Exploring serum sST2 and miR-223 as potential biomarkers of disease activity and clinical response to SLIT in HDM-induced allergic rhinitis

Abstract

Soluble suppression of tumorigenicity 2 (sST2), a decoy receptor of interleukin-33 (IL-33), is involved in allergic inflammation and may serve as a biomarker in allergic rhinitis (AR). MicroRNA-223 (miR-223) has also been implicated in allergic inflammation. To investigate serum sST2 and miR-223 as potential biomarkers of disease activity and treatment-associated changes in patients with house dust mite (HDM)-induced moderate-to-severe allergic rhinitis (MSAR) undergoing sublingual immunotherapy (SLIT). This study included 54 patients with moderate-to-severe AR (MSAR) and 54 healthy controls (HC). Serum sST2, total IgE, HDM-specific IgE, and eosinophil counts were measured. Serum miR-223 relative expression was assessed using real-time PCR. Clinical severity was assessed using the total nasal symptom score (TNSS) and visual analogue scale (VAS). MSAR patients received SLIT for 6 months. Treatment response was defined as a ≥ 30% reduction in TNSS from baseline. Serum sST2 and miR-223 were significantly higher in MSAR patients than in healthy controls (p < 0.001). sST2 correlated positively with HDM-specific IgE and eosinophil counts and decreased significantly after six months of treatment (p < 0.001). Post-treatment sST2 was lower in responders, whereas baseline sST2 did not differ significantly between groups. ROC analysis showed good discrimination of MSAR from HC for sST2 (AUC = 0.933) and miR-223 (AUC ≈ 0.96). Serum sST2 and miR-223 show promising potential as candidate biomarkers of HDM-induced MSAR. The decrease in sST2 after treatment suggests its potential value for monitoring treatment-associated changes. Further studies are needed to validate these findings.

Trial registration Retrospectively registered on ClinicalTrials.gov, registered at 25/2/ 2026, Identifier (NCT07436208). Registered at: https://clinicaltrials.gov/study/NCT07436208.

Introduction

House dust mite (HDM)-induced allergic rhinitis (AR) is one of the most prevalent chronic respiratory conditions globally, significantly impacting quality of life and imposing a substantial healthcare burden¹. Sensitization to HDM allergens, particularly Dermatophagoides pteronyssinus and Dermatophagoides farinae, is common in both adults and children, contributing to persistent nasal inflammation and symptoms such as rhinorrhea, sneezing, nasal congestion, and itching of the eyes and palate².

Pharmacologic treatments such as antihistamines and intranasal corticosteroids offer symptomatic relief without modifying the underlying immune response to allergens. Allergen-specific immunotherapy (AIT) is a disease-modifying treatment for AR. In particular, sublingual immunotherapy (SLIT) using HDM extracts has demonstrated clinical efficacy, safety, and convenience compared with subcutaneous approaches^3,4.

Despite its benefits, SLIT responses vary among patients. Identifying reliable biomarkers that can predict treatment efficacy and monitor disease activity remains a priority in personalizing immunotherapy strategies⁵.

Serum soluble suppression of tumorigenicity 2 (sST2), a soluble form of the ST2 receptor that can act as a decoy receptor for interleukin-33 (IL-33), has been investigated as a potential biomarker in allergic conditions, including asthma and allergic rhinitis. However, available evidence regarding serum sST2 levels in allergic rhinitis is limited and appears variable⁶. Data on its role in monitoring treatment-associated changes following sublingual immunotherapy (SLIT) in HDM-induced AR are also limited⁷.

MicroRNAs (miRNAs) are small, single-stranded, non-coding RNAs, typically 18–24 nucleotides in length, that regulate gene expression primarily by binding to complementary sequences within the 3′-untranslated regions of target messenger RNAs (mRNAs), leading to mRNA degradation or inhibition of protein translation⁸.

