Exploration of cytokines relating to masked obesity in Japanese women

Abstract

Background

Masked obesity (MO) has been associated with chronic inflammation; however, its immunological profile has not been extensively studied.

Objective

This exploratory study aimed to investigate immunological profile in MO by examining serum cytokine levels and clinical characteristics in individuals with reduced muscle mass and increased fat.

Methods

Body composition was assessed in 432 female employees of The University of Osaka (aged 30–59 years) using InBody270 during health examinations. Participants with skeletal muscle mass index (SMI) ≥ 5.7 kg/m² and percent body fat (PBF) < 30% were classified as the healthy (H) group, while those with SMI < 5.7 kg/m² and PBF ≥ 30% were classified as the MO group. Serum cytokine concentrations were measured in randomly selected participants, 34 from the H group and 24 from the MO group.

Results

Compared to the H group, the MO group had significantly higher triglyceride (p < 0.0001), low-density lipoprotein cholesterol (p = 0.0047), and insulin (p = 0.0008) levels, and significantly lower grip strength (p < 0.0001). Interleukin-17 C levels (unadj. p = 0.017, q = 0.33), colony-stimulating factor-1 (unadj. p = 0.051, q = 0.33) and lymphotoxin-alpha (unadj. p = 0.053, q = 0.33) levels tended to be lower, and interleukin-6 (unadj. p = 0.049, q = 0.33), C-C motif chemokine ligand (CCL) 8 (unadj. p = 0.064, q = 0.33) and CCL2 (unadj. p = 0.066, q = 0.33) levels tended to be higher in the MO group than in the H group.

Conclusion

This exploratory study demonstrates that MO might be associated with circulating levels of some cytokines, suggesting that these cytokines could be potential biomarkers for MO.

Supplementary Information

The online version contains supplementary material available at https://doi.org/10.1007/s12020-026-04788-0.

Introduction

Masked obesity is a concept similar to normal-weight obesity, characterized by a condition in which body fat has accumulated without apparent obesity. This condition has been linked to an increased risk of metabolic and cardiovascular diseases [1, 2]. Masked obesity is particularly common among Asian populations [3], who tend to have a higher body fat percentage at equivalent BMI levels compared with non-Asian. A normal BMI in the presence of increased body fat often reflects reduced skeletal muscle mass. Previous studies have reported decreased skeletal muscle mass [4], impaired physical function [5], and a higher prevalence of sarcopenia [6] in individuals with masked obesity. Sarcopenia is defined as the loss of skeletal muscle mass accompanied by decreased muscle strength [7]. A decline in muscle strength is known to precede in muscle mass loss [8]. Impairments in neural activation such as changes in motor unit firing pattern [9] and neuromuscular junction transmission failure [10] are contributing factors of muscle strength loss. Since reduced muscle mass is not the primary cause of muscle weakness, early detection of both muscle strength loss and muscle mass loss is thus essential for preventive interventions.

Both sarcopenia and obesity share chronic inflammation as a common pathophysiological feature. In obesity, adipokines and cytokines secreted by adipocytes and infiltrating immune cells contribute to chronic low-grade inflammation [11, 12]. Individuals with sarcopenia also exhibit persistent low-grade inflammation, with increased infiltration of immune cells such as macrophages into skeletal muscle tissue [13]. Inflammatory cytokines inhibit protein synthesis and promote skeletal muscle catabolism [14], further exacerbating muscle loss. Elevated circulating levels of tumor necrosis factor (TNF) [15–18], interleukin-6 (IL-6) [15, 17–19], and interleukin-1 beta (IL-1β) [18, 20] have been reported in individuals with sarcopenia. However, cytokine alterations during the pre-sarcopenic stage, characterized by reduced skeletal muscle mass and increased fat accumulation (as seen in masked obesity), remain poorly understood. Because cytokines are highly sensitive indicators of disease status, changes in their circulating levels may be detectable even at early disease stages. Therefore, measuring cytokine levels in masked obesity may provide a useful biomarker for predicting disease progression and evaluating therapeutic interventions. This study was designed as an exploratory study to extract potential biomarkers for MO by performing a comprehensive analysis of blood cytokine concentration. In this study, we focused on masked obesity accompanied by skeletal muscle mass reduction. It is known that women are more likely to have reduced skeletal muscle mass and accumulated body fat than men [21]. Additionally, a previous report showed that a decrease in skeletal muscle mass was associated with an increased risk of insulin resistance only in women [22]. As, investigating MO is particularly important in women, we focused on women in this study.

