Aging considerations of cytokines (IL-18, IL-36α, IL-37, and IL-38) associated with rheumatoid arthritis

ABSTRACT

The pro-inflammatory cytokines IL-18 and IL-36α and the anti-inflammatory cytokines IL-37 and IL-38 are proposed to have a pathophysiological role in rheumatoid arthritis (RA). However, this role has not been well investigated in relation to aging. In this study, serum IL-18, IL-36α, IL-37, and IL-38 levels were examined in 100 RA patients who were classified as young adults (<40 years) and elderly (>60 years). A similar classification was applied to 100 healthy controls. Serum IL-18, IL-36α, and IL-38 levels were significantly elevated in all RA patients compared to controls. When comparing young adult patients to elderly patients, only IL-18 showed significantly elevated levels in elderly patients. IL-18, IL-36α, and IL-38 showed excellent performance in differentiating between young adult patients and young adult controls, whereas in elderly patients, only IL-37 retained this performance. IL-18, IL-36α, and IL-38 were associated with an increased risk of RA in young adults, whereas in the elderly, only IL-37 was associated with an increased risk of RA. In conclusion, aging is an important factor proposed to influence the association pattern of IL-18, IL-36α, IL-37, and IL-38 in RA. These findings may reshape our understanding of the role of cytokines in the pathophysiology of RA.

KEYWORDS:

• Aging
• cytokines, pro-inflammatory
• anti-inflammatory
• ROC curve
• logistic regression

Introduction

The chronic inflammatory autoimmune disease rheumatoid arthritis (RA) is one of the most common rheumatic musculoskeletal disorders, with an estimated global prevalence and incidence of 224.3 and 12.2 cases per 100,000 population, respectively [1]. Progressive articular inflammation (chronic synovitis) and subsequent joint destruction are prominent features of RA. However the etiology and pathophysiology of the disease is complex and many factors are proposed to participate in mediating its pathogenesis, including genetic, epigenetic, hormonal, infectious, and environmental factors [2]. Although the mechanism by which these factors contribute to the initiation and progression of RA is not well resolved, there is compelling evidence that their interactions can cause immune dysregulation, which is proposed to be a key driver of disease risk [3]. Indeed, studies aimed at understanding the pathologic mechanisms in RA have shown that most cellular (neutrophils, monocytes/macrophages, T and B lymphocytes) and humoral (immunoglobulins, cytokines and chemokines) aspects of innate and adaptive immunity are dysregulated [4].

Increasing evidence suggests that stimulation of synovial cell proliferation occurs via a network of different cytokines that have also been indicated to be involved in the pathogenesis of cartilage and bone damage [5]. Among the cytokines proposed to play a role in the pathogenesis of RA are tumor necrosis factor (TNF)-α, granulocyte macrophage colony-stimulating factor (GM-CSF), interleukin (IL)-1, IL-2, IL-6, IL-7, IL-17, IL-18, IL-21, IL-23, and IL-33 [6]. Cytokines are grouped into families of functionally related members, one of which is the IL-1 cytokine family, which has attracted the attention of researchers to study the role of its cytokines in the pathogenesis of RA [7]. Interestingly, the IL-1 family includes both pro-inflammatory (IL-1α, IL-1β, IL-18, IL-33, and IL-36) and anti-inflammatory (IL-37 and IL-38) cytokines, and this reinforces its importance not only in RA, but also in most autoimmune diseases, which are pathologically characterized by an imbalance of both types of cytokines [8]. Our focus was on one classic member of this family, IL-18, and three others discovered in recent years, IL-36α, IL-37, and IL-38, which are proposed to play a key role in mediating the dysregulated inflammatory response in RA [9–12].