MiRNAs have been implicated in various inflammatory and allergic diseases through the regulation of immune responses, including Th1/Th2 polarization and persistent epithelial inflammation mediated by innate immune cells⁹. Because circulating miRNAs are relatively stable in biological fluids and may reflect disease-associated molecular changes, they have emerged as promising non-invasive biomarkers¹⁰.

Among these miRNAs, miR-223 has been implicated in allergic inflammation and may contribute to the pathogenesis of AR. Zhou et al. reported that inhibition of miR-223-3p using an antagomir attenuated pathological changes, whereas administration of a miR-223-3p agomir promoted eosinophilic infiltration of the nasal mucosa in a mouse model of AR¹¹. These findings suggest that miR-223 may play a role in the regulation of allergic inflammation and could have potential as a biomarker in AR.

This study aimed to investigate serum sST2 and miR-223 as potential biomarkers of disease activity and treatment-associated changes in patients with HDM-induced MSAR undergoing SLIT.

Materials and methods

Study design and setting

The study was conducted in two phases. The first phase was a case-control study carried out over three months at the Department of Medical Microbiology and Immunology, Faculty of Medicine, Zagazig University, in collaboration with the Allergy Unit at Zagazig University Hospitals, from February to May 2025. The second phase was a quasi-experimental interventional study involving MSAR patients, conducted over six months. The study flow is illustrated in the study flow diagram (Fig. 1).

Fig. 1.

Study flow diagram.

Study population, inclusion and exclusion criteria

Eligible participants were aged 14 years or older and diagnosed with house dust mite (HDM)-induced allergic rhinitis according to the Allergic Rhinitis and its Impact on Asthma (ARIA) guidelines¹². Included patients had persistent moderate-to-severe symptoms for at least two years and were willing to initiate and adhere to sublingual immunotherapy (SLIT) for six months. Exclusion criteria comprised autoimmune or chronic inflammatory diseases, severe asthma or other pulmonary conditions, recent use (within four weeks) of systemic corticosteroids or immunosuppressive therapy, pregnancy or lactation, and refusal to provide informed consent. Age- and sex-matched healthy controls without a history of allergic disease and with negative skin prick test (SPT) results were also recruited.

Sample size

The sample size was calculated using OpenEpi, version 3.01, using the Sample Size for Comparing Two Means module. Based on a previous study¹³, the expected mean serum sST2 levels were 19.8 ± 7.5 ng/mL in patients with allergic rhinitis and 14.3 ± 6.9 ng/mL in healthy controls, corresponding to an expected mean difference of 5.5 ng/mL. Using a two-sided 95% confidence level (α = 0.05), 80% power, and a 1:1 allocation ratio, the minimum required sample size was 27 participants per group (54 participants in total). We enrolled 54 participants in each group (108 in total) to increase the precision of the estimates and support exploratory analyses.

Clinical and laboratory procedures

Baseline assessment

All participants were subjected to a comprehensive clinical evaluation that included demographic data, disease duration, body mass index (BMI), and prior treatment history.

Symptom scoring

Symptom severity was assessed using the standard four-symptom Total Nasal Symptom Score (TNSS), comprising nasal congestion, rhinorrhea, nasal itching, and sneezing. Each symptom was scored from 0 (no symptoms) to 3 (severe symptoms), yielding a total score ranging from 0 to 12, with higher scores indicating greater symptom severity¹⁴. A 10-cm Visual Analogue Scale (VAS) was also used to assess overall symptom burden¹⁵.

Skin prick testing (SPT)

Skin prick testing (SPT) was performed using standardized allergen extracts, including Dermatophagoides pteronyssinus and D. farinae. Histamine and saline were used as positive and negative controls, respectively^16,17. A wheal diameter ≥ 3 mm greater than that of the negative control was considered a positive result.