Methods

Study participants

This study included 432 women aged 30–59 years who underwent health examinations at The University of Osaka Health and Counseling Center between May 2022 and October 2022. All participants provided informed consent for the use of surplus blood samples and for additional assessments, including body composition analysis (InBody270; InBody Japan, Tokyo) and grip strength measurement. The participants were classified into two groups based on skeletal muscle mass index (SMI) and body fat percentage (PBF). Criteria for skeletal muscle mass reduction were defined as SMI < 5.7 kg/m² according to the Asian Working Group for Sarcopenia (AWGS) standards, derived from the lower two standard deviations (SD) of young adult averages [23, 24]. Body fat accumulation was defined as PBF ≥ 30%. Participants with SMI ≥ 5.7 kg/m² and PBF < 30% were classified as the healthy (H) group, while those with SMI < 5.7 kg/m² and PBF ≥ 30% were classified as the masked obesity (MO) group. Serum cytokine levels were measured in a randomly selected subgroup of 60 participants: 35 from the H group and 25 from the MO group.

Clinical parameters

Clinical parameters, including age, height, body weight, BMI, waist circumference (WC), systolic blood pressure (SBP), diastolic blood pressure (DBP), white blood cell count (WBC), hemoglobin (Hb), platelet count (Plt), aspartate aminotransferase (AST), alanine aminotransferase (ALT), γ-glutamyl transpeptidase (γGTP), blood urea nitrogen (BUN), creatinine (Cr), uric acid (UA), total cholesterol (T-Cho), triglycerides (TG), high-density lipoprotein cholesterol (HDL-C), low-density lipoprotein cholesterol (LDL-C), glucose (Glu), and hemoglobin A1c (HbA1c), were obtained from the health examination data. Body composition, including PBF and SMI, was measured using a bioelectric impedance analyzer (InBody 270; InBody Japan, Tokyo). Grip strength was measured using a Smedley-type hand dynamometer (TKK5401; Takei Scientific Instruments Co., Ltd., Niigata, Japan). Serum insulin concentrations were determined by chemiluminescent enzyme immunoassay (Lumipulse Presto insulin, Fujirebio, Tokyo, Japan) using residual serum samples. The waist-to-height ratio (WHtR) was calculated as WC/Height, and A Body Shape Index (ABSI) was calculated as WC/ (BMI^2/3 × height^1/2) [m^11/6 kg^{− 2/3}] [25]. The homeostasis model assessment of insulin resistance (HOMA-R) was calculated as fasting plasma glucose (mmol/L) * fasting immunoreactive insulin (µU/mL) / 22.5 [26].

Cytokine measurement

Cytokines were measured using the Olink^® Target 48 Panel (Olink Proteomics AB, Uppsala, Sweden) employing Proximity Extension Assay (PEA) technology in 35 participants from the H group and 25 from the MO group. The panel comprised 45 cytokines (listed in Supplementary Table S1). In this assay, paired oligonucleotide-labeled antibodies bind to specific epitopes, undergo hybridization, and are then amplified and quantified by real-time PCR. Internal and external spike-in controls were used for quality control of each sample plate. Relative quantitative values were normalized using calibrator data and converted to absolute concentrations (pg/mL). The analyses were conducted by the APRO Science Group/Pharma Foods International Co., Ltd. (Japan). Because each panel could accommodate up to 40 samples, measurements were performed across two panels. Two samples with values comparable to negative controls across all cytokines were excluded. Of the 45 cytokines, colony-stimulating factor-2 (CSF-2), interleukin-2 (IL-2), interleukin-4 (IL-4), interleukin-13 (IL-13), interleukin-33 (IL-33), and thymic stromal lymphopoietin (TSLP) were excluded from the analysis because their concentrations were below the detection limit in more than half of the samples. IL-17 A was also excluded because standard sample values deviated by more than ± 30% from prescribed reference values on one of the panels.

