Interplay of NF-κB and PPAR-γ transcription factors in patients with juvenile systemic lupus erythematosus

Abstract

Objective

Juvenile SLE (jSLE) is an autoimmune disease characterised by the presence of high levels of autoantibodies, predominantly targeting nuclear antigens, resulting in a breakdown of self-tolerance. However, its pathogenesis is multifactorial and poorly understood. The aim of this study was to evaluate the potential of nuclear factor-kappa B (NF-κB) and peroxisome proliferator-activated receptor-gamma (PPAR-γ) as biomarkers for jSLE.

Methods

In this study, serum NF-κB and PPAR-γ levels were determined by immunoassay in 42 patients with jSLE. In addition, 19 juvenile systemic sclerosis (jSSc) and 25 age-matched healthy children were selected as patient control and healthy control, respectively.

Results

Serum NF-κB levels in patients with jSLE demonstrated a positive trend towards elevation compared with the controls with no significant difference (p=0.030). In addition, serum NF-κB levels in patients with jSSc were significantly higher than that of the healthy controls (p=0.005). Serum PPAR-γ levels were tend to be lower in both patients with jSLE and jSSc compared with the controls, with no significant difference. Specifically, NF-κB levels were significantly higher in patients with jSLE with cumulative damage (PedSDI≥1) compared with those without, at p=0.044. Logistic regression showed that PPAR-γ levels lower than 2.42 ng/mL were associated with the development of jSLE (OR 7.59) and lower than 2.16 ng/mL for jSSc (OR 10.90). The combined high levels of NF-κB with low PPAR-γ increased the risk of developing jSSc by 21.33-fold.

Conclusions

The observed trend of elevated NF-κB levels and decreased PPAR-γ levels in our study suggests their potential as biomarkers associated with increased proinflammatory signalling in jSLE and jSSc. However, our findings must be regarded as hypothesis-generating and confirmed in larger datasets. Moreover, their roles in monitoring the course of a disease and guiding therapeutic strategies in juvenile systemic autoimmune diseases need to be clearly investigated. Further extension of these findings may lead to better management and improvement in the outcomes of such patients.

WHAT IS ALREADY KNOWN ON THIS TOPIC

• The pathogenesis of juvenile SLE (jSLE), which is rarer than SLE, is multifactorial and poorly understood.

• This leads to an urgent need to identify non-invasive biochemical markers for early diagnosis and prediction of disease activity and response to treatment.

WHAT THIS STUDY ADDS

• Our analysis revealed a trend towards elevated nuclear factor-kappa B (NF-κB) levels in patients with jSLE and jSSc compared with healthy controls, while serum peroxisome proliferator-activated receptor-gamma (PPAR-γ) levels showed a downward trend.

• We identified that NF-κB and PPAR-γ levels did not exhibit disease-specific differences between the SLE and jSSc groups.

• A negative correlation was observed between serum NF-κB levels and serum PPAR-γ levels in patients with jSSc but not in patients with jSLE or healthy controls.

HOW THIS STUDY MIGHT AFFECT RESEARCH, PRACTICE OR POLICY

• A greater understanding of the potential of NF-κB and PPAR-γ as biomarkers in jSLE and jSSc could inform future research efforts, aiding in disease monitoring and potentially guiding the development of targeted therapies.

Introduction

Juvenile SLE (jSLE) is a rare autoimmune disease that affects children and is associated with a life-threating multiple organ inflammation.^{1 2} The hallmarks of jSLE is the production of autoantibodies directed against cellular and nuclear components involved in pathogenic immune complex (IC) formation and loss of B cell tolerance. Deposition of ICs and inflammation within the affected tissue caused multiorgan damage.^{3 4} Although the morbidity and mortality rate in patients has decreased with the developments in recent years, there is still a pressing need to identify non-invasive biochemical markers that will enable early detection of jSLE, as well as predict disease activity and response to treatment.

