0026thinsp;0.41, p\u0026thinsp;=\u0026thinsp;0.017). None remained significant after correction for multiple comparisons, and no consistent associations were seen across timepoints, suggesting that biomarker profiles do not robustly predict hospital stay in this cohort.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec12\" class=\"Section2\"\u003e \u003ch2\u003e2.10 Biomarker differences by clinical phenotype\u003c/h2\u003e \u003cp\u003eFinally, subgroup analyses revealed phenotype-specific biomarker associations, with 157 significant differences identified across 1,646 tested. Renal and cardiovascular involvement showed the densest signal clusters. Children with renal involvement demonstrated marked elevations of EMMPRIN (median 14 vs 8 ng/mL, p\u0026thinsp;=\u0026thinsp;1.7 \u0026times; 10⁻⁵) and TIMP-1 (median 680 vs 329 ng/mL, p\u0026thinsp;=\u0026thinsp;0.0001) at baseline, while shock on admission was similarly associated with high TIMP-1 and total TIMP concentrations. Cardiac complications also exhibited distinct protease\u0026ndash;inhibitor profiles: pericardial effusion was linked to differential EMMPRIN dynamics post-treatment (median fold change 0.37 vs 0.81, p\u0026thinsp;=\u0026thinsp;0.0008), and reduced left ventricular contractility showed 14 significant biomarker associations. Gastrointestinal disease and hypotension each displayed around 15 associations, again enriched for TIMP and EMMPRIN changes. In contrast, neurological, pulmonary, and hematological features were associated with fewer or less consistent biomarker differences. Together, these findings highlight TIMP-1 and EMMPRIN as recurring signals across multiple severe phenotypes, suggesting that dysregulation of the protease\u0026ndash;inhibitor axis underlies multi-organ involvement in MIS-C/KD. Detailed results are provided in Supplementary Table \u003cspan refid=\"MOESM6\" class=\"InternalRef\"\u003eS6\u003c/span\u003e.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec13\" class=\"Section2\"\u003e \u003ch2\u003e2.11 Comparison of MIS-C and Kawasaki disease\u003c/h2\u003e \u003cp\u003eAlthough MIS-C patients generally presented with more severe cardiovascular involvement and higher systemic inflammation (Supplementary Table \u003cspan refid=\"MOESM7\" class=\"InternalRef\"\u003eS7\u003c/span\u003e), direct biomarker comparisons with Kawasaki disease revealed only modest differences. At the uncorrected level, several analytes showed higher concentrations in KD, including MMP-7 (notably at follow-up), MMP-10 at baseline, and their ratios to TIMPs (MMP-7/TIMP-1, MMP-7/TIMP-2, MMP-10/TIMP-1, MMP-10/TIMP-4). In contrast, MMP-1 tended to be lower in KD than MIS-C. However, none of these associations remained significant after correction for multiple comparisons, reflecting both the small KD sample size and the substantial overlap between groups. Because MIS-C and KD overlap clinically, particularly after 2020, we also analyzed biomarker profiles according to the evidence of of SARS-CoV-2 recent infection. Several markers, including MMP-13 (0.98 vs 1.30 ng/mL) and related ratios (MMP-13/TIMP-1, MMP-13/TIMP-2, MMP-13/TIMP-4), as well as MMP-10/TIMP-1, were nominally lower in virus-exposed children, but none remained significant after correction. Importantly, the major discriminatory signals identified in our primary analysis \u0026mdash; TIMP-1, MMP-8, MMP-3, TNF-α, and MMP-8/TIMP ratios \u0026mdash; did not differ between MIS-C and KD, underscoring the shared protease\u0026ndash;inhibitor dysregulation that defines hyperinflammatory state in both conditions.\u003c/p\u003e \u003c/div\u003e"},{"header":"3. Discussion","content":"\u003cp\u003eThis study shows that children with MIS-C and Kawasaki disease (KD) share a common protease\u0026ndash;inhibitor signature that clearly distinguishes them from febrile and healthy controls. The profile is dominated by elevated TIMP-1, MMP-8, MMP-3, and TNF-α, together with shifts in MMP/TIMP ratios indicating a net proteolytic bias. These findings place dysregulation of the protease\u0026ndash;inhibitor system at the center of pediatric hyperinflammatory conditions and extend prior observations from KD into MIS-C. Recent reviews underscore endothelial injury and microvascular dysfunction as key elements of MIS-C pathobiology, with variable benefits of IVIG and corticosteroids across studies [\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e]. Our results also align with prior pediatric COVID-19 work showing elevations of MMP-3, MMP-8, and MMP-9 compared with healthy children and lower levels than in adults [\u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eIndependent cohorts document endothelial dysfunction and arterial stiffness in MIS-C, and endothelial biomarkers such as endocan (ESM-1) are elevated\u0026mdash;consistent with our observation that the upstream inducer EMMPRIN decreased after therapy [\u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e26\u003c/span\u003e, \u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e27\u003c/span\u003e]. The strong signal from neutrophil-associated proteases, particularly MMP-8 with its correlated cluster (MMP-12, and MMP-13 which can be activated by neutrophil elastase), underscores innate immune activation in MIS-C and KD; many of these signals declined with treatment [\u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e28\u003c/span\u003e, \u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e29\u003c/span\u003e]. In KD, elevated MMP-8 and MMP-9 have long been linked to coronary artery damage and aneurysm formation [\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e, \u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e30\u003c/span\u003e], while in MIS-C, MMPs are among the most discriminatory serum proteins differentiating cases from acute COVID-19 or dengue [\u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e]. Our ROC analyses reinforce these observations: TIMP-1, MMP-3, and MMP-8 each achieved AUCs\u0026thinsp;\u0026gt;\u0026thinsp;0.9, suggesting promising diagnostic utility that warrants validation in larger and more diverse cohorts, particularly where overlapping infections complicate clinical diagnosis.\u003c/p\u003e \u003cp\u003eTIMP-1 emerged as the biomarker most consistently associated with severity, showing the strongest correlations with a composite severity score and a clear stepwise increase across mild, moderate, and severe disease. Across clinical phenotypes, TIMP-1 and EMMPRIN were the most recurrent signals, suggesting a shared upstream axis tied to multi-organ involvement. This is compatible with the multifunctional role of TIMP-1 as both an MMP inhibitor and an acute-phase, cytokine-like protein [\u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e31\u003c/span\u003e]. Notably, the most severe cases exhibited declining MMP/TIMP ratios, indicating relative inhibitor dominance despite escalating inflammation\u0026mdash;potentially reflecting compensatory TIMP upregulation to restrain proteolysis. Similar patterns have been reported in critically ill MIS-C patients in India [\u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e], supporting the idea that the balance between protease activity and TIMP response, rather than absolute MMP abundance alone, may track clinical severity. Conversely, lower post-treatment TIMP-1 has been associated with persistent coronary artery lesions in KD, raising the possibility that TIMP-1 contributes to vascular healing [\u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e28\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eClassical inflammatory markers (CRP, ESR, ferritin, procalcitonin) remain central to MIS-C and KD assessment, but they primarily reflect systemic inflammation. By contrast, MMPs and TIMPs capture extracellular matrix remodeling, neutrophil and macrophage activation, and vascular injury. The robust elevation of TIMP-1 and MMP-8 in our cohort highlights how matrix-directed biomarkers may complement conventional laboratory tests by reporting a distinct dimension of the hyperinflammatory response [\u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e32\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eTreatment dynamics provide further insight into regulation of this system. Immunomodulation was associated with a marked reduction in EMMPRIN, a known inducer of MMP expression [\u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e33\u003c/span\u003e], indicating upstream suppression of protease activation. In parallel, MMP-1 concentrations rose with corresponding increases in MMP-1/TIMP ratios. This rebound may reflect activation of collagenase-dependent remodeling during vascular repair, consistent with reported roles of MMP-1 in angiogenesis and endothelial permeability [\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e]. Overall, rather than uniform suppression, therapy appears to selectively rebalance individual MMP/TIMP axes with potential implications for vascular recovery [\u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e34\u003c/span\u003e]. Within this context, our exploratory dose\u0026ndash;response analysis\u0026mdash;situated among heterogeneous real-world immunomodulation data [\u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e35\u003c/span\u003e] \u0026mdash;suggested that high-dose pulse steroids blunted the post-treatment decline of neutrophil-related proteases (MMP-8 and MMP-13). Although limited by sample size, these results raise the hypothesis that very high steroid exposure may incompletely suppress neutrophil protease activity, potentially prolonging vascular injury. Prior KD studies have similarly reported variable effects of IVIG and steroids on MMP expression [\u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e36\u003c/span\u003e], underscoring the need to evaluate treatment effects on protease biology systematically.