Statistical analyses

Continuous variables as expressed as median (first quartile, third quartile). Statistical significance was set at p < 0.05. The Wilcoxon test was used to compare continuous variables between groups, and effect size was expressed as the Hodges–Lehmann estimate of the median difference with 95% confidence interval (CI). P-values were adjusted for multiple comparisons using the false discovery rate (FDR) method according to Storey’s approach and adjusted value was provided as q-values. Statistical significance was set at q < 0.1. Associations between serum cytokine levels and clinical characteristics were evaluated using Pearson’s correlation coefficient. Analysis of covariance (ANCOVA) was performed to compare cytokine levels between groups adjusted for relevant factors. For variables that exhibited non-normal distributions, a logarithmic transformation was applied before performing Pearson’s correlation analyses or ANCOVA. Log-transformed values were expressed with ‘ln’ at the beginning. All statistical analyses were performed using JMP^®18 software (SAS Institute Inc., Cary, NC, USA).

Results

Clinical characteristics of masked obesity

Of the 432 participants, 193 were classified into the H group and 27 into the MO group. Thirty-five participants from the H group and 25 from the MO group were randomly selected for detailed analysis. The clinical characteristics of these participants are presented in Table 1. All participants included in the cytokine analysis had a BMI < 25 kg/m². Compared with the H group, the MO group had significantly higher WHtR (p = 0.0020), ABSI (p = 0.016), PBF (p < 0.0001), Hb (p = 0.0020), T-Cho (p = 0.042), TG (p < 0.0001), LDL-C (p = 0.0047), insulin (p = 0.0008), and HOMA-R (p = 0.0006) values, whereas SMI (p < 0.0001), grip strength (p < 0.0001), and Cr (p = 0.0084) were significantly lower. No significant differences in BMI were observed between the two groups (Table 1).

Table 1.

Clinical characteristics of study participants

	Healthy group (n = 35)	Masked obesity group (n = 25)	p-value
Age, years	47 (44–52)	48 (44–54)	0.71
BMI, kg/m2	20.6 (19.5–21.6)	20.6 (19.8–21.3)	0.82
WC, cm	71.5 (66.0-76.5)	72.5 (70.0-76.5)	0.33
WHtR	0.44 (0.41–0.46)	0.47 (0.45–0.49)	0.0020
ABSI, *10− 3 m11/6 kg− 2/3	75 (73–78)	78 (75–82)	0.016
PBF, %	22.3 (20.4–26.4)	32.8 (31.2–34.6)	< 0.0001
SMI, kg/m2	6.3 (6.0-6.6)	5.3 (5.2–5.5)	< 0.0001
Grip strength, kg	26.0 (23.5–28.5)	20.5 (18.0-22.3)	< 0.0001
SBP, mmHg	111 (103–121)	115 (107–126)	0.22
DBP, mmHg	69 (63–74)	74 (67–76)	0.099
WBC, *103/µL	5.1 (4.8–5.8)	4.8 (4.3–5.9)	0.38
Hb, g/dL	13.0 (12.4–13.4)	13.8 (12.7–14.5)	0.0020
Plt *104, /µL	23.3 (19.4–27.2)	22.2 (20.7–26.4)	0.76
AST, U/L	19 (17–23)	20 (17–22)	0.60
ALT, U/L	13 (11–18)	13 (11–16)	0.65
γGTP, U/L	16 (12–23)	17 (13–23)	0.80
BUN, mg/dL	11.1 (9.6–12.6)	11.3 (9.7–13.4)	0.75
Cr, mg/dL	0.66 (0.63–0.73)	0.61 (0.55–0.68)	0.0084
UA, mg/dL	4.3 (3.7–4.9)	4.4 (3.9–5.4)	0.44
T-Cho, mg/dL	191 (169–209)	215 (180–230)	0.042
TG, mg/dL	47 (38–61)	76 (59–95)	< 0.0001
HDL-C, mg/dL	74 (60–95)	74 (56–82)	0.25
LDL-C, mg/dL	105 (84–118)	119 (106–147)	0.0047
Glu, mg/dL	84 (81–89)	87 (82–91)	0.12
HbA1c, %	5.3 (5.1–5.5)	5.2 (5.1–5.4)	0.92
insulin, µU/mL	2.9 (1.8-4.0)	4.3 (3.0-4.7)	0.0008
HOMA-R	0.57 (0.40–0.83)	0.87 (0.67-1.0)	0.0006