Nuclear factor-kappa B (NF-κB) and peroxisome proliferator-activated receptor-gamma (PPAR-γ) are two important transcription factors involved in regulating various cellular processes, including immune responses and inflammation.⁵ NF-κB is generally considered a proinflammatory transcription factor, as its activation leads to the expression of proinflammatory genes and the production of inflammatory cytokines.^{5 6} On the other hand, PPAR-γ is considered an anti-inflammatory transcription factor by inhibiting the expression of proinflammatory mediators and enhancing anti-inflammatory-related gene expression.⁷ There is a complex crosstalk between NF-κB and PPAR-γ. The activation of PPAR-γ can inhibit NF-κB’s proinflammatory actions, and conversely, NF-κB activation can also interfere with PPAR-γ’s anti-inflammatory effects.⁸ Often in patients with SLE, aberrant NF-κB activity contributes to the overproduction of proinflammatory cytokines, such as tumour necrosis factor-alpha (TNF-α) and interleukin-6 (IL-6).⁹ The sustained inflammatory response is a hallmark of SLE and contributes to tissue damage.¹⁰ PPAR-γ agonists are thought to have anti-inflammatory and immunomodulatory effects and have a role in SLE.11,13 The rarity of jSLE relative to SLE and the complexity of the disease have contributed to the limited number of specific studies focusing exclusively on NF-κB and PPAR-γ in patients with jSLE.

The present study investigated NF-κB and PPAR-γ as potential biomarkers for jSLE and jSSc. In recent years, an important relationship between increased NF-κB activation and decreased PPAR-γ expression has been identified in the pathogenesis of systemic sclerosis (SSc). This imbalance has been associated with an increase in proinflammatory cytokines (IL-6, TNF-α and CCL5) and has been suggested to contribute to the specific phenotype of the disease.¹⁴ In this context, when investigating these biomarkers in SLE, comparison with patients with SSc was considered a suitable approach to better understand whether these molecules are disease specific. Furthermore, we sought to investigate the correlation between these factors and disease activity, organ involvement, acute phase markers and autoantibodies in jSLE.

The potential of these biomarkers for the assessment of disease activity was also examined. Thus, we anticipate that it will shed light on future studies by providing preliminary data on whether it can be a focus for future diagnostic and therapeutic strategies.

Materials and methods

Study groups

The study population included 42 patients with jSLE, 19 age-matched patients with juvenile systemic sclerosis (jSSc) as a disease control group and 25 age-matched healthy controls. Prior to the study, written consent was acquired from all participants. All patients with jSLE and jSSc fulfilled the American College of Rheumatology (ACR) revised criteria for the classification of SLE¹⁵ and Pediatric Rheumatology European Society (PReS)/American College of Rheumatology (ACR)/European Alliance of Associations for Rheumatology (EULAR) provisional classification criteria for jSSc,¹⁶ respectively. Patients with jSSc were selected as disease controls because both jSSc and jSLE have a similar pathogenesis based on systemic involvement such as connective tissue disorders. This choice is further supported by evidence in the literature indicating the involvement of the NF-κB/PPAR-γ pathway in jSSc and its role in systemic inflammation and fibrosis, which are hallmarks of the disease.^{14 17 18} Therefore, including patients with jSSc as disease controls allowed us to explore the potential of these transcription factors as biomarkers and to distinguish their disease-specific roles. Individuals in the healthy control group were selected from individuals who did not have any rheumatological, infectious, immunological, haematological, cardiovascular, renal, neuropsychiatric and endocrine diseases and had similar demographic characteristics to the patient groups.

The SLE Disease Activity Index 2000 (SLEDAI-2K)¹⁹ and Juvenile Systemic Sclerosis Severity Score (J4S)²⁰ were used to assess disease activity for jSLE and jSSc, respectively. Additionally, the cumulative damage of patients with SLE was calculated according to the paediatric version of the Systemic Lupus International Collaborating Clinics/American College of Rheumatology Damage Index (PedSDI).²¹ Active organ involvement, current clinical findings and medication use were documented based on clinical assessment at the time of sample collection.

Sample collection and analytical assays

Peripheral blood samples were obtained from all participants after a 12-hour fast and were centrifuged for 15 min at 3000 RCF at 4°C. Routine clinical chemistry analyses were then performed immediately. For the analysis of NF-κB and PPAR-γ, serum samples were frozen and stored at –80°C for subsequent experiments.