IL-18 is a cytokine that has pro-inflammatory functions and also exhibits immunomodulatory effects. In particular, it is involved in promoting the differentiation of natural killer (NK) cells and T lymphocytes, along with its effects on stimulating the production of other types of cytokines such as IFN-γ [13]. IL-36 is a cytokine that encompasses three agonist ligands, namely IL-36α, IL-36β and IL-36γ, and an antagonist ligand, IL-36Ra. IL-36 cytokines have rapidly emerged as IL-1 family cytokines involved in regulating the immune response in terms of promoting resolution of infection or inflammation [14]. The expression and action of IL-36 cytokines are well regulated and have a role in maintaining homeostatic conditions. Conversely, dysregulated expression or activation of IL-36 cytokines can promote a pathological inflammatory response [15]. IL-37 is credited with anti-inflammatory and immunosuppressive functions through which IL-37 can prevent excess inflammation and prevent inflammation-induced tissue damage [16]. Therefore, IL-37 is proposed to play a prominent functional role in balancing immune homeostasis [17]. IL-38 is the least investigated member of the IL-1 cytokine family and since its discovery in 2001, studies have linked this cytokine to the pathophysiology of various diseases, especially autoimmune and inflammatory diseases [18]. Anti-inflammatory functions of IL-38 have been described and available data suggest a regulatory role for IL-38 on the axis of T helper 17 cell. Additional evidence highlights the tolerogenic effects of IL-38 on T regulatory cells, dendritic cells, and macrophages [19]. Dysregulated production of the four cytokines (IL-18, IL-36α, IL-37, and IL-38) has been associated with the pathogenesis of RA and their utility as biomarkers and therapeutic targets in the disease has been suggested [6,7,9,12,20,21].

Cytokines have been shown to be influenced aging, which has a prominent feature of dysregulated production of inflammatory cytokines. Elevated concentrations of pro-inflammatory cytokines and other inflammatory markers have been identified in the circulation of older adults and are associated with an increased risk of developing age-related diseases. Recognized inflammatory mediators in this context include TNF-α, IL-1, IL-8, IFN-γ, and acute phase proteins [22]. In addition, cytokines that have a regulatory role in immune response and inflammation, anti-inflammatory cytokines such as transforming growth factor (TGF)-β and IL-10, have also been shown to be dysregulated during aging [23]. Regarding RA, the disease affects adults of all ages, but is most common among older people over 65 years of age [24]. Since cytokines exhibit dysregulated levels during aging, their serum profile in elderly RA patients may differ from that in young adult patients. In fact, studies have not covered this issue extensively, and thus it may be important to determine the effect of aging on cytokine levels in RA patients in order to achieve a better understanding of their role in the pathogenesis of the disease.

In this study, two pro-inflammatory cytokines (IL-18 and IL-36α) and two anti-inflammatory cytokines (IL-37 and IL-38) of the IL-1cytokine family were explored in the serum of two groups of RA patients; young adults under 40 years of age and elderly over 60 years of age. The aim was to understand the pattern of association of these cytokines with RA in young adult and elderly patients, as well as to evaluate their relationship with clinical indicators of the disease in both groups of patients. To the researchers’ knowledge, the pattern of association of current cytokines with RA during aging has not been well studied.

Materials and methods

Patients

From August 2022 to March 2023, a case-control study was conducted on 100 patients with RA (mean age ± standard deviation [SD]: 48 ± 18 years; 50 males, 50 females). Patients were classified into two age groups. The first group included 50 young adult RA patients under the age of 40 years (age range: 18–39 years). The second group included 50 elderly RA patients over 60 years of age (age range: 61–74 years). Patients were referred to rheumatology units in two hospitals in Baghdad (Baghdad and Kadhimiya teaching hospitals). Diagnosis of RA was according to the 2010 criteria of the American College of Rheumatology/European League Against Rheumatism (ACR/EULAR) collaborative initiative [25]. Patients included were those who followed the diagnostic criteria and whose ages were consistent with the age specifications in the first and second groups. Patients suffering from chronic diseases such as diabetes and cancer, as well as pregnant women, were excluded. The disease activity score 28 (DAS28) based on erythrocyte sedimentation rate (ESR) was used to evaluate RA activity. Accordingly, RA patients were classified into four groups in terms of disease activity: remission (score: < 2.6), low activity (score: ≥ 2.6 - < 3.1), moderate activity (score: > 3.1 - < 5.1), and high activity (score: ≥ 5.1) [26]. Data on age, gender, disease duration, medication, ESR, anti-cyclic citrullinated peptide (ACCP), rheumatoid factor (RF) were collected from the hospital patient registry. According to disease duration, patients were classified into three groups: 1–5, 6–10, and >10 years. For medication, all patients were on treatment but under four protocols: anti-TNF, methotrexate (MTX), anti-TNF+MTX, or anti-TNF+corticosteroids (CORT).