Immunological marker and sST2 measurement

Venous blood samples were obtained from all participants at baseline. Peripheral blood eosinophil counts were determined as part of the complete blood count and reported as cells/µL. Serum total immunoglobulin E (IgE) and house dust mite–specific IgE (HDM-IgE, kU/L) were measured using enzyme-linked immunosorbent assay (ELISA) according to the manufacturer’s instructions (Chemus Bioscience, USA).

Serum sST2 levels were quantified using a commercially available ELISA kit (Elabscience, Cat. No. E-EL-H6082). For patients receiving sublingual immunotherapy (SLIT), additional serum samples were obtained after six months of treatment. All assays were conducted in duplicate, and laboratory personnel were blinded to both clinical data and sampling time.

Detection of miR-223 expression levels by quantitative reverse transcription polymerase chain reaction (qRT-PCR)

Total RNA enriched in microRNA was extracted by commercial miRNeasy Serum/Plasma Kits (Qiagen, Germany), according to the manufacturer’s instructions. After extraction, microRNA was reversibly transcribed by reverse transcription kits (miScript II RT Kit, Qiagen, Germany) according to the manufacturer’s instructions. Then, it was incubated in a dry bath block (Rocker Sahara 320 dry bath heat block, Rocker Scientific, Taiwan), initially at 37 °C for 60 min, followed by a second incubation at 95 °C for 5 min to inactivate miScript RT. After that, the reaction mixture was subsequently stored at − 80 °C. Quantitative analysis of mature miR-223 was performed using target-specific miScript Primer Assays with a commercial miScript SYBR Green real-time PCR kit (Qiagen, Germany), in accordance with the manufacturer’s instructions. Human RNU6B (RNU6-2) served as the housekeeping gene. Following an initial activation step of HotStarTaq DNA Polymerase at 95 °C for 15 min, real-time PCR amplification was carried out for 40 cycles, each consisting of denaturation at 95 °C for 15 s, annealing at 55 °C for 30 s, and extension at 70 °C for 30 s. Threshold cycle (Ct) values were determined using the Stratagene Mx3005P system software (Agilent Technologies, USA). Relative miR-223 expression levels were calculated using the comparative Ct (2^−ΔΔCt) method¹⁸.

Intervention stage for (MSAR patients) sublingual immunotherapy (SLIT) protocol

Patients received standardized daily sublingual immunotherapy (SLIT) with house dust mite (HDM) extracts for six months. The allergen extracts were diluted in 50% glycerin and dispensed in 20-mL bottles fitted with glass droppers. The extracts were administered as sublingual drops in the morning on an empty stomach, and patients were instructed to retain the drops under the tongue for two minutes before swallowing. The SLIT protocol consisted of two phases: a build-up phase and a maintenance phase. During the build-up phase, three increasing allergen concentrations (1/200 W/V, 1/100 W/V, and 1/50 W/V) were administered over the first three months. The maintenance phase started in the fourth month using the 1/50 W/V concentration and continued until the end of the six-month treatment period¹⁹. Treatment was initiated under medical supervision, and patients were followed monthly to assess treatment adherence and monitor adverse events. Treatment response was evaluated after six months based on a ≥ 30% reduction in TNSS from baseline. Patients meeting this criterion were classified as responders, whereas those who did not meet it were classified as non-responders. Changes in TNSS, VAS, serum sST2, and miR-223 observed after six months of treatment were subsequently compared between responders and non-responders²⁰.

Individual TNSS reduction (%) was calculated as: [(baseline TNSS − 6-month TNSS)/baseline TNSS] × 100. The ≥ 30% responder criterion was applied at the individual-patient level. For descriptive reporting in Table 2, the percentage reduction in group mean TNSS was calculated from the reported group mean scores.

Table 2.

Comparison of demographic and clinical characteristics between SLIT responders and non-responders (n = 54).