\u003c/p\u003e \u003cp\u003ePhenotype-stratified analyses indicated that biomarker changes mirror organ involvement. Liver disease associated with broad upregulation of TIMPs and neutrophil proteases, whereas gastrointestinal presentations showed selective reductions in MMP-1/TIMP ratios. These patterns are consistent with literature framing extracellular matrix remodeling as a driver of organ injury in systemic inflammation [\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e]. Recent proteomic work likewise separates MIS-C and KD from other pediatric hyperinflammatory states and highlights cytokine/immune-activation axes that co-vary with our MMP/TIMP signature [\u003cspan citationid=\"CR37\" class=\"CitationRef\"\u003e37\u003c/span\u003e, \u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e38\u003c/span\u003e]. Direct MIS-C vs KD comparisons in our cohort revealed only modest differences that did not survive correction for multiple testing, supporting the view that the protease\u0026ndash;inhibitor signature reflects a shared hyperinflammatory vasculopathy phenotype rather than the infectious trigger per se and that MIS-C and KD occupy overlapping immunopathologic space [\u003cspan additionalcitationids=\"CR40\" citationid=\"CR39\" class=\"CitationRef\"\u003e39\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR41\" class=\"CitationRef\"\u003e41\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eThis study has limitations. It is single-center, the KD group was relatively small (limiting power for disease-specific contrasts), and we measured protein concentrations rather than enzymatic activity (which may diverge due to post-translational regulation and endogenous inhibitors). Treatment heterogeneity and the observational design also constrain causal inference about therapy effects. Nevertheless, strengths include a comparatively large MIS-C cohort, inclusion of both febrile and healthy controls, broad multiplex profiling of MMPs, TIMPs, and inducers, and paired sampling before and after immunomodulation\u0026mdash;features that together provide an integrated view of biomarkers at presentation, their diagnostic performance, their relation to clinical severity and organ involvement, and their evolution with treatment.\u003c/p\u003e \u003cp\u003eIn summary, dysregulation of the MMP/TIMP system is a defining feature of pediatric hyperinflammatory vasculopathy. TIMP-1 and MMP-8 were the most informative markers, capturing diagnostic separation and severity associations, while treatment produced selective rebalancing of the protease\u0026ndash;inhibitor network. Pooling MIS-C and KD for primary analyses, with disease-specific and serology-based contrasts treated as exploratory, allowed us to interrogate shared and divergent biology across this spectrum. These data support further evaluation of the protease\u0026ndash;inhibitor axis as a mechanistic and biomarker framework for MIS-C and KD, with prospective validation needed to define clinical utility.\u003c/p\u003e"},{"header":"4. Methods","content":"\u003cdiv id=\"Sec16\" class=\"Section2\"\u003e \u003ch2\u003e\u003cb\u003e4.1\u003c/b\u003e Study Design and Participants\u003c/h2\u003e \u003cp\u003eThis was a prospective observational study with both cross-sectional and longitudinal components, conducted at the University Children\u0026rsquo;s Clinical Hospital of the Medical University of Bialystok, Poland. Four cohorts were included: children with multisystem inflammatory syndrome in children (MIS-C), Kawasaki disease (KD), febrile pediatric controls (FC), and healthy pediatric controls (HC). Patients with MIS-C and KD were diagnosed according to the Centers for Disease Control and Prevention (CDC) and American Heart Association (AHA) criteria, respectively. Because MIS-C and KD share overlapping clinical phenotypes and host responses, and differentiation can be challenging in children with positive SARS-CoV-2 IgG antibodies, we prespecified pooled primary analyses (MIS-C/KD vs. FC/HC), with MIS-C versus KD comparisons treated as exploratory.\u003c/p\u003e \u003cp\u003eBlood samples for MIS-C and KD were collected prospectively at baseline (T1, before treatment) and 4\u0026ndash;5 days after defervescence (T2, following IVIG and/or corticosteroid therapy). Controls were matched for sex and age when feasible. Clinical laboratory information was abstracted retrospectively from electronic medical records on a standardized case report form without access to the outcome of the biomarkers.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec17\" class=\"Section2\"\u003e \u003ch2\u003e4.2 Sample Collection and Storage\u003c/h2\u003e \u003cp\u003ePeripheral blood samples were collected into serum tubes, centrifuged within 2 hours, and stored at \u0026minus;\u0026thinsp;80\u0026deg;C until analysis. All samples were processed using standardized protocols to minimize pre-analytical variability.