Two groups of healthy individuals were included as controls (HC groups) matched to patients for age and gender (mean age ± SD: 45 ± 18 years; 50% males, 50% females). The first group included 50 young adults under the age of 40 years (age range: 20–38 years). They were blood donors and their serum profile for the panel of antibodies at the National blood Transfusion Centre (Baghdad) was normal. The second group included 50 clinically disease-free elderly people over 60 years of age (age range: 61–73 years). They were health service workers and university employees and did not suffer from infectious or chronic disease during the past 12 months.

Approval was obtained from the institutional ethics committee to conduct the study (Ethics Committee, Mustansiriyah University; Reference number: BCSMU/0822/00030Z dated 20 August 2022), which relied on the approval of the Training and Human Development Center of the Baghdad Medical Educational City (Ministry of Health) on 20 June 2022 (Reference number: 24918).

Cytokine immunoassays

Serum concentrations of IL-18, IL-36α, IL-37, and IL-38 were measured using enzyme-linked immunosorbent assay (ELISA) kits provided by Cloud-Clone Corporation, USA (Catalogue Number: SEA064Hu, SEE843Hu, SEE842Hu, and SEQ458Hu, respectively). These kits were based on the principles of sandwich ELISA. The detection range for the kits was 15.6–1000 pg/mL (IL-18), 15.6–1000 pg/mL (IL-36α), 7.8–500 pg/mL (IL-37), and 7.8–500 pg/mL (IL-38). Procedures recommended by the manufacturer were followed.

Statistical analysis

Categorical variables were given as number and percentage and Pearson chi-square test was used to assess significant differences. Continuous variables were given as mean ± SD or median and interquartile range (IQR: 25%–75%) based on normal distribution tests (Shapiro-Wilk and Kolmogorov-Smirnov tests). Where applicable, significance was assessed using the one-way analysis of variance (ANOVA) test, Mann-Whitney U test (comparing two groups), or the Kruskal-Wallis test (comparing more than two groups). Receiver operating characteristic (ROC) curve analysis (patients versus HC) was performed to estimate the area under the curve (AUC) and 95% confidence interval (CI) associated with each cytokine in each group examined. Likewise, multinomial logistic regression analysis was used to calculate odds ratio 95% CI for each cytokine. The pairwise correlation coefficient (r) between cytokines was determined with Spearman’s rank-order correlation analysis. A probability level (p) < 0.05 was considered statistically significant. IBM SPSS Statistics 25.0 (Armonk, NY: IBM Corp.) was used to carry out statistical analysis. G*power software version 9.3.1.7 was used to estimate power of sample size [27].

Results

Sample size power

Sample size power was estimated using the software G*power, which was given the following inputs: two-tailed α error p = 0.05, effect size d = 0.5, RA sample size = 100, and HC sample size = 100. The estimated power (1-β error p) was 94%, which is higher than the accepted power of 80%, and this may support the statistical power of the sample size of the current patients and HC.

Baseline characteristics

Duration of RA was significantly higher in elderly patients compared to young adult patients (p = 2.9 × 10⁻⁸), as were ESR (p = 7.7 × 10⁻²⁴) and DA28 (p = 0.001), while ACCP and RF antibody titers showed no significant differences (p = 0.073 and 0.104, respectively). The frequency of medication groups also showed no significant differences between young adult and elderly RA patients (p = 0.274) (Table 1).

Table 1. Baseline characteristics of patients with rheumatoid arthritis.

Characteristic; mean ± SD or n (%)	Young adults; n = 50	Elderly; n = 50	p-value
Disease duration; year	5 ± 3	12 ± 7	2.9 × 10^−8
1–5	28 (56.0)	9 (18.0)	5.8 × 10^−6
6–10	17 (34.0)	15 (30.0)
>10	5 (10.0)	26 (52.0)
ESR; mm/h	35 ± 18	58 ± 30	7.7 × 10^−24
ACCP; U/L	64 ± 27	74 ± 28	0.073
RF; IU/L	39 ± 18	44 ± 16	0.104
DAS28	4 ± 1	5 ± 1	0.001
Remission	6 (12.0)	2 (4.0)	0.005
Low activity	8 (16.0)	0 (0.0)
Moderate activity	23 (46.0)	25 (50.0)
High activity	13 (26.0)	23 (46.0)
Medication
Anti-TNF	18 (36.0)	15 (30.0)	0.274
MTX	20 (40.0)	14 (28.0)
Anti-TNF+MTX	7 (14.0)	11 (22.0)
Anti-TNF+CORT	5 (10.0)	10 (20.0)