Variable	Responders (n = 32)	Non-responders (n = 22)	P-value
Sex (Male %)	18 (56.25%)	13 (59.1%)	0.836 b
Age (years)	28.8 ± 5.88	28.8 ± 5.61	0.966 c
BMI (kg/m²)	22.8 (22.4–23.4)	22.5 (22.3–23.1)	0.355 a
Disease duration (years)	5 (4–5.25)	4 (3–4)	0.003 ** a
Total IgE (IU/ml)	294.2 (283.5–313.4)	347.3 (333.2–442.1)	0.005 ** a
HDM-specific IgE (kU/l)	18.4 (17.1–18.95)	23.5 (17.9–25.7)	0.013 * a
Blood eosinophils (cells/µl)	311.8 (284.5–329.5)	389.15 (356.5–479.1)	< 0.001 ** a
TNSS (baseline)	8.97 ± 1.99	9.27 ± 1.93	0.586 c
TNSS (6 months)	5.05 ± 1.19	7.77 ± 1.63	< 0.001 ** c
Reduction in group mean TNSS (%)	43.7%	16.2%	–
VAS	7.9 (7.2–8.9)	8.9 (8–9.2)	0.028 * a
sST2 (pre-SLIT, ng/ml)	19.4 ± 4.55	21.6 ± 7.87	0.244 c
sST2 (post-SLIT, ng/ml)	13.1 ± 2.66	19.9 ± 5.94	< 0.001 ** c
miR-223 relative expression (pre-SLIT)	2.98 ± 0.37	3.88 ± 0.06	< 0.001 ** c
miR-223 relative expression (post-SLIT)	0.86 ± 0.11	2.99 ± 0.73	< 0.001 ** c

Values are presented as mean ± SD for normally distributed variables and median (interquartile range) for skewed variables. Categorical data are presented as number (%). Comparisons were performed using the Mann–Whitney U test (a), Chi-square test (b), or independent-samples t-test (c), as appropriate. Statistical significance: p < 0.05 (*) and p < 0.001 (**). The TNSS reduction row is a descriptive percentage calculated from the reported group mean TNSS values and is not the mean of individual percentage changes.

Statistical analysis

Statistical analysis was performed using IBM SPSS for Windows (version 27.0; SPSS Inc., Chicago, IL, USA). The normality of quantitative variables was evaluated using the Shapiro–Wilk test. Data were presented as mean ± standard deviation (SD) or median with interquartile range (IQR), as appropriate. Categorical variables were expressed as frequencies and percentages. Comparisons between groups were conducted using the independent samples t-test or Mann–Whitney U test, while within-group comparisons were analyzed using the paired t-test or Wilcoxon signed-rank test. The chi-square test was applied for categorical data. Correlations between sST2 levels and clinical symptom scores were assessed using Spearman’s correlation coefficient. A p-value < 0.05 was considered statistically significant.

Receiver operating characteristic (ROC) curve analysis was performed to evaluate the discriminatory ability of serum sST2 and miR-223 to distinguish MSAR patients from healthy controls. The area under the curve (AUC), sensitivity, specificity, and optimal cutoff values were calculated. Positive predictive value (PPV), negative predictive value (NPV), and overall diagnostic accuracy were derived from the corresponding 2 × 2 classification table at the selected cutoff.

Results

Baseline characteristics of the study population

A total of 108 participants were enrolled in this study, comprising 54 in the HC group and 54 patients with moderate to severe AR (MSAR group). The demographic characteristics between the two groups were comparable, with no statistically significant differences observed in age [27 (24.5–30.5) years in HC vs. 28 (25–33.8) years in MSAR, p = 0.655], sex distribution (48.1% male in HC vs. 57.4% in MSAR, p = 0.430), or BMI [22.6 (22–23.3) kg/m² in HC vs. 22.7 (22.3–23.4) kg/m² in MSAR, p = 0.411]. Among MSAR patients, the median disease duration was 4.5 years (IQR: 4–5 years) (Table 1).

Table 1.