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec18\" class=\"Section2\"\u003e \u003ch2\u003e4.3 Protein Quantification\u003c/h2\u003e \u003cp\u003eSerum concentrations of 9 matrix metalloproteinases (MMP-1, -2, -3, -7, -8, -9, -10, -12, -13), 4 tissue inhibitors of metalloproteinases (TIMP-1, -2, -3, -4), TNF-α, and EMMPRIN were measured using custom and commercially available Luminex\u0026reg; multiplex bead-based immunoassays (R\u0026amp;D Systems/Bio-Techne, Minneapolis, MN).\u003c/p\u003e \u003cp\u003eSamples were assayed in three batches over two years; batch-to-batch variation was addressed by normalization using internal control samples repeated across batches and by median ratio correction for each marker.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec19\" class=\"Section2\"\u003e \u003ch2\u003e4.4 Clinical severity score\u003c/h2\u003e \u003cp\u003eTo capture the extent of clinical involvement in MIS-C and Kawasaki disease, we developed a composite severity score integrating organ dysfunction and standard laboratory indices, adapted from commonly used severity markers in pediatric inflammatory syndromes.[\u003cspan additionalcitationids=\"CR43 CR44\" citationid=\"CR42\" class=\"CitationRef\"\u003e42\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR45\" class=\"CitationRef\"\u003e45\u003c/span\u003e] Twelve domains were included (Supplementary table \u003cspan refid=\"MOESM8\" class=\"InternalRef\"\u003eS8\u003c/span\u003e). Each domain was assigned a score reflecting the degree of abnormality, and the total score ranged from 0 to 20, with higher values indicating greater severity. The domains were: (i) hemodynamic instability/shock, (ii) cardiac involvement, (iii) respiratory support, (iv) neurological involvement, (v) gastrointestinal symptoms, (vi) inflammatory markers (CRP, PCT), (vii) coagulopathy (D-dimer, APTT/INR), (viii) liver function and albumin, (ix) hyponatremia, (x) lymphopenia, (xi) prolonged fever, and (xii) ICU admission.\u003c/p\u003e \u003cp\u003eThis score was used both categorically (mild, moderate, severe) and continuously. For categorical analysis, children were initially stratified into mild (\u0026le;\u0026thinsp;7 points), moderate (8\u0026ndash;11 points), and severe (\u0026ge;\u0026thinsp;12 points). To minimize information loss, correlations between continuous severity scores and biomarker concentrations were also performed.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec20\" class=\"Section2\"\u003e \u003ch2\u003e4.5 Statistical Analysis\u003c/h2\u003e \u003cp\u003eBecause biomarker distributions were non-normal, non-parametric tests were used throughout. Overall comparisons between MIS-C/KD, febrile controls, and healthy controls were performed with the Kruskal\u0026ndash;Wallis test, followed by post hoc pairwise Mann\u0026ndash;Whitney U tests. Effect sizes were reported as η\u0026sup2; for Kruskal\u0026ndash;Wallis tests and rank-biserial correlation (r) for Mann\u0026ndash;Whitney and Wilcoxon signed-rank tests. Where parametric comparisons or propensity score\u0026ndash;matched analyses were appropriate, effect sizes were expressed as Cohen\u0026rsquo;s d. To control for multiple comparisons, Bonferroni correction was applied; unadjusted p-values are presented for exploratory analyses. Diagnostic performance of candidate biomarkers was assessed using receiver operating characteristic (ROC) curves. Area under the curve (AUC) values were reported with 95% confidence intervals, with thresholds of \u0026ge;\u0026thinsp;0.7, \u0026ge;\u0026thinsp;0.8, and \u0026ge;\u0026thinsp;0.9 interpreted as acceptable, good, and excellent discrimination, respectively. Optimal cut-off values were defined by the Youden index, and corresponding sensitivity and specificity values were calculated. To evaluate the impact of steroid treatment, fold changes (FC) were calculated as the ratio of follow-up (T2) to baseline (T1) concentrations. Propensity score matching (PSM) was used to minimize confounding by disease severity, with propensity scores estimated from logistic regression models including the calculated clinical severity score. One-to-one nearest-neighbor matching without replacement was performed, and covariate balance was verified using standardized mean differences (SMD), with SMD\u0026thinsp;\u0026lt;\u0026thinsp;0.1 indicating adequate balance. Biomarker fold changes were compared between matched groups using Mann\u0026ndash;Whitney U tests. All analyses were two-sided with a significance threshold of 0.05. Missing data were handled by pairwise deletion without imputation. Statistical analyses were performed in Python (SciPy, scikit-learn), GraphPad Prism (GraphPad Software, San Diego, CA), and Statistica (TIBCO Software Inc., v14).