SD: Standard deviation; DAS28: Erythrocyte sedimentation rate; Disease activity score 28; ESR: ACCP: Anti-cyclic citrullinated peptide; RF: Rheumatoid factor; TNF: Tumor necrosis factor; MTX: Methotrexate; CORT: Corticosteroids; p: Probability (significant p-value is indicated by bold). Pearson chi-square test (categorical variable) or one-way analysis of variance test (continuous variable) was used to assess significant differences.

Serum cytokine levels

Serum cytokine levels were initially compared between RA patients and HC regardless of age group. Results revealed that IL-18, IL-36α, and IL-38 levels were significantly higher in patients than in HC (p = 1.1 × 10⁻⁵, 1.0 × 10⁻⁶, and 8.0 × 10⁻⁸, respectively). For IL-37 levels, there was no significant difference between patients and HC (p = 0.374). Next, cytokine levels were compared between young adult patients and elderly patients. This time, IL-18 levels were significantly elevated in elderly patients compared to young adult patients (p = 0.007). The levels of remaining cytokines (IL-36α, IL-37, and IL-38) showed no significant differences between the two groups of RA patients (p = 0.759, 0.363, and 0.612, respectively) (Table 2).

Table 2. Serum cytokine levels in patients with rheumatoid arthritis and healthy control subjects, as well as in patients classified as young adults and elderly.

Cytokine; median (IQR)	All; n = 200		p-value	RA; n = 100		p-value
	RA; n = 100	HC; n = 100		Young adults; n = 50	Elderly; n = 50
IL-18; pg/mL	89 (78–172)	80 (64–112)	1.1 × 10^−5	82 (76–159)	102 (81–178)	0.007
IL-36α; pg/mL	92 (72–99)	72 (63–84)	1.0 × 10^−6	92 (64–99)	92 (76–101)	0.759
IL-37; pg/mL	106 (86–132)	111 (72–127)	0.374	110 (91–133)	104 (80–129)	0.363
IL-38; pg/mL	89 (80–95)	76 (62–88)	8.0 × 10^−8	104 (80–129)	87 (79–95)	0.612

IQR: Interquartile range; RA: Rheumatoid arthritis; HC: Healthy controls; IL: Interleukin; p: Probability (significant p-value is indicated by bold). Mann-Whitney U test was used to assess significant differences.

Cytokines levels and RA characteristics

Serum levels of IL-18, IL-36α, IL-37, and IL-38 were stratified by gender, disease duration (1–5, 6–10, and >10 years), DAS28 (remission, low activity, moderate activity, and high activity) and medication (anti-TNF, MTX, anti-TNF+MTX, and anti-TNF+CORT) in young adult and elderly RA patients. Both groups of patients shared significantly elevated levels of IL-36α (p = 2.4 × 10⁻⁹ and 3.0 × 10⁻⁸, respectively) and IL-38 (p = 3.7 × 10⁻⁷ and 0.004, respectively) in male patients compared to female patients. Conversely, IL-37 levels were significantly elevated in females compared to males in both young adult patients and elderly patients (p = 2.3 × 10⁻⁴ and 0.003, respectively). Elderly RA patients showed two additional significant differences. First, IL-37 levels were significantly different between disease duration groups, and patients with disease duration >10 years showed the highest levels compared with patients with shorter disease durations (p = 0.003). Second, IL-38 levels were significantly different between medication groups, and patients treated with MTX showed the highest levels compared to patients treated with other protocols (p = 0.004) (Tables 3 and 4).

Table 3. Serum cytokine levels in young adult rheumatoid arthritis patients classified according to some of their characteristics.