Serum NF-κB/p65 subunit and serum PPAR-γ levels were measured with a sandwich ELISA kit following the instructions provided by the manufacturer (Human NF-κB-p65 kit, Cat. No. E-EL-H1388, and Human PPAR-γ kit, Cat. No. E-EL-H1361, respectively, Elabscience Biotechnology, USA). The sensitivity of these commercial assay kits was 0.10 ng/mL. The intra-assay and inter-assay coefficients of variation for NF-κB were <5.56% and <5.77%, respectively. Both the coefficients of intra-assay and inter-assay variations for PPAR-γ were less than 10.

Statistical analysis

All statistical analysis was performed with SPSS V.26.0 (IBM SPSS Statistics for Windows, V.20.0, IBM, Armonk, New York, USA). Categorical variables were presented as percentages or frequencies and were compared using Pearson’s χ² tests. Continuous variables were described as mean values along with their corresponding SD. The normal distribution of the variables was determined by histogram and/or by Kolmogrov-Smirnov/Shapiro-Wilk test. The statistical significance of the mean differences between the parameters of the three groups was determined by analysis of variance, followed by post hoc Bonferonni correction for multiple comparisions. The univariate associations were evaluated by Spearman’s rank correlation analysis. The study assessed the sensitivity and specificity of the measured variables as biomarkers through a receiver operating characteristic (ROC) curve analysis. Cut-off values were determined by using the Youden’s J Index. A priori power analysis indicated that a sample size of 25 individuals for the control group and 31 individuals for the SLE patient group was sufficient for PPAR-γ analysis. For NF-κB, a sample size of 41 patients with SLE was deemed necessary. In the post hoc power analysis of our study, the achieved statistical power was 84.5% with an alpha level of 0.05, indicating that our study was adequately powered to detect significant differences. As a result of the Bonferroni correction for multiple comparisons, a p value <0.017 was considered as statistically significant. For other analyses, a p value <0.05 was regarded as statistically significant.

Results

Analysis of demographics, clinical features and laboratory findings in a patient cohorts and controls

The demographic, clinical characteristics and laboratory evaluations of patients with jSLE and jSSc are summarised in tables1 2, correspondingly. A total of 42 patients with jSLE and age-matched 19 disease control patients with jSSc and 25 healthy controls were included in the study. The mean ages of patients with jSLE, patients with jSSc and controls were 15.90±2.80, 17.47±2.39 and 15.04±3.06 years, respectively. The female/male ratios for jSLE, jSSc and controls were 5:1 (35 female/7 male), 8.5:1 (17 female/2 male) and 1.5:1 (15 female/10 male), respectively.

Table 1. Demographic, clinical features and laboratory assessments of patients with jSLE (n=42) at study time.

Demographic characteristics
Mean age (mean±SD) (years)	15.90±2.80 years
Female/male ratio (n)	5/1 (35 female/7 male)
Mean disease duration (range) (years)	4.17±2.36 years (range 0–10.3 years)
Mean disease onset (range) (years)	11.73±2.74 years (range 6.0–17.5 years)
Clinical features
Major organ involvement
Mucocutaneous involvement	15 (35.7%)
Musculoskeletal involvement	2 (4.8 %)
Neuropsychiatric involvement	3 (7.1 %)
Haematological involvement	2 (4.8 %)
Pulmonary involvement	1 (2.4 %)
Renal involvement	5 (11.9 %)
Serositis	0 (0)
Raynaud’s syndrome	5 (11.9 %)
Disease activity and damage scores
Mean SLEDAI score (range)	5.64±5.41 (0–18)
Mean PedSDI score	0.36±0.49
Serological tests,n (%)
Antinuclear antibody positivity	41 (97.6)
Anti-ds DNA positivity	34 (81.0)
Anticardiolipin antibody (aCL) positivity	3 (7.1)
Laboratory assessments
Mean erythrocyte sedimentation rate (mm/hour)	24.07±19.12
Mean C-reactive protein level (mg/dL)	0.43±0.82
Mean procalcitonin level (ng/mL)	0.07±0.12
Mean Complement C3 level (g/L)	0.96±0.28
Mean Complement C4 level (g/L)	0.16±0.10

PedSDIpaediatric version of the Systemic Lupus International Collaborating Clinics/American College of Rheumatology Damage IndexSLEDAI, Systemic Lupus Erythematosus Disease Activity Index

Table 2. Demographic, clinical features and laboratory assessments of disease control patients with jSSc at study time (n=19).