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec21\" class=\"Section2\"\u003e \u003ch2\u003e4.6 Ethics Statement\u003c/h2\u003e \u003cp\u003e This study was conducted according to the declaration of Helsinki and approved by the Ethics Committee of Medical University of Bialystok (approval no. APK.002.259.2020 and APK.002.428.2022). A written informed consent was obtained from all studied patients and/or patient parents/caregivers.\u003c/p\u003e \u003c/div\u003e"},{"header":"Declarations","content":"\u003cp\u003eAuthor contributions\u003c/p\u003e\n\u003cp\u003eConceptualization: KT, AS; Data curation: DL, DM; Formal Analysis: KT, DM; Funding acquisition: KT; Investigation: KT, DM; Project administration: AS; Resources: KT, DL, AS;; Supervision: AS; Visualization: KT; Writing \u0026ndash; original draft: KT; Writing \u0026ndash; review \u0026amp; editing: KT, DL, DM, AS\u003c/p\u003e\n\u003cp\u003eAcknowledgements\u003c/p\u003e\n\u003cp\u003eWe thank Prof. Robert Flisiak, Head of the Department of Infectious Diseases and Hepatology at the Medical University of Białystok, Poland, for providing institutional support and maintaining the laboratory environment that enabled this work.\u003c/p\u003e\n\u003cp\u003eArtificial intelligence\u0026ndash;assisted tools (ChatGPT, OpenAI) were used to improve the clarity and grammar of the manuscript text. The authors reviewed and take full responsibility for all content.\u003c/p\u003e\n\u003cp\u003eData availability statement\u003c/p\u003e\n\u003cp\u003eThe datasets generated and analyzed during the current study are available in the Harvard Dataverse repository, https://doi.org/10.7910/DVN/XAHZHY. All other data supporting the findings of this study are provided within the article and its Supplementary Information files.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCompeting interests\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eKT, DL, and AS were employed by Pfizer in a clinical trial program evaluating a Lyme disease vaccine. KT has received honoraria from GSK for lectures on meningococcal disease and vaccines. The authors declare no other competing interests.\u003c/p\u003e\n\u003cp\u003eFunding\u003c/p\u003e\n\u003cp\u003eThis work was supported by the Medical University of Bialystok, Poland (Grants No. B.SUB.25.440 and B.SUB.24.448).\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n \u003cli\u003eMultisystem Inflammatory Syndrome in Children (MIS-C) Associated with SARS-CoV-2 Infection 2023 Case Definition | CDC. (2024). at \u0026lt;https://ndc.services.cdc.gov/case-definitions/multisystem-inflammatory-syndrome-in-children-mis-c-2023/\u0026gt;\u003c/li\u003e\n \u003cli\u003eConsiglio, C. R. \u003cem\u003eet al.\u003c/em\u003e The Immunology of Multisystem Inflammatory Syndrome in Children with COVID-19. \u003cem\u003eCell\u003c/em\u003e \u003cstrong\u003e183,\u003c/strong\u003e 968-981.e7 (2020).\u003c/li\u003e\n \u003cli\u003eHoste, L., Van Paemel, R. \u0026amp; Haerynck, F. Multisystem inflammatory syndrome in children related to COVID-19: a systematic review. \u003cem\u003eEur J Pediatr\u003c/em\u003e \u003cstrong\u003e180,\u003c/strong\u003e 2019\u0026ndash;2034 (2021).\u003c/li\u003e\n \u003cli\u003eJiang, L. \u003cem\u003eet al.\u003c/em\u003e COVID-19 and multisystem inflammatory syndrome in children and adolescents. \u003cem\u003eLancet Infect Dis\u003c/em\u003e \u003cstrong\u003e20,\u003c/strong\u003e e276\u0026ndash;e288 (2020).\u003c/li\u003e\n \u003cli\u003eRamcharan, T. \u003cem\u003eet al.\u003c/em\u003e Paediatric Inflammatory Multisystem Syndrome: Temporally Associated with SARS-CoV-2 (PIMS-TS): Cardiac Features, Management and Short-Term Outcomes at a UK Tertiary Paediatric Hospital. \u003cem\u003ePediatr Cardiol\u003c/em\u003e \u003cstrong\u003e41,\u003c/strong\u003e 1391\u0026ndash;1401 (2020).\u003c/li\u003e\n \u003cli\u003eLampidi, S. \u003cem\u003eet al.\u003c/em\u003e Multisystem inflammatory syndrome in children (MIS-C): A nationwide collaborative study in the Greek population. \u003cem\u003eEur J Pediatr\u003c/em\u003e \u003cstrong\u003e183,\u003c/strong\u003e 1693\u0026ndash;1702 (2024).\u003c/li\u003e\n \u003cli\u003eKawasaki, T. [Acute febrile mucocutaneous syndrome with lymphoid involvement with specific desquamation of the fingers and toes in children]. \u003cem\u003eArerugi\u003c/em\u003e \u003cstrong\u003e16,\u003c/strong\u003e 178\u0026ndash;222 (1967).\u003c/li\u003e\n \u003cli\u003eBurns, J. C. \u0026amp; Glod\u0026eacute;, M. P. Kawasaki syndrome. \u003cem\u003eThe Lancet\u003c/em\u003e \u003cstrong\u003e364,\u003c/strong\u003e 533\u0026ndash;544 (2004).\u003c/li\u003e\n \u003cli\u003eOnouchi, Y. The genetics of Kawasaki disease. \u003cem\u003eInt J Rheum Dis\u003c/em\u003e \u003cstrong\u003e21,\u003c/strong\u003e 26\u0026ndash;30 (2018).\u003c/li\u003e\n \u003cli\u003eOlejarz, W., Łacheta, D. \u0026amp; Kubiak-Tomaszewska, G. Matrix Metalloproteinases as Biomarkers of Atherosclerotic Plaque Instability. \u003cem\u003eInternational Journal of Molecular Sciences\u003c/em\u003e \u003cstrong\u003e21,\u003c/strong\u003e 3946 (2020).