Characteristic	Group	Median (IQR); pg/mL
		IL-18	IL-36α	IL-37	IL-38
Gender	Male; n = 25	92 (74–171)	99 (97–119)	95 (78–104)	94 (90–103)
	Female; n = 25	80 (77–96)	64 (56–82)	133 (124–139)	86 (78–90)
	p-value	0.455	2.4 × 10^−9	3.7 × 10^−7	2.3 × 10^−4
Disease duration; year	1–5; n = 28	79 (73–146)	91 (75–98)	107 (87–134)	91 (85–96)
	6–10; n = 17	82 (78–122)	83 (60–98)	113 (101–128)	86 (72–92)
	>10; n = 5	96 (82–172)	105 (97–110)	100 (71–118)	92 (87–95)
	p-value	0.397	0.391	0.429	0.172
DAS28	Remission; n = 6	90 (70–176)	86 (61–98)	107 (95–133)	92 (86–99)
	Low activity; n = 8	76 (73–151)	87 (60–101)	115 (73–132)	91 (87–95)
	Moderate activity; n = 23	80 (76–92)	94 (64–110)	112 (86–133)	89 (72–95)
	High activity; n = 13	96 (88–171)	92 (79–99)	102 (98–133)	87 (79–95)
	p-value	0.129	0.869	0.990	0.684
Medication	Anti-TNF; n = 18	79 (74–88)	91 (60–119)	118 (102–133)	87 (82–99)
	MTX; n = 20	130 (77–173)	97 (84–106)	99 (77–127)	93 (85–97)
	Anti-TNF+MTX; n = 7	100 (78–159)	70 (59–84)	133 (95–139)	86 (78–90)
	Anti-TNF+CORT; n = 5	82 (78–122)	64 (53–93)	112 (105–135)	89 (89–91
	p-value	0.274	0.038	0.171	0.304

IQR: Interquartile range; IL: Interleukin; DAS28: Disease activity score 28; TNF: Tumor necrosis factor; MTX: Methotrexate; CORT: Corticosteroids; p: Probability (significant p-value is indicated in bold). Significance was determined using the Mann-Whitney U test (comparing two groups), or the Kruskal-Wallis test (comparing more than two groups).

Table 4. Serum cytokine levels in elderly rheumatoid arthritis patients classified according to some of their characteristics.

Characteristic	Group	Median (IQR); pg/mL
		IL-18	IL-36α	IL-37	IL-38
Gender	Male; n = 25	155 (82–179)	101 (94–118)	98 (78–109)	93 (84–110)
	Female; n = 25	88 (81–178)	76 (65–82)	129 (98–136)	84 (78–88)
	p-value	0.264	3.0 × 10^−8	0.003	0.004
Disease duration; year	1–5; n = 9	94 (82–137)	94 (81–107)	78 (73–89)	94 (85–111)
	6–10; n = 15	178 (80–188)	94 (72–101)	98 (86–119)	88 (82–95)
	>10; n = 26	94 (81–172)	87 (73–98)	115 (103–133)	87 (78–92)
	p-value	0.120	0.662	0.003	0.217
DAS28	Remission; n = 2	108 (79–137)	109 (94–124)	91 (78–103)	76 (73–78)
	Low activity; n = 0	-	-	-	-
	Moderate activity; n = 25	103 (81–179)	79 (67–97)	103 (86–130)	87 (83–94)
	High activity; n = 23	99 (80–175)	94 (82–108)	108 (79–128)	89 (79–100)
	p-value	0.625	0.088	0.659	0.185
Medication	Anti-TNF; n = 15	94 (80–175)	78 (67–107)	98 (86–130)	87 (84–91)
	MTX; n = 14	146 (82–185)	95 (93–97)	105 (78–111)	98 (92–111)
	Anti-TNF+MTX; n = 11	82 (79–172)	98 (82–124)	103 (86–114)	77 (68–87)
	Anti-TNF+CORT; n = 10	139 (83–180)	78 (67–94)	133 (77–137)	85 (82–92)
	p-value	0.323	0.082	0.404	0.004

IQR: Interquartile range; IL: Interleukin; DAS28: Disease activity score 28; TNF: Tumor necrosis factor; MTX: Methotrexate; CORT: Corticosteroids; p: Probability (significant p-value is indicated in bold). Significance was determined using the Mann-Whitney U test (comparing two groups), or the Kruskal-Wallis test (comparing more than two groups).