Demographic, clinical, and laboratory characteristics of the study groups (HC and MSAR) (n = 108).

Variable	HC (n = 54)	MSAR (n = 54)	P-value
Age (years)	27 (24.5–30.5)	28 (25–33.8)	0.655 a
Sex (Male %)	26/54 (48.1%)	31/54 (57.4%)	0.430 b
BMI (kg/m²)	22.6 (22–23.3)	22.7 (22.3–23.4)	0.411 a
Disease Duration (yrs)	Not applicable	4.5 (4–5)	–
Total IgE (IU/ml)	44.5 (37.9–51.1)	307 (284–357)	< 0.001** a
HDM-IgE (kU/l)	0.20 (0.05–0.30)	18.8 (17–24)	< 0.001** a
Eosinophils (cells/µl)	140 (80–200)	324 (292–389)	< 0.001 ** c
TNSS	1 (0–2)	9 (8–11)	< 0.001 ** a
VAS	1.1 (0.85–1.3)	8.4 (7.7–9.2)	< 0.001 ** a
sST2 (pre-SLIT, ng/ml)	8.84 ± 4.16	20.3 ± 6.15	< 0.001 ** c
sST2 (post-SLIT, ng/ml)	–	14.2 (12.4–19.7)	–
miR-223 relative expression	1.06 ± 0.31	3.35 ± 0.53	< 0.001 ** c
Responders (n)	–	32/54 (59.25%)	–
Non-responders (n)	–	22/54 (40.75%)	–

Numerical data are presented as mean ± SD for normally distributed variables and median (interquartile range) for skewed data, while categorical data are presented as number (%). a: Mann–Whitney U test; b: Chi-square test; c: independent-samples t-test. Statistical significance: p < 0.05 (*) and p < 0.001 (**).

Immunological and clinical characteristics

Compared to the HC group, MSAR patients exhibited significantly elevated levels of immunological markers. Serum total IgE levels were markedly increased in the MSAR group [307 IU/ml (284–357)] compared with the HC group [44.5 IU/ml (37.9–51.1), p < 0.001]. Likewise, HDM-specific IgE concentrations were significantly higher in MSAR patients [18.8 kU/l (17–24)] than in the HC group [0.20 kU/l (0.05–0.30), p < 0.001]. Peripheral blood eosinophil counts were also significantly elevated in the MSAR group [324 cells/µl (292–389)] compared with the HC group [140 cells/µl (80–200), p < 0.001].

Symptom burden was significantly greater among MSAR patients. The TNSS was substantially higher in the MSAR group [9 (8–11)] than in the HC group [1 (0–2), p < 0.001]. Similarly, VAS scores for symptom severity were significantly elevated in the MSAR group [8.4 (7.7–9.2)] compared with the HC group [1.1 (0.85–1.3), p < 0.001] (Table 1).

Clinical response to SLIT

Following SLIT, 32 of 54 MSAR patients (59.25%) achieved a clinical response based on the predefined clinical response criteria and were classified as responders, whereas 22 patients (40.75%) were classified as non-responders. Clinical response was defined as a ≥ 30% reduction in TNSS from baseline after six months of treatment.

Comparison of demographic, laboratory, and clinical characteristics between SLIT responders and non-responders

To evaluate factors associated with treatment efficacy, clinical and demographic characteristics were compared between SLIT responders (n = 32) and non-responders (n = 22) (Table 2; Fig. 2). No significant differences were observed in sex distribution (56.25% vs. 59.1% male; p = 0.836), age (28.8 ± 5.88 vs. 28.8 ± 5.61 years; p = 0.966), or BMI [22.8 (22.4–23.4) vs. 22.5 (22.3–23.1) kg/m²; p = 0.355] between groups.

Fig. 2.