Data are expressed as median (interquartile range). BMI body mass index, WC waist circumference, WHtR waist-to-height ratio, ABSI a body shape index, PBF percent body fat, SMI skeletal muscle mass index, SBP systolic blood pressure, DBP diastolic blood pressure, WBC white blood cell count, Hb hemoglobin, Plt platelet, AST aspartate aminotransferase, ALT alanine aminotransferase, γGTP γ-glutamyl transpeptidase, BUN blood urea nitrogen, Cr creatinine, UA uric acid, T-Cho total cholesterol, TG triglyceride, HDL-C high-density lipoprotein cholesterol, LDL-C low-density lipoprotein cholesterol, Glu glucose, HbA1c hemoglobin A1c, HOMA-R homeostatic Model Assessment of Insulin Resistance. The Wilcoxon test was used to compare continuous variables between groups. Bold p-values indicate statistical significance

Comparison in cytokines between healthy and masked obesity group

Cytokine profiles were compared between the H and MO groups. Cytokine analyses were performed in 58 cases, excluding the two cases where measurement was not possible as mentioned previously. In the MO group, interleukin-17 C (IL-17 C) levels were lower with a median difference of 2.1 (0.4–4.4) pg/mL (unadjusted: unadj. p = 0.017, q = 0.33), whereas IL-6 levels were higher with a median difference of 0.41 (0.00–0.91) pg/mL (unadj. p = 0.049, q = 0.33) compared with the H group (Fig. 1). Additionally, the MO group showed a trend toward lower levels of colony-stimulating factor-1 (CSF-1) with a median difference of 12 (0–23) pg/mL (unadj. p = 0.051, q = 0.33) and lymphotoxin-alpha (LTA) with a median difference of 1.3 (0.0–2.5) pg/mL (unadj. p = 0.053, q = 0.33), and a trend toward higher levels of C-C motif chemokine ligand 8 (CCL8) with a median difference of 23 (-2–50) pg/mL (unadj. p = 0.064, q = 0.33) and C-C motif chemokine ligand 2 (CCL2) with a median difference of 52 (-5–113) pg/mL (unadj. p = 0.066, q = 0.33). No differences were observed between the two groups for the other parameters before applying the FDR correction (Supplementary Table S2).

Fig. 1.

Difference in cytokine levels between the healthy group and the masked obesity group. Box plots show the minimum, lower quartile, median, upper quartile, and maximum cytokine levels. In the masked obesity group, interleukin-17 C (IL-17 C) levels were lower with a median difference of 2.1 (0.4–4.4) pg/mL (unadjusted: unadj. p = 0.017, q = 0.33) (A), whereas interleukin-6 (IL-6) levels were higher with a median difference of 0.41 (0.00–0.91) pg/mL (unadj. p = 0.049, q = 0.33) (B) compared with the H group. In addition, the masked obesity group showed a trend toward lower levels of colony-stimulating factor-1 (CSF-1) (unadj. p = 0.051, q = 0.33) (C) and lymphotoxin-alpha (LTA) (unadj. p = 0.053, q = 0.33) (D), and a trend toward higher levels of C-C motif chemokine ligand 8 (CCL8) (unadj. p = 0.064, q = 0.33) (E) and C-C motif chemokine ligand 2 (CCL2) (unadj. p = 0.066, q = 0.33) (F). It was confirmed that there were no significant differences after FDR correction. The Wilcoxon test was used to compare cytokine levels between the two groups. The false discovery rate (FDR) method was used to adjust the p-values for multiple comparisons according to Storey’s approach, and the adjusted values were provided as q-values. Healthy group: n = 34, masked obesity group: n = 24