ROC curve analysis

ROC curve analysis demonstrated that IL-18, IL-36α, and IL-38 were excellent in distinguishing between young adult RA patients and young adult HC (AUC = 0.903, 0.808, and 0.911; p = 3.6 × 10⁻¹², 1.1 × 10⁻⁷, and 1.3 × 10⁻¹², respectively), while IL-37 was less significant in distinguishing between young adult patients and young adult HC (AUC = 0.622; p = 0.035). In elderly RA patients, only IL-37 retained acceptable discrimination ability from elderly HC (AUC = 0.720; p = 1.5 × 10⁻⁴), whereas the discrimination performance of IL-18, IL-36α and IL-38 between elderly patients and elderly HC (AUC = 0.51, 0.548, and 0.511; p = 0.983, 0.408, and 0.850, respectively) was poor (Table 5).

Table 5. Reciever operating characteristic curve analysis of cytokines in young adult and elederly rheumatoid arthritis patients versus their corresponding control groups.

Cytokine	Young adult RA patients			Elderly RA patients
	AUC	95% CI	p-value	AUC	95% CI	p-value
IL-18	0.903	0.847–0.960	3.6 × 10^−12	0.501	0.372–0.631	0.983
IL-36α	0.808	0.713–0.902	1.1 × 10^−7	0.548	0.430–0.666	0.408
IL-37	0.622	0.503–0.742	0.035	0.720	0.615–0.824	1.5 × 10^−4
IL-38	0.911	0.856–0.967	1.3 × 10^−12	0.511	0.397–0.625	0.850

RA: Rheumatoid arthritis; AUC: Area under the curve; CI: 95% confidence interval; IL: Interleukin; p: Probability (significant p-value is indicated in bold).

Logistic regression analysis

Multinomial logistic regression analysis adjusted for gender was performed using IL-18, IL-36α, IL-37, and IL-38 as independent predictors of RA in young adult patients versus young adult HC and elderly patients versus elderly HC. The analysis revealed that elevated levels of IL-18, IL-36α, and IL-38 were associated with an increased risk of RA among young adults (OR = 1.24, 1.08, and 1.26; p = 0.008, 0.017, and 0.002, respectively), whereas in elderly people, only IL-37 was associated with an increased risk of developing the disease (OR = 1.03; p = 1.8 × 10⁻⁴) (Table 6).

Table 6. Gender-adjusted multinomial regression analysis of cytokines in young adult and elederly rheumatoid arthritis patients versus their corresponding control groups.

Cytokine	Young adult RA patients			Elderly RA patients
	OR	95% CI	p-value	OR	95% CI	p-value
IL-18	1.24	1.06–1.44	0.008	1.01	1.00–1.02	0.224
IL-36α	1.08	1.01–1.15	0.017	1.02	0.99–1.05	0.205
IL-37	0.92	0.83–1.01	0.086	1.03	1.01–1.05	1.8 × 10^−4
IL-38	1.26	1.09–1.46	0.002	1.01	0.97–1.05	0.589

RA: Rheumatoid arthritis; OR: Odds ratio; CI: 95% confidence interval; IL: Interleukin; p: Probability (significant p-value is indicated in bold).

Correlation analysis

Spearman’s rank-order correlation analysis between IL-1 family cytokines (IL-18, IL-36α, IL-37, and IL-38) and clinical indicators of RA (ESR, DAS28, ACCP, and RF) was performed independently in young adult and elderly RA patient groups. The correlation pattern was different in each group of patients. In young adult patients, IL-36α negatively correlated with IL-37 (r = −0.578; p = 1.1 × 10⁻⁵) and positively correlated with IL-38 (r = 0.602; p = 4.0 × 10⁻⁶) and RF (r = 0.290; p = 0.041). In addition, IL-37 negatively correlated with IL-38 (r = −0.519; p = 1.1 × 10⁻⁴). In elderly patients, IL-18 negatively correlated with ESR (r = −0.310; p = 0.029), IL-36α positively correlated with IL-38 (r = 0.410; p = 0.003), and ESR positively correlated with DAS28 (r = 0.347; p = 0.014) (Table 7).

Table 7. Correlation coefficient of the analysis between cytokines and clinical indicators of rheumatoid arthritis among young adult and elderly patients.