Clinical and immunological characteristics of SLIT responders and non-responders (n = 54). Box plots illustrate comparisons between responders and non-responders: (A) total IgE levels, (B) HDM-specific IgE levels, (C) peripheral blood eosinophil counts, (D) Total Nasal Symptom Score (TNSS), (E) Visual Analogue Scale (VAS) scores, (F) serum sST2 levels before SLIT, (G) serum sST2 levels after SLIT, and (H) miR-223 relative expression.

In contrast, responders exhibited significantly longer disease duration compared with non-responders [5 (4–5.25) vs. 4 (3–4) years; p = 0.003]. Immunological parameters also differed significantly between the two groups, with responders having lower levels of total IgE [294.2 (283.5–313.4) vs. 347.3 (333.2–442.1) IU/ml; p = 0.005], HDM-specific IgE [18.4 (17.1–18.95) vs. 23.5 (17.9–25.7) kU/l; p = 0.013], and peripheral eosinophil counts [311.8 (284.5–329.5) vs. 389.15 (356.5–479.1) cells/µl; p < 0.001].

Baseline TNSS did not differ significantly between the two groups. Responders had a baseline TNSS of 8.97 ± 1.99 compared with 9.27 ± 1.93 in non-responders (p = 0.586). In contrast, responders had significantly lower baseline VAS scores than non-responders [7.9 (7.2–8.9) vs. 8.9 (8–9.2), p = 0.028]. At six months, TNSS decreased to 5.05 ± 1.19 in responders and 7.77 ± 1.63 in non-responders (p < 0.001). For descriptive purposes, the reduction in group mean TNSS was 43.7% among responders and 16.2% among non-responders (Table 2).

Baseline sST2 levels did not differ significantly between responders and non-responders (19.4 ± 4.55 vs. 21.6 ± 7.87 ng/ml, p = 0.244). In contrast, post-SLIT sST2 levels were significantly lower in responders than in non-responders (13.1 ± 2.66 vs. 19.9 ± 5.94 ng/ml, p < 0.001).

Comparison of miR-223 relative expression between SLIT responders and non-responders demonstrated a statistically significant difference between the two groups. miR-223 relative expression was significantly lower in responders than in non-responders both before SLIT (2.98 ± 0.37 vs. 3.88 ± 0.06, p < 0.001) and after SLIT (0.86 ± 0.11 vs. 2.99 ± 0.73, p < 0.001) (Table 2 and Fig. 2).

Changes in sST2 levels before and after SLIT

Serum sST2 levels were significantly higher in MSAR patients than in healthy controls (20.3 ± 6.15 vs. 8.84 ± 4.16 ng/mL, respectively; mean difference = 11.46 ng/mL, 95% CI: 9.45–13.47; p < 0.001). Following six months of SLIT, serum sST2 levels decreased to 14.2 (12.4–19.7) ng/mL (Table 1).

As shown in Fig. 3, serum sST2 concentrations decreased significantly following six months of SLIT compared with baseline among the 54 patients with paired measurements (t(53) = 9.01, p < 0.001). The mean paired difference was 4.41 ng/mL (SE = 0.489; 95% CI: 3.43–5.39 ng/mL) (Fig. 3).

Fig. 3.

Paired comparison of serum sST2 levels before and after six months of SLIT in 54 patients with MSAR. All 54 patients had complete pre- and post-treatment sST2 measurements.

Serum miR-223 relative expression in MSAR patients and healthy controls

Serum miR-223 relative expression was also significantly higher in MSAR patients than in healthy controls (3.35 ± 0.53 vs. 1.06 ± 0.31, respectively; mean difference = 2.29, 95% CI: 2.12–2.45; p < 0.001) (Table 1).

Correlations of sST2 and miR-223 with clinical and immunological parameters

Spearman’s rank correlation analysis was performed to assess the associations of serum sST2 levels with clinical and immunological parameters before and after SLIT (Table 3), while correlations of miR-223 relative expression with selected clinical and immunological parameters were assessed separately (Fig. 4).

Table 3.