\u003c/li\u003e\n \u003cli\u003eVisse, R. \u0026amp; Nagase, H. Matrix metalloproteinases and tissue inhibitors of metalloproteinases: structure, function, and biochemistry. \u003cem\u003eCirc Res\u003c/em\u003e \u003cstrong\u003e92,\u003c/strong\u003e 827\u0026ndash;839 (2003).\u003c/li\u003e\n \u003cli\u003eParks, W. C., Wilson, C. L. \u0026amp; L\u0026oacute;pez-Boado, Y. S. Matrix metalloproteinases as modulators of inflammation and innate immunity. \u003cem\u003eNat Rev Immunol\u003c/em\u003e \u003cstrong\u003e4,\u003c/strong\u003e 617\u0026ndash;629 (2004).\u003c/li\u003e\n \u003cli\u003eBonnans, C., Chou, J. \u0026amp; Werb, Z. Remodelling the extracellular matrix in development and disease. \u003cem\u003eNat Rev Mol Cell Biol\u003c/em\u003e \u003cstrong\u003e15,\u003c/strong\u003e 786\u0026ndash;801 (2014).\u003c/li\u003e\n \u003cli\u003eLu, P., Takai, K., Weaver, V. M. \u0026amp; Werb, Z. Extracellular Matrix Degradation and Remodeling in Development and Disease. \u003cem\u003eCold Spring Harb Perspect Biol\u003c/em\u003e \u003cstrong\u003e3,\u003c/strong\u003e a005058 (2011).\u003c/li\u003e\n \u003cli\u003eGalis, Z. S. \u0026amp; Khatri, J. J. Matrix metalloproteinases in vascular remodeling and atherogenesis: the good, the bad, and the ugly. \u003cem\u003eCirc Res\u003c/em\u003e \u003cstrong\u003e90,\u003c/strong\u003e 251\u0026ndash;262 (2002).\u003c/li\u003e\n \u003cli\u003eLenglet, S., Mach, F. \u0026amp; Montecucco, F. Role of Matrix Metalloproteinase-8 in Atherosclerosis. \u003cem\u003eMediators Inflamm\u003c/em\u003e \u003cstrong\u003e2013,\u003c/strong\u003e 659282 (2013).\u003c/li\u003e\n \u003cli\u003eLee, E.-J. \u003cem\u003eet al.\u003c/em\u003e Regulation of neuroinflammation by matrix metalloproteinase-8 inhibitor derivatives in activated microglia and astrocytes. \u003cem\u003eOncotarget\u003c/em\u003e \u003cstrong\u003e8,\u003c/strong\u003e 78677\u0026ndash;78690 (2017).\u003c/li\u003e\n \u003cli\u003eChua, P. K. \u003cem\u003eet al.\u003c/em\u003e Elevated Levels of Matrix Metalloproteinase 9 and Tissue Inhibitor of Metalloproteinase 1 during the Acute Phase of Kawasaki Disease. \u003cem\u003eClin Vaccine Immunol\u003c/em\u003e \u003cstrong\u003e10,\u003c/strong\u003e 308\u0026ndash;314 (2003).\u003c/li\u003e\n \u003cli\u003eGavin, P. J., Crawford, S. E., Shulman, S. T., Garcia, F. L. \u0026amp; Rowley, A. H. Systemic arterial expression of matrix metalloproteinases 2 and 9 in acute Kawasaki disease. \u003cem\u003eArterioscler Thromb Vasc Biol\u003c/em\u003e \u003cstrong\u003e23,\u003c/strong\u003e 576\u0026ndash;581 (2003).\u003c/li\u003e\n \u003cli\u003eSekiguchi, K. \u003cem\u003eet al.\u003c/em\u003e The Imbalance between Matrix Metalloproteinase-9, -2 and Tissue Inhibi tor of Metalloproteinases-1 in Acute Phase Kawasaki Disease. \u003cem\u003ePediatr Res\u003c/em\u003e \u003cstrong\u003e53,\u003c/strong\u003e 170\u0026ndash;170 (2003).\u003c/li\u003e\n \u003cli\u003eRies, C. Cytokine functions of TIMP-1. \u003cem\u003eCell Mol Life Sci\u003c/em\u003e \u003cstrong\u003e71,\u003c/strong\u003e 659\u0026ndash;672 (2014).\u003c/li\u003e\n \u003cli\u003eMoore, C. S. \u0026amp; Crocker, S. J. An alternate perspective on the roles of TIMPs and MMPs in pathology. \u003cem\u003eAm J Pathol\u003c/em\u003e \u003cstrong\u003e180,\u003c/strong\u003e 12\u0026ndash;16 (2012).\u003c/li\u003e\n \u003cli\u003ePavan Kumar, N. \u003cem\u003eet al.\u003c/em\u003e Role of matrix metalloproteinases in multi-system inflammatory syndrome and acute COVID-19 in children. \u003cem\u003eFront Med (Lausanne)\u003c/em\u003e \u003cstrong\u003e9,\u003c/strong\u003e 1050804 (2022).\u003c/li\u003e\n \u003cli\u003eNewby, A. C. Dual role of matrix metalloproteinases (matrixins) in intimal thickening and atherosclerotic plaque rupture. \u003cem\u003ePhysiol Rev\u003c/em\u003e \u003cstrong\u003e85,\u003c/strong\u003e 1\u0026ndash;31 (2005).\u003c/li\u003e\n \u003cli\u003eToczyłowski, K. \u003cem\u003eet al.\u003c/em\u003e Differential Inflammatory Responses in Adult and Pediatric COVID-19 Patients: Implications for Long-Term Consequences and Anti-Inflammatory Treatment. \u003cem\u003eMed Sci Monit\u003c/em\u003e \u003cstrong\u003e30,\u003c/strong\u003e e944052 (2024).\u003c/li\u003e\n \u003cli\u003eCannavo, A. \u003cem\u003eet al.\u003c/em\u003e Serum endocan (ESM-1) as diagnostic and prognostic biomarker in Multisystem inflammatory syndrome in children (MIS-C). \u003cem\u003eCytokine\u003c/em\u003e \u003cstrong\u003e184,\u003c/strong\u003e 156797 (2024).\u003c/li\u003e\n \u003cli\u003eAhmadi, A. \u003cem\u003eet al.\u003c/em\u003e Vascular function and arterial stiffness in multisystem inflammatory syndrome in children with Covid-19. \u003cem\u003eARYA Atheroscler\u003c/em\u003e \u003cstrong\u003e21,\u003c/strong\u003e 54\u0026ndash;62 (2025).