Variable	Analysis	IL-18	IL-36α	IL-37	IL-38	ESR	DAS28	ACCP	RF
Young adult rheumatoid arthritis patients; n = 50
IL-18	r	1.000
	p-value	1.000
IL-36α	r	−0.012	1.000
	p-value	0.933	1.000
IL-37	r	−0.099	−0.578	1.000
	p-value	0.492	1.1 × 10^−5	1.000
IL-38	r	0.130	0.602	−0.519	1.000
	p-value	0.368	4.0 × 10^−6	1.1 × 10^−4	1.000
ESR	r	−0.156	−0.198	0.183	−0.062	1.000
	p-value	0.280	0.169	0.203	0.671	1.000
DAS28	r	0.244	0.162	−0.007	0.018	0.119	1.000
	p-value	0.087	0.260	0.962	0.904	0.410	1.000
ACCP	r	−0.061	−0.022	0.092	−0.063	−0.145	0.268	1.000
	p-value	0.675	0.877	0.527	0.666	0.315	0.059	1.000
RF	r	0.213	0.290	−0.097	0.254	−0.186	0.109	−0.069	1.000
	p-value	0.138	0.041	0.503	0.075	0.196	0.451	0.635	1.000
Elderly rheumatoid arthritis patients; n = 50
IL-18	r	1.000
	p-value	1.000
IL-36α	r	0.185	1.000
	p-value	0.198	1.000
IL-37	r	0.066	−0.265	1.000
	p-value	0.648	0.063	1.000
IL-38	r	0.084	0.410	−0.159	1.000
	p-value	0.563	0.003	0.270	1.000
ESR	r	−0.310	−0.127	0.038	0.025	1.000
	p-value	0.029	0.380	0.793	0.863	1.000
DAS28	r	−0.056	0.161	0.098	0.137	0.347	1.000
	p-value	0.701	0.264	0.500	0.344	0.014	1.000
ACCP	r	−0.119	0.086	0.164	−0.243	0.101	0.146	1.000
	p-value	0.411	0.552	0.254	0.090	0.483	0.313	1.000
RF	r	0.049	−0.266	0.163	−0.096	0.116	0.114	0.196	1.000
	p-value	0.737	0.062	0.258	0.509	0.422	0.429	0.172	1.000

IL: Interleukin; ESR: Erythrocyte sedimentation rate; DAS28: Disease activity score 28; ACCP: Anti-cyclic citrullinated peptide; RF: Rheumatoid factor; r: Correlation coefficient; p: Probability (significant correlation is indicated in bold).

Discussion

In the current study, serum levels of two pro-inflammatory cytokines (IL-18 and IL-36α) and two anti-inflammatory cytokines (IL-37 and IL-38) were examined in RA patients younger than 40 years (young adult patients) and older than 60 years (elderly patients). IL-18, IL-36α, and IL-38 showed significantly elevated levels in RA patients compared with HC regardless of age, while IL-37 levels were decreased in patients compared with HC but the difference was not significant. However, the levels of these cytokines showed no significant differences between young adult patients and elderly patients, except for IL-18 levels, which were significantly elevated in elderly patients compared to young adult patients. ROC curve analysis demonstrated that the discriminatory performance of IL-18, IL-36α, and IL-38 was different in each group of patients. IL-18, IL-36α, and IL-38 showed excellent differential performance between young adult patients and young adult HC (AUC >0.8). In the elderly, only IL-37 showed acceptable discrimination performance (AUC = 0.72) between elderly patients and elderly HC. The association of cytokines with RA risk was also affected by aging. IL-18, IL-36α, and IL-38 levels were associated with an increased risk of developing RA in young adults (1.24-fold, 1.08-fold, and 1.26-fold, respectively), while IL-37 showed no association with RA risk in young adults. In contrast, IL-37 was associated with an increased risk of RA in the elderly, whereas IL-18, IL-36α, and IL-38 did not show this association. These results suggest that aging may be considered an important modulator of the serum profile of IL-18, IL-36α, IL-37, and IL-38 in RA patients. Therefore, aging may have a role in altering the differential performance of these cytokines as biomarkers associated with RA risk.

The aging process is a complex biological event accompanied by changes characterized by a gradual decline in physiological capabilities and a general deterioration in body functions. Consequently, the human body’s ability to adapt to these changes and maintain biological homeostasis is diminished [28]. One of the biological systems affected by the aging process is the immune system, and studies have indicated that most of its humoral and cellular components deteriorate functionally as a result of aging [29,30]. In this context two terms have been coined, immunosenescence and inflammaging. The first describes age-associated alternations in the immune system that may lead to functional dysregulation of the innate and acquired immunity in older people. The second refers to a chronic inflammatory state of a low grade that parallels aging in the absence of overt disease [31]. Both immunosenescence and inflammaging are proposed to be the root cause of most aging-related diseases, such as chronic inflammatory diseases and autoimmune disorders, as well as cancer [32]. Evidence suggests that a major contributor to immunosenescence and inflammaging is dysfunction of cells involved in innate and acquired immune responses, including neutrophils, monocytes, macrophages, NK cells, T cells and B cells. During aging, these cells synthesize cytokines with an irregular balance between those with pro-inflammatory and anti-inflammatory functions [33].