Spearman’s correlation analysis between serum sST2 levels and clinical markers in MSAR patients before and after SLIT (n = 54).

Marker	Pre-SLIT Spearman ρ	P-value	Post-SLIT Spearman ρ	P-value
Total IgE	−0.027	0.847	0.064	0.645
HDM-specific IgE	0.671	< 0.001**	0.620	< 0.001**
Eosinophils	0.398	0.003**	0.502	< 0.001**
TNSS	0.213	0.123	0.305	0.025 *
VAS	0.224	0.104	0.269	0.049 *

Spearman’s rank correlation analysis was used. SLIT Sublingual immunotherapy, TNSS Total Nasal Symptom Score, VAS Visual Analogue Scale, HDM House Dust Mite. Statistical significance: p < 0.05 (*) and p < 0.001 (**).

Fig. 4.

Scatter plots showing the correlations between miR-223 relative expression and selected clinical and immunological parameters in patients with MSAR (n = 54): (A) total IgE (r = 0.871, p < 0.001), (B) HDM-specific IgE (r = 0.842, p < 0.001), (C) pre-SLIT serum sST2 levels (r = 0.906, p < 0.001), (D) Total Nasal Symptom Score (TNSS) (r = 0.789, p < 0.001), and (E) peripheral blood eosinophil count (r = 0.881, p < 0.001).

At baseline (pre-SLIT), serum sST2 levels demonstrated a strong positive correlation with HDM-specific IgE (ρ = 0.671, p < 0.001) and a moderate positive correlation with peripheral blood eosinophil counts (ρ = 0.398, p = 0.003). In contrast, no statistically significant correlations were observed between sST2 levels and total IgE (ρ = −0.027, p = 0.847), TNSS (ρ = 0.213, p = 0.123), or VAS (ρ = 0.224, p = 0.104) (Table 3).

Correlation analysis revealed significant positive associations between miR-223 relative expression and multiple immunological and clinical markers. miR-223 expression showed positive correlations with total IgE levels (r = 0.871, p < 0.001) and HDM-specific IgE (r = 0.842, p < 0.001). Similarly, a positive correlation was observed between miR-223 expression and peripheral blood eosinophil count (r = 0.881, p < 0.001). In addition, miR-223 expression was positively correlated with clinical disease severity, as assessed by the TNSS (r = 0.789, p < 0.001). The strongest association was observed between miR-223 expression and serum sST2 levels measured prior to SLIT initiation (r = 0.906, p < 0.001), indicating a strong positive correlation between the two biomarkers (Fig. 4).

Following SLIT (post-SLIT), sST2 levels remained positively correlated with HDM-specific IgE (ρ = 0.620, p < 0.001) and peripheral blood eosinophil counts (ρ = 0.502, p < 0.001), indicating moderate positive correlations. In contrast, weak positive correlations were observed between sST2 levels and TNSS (ρ = 0.305, p = 0.025) and VAS (ρ = 0.269, p = 0.049). No statistically significant correlation was observed between sST2 levels and total IgE (ρ = 0.064, p = 0.645).

Diagnostic performance of sST2 in differentiating MSAR from healthy controls

Receiver operating characteristic (ROC) curve analysis was performed to evaluate the diagnostic performance of serum sST2 levels measured before SLIT (pre-SLIT) in differentiating patients with MSAR from healthy controls (HC) (Fig. 5). The area under the curve (AUC) was 0.933 (95% CI: 0.883–0.983; p < 0.001), indicating excellent discriminative ability. At the optimal cutoff value of ≥ 14.3 ng/mL, the 2 × 2 classification table included 48 true-positive, 6 false-negative, 50 true-negative, and 4 false-positive classifications (Table 4). Accordingly, serum sST2 demonstrated a sensitivity of 88.9%, specificity of 92.6%, positive predictive value (PPV) of 92.3%, negative predictive value (NPV) of 89.3%, and overall diagnostic accuracy of 90.7%. These findings demonstrate that serum sST2 has excellent diagnostic performance in distinguishing patients with MSAR from healthy controls. In addition, miR-223 showed similarly high discriminative performance, with an AUC of 0.960 (95% CI: 0.922–0.998; p < 0.001), sensitivity of 94.0%, and specificity of 90.4%.