Demographic characteristics
Mean age (mean±SD) (years)	17.47±2.39 years
Female/male ratio (%, n)	8.5/1 (17 female/2 male)
Mean disease duration (years) (range)	6.89±3.54 years (range 1.0–18.0 years)
Mean disease onset (years) (range)	10.42±3.31 years (range 2.0–17.0 years)
Clinical features, n (%)
Sclerosis and induration	19 (100%)
Sclerodactyly	19 (100%)
Raynaud’s syndrome	14 (73.7%)
Telangiectasias	13 (68.4%)
Pulmonary hypertension	3 (15.8%)
Arthalgia	14 (73.7%)
Arthritis	6 (31.6%)
Dysphagia	5 (26.3%)
Ocular involvement	9 (47.4%)
Disease activity
The mean J4S	6.62±4.84
Laboratory assessments
Anti Scl-70 (IU/mL)	56.56±92.12
Hgb (g/dL)	12.45±1.31

jSScjuvenile systemic sclerosis

The mean age of patients with jSLE at diagnosis was 11.73±2.74 years. The mean disease duration was 4.17±2.36 years. The mean age of patients with jSSc at diagnosis was 10.42±3.31 years. The mean disease duration was 6.89±3.54 years.

The distribution of organ involvements present at the time of sample collection in patients with SLE is summarised in table 1. During the study visit, mucocutaneous involvement was the most common manifestation, which was observed in 15 patients with jSLE (35.7%). Renal involvement and Raynaud’s syndrome were present in five patients (11.9 %). Neuropsychiatric involvement was observed in three patients (7.1%), musculoskeletal and haematological involvement were present in two patients (4.8%) (table 1). The mean SLEDAI-2K score was 5.64±5.41, ranging from 0 to 18 at the time of the study (table 1). In total, 19 individuals had organ involvements shown in table 1, while 23 patients with jSLE had no organ involvement.

Sclerosis, induration and sclerodactyly were present in all patients with jSSc. Raynaud’s syndrome and telangiectasias were also common, observed in 14 (73.7 %) and 13 (68.4 %) patients, respectively. The mean J4S score was 6.62±4.84, ranging from 0 to 17 at the time of the study (table 2).

Analysis of serum NF-κB and PPAR-γ levels

Our findings indicate a trend towards elevated serum NF-κB (p65) levels in patients with jSLE compared with healthy controls (mean±SD=1.87±1.00 ng/mL vs 1.25±0.70 ng/mL, respectively, p=0.030); however, this difference does not meet the Bonferroni-adjusted significance threshold of p<0.017. Patients with jSSc had serum NF-κB levels of 2.17±0.98 ng/mL. The serum NF-κB levels in patients with jSSc were also significantly higher than those in healthy controls (p=0.005), indicating a statistically significant difference despite the correction. No significant difference was observed between jSLE and jSSc groups (figure 1).

Figure 1. Comparison of serum NF-κB levels between patients with jSLE and their disease controls (patients with jSSc) and healthy controls. a: vs healthy control, *p<0.017. jSLE, juvenile SLE; jSSc, juvenile systemic sclerosis; NF-κB, nuclear factor-kappa B.

Additionally, serum PPAR-γ levels in patients with jSLE showed a downward trend compared with the healthy controls (1.58±0.60 vs 2.03±0.89, p=0.028), with a similar trend observed in patients with jSSc (1.52±0.49 ng/mL, p=0.046) compared with the controls. However, these differences also do not reach the significant threshold adjusted by Bonferroni correction. The PPAR-γ levels in patients with jSLE and jSSc were found to be very similar and remained well above the threshold for statistical significance (figure 2).

Figure 2. Comparison of serum PPAR-γ levels between patients with jSLE and their disease controls (patients with jSSc) and healthy controls. jSLE, juvenile SLE; jSSc, juvenile systemic sclerosis; PPAR-γ, peroxisome proliferator-activated receptor-gamma.