Due to these aging-associated alternations in immune function, it might be expected that serum levels of IL-18, IL-36α, IL-37, and IL-38 would undergo changes in elderly RA patients. Up-regulated serum/plasma concentrations of IL-18 have been associated with the pathogenesis of several aging-related disorders and inflammatory conditions such as atherosclerosis, myocardial infarction, RA, obesity, insulin resistance, and hypertension, as well as neurodegeneration [34]. It has also been observed that aging is characterized by higher plasma IL-18 concentrations that closely parallel poor physical performance in older adults [23]. Regarding IL-36, although it has not been well studied, recent experimental evidence has demonstrated the presence of IL-36α, IL-36β, and IL-36 R (IL-36 receptor) in the mouse ischemia/reperfusion-injured heart. In addition, IL-36α and IL-36β expression increased with aging [35]. For IL-37, a recent study reported that plasma IL-37 concentrations were lower in elderly people (free of overt clinical disease) than in healthy young adults [36]. The relationship between IL-38 and aging is not well understood, but it has been suggested that this cytokine may exhibit dysregulated levels in older adults [37]. Furthermore, dysregulated production of IL-38 has been linked to the pathogenesis of age-related illnesses such as RA, type 2 diabetes mellitus, myocardial infarction and degenerative musculoskeletal diseases [38].

Analysis of IL-18, IL-36α, IL-37, and IL-38 in relation to gender, disease duration, DAS28, and type of treatment revealed that male RA patients, young adults and elderly, shared significantly elevated levels of IL-36α and IL-38 compared to females. Conversely, IL-37 levels were significantly increased in females compared to males in both groups of patients. In elderly patients, two additional significant differences were found. First, IL-36α showed elevated levels that paralleled increased disease duration. Second, elevated levels of IL-38 were more evident in patients receiving MTX. Although there is no direct evidence to support these findings, it has recently been demonstrated that levels of the proatherogenic cytokines IL-1β, IL-6, and TNF-α were significantly higher in healthy males compared to females [39]. Moreover, it has also been shown that the production of monocyte-derived cytokines (IL-1β, IL-1RA, IL-6, IL-10, IL-12, and TNF-α) in response to lipopolysaccharide was stronger in healthy males than in females. Interestingly, these sex differences in cytokine responses were independent of age [40]. Sex hormones, such as testosterone, estrogen, and progesterone, may be responsible for sex differences in cytokine production as these hormones have been shown to significantly influence both types of immunity, innate and adaptive [41].

Correlation analysis indicated that the relationship between cytokines and clinical indicators of RA could be influenced by aging. For example, IL-37 showed a strong negative correlation with IL-36α and IL-38 in young adult patients, whereas these correlations were not found in elderly patients. Meanwhile, both groups of patients shared a strong positive correlation between IL-36α and IL-38. As discussed earlier, the aging process is associated with dysregulated production and balance of pro-inflammatory and anti-inflammatory cytokines, and thus the functional relationship between these cytokines may be affected [33].

Although the sample size of the current study had acceptable sample size power, the number of young adult and elderly patients included, as well as controls was relatively small and this may represent an important limitation of the study. Similar limitations apply to the number of males and females in each group studied.

Conclusions

Aging is an important factor proposed to influence the association pattern of IL-18, IL-36α, IL-37, and IL-38 in RA. Moreover, the significance of these cytokines in differentiating between RA patients and HC was also affected by the aging process. These findings may reshape our understanding of the role of cytokines in the pathophysiology of RA and may pave the way for further studies.

Acknowledgments

The authors express their gratitude to the medical staff at Baghdad and Kadhimiya Teaching Hospitals and the National Blood Transfusion Center (Baghdad) for their assistance. We extend our thanks to the people who volunteered to participate in this study.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Additional information

Funding

This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.