Relationship between cytokines and clinical characteristics

We next examined the relationships between cytokines that differ significantly between the H and MO groups and various clinical indicators. IL-17 C levels were positively correlated with SMI (r = 0.28, p = 0.030) and Cr (r = 0.34, p = 0.0086) levels, whereas IL-6 levels were positively correlated with WHtR (r = 0.29, p = 0.030) and PBF (r = 0.31, p = 0.020) and negatively correlated with grip strength (r = -0.29, p = 0.025) (Fig. 2). CSF-1 levels were positively correlated with age, AST, Cr, and HbA1c levels and negatively correlated with PBF. LTA levels were positively correlated with age, AST, BUN, Cr, and HbA1c levels. CCL2 levels were positively correlated with ABSI, LDL-C, and HbA1c and negatively correlated with grip strength and HDL-C (Supplementary Table S3). We confirmed between-group differences in these cytokines using models adjusted for clinical parameters closely related to these cytokines, age, WHtR, grip strength, creatinine, HbAlc and HOMA-R. The difference in IL-6 between groups got small after adjusting for WHtR and grip strength (Supplementary Table S4).

Fig. 2.

Pearson’s correlation analyses between serum cytokine levels and clinical characteristics (n = 58). Since interleukin-17 C (IL-17 C) and interleukin-6 (IL-6) exhibited a non-normal distribution, a logarithmic transformation was applied prior to analysis. The vertical axis shows the values after this transformation. IL-17 C levels were positively correlated with SMI (r = 0.28, p = 0.030) (A) and Cr (r = 0.34, p = 0.0086) (B) levels, whereas IL-6 levels were positively correlated with waist-to-height ratio (WHtR) (r = 0.29, p = 0.030) (C) and percent body fat (PBF) (r = 0.31, p = 0.020) (D) and negatively correlated with grip strength (r = -0.29, p = 0.025) (E)

Discussion

This is an exploratory study characterized by the blood cytokine profile in individuals with MO. The MO group in the study is not strictly identical to population of general normal-weight obesity, which is defined by BMI and body fat. Therefore, the prevalence of MO (6.3%: 27 out of 432 participants) might have been lower than those in the previous reports: 25.0% [27] − 31.6% [28] on Asian female population. We revealed that IL-6 had a tendency to elevate even in individuals exhibiting only minor abnormalities, which is characterized by mild skeletal muscle mass reduction and fat accumulation. IL-6 is known to play a key role in various chronic inflammatory diseases [29, 30]. Earlier studies have shown that chronic inflammatory states associated with sarcopenia [15, 17–19], obesity [11, 12], and sarcopenic obesity [31], are marked by increased secretion of IL-6 from inflammatory cells and adipocytes, leading to elevated circulating IL-6 levels. Therefore, the tendency for the elevation of IL-6 in MO is consistent with the previous reports.

As shown in Table 1, participants in the MO group also demonstrated significantly lower grip strength than those in the H group, suggesting not only a reduction in skeletal muscle mass but also a potential risk of future progression to sarcopenia. In addition, the MO group exhibited a higher WHtR than the H group, indicating possible visceral fat accumulation, an important driver of inflammatory cytokine production alongside PBF. The results that IL-6 were positively correlated with WHtR and PBF and negatively correlated with grip strength, which is shown in Fig. 2, also support the finding that IL-6 tended to be higher in the MO group than in the H group. The difference in IL-6 between groups decreased after adjusting for WHtR and grip strength (Supplementary Table S4), suggesting that high IL-6 levels may reflect the combined effects of both abdominal obesity and muscle weakness, rather than being a characteristic specific to MO. Elevated IL-6 is known to promote skeletal muscle breakdown [14], reduce basal metabolic rate, and promote fat accumulation. The result that IL-6 levels had a tendency to increase in MO suggests that the manifestations of reduced skeletal muscle mass and increased fat accumulation may progress further in the future. CCL2 and CCL8, also known as MCP-1 and MCP-2, respectively, share 62–71% sequence similarity [32]. Both chemokines recruit immune cells such as monocytes, lymphocytes, and eosinophils to inflammatory sites and play crucial roles in immune responses [33, 34]. In this study, both CCL2 and CCL8 tended to be higher in the MO group, reinforcing the potential link between MO and low-grade inflammation.