In addition, NF-kB and PPAR-γ levels were analysed in subgroups formed according to the presence of organ involvement, disease activity and cumulative damage. As shown in table 3, neither NF-kB nor PPAR-γ levels were significantly different between patients with jSLE with and without organ involvement. Similarly, in the disease activity subgroups formed using the SLEDAI-2K score, no significant difference was found between patients with jSLE with no or mild activity and those with moderate or high concerning these parameters. However, NF-kB levels were higher in patients with jSLE with cumulative damage (group II, PedSDI score ≥1) compared with those without cumulative damage (table 3).

Table 3. Relationship between NF-kB and PPAR-γ expression levels and major organ involvement, disease activity and cumulative damage.

	NF-kB	P value	PPAR-γ	P value
Major organ involvement		0.247		0.416
No (n=23)	1.76±0.84		1.56±0.48
Yes (n=19)	1.99±1.19		1.60±0.73
Disease activity*		0.161		0.169
No or mild activity (n=28)	1.74±0.84		1.65±0.51
Moderate or high activity (n=14)	2.12±1.28		1.43±0.74
Cumulative damage†		0.044		0.295
Group I (n=27)	1.64±0.77		1.54±0.59
Group II (n=15)	2.28±1.26		1.65±0.63

^*Disease Aactivity subgroups were formed according to the SLEDAI-2K score. No activity: SLEDAI=0,; Mmild activity: SLEDAI 1–5,; Mmoderate activity: SLEDAI 6–10,; Hhigh activity: SLEDAI 11–19; SLEDAI-2K: The SLE Disease Activity Index 2000;.

^†The presence of cumulative damage was categorizedcategorised using the PedSDI (the pediatricpaediatric version of the Systemic Lupus International Collaborating Clinics/American College of Rheumatology Damage Index) score. Group 1: Patients with a PedSDI score of 0,. Group 2: Patients with a PedSDI score of 1 or higher.

SLEDAI-2KThe SLE Disease Activity Index 2000

In our study, no correlation was observed between NF-kB and PPAR-γ with acute phase markers (erythrocyte sedimentation rate (ESR), C-reactive protein (CRP) and procalcitonin (PCT)) or autoantibodies (ANA, anti-dsDNA and aCL) in patients with jSLE. In patients with jSLE, a weak negative correlation was observed between NF-κB and both age at disease onset and age at diagnosis. The correlation coefficients were −0.326 (p=0.018) and −0.286 (p=0.033), respectively. In patients with jSSc, a moderate negative correlation was observed between NF-κB and PPAR-γ (r=−0.49; p=0.032). No correlation was identified between these biomarkers and the other clinical parameters under study.

ROC and logistic regression analysis

To investigate the potential of serum NF-κB and PPAR-γ levels as biomarkers for jSLE, ROC analysis was performed, and AUC (area under the curve) was calculated. The analysis revealed that serum NF-κB levels could be used to distinguish both patients with jSLE (AUC=0.671, 52% sensitivity and 76% specificity with cut-off value=1.77 ng/mL; p=0.020) and patients with jSSc (AUC=0.787, 84% sensitivity and 68% specificity with cut-off value=1.52 ng/mL; p=0.001) from healthy controls (figure 3B,A, respectively). PPAR-γ demonstrated the ability to differentiate patients with jSLE from healthy individuals with 93% sensitivity and 40% specificity (AUC=0.655, p=0.035) for values below 2.42 ng/mL (figure 3B). Additionally, it showed the ability to differentiate patients with jSSc and healthy individuals with 95% sensitivity and 52% specificity (AUC=0.691, p=0.032) for values below 2.16 ng/mL (figure 3A).

Figure 3. Comparisons of the sensitivity and specificity of the diagnosis by serum NF-κB and PPAR-γ. The ROC curve of (A) jSSc vs healthy controls; (B) jSLE vs healthy controls; (C) jSLE vs jSSc. To display both decreasing and increasing parameters within the same figure, decreasing parameters have been transformed into negative values. NF-κB, nuclear factor-kappa B; jSLE, juvenile SLE; jSSc, juvenile systemic sclerosis; PPAR-γ, peroxisome proliferator-activated receptor-gamma; ROC, receiver operating characteristic.