\u003c/li\u003e\n \u003cli\u003eLi, Y. \u003cem\u003eet al.\u003c/em\u003e Dynamic changes of serum matrix metalloproteinases and tissue inhibitors in Kawasaki disease: implications for coronary artery lesions persistence. \u003cem\u003eItal J Pediatr\u003c/em\u003e \u003cstrong\u003e51,\u003c/strong\u003e 169 (2025).\u003c/li\u003e\n \u003cli\u003eVandooren, J., Van den Steen, P. E. \u0026amp; Opdenakker, G. Biochemistry and molecular biology of gelatinase B or matrix metalloproteinase-9 (MMP-9): the next decade. \u003cem\u003eCrit Rev Biochem Mol Biol\u003c/em\u003e \u003cstrong\u003e48,\u003c/strong\u003e 222\u0026ndash;272 (2013).\u003c/li\u003e\n \u003cli\u003eSenzaki, H. The pathophysiology of coronary artery aneurysms in Kawasaki disease: role of matrix metalloproteinases. \u003cem\u003eArch Dis Child\u003c/em\u003e \u003cstrong\u003e91,\u003c/strong\u003e 847\u0026ndash;851 (2006).\u003c/li\u003e\n \u003cli\u003eJackson, H. W., Defamie, V., Waterhouse, P. \u0026amp; Khokha, R. TIMPs: versatile extracellular regulators in cancer. \u003cem\u003eNat Rev Cancer\u003c/em\u003e \u003cstrong\u003e17,\u003c/strong\u003e 38\u0026ndash;53 (2017).\u003c/li\u003e\n \u003cli\u003eHenderson, L. A. \u003cem\u003eet al.\u003c/em\u003e American College of Rheumatology Clinical Guidance for Multisystem Inflammatory Syndrome in Children Associated With SARS\u0026ndash;CoV-2 and Hyperinflammation in Pediatric COVID-19: Version 2. \u003cem\u003eArthritis \u0026amp; Rheumatology\u003c/em\u003e \u003cstrong\u003e73,\u003c/strong\u003e e13\u0026ndash;e29 (2021).\u003c/li\u003e\n \u003cli\u003evon Ungern-Sternberg, S. N. I., Zernecke, A. \u0026amp; Seizer, P. Extracellular Matrix Metalloproteinase Inducer EMMPRIN (CD147) in Cardiovascular Disease. \u003cem\u003eInt J Mol Sci\u003c/em\u003e \u003cstrong\u003e19,\u003c/strong\u003e 507 (2018).\u003c/li\u003e\n \u003cli\u003eTian, F. \u003cem\u003eet al.\u003c/em\u003e Correlation Between Matrix Metalloproteinases With Coronary Artery Lesion Caused by Kawasaki Disease. \u003cem\u003eFront Pediatr\u003c/em\u003e \u003cstrong\u003e10,\u003c/strong\u003e 802217 (2022).\u003c/li\u003e\n \u003cli\u003eMcArdle, A. J. \u003cem\u003eet al.\u003c/em\u003e Immunomodulatory Treatment of Multisystem Inflammatory Syndrome in Children. \u003cem\u003eThe New England journal of medicine\u003c/em\u003e \u003cstrong\u003e385,\u003c/strong\u003e 11 (2021).\u003c/li\u003e\n \u003cli\u003eSakata, K. \u003cem\u003eet al.\u003c/em\u003e Matrix Metalloproteinase-9 in Vascular Lesions and Endothelial Regulat ion in Kawasaki Disease. \u003cem\u003eCirc J\u003c/em\u003e \u003cstrong\u003e74,\u003c/strong\u003e 1670\u0026ndash;1675 (2010).\u003c/li\u003e\n \u003cli\u003eReiter, A. \u003cem\u003eet al.\u003c/em\u003e Proteomic mapping identifies serum marker signatures associated with MIS-C specific hyperinflammation and cardiovascular manifestation. \u003cem\u003eClinical Immunology\u003c/em\u003e \u003cstrong\u003e264,\u003c/strong\u003e 110237 (2024).\u003c/li\u003e\n \u003cli\u003eDourdouna, M.-M., Tatsi, E.-B., Syriopoulou, V. \u0026amp; Michos, A. Proteomic Signatures of Multisystem Inflammatory Syndrome in Children (MIS-C) Associated with COVID-19: A Narrative Review. \u003cem\u003eChildren (Basel)\u003c/em\u003e \u003cstrong\u003e11,\u003c/strong\u003e 1174 (2024).\u003c/li\u003e\n \u003cli\u003eTsoukas, P. \u0026amp; Yeung, R. S. M. Kawasaki disease and MIS-C share a host immune response. \u003cem\u003eNat Rev Rheumatol\u003c/em\u003e \u003cstrong\u003e18,\u003c/strong\u003e 555\u0026ndash;556 (2022).\u003c/li\u003e\n \u003cli\u003eGhosh, P. \u003cem\u003eet al.\u003c/em\u003e An Artificial Intelligence-guided signature reveals the shared host immune response in MIS-C and Kawasaki disease. \u003cem\u003eNat Commun\u003c/em\u003e \u003cstrong\u003e13,\u003c/strong\u003e 2687 (2022).\u003c/li\u003e\n \u003cli\u003eNoval Rivas, M. \u0026amp; Arditi, M. Kawasaki Disease and Multisystem Inflammatory Syndrome in Children. \u003cem\u003eRheum Dis Clin North Am\u003c/em\u003e \u003cstrong\u003e49,\u003c/strong\u003e 647\u0026ndash;659 (2023).\u003c/li\u003e\n \u003cli\u003eMolloy, M. J. \u003cem\u003eet al.\u003c/em\u003e Epidemiology and Severity of Illness of MIS-C and Kawasaki Disease During the COVID-19 Pandemic. \u003cem\u003ePediatrics\u003c/em\u003e \u003cstrong\u003e152,\u003c/strong\u003e e2023062101 (2023).\u003c/li\u003e\n \u003cli\u003eSavorgnan, F. \u003cem\u003eet al.\u003c/em\u003e Physiologic profile associated with severe multisystem inflammatory syndrome in children: a retrospective study. \u003cem\u003ePediatr Res\u003c/em\u003e \u003cstrong\u003e93,\u003c/strong\u003e 102\u0026ndash;109 (2023).\u003c/li\u003e\n \u003cli\u003eTran, D. M. \u003cem\u003eet al.\u003c/em\u003e Severity predictors for multisystemic inflammatory syndrome in children after SARS-CoV-2 infection in Vietnam. \u003cem\u003eSci Rep\u003c/em\u003e \u003cstrong\u003e14,\u003c/strong\u003e 15810 (2024).