Fig. 5.

Receiver operating characteristic (ROC) curves for pre-SLIT serum sST2 and miR-223 relative expression in discriminating patients with MSAR from healthy controls (HC) (n = 108). Serum sST2 showed excellent discriminative performance, with an AUC of 0.933 (95% CI: 0.883–0.983; p < 0.001). At the cutoff of ≥ 14.3 ng/mL, sensitivity was 88.9%, specificity 92.6%, PPV 92.3%, NPV 89.3%, and overall accuracy 90.7%. miR-223 showed similarly high discriminative performance, with an AUC of 0.960 (95% CI: 0.922–0.998; p < 0.001), sensitivity of 94.0%, and specificity of 90.4%.

Table 4.

Classification table for pre-SLIT serum sST2 at the optimal cutoff of ≥ 14.3 ng/mL (n = 108).

Reference diagnosis	sST2 ≥ 14.3 ng/mL (Test positive)	sST2 < 14.3 ng/mL (Test negative)	Total
MSAR	48	6	54
Healthy controls	4	50	54
Total	52	56	108

At this cutoff: sensitivity = 48/54 = 88.9%; specificity = 50/54 = 92.6%; PPV = 48/52 = 92.3%; NPV = 50/56 = 89.3%; overall accuracy = 98/108 = 90.7%.

Discussion

sST2 is a soluble form of the ST2 receptor and a component of the IL-33/ST2 signaling pathway, which plays an important role in allergic inflammation²¹. It has been identified as a potential biomarker in allergic diseases, including asthma and AR⁴. However, limited data are available regarding its role in reflecting disease activity and treatment-associated changes in HDM-induced AR⁵. Therefore, this study investigated the potential role of sST2 and miR-223 as candidate biomarkers of disease activity and treatment-associated changes in patients with HDM-induced MSAR undergoing SLIT.

Previous studies have reported variable findings regarding serum sST2 levels in allergic rhinitis, with some studies demonstrating increased levels and others reporting no significant differences compared with healthy controls. These discrepancies may be attributable to differences in age, disease phenotype, disease severity, study populations, or assay characteristics¹³. In the present study, serum sST2 levels were significantly higher in patients with HDM-induced MSAR than in healthy controls and showed positive correlations with HDM-specific IgE and peripheral blood eosinophil counts. These findings support the potential involvement of sST2 in the allergic inflammatory response in HDM-induced MSAR and are consistent with its proposed role in type 2 immune responses.

The current study demonstrated that serum miR-223 relative expression was significantly higher in patients with HDM-induced MSAR than in healthy controls, consistent with the findings reported by Zhou et al.¹¹. They reported significantly increased serum miR-223-3p expression in patients with AR, as well as increased miR-223-3p expression in an AR mouse model compared with controls. Similarly, Ruan et al.²² reported significantly higher median serum miR-223 expression in patients with AR than in healthy controls. Collectively, these findings support the potential involvement of miR-223 in the pathophysiology of AR and its potential utility as a biomarker of allergic inflammation.

Importantly, the present study also demonstrated a strong positive correlation between serum miR-223 expression and sST2 levels before SLIT (r = 0.906, p < 0.001). This close association suggests that miR-223 and sST2 may reflect related aspects of the allergic inflammatory response in HDM-induced AR. Given the reported involvement of miR-223 in eosinophilic inflammation and the role of sST2 within the IL-33/ST2 pathway, their strong correlation may indicate a potential association between these inflammatory pathways. However, the present study was observational and did not investigate the underlying molecular mechanisms; therefore, this association