Additionally, to emphasise the potential importance of these biomarkers, logistic regression analysis was performed. According to the cut-off values obtained from the ROC analysis, a PPAR-γ level below 2.42 ng/mL increased the risk of SLE by 7.59 times, while a level below 2.16 ng/mL increased the risk of jSSc by 10.90 times. Although there was no correlation between elevated NF-κB levels and the risk of SLE, an NF-κB level exceeding 1.52 ng/mL was found to increase the risk of jSSc by 6.75 times. In addition, when both NF-κB levels exceeded 1.52 ng/mL and PPAR-γ levels were below 2.16 ng/mL, the risk of jSSc increased by 21.33 times (table 4).

Table 4. Logistic regression analysis of NF-κB and PPAR-γ levels in jSLE and jSSc vs control.

Parameters	B	SE	Wald	df	P value	Exp(B)	95% CI for EXP(B)
							Lower	Upper
Logistic regression analysis data for jSLE vs control
NF-κB >1.77 ng/mL	1.084	0.599	3.272	1	0.070	2.956	0.913	9.564
PPAR-γ <2.42 ng/mL	2.027	0.743	7.451	1	0.006	7.594	1.771	32.562
Logistic regression analysis data for jSSc vs control
NF-κB >1.52 ng/mL	1.909	0.820	5.428	1	0.020	6.748	1.354	33.633
PPAR-γ <2.16 ng/mL	2.389	1.156	4.269	1	0.039	10.900	1.131	105.075
NF-κB >1.52 ng/mL and PPAR-γ <2.16 ng/mL	3.060	0.804	14.501	1	<0.001	21.333	4.416	103.066

jSLEjuvenile SLEjSScjuvenile systemic sclerosisNF-κBnuclear factor-kappa BPPAR-γperoxisome proliferator-activated receptor-gamma

Discussion

jSLE primarily arises from dysregulation of cellular and humoral immune responses, and its progression can result in multiorgan dysfunction or, in severe cases, mortality. Thus, unravelling the pathogenesis of SLE remains a central focus in the clinical prevention and management of the disease.⁹ Typically, jSLE constitutes approximately 15–20% of all diagnosed SLE cases, and it is often characterised by a more severe disease presentation compared with adult-onset cases.²² Our previous studies have shown that pentraxin-3 may be a biomarker associated with juvenile scleroderma and SSc.^{23 24} Although the functions of NF-κB and PPAR-γ have been generally understood, their exact roles as biomarkers in the context of jSLE are still poorly explained. In this study, we measured the levels of NF-κB and PPAR-γ in serum by determining their potential use as biomarkers. Serum biomarkers are less invasive and more standardised approach than tissue samples, which is particularly important for our young subject population. Minimising procedural discomfort and using smaller volumes of blood make this approach quite suitable for studying paediatric patients.

As it is known, the NF-κB signalling pathway has been noted to be triggered in SLE, and its overstimulation has been associated with the pathogenesis and progression of the disease.²⁵ RelA/p65, the most abundant subunit of active NF-κB, has been shown in several studies to be closely associated with NF-κB activation.^{26 27} Moreover, experimental research has demonstrated a correlation between p65 expression levels in tissues/cells and serum p65 levels, paralleling findings in PPAR-γ where similar relationships are observed.28,30 In the literature, various studies have been published similar to our findings. For instance, transgenic mice carrying a mutation in ABIN1 that disrupts the inhibition of NF-κB, spontaneously and progressively developed various SLE-associated phenotypes, such as splenomegaly and an increase in circulating ICs, as well as pathogenic ANA and anti-ds DNA autoantibodies.³¹ In 2023, Zhang et al showed that NF-κB increased in both protein and mRNA levels in the circulation of patients with SLE (especially in active state) compared with healthy controls. In the same study, researchers reported that IL-38 may play a role in the increase in NF-κB observed in patients with SLE through in vitro and in vivo analyses.²⁵ While our study found that NF-κB levels were not significantly correlated with disease activity, organ involvement, or ANA and anti-dsDNA levels in patients with jSLE, it is important to consider the limitations of our sample size. Although our power analysis indicated that our sample size was sufficient, with a post hoc power of 84.5% (alpha=0.05), the small sample size might still influence the detection and interpretation of subtle correlations. Future studies with larger populations would be beneficial to confirm and expand on these findings. Nevertheless, a significant correlation was observed between elevated NF-κB levels and the cumulative damage index in patients with jSLE. This suggests that elevated NF-κB levels may be indicative of cumulative damage and may potentially reflect the long-term impact of the disease. Further studies with larger sample sizes and improved statistical power are required to confirm these findings and evaluate the potential role of NF-κB in the clinical manifestations of jSLE. Experimental studies carried out in recent years provide data that molecules that inhibit the activation of the NF-κB pathway can be used as a therapeutic target. As it is known, NF-κB has two distinct pathways, one is canonical and the other one is non-canonical. In the literature, Yang et al³² found that agonists of the V-domain immunoglobulin suppressor of T cell activation (VISTA) attenuated lupus-like disease progression by inhibiting the activation of IFN-I and the non-canonical NF-κB pathway.

Our ROC analysis results showed that NF-κB could be a valuable diagnostic biomarker for patients with SLE. To provide further support for the significance of our findings, we conducted a logistic regression analysis in addition to a ROC curve analysis. Logistic regression analysis also showed that an NF-κB level exceeding 1.52 ng/mL was associated with a 6.75-fold increased risk for jSSc. It is noteworthy that the risk of jSSc increased by 21.33-fold when both NF-κB levels exceeded 1.52 ng/mL and PPAR-γ levels were below 2.16 ng/mL. While these results support the potential of NF-κB and PPAR-γ as biomarkers, further validation is required to ascertain their clinical utility in distinguishing between jSLE and jSSc.

It is essential to identify well the modulators of the NF-κB signalling pathway in jSLE for the identification and development of novel therapeutic target molecules. For this purpose, in recent years, it has been reported that PPAR-γ is one of the negative modulators of NF-κB.³³ However, to the best of our knowledge, the effect of PPAR-γ on the activation of NF-κB in patients with jSLE or SLE is not yet known. In our study, in patients with jSLE, the decrease in PPAR-γ levels was not found to be associated with the increase in NF-κB levels, whereas it was determined that they were negatively correlated in patients with jSSc. This might be due to the limited sample size of patients. PPAR-γ is considered a transcription factor implicated in SLE; therefore, its downregulation could be linked to the activity of the disease. Based on our findings, one could consider PPAR-γ as a potential biomarker in SLE, supporting further studies to establish the clinical relevance of this pathway. Although our study did not reach the threshold for statistical significance, recent studies supporting the downward trend of PPAR-γ have reported that PPAR-γ expression is decreased, especially in patients with SLE with skin lesions, and increased PPAR-γ may play a protective role in SLE.⁴ Additionally, data have been presented indicating that the protective role of PPAR-γ in SLE pathogenesis may occur through the TLR2/Sirt1/PPAR-γ signalling pathway.⁴ Even though, unlike our study, one study reported that protein and mRNA levels of PPAR-γ increased, especially in patients with active SLE and this increase may be due to its feedback regulatory role.³⁴ In another experimental study, it was reported that with macrophage-specific deletion of PPAR-γ, antinuclear antibody production increased, which is one of the characteristic features of SLE, and glomerulonephritis was observed.³⁵ Even so, the use of PPAR-γ agonists has been reported to significantly suppress the development of SLE and SLE-related atherosclerosis in mouse studies.^{13 36} Similarly, in a small-scale randomised controlled study involving patients diagnosed with SLE, it was observed that CRP levels significantly reduced after 4 weeks of treatment with PPAR-γ agonist.³⁷

Our study suggests that NF-κB and PPAR-γ could be used as potential biomarkers for jSLE; however, these findings should be interpreted cautiously. Our findings must be regarded as hypothesis-generating and confirmed in larger datasets. Further elucidation of these factors may prove beneficial in monitoring the disease process and may contribute to the development of a future strategy with the objective of achieving more favourable outcomes in the management of patients with jSLE.

Data availability statement

All data relevant to the study are included in the article or uploaded as supplementary information.