\u003c/li\u003e\n \u003cli\u003eMartin, B., Rao, S. \u0026amp; Bennett, T. D. Disparities in Multisystem Inflammatory Syndrome in Children and COVID-19 Across the Organ Dysfunction Continuum. \u003cem\u003eJAMA Netw Open\u003c/em\u003e \u003cstrong\u003e6,\u003c/strong\u003e e2249552 (2023).\u003c/li\u003e\n\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"scientific-reports","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"scirep","sideBox":"Learn more about [Scientific Reports](http://www.nature.com/srep/)","snPcode":"","submissionUrl":"","title":"Scientific Reports","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Scientific Reports","inReviewEnabled":true,"inReviewRevisionsEnabled":true},"keywords":"Multisystem Inflammatory Syndrome in Children, Kawasaki Disease, Matrix Metalloproteinases, Tissue Inhibitor of Metalloproteinases, Vasculitis, Biomarkers","lastPublishedDoi":"10.21203/rs.3.rs-7754190/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-7754190/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eMultisystem inflammatory syndrome in children (MIS-C) and Kawasaki disease (KD) are pediatric hyperinflammatory conditions with overlapping features, but their molecular basis remains poorly defined. Matrix metalloproteinases (MMPs) and their endogenous tissue inhibitors of metalloproteinases \u0026nbsp;(TIMPs), and inducers like EMMPRIN, are a tightly regulated proteolytic system that controls extracellular matrix ECM homeostasis and modulate immune responses. We profiled serum concentrations of MMPs, TIMPs, and related mediators (EMMPRIN and TNF-α) in 42 children with MIS-C, 5 with KD, 41 febrile controls, and 25 healthy controls. At presentation, MIS-C/KD showed a distinct protease–inhibitor signature characterized by markedly elevated TIMP-1, MMP-8, MMP-3, and TNF-α, with ratios indicating a net proteolytic bias despite compensatory TIMP upregulation. TIMP-1 and MMP-8 demonstrated excellent diagnostic performance (AUC \u0026gt;0.94 vs controls), and TIMP-1 correlated most strongly with clinical severity (ρ=0.55). Treatment with IVIG and corticosteroids reduced EMMPRIN but was paradoxically associated with rebound increases in MMP-1 activity. High-dose steroids blunted the post-treatment decline of neutrophil-derived proteases (MMP-8 and MMP-13). Subgroup analyses linked renal and cardiovascular involvement, shock, and gastrointestinal disease to the most pronounced biomarker alterations, driven by TIMP-1 and EMMPRIN elevations. Direct comparisons between MIS-C and KD revealed only modest, non-significant differences after correction, while SARS-CoV-2 serology status did not affect the main biomarker signals. These findings identify dysregulated MMP/TIMP balance as a central feature of pediatric hyperinflammatory conditions and highlight TIMP-1 as a marker of disease severity.\u003c/p\u003e","manuscriptTitle":"Matrix metalloproteinase–inhibitor imbalance in Kawasaki disease and multisystem inflammatory syndrome in children","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2026-04-10 16:51:35","doi":"10.21203/rs.3.rs-7754190/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"editorInvitedReview","content":"","date":"2026-04-15T08:05:28+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"243079770903679070841278567198826517697","date":"2026-04-06T10:20:01+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2026-04-06T08:51:43+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2026-02-03T16:03:47+00:00","index":"","fulltext":""},{"type":"editorInvited","content":"","date":"2025-10-17T08:17:14+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2025-10-09T12:05:44+00:00","index":"","fulltext":""},{"type":"submitted","content":"Scientific Reports","date":"2025-10-06T10:59:28+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"scientific-reports","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"scirep","sideBox":"Learn more about [Scientific Reports](http://www.nature.com/srep/)","snPcode":"","submissionUrl":"","title":"Scientific Reports","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Scientific Reports","inReviewEnabled":true,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"c04bda8c-288f-426b-a97c-e1d1b6f35ef8","owner":[],"postedDate":"April 10th, 2026","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"under-review","subjectAreas":[{"id":65812120,"name":"Health sciences/Biomarkers"},{"id":65812121,"name":"Health sciences/Diseases"},{"id":65812122,"name":"Biological sciences/Immunology"},{"id":65812123,"name":"Health sciences/Medical research"}],"tags":[],"updatedAt":"2026-04-10T16:51:35+00:00","versionOfRecord":[],"versionCreatedAt":"2026-04-10 16:51:35","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-7754190","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-7754190","identity":"rs-7754190","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}