Conversely, IL-17 C, CSF-1, and LTA had a tendency to decrease in the MO group. IL-17 C, a member of the IL-17 cytokine family (A-F) [35], is primarily expressed by epithelial cells at mucosal and barrier surfaces such as the gastrointestinal tract, respiratory tract, and skin [36, 37]. It has been implicated in psoriasis [38] and inflammatory bowel disease [37]. IL-17 C is induced by bacterial infection or inflammatory cytokine stimulation and promotes production of cytokines such as TNF-α and IL-1β [39, 40]. To our knowledge, no prior studies have reported an association between IL-17 C and either sarcopenia or obesity. CSF-1, also known as M-CSF, is primarily expressed in mesenchymal cells, while its receptor is expressed in monocyte-derived cells [41, 42]. CSF-1 signaling is essential for the differentiation and survival of macrophages and osteoclasts [43] and stimulates production of inflammatory cytokines such as IL-6 and TNF [41]. Although previous studies have reported a negative correlation between serum CSF-1 and appendicular lean mass [44, 45] and increased CSF-1 expression in adipose tissue of obese individuals [46], our results contradict these findings. Similarly, LTA (TNF-β), a cytotoxic factor produced by activated T cells [47], forms a heterotrimeric structure with lymphotoxin β (LTβ) and binds to the LTβ receptor (LTβR), activating a signal transduction pathway that is crucial for normal immune function [48]. LTA is associated with a higher risk of frailty [44] and has been reported to be higher in obese individuals [49].

IL-17 C, CSF-1, and LTA were correlated with each other (Supplementary Table S5) and were positively correlated with, or tend to correlate with SMI and Cr, parameters reflecting skeletal muscle mass (Fig. 2, Supplementary Table S3). CSF-1 has been reported to be secreted from the skeletal muscle [50], suggesting that serum concentrations of these cytokines may partially reflect skeletal muscle. Considering the possibility of skeletal muscle mass influence, these cytokines were further assessed after adjusting for creatinine, which is an indicator of skeletal muscle mass (Supplementary Table S5). These factors remained lower in the MO group compared with the H group, suggesting that the differences were not solely attributable to skeletal muscle mass.

One possible hypothesis is involvement of intestinal immunity. Interestingly, IL-17 C [36, 51], CSF-1 [52], and LTA [53] are all implicated in intestinal immunity. IL-17 C upregulates antimicrobial peptides, pro-inflammatory molecules [36], and tight junction proteins [51] in the intestinal epithelium. CSF-1 signaling contributes to Paneth cell differentiation and antimicrobial peptide secretion [52]. LTA is essential for the development and maintenance of Peyer’s patches and lymphoid tissue [53]. Reduced intestinal bacterial diversity has been linked to impaired intestinal barrier function in sarcopenia [54] and obesity [55]. Thus, reductions in these cytokines may reflect decreased microbial diversity or compromised intestinal barrier function. Improving the intestinal microbiome may represent a novel strategy for addressing MO.

Another possible hypothesis is that adaptive mechanisms may be involved. We previously reported that circulating myostatin, which is considered to increase in obesity and promote obesity, is low in MO, suggesting a compensatory response [56]. Similar to myostatin, these pro-inflammatory cytokines such as IL-17 C, CSF-1 and LTA may be downregulated in MO as part of a compensatory feedback mechanism that slows disease progression. In addition, although this remains speculative, the MO group is inferred to have lower bone density compared to the H group. Given that IL-17 C [57], CSF-1 [58], and LTA [59] are reported to promote bone resorption and reduce bone density, the negative feedback mechanism in the MO group might have acted to suppress bone density loss.

This study has several limitations. First, none of the associations remained statistically significant after FDR correction. Although the lack of statistical power due to the relatively small sample size may have contributed to this finding, the results should be interpreted with caution, and definitive conclusions cannot be drawn. Second, the study included only middle-aged women from a single facility. In addition, the group with preserved skeletal muscle and high body fat, as well as the group with both low skeletal muscle and low body fat, were excluded from the study population; therefore, the results may not reflect broad biological conclusions. Future multicentered larger-cohort studies with analyses for all the body composition groups are needed to validate these findings. Third, certain factors or cases had to be excluded from the analyses in assay-based problems. For some cytokines, including IL-1β, the analyses were performed using a small sample size due to missing values caused by levels below the detection limit, reducing the ability to draws definitive conclusions. Also, IL-17 A was excluded from the analyses due to a significant inter-panel deviation, suggesting a limitation of multi-panel design. Fourth, body composition analysis using BIA can be affected by hydration status. To minimize this impact, the measurements were performed under fasting at roughly in same time each morning, and participants were confirmed to have no conditions that could cause edema. Finally, this study did not account for several important factors, such as physical activity, dietary habits, menopausal status, and systemic inflammation markers. Thus, the observed association may be modified by these factors.

Conclusion

This exploratory study indicated that some cytokines, such as IL-17 C and IL-6, could be potential biomarkers for MO. Even minor abnormalities, such as MO, suggested to associate with circulating IL-6 levels. Our findings also suggest that IL-17 C behaves differently from typical chronic inflammatory markers in early disease states such as masked obesity. To improve our understanding of MO and its relationship with cytokines, it is necessary to conduct future studies involving larger and more diverse populations using both cross-sectional and longitudinal methods.

Supplementary information

Below is the link to the electronic supplementary material.

Glossary

ANCOVA

analysis of covariance

AWGS

Asian Working Group for Sarcopenia

ALT

alanine aminotransferase

AST

aspartate aminotransferase

BMI

body mass index

BUN

blood urea nitrogen

CCL2

C-C motif chemokine ligand 2

CCL8

C-C motif chemokine ligand 8

Cr

creatinine

CSF-1

colony-stimulating factor-1

CSF-2

colony-stimulating factor-2

DBP

diastolic blood pressure

FDR

False discovery rate

Glu

glucose

Hb

hemoglobin

HbA1c

hemoglobin A1c

HDL-C

high-density lipoprotein cholesterol

HOMA-R

homeostasis model assessment of insulin resistance

IL-13

interleukin-13

IL-17C

interleukin-17C

IL-1β

interleukin-1 beta

IL-2

interleukin-2

IL-33

interleukin-33

IL-4

interleukin-4

IL-6

interleukin-6

LDL-C

low-density lipoprotein cholesterol

LTA

lymphotoxin-alpha

LTβ

lymphotoxin β

LTβR

LTβ receptor

PBF

body fat percentage

Plt

platelet count

SBP

systolic blood pressure

SMI

skeletal muscle mass index

T-Cho

total cholesterol

TG

triglycerides

TNF

tumor necrosis factor

TNF-β

LTA cytotoxic factor produced by activated T cells

TSLP

thymic stromal lymphopoietin

UA

uric acid

WBC

white blood cell count

WC

waist circumference

γGTP

γ-glutamyl transpeptidase

Author contributions

C.I., K.N., I.N. and K.Y.-T. designed the study. C.I. collected and analyzed the data and wrote the manuscript. K.N. supervised the study and critically revised the manuscript. I.N. and K.Y.-T. provided guidance on study development and reviewed the manuscript. D.K. contributed to preservation the serum samples. M.N., K.M., M.S., R.Y. and T.M. contributed to the discussion. All the authors read and approved the final version of the manuscript.

Funding

Open Access funding provided by The University of Osaka. This work was supported by JSPS KAKENHI Grant Number 22K17775.

Data availability

The authors confirm that the data supporting the results of this study are available upon reasonable request.

Declarations

Ethical approval

This was approved by the Ethics Committee of the Health and Counseling Center, The University of Osaka, and conducted in accordance with the principles of the Declaration of Helsinki. Written informed consent was obtained from all participants prior to enrollment.

Consent to participate

Informed consent was obtained from all individual participants included in the study.

Competing interests

The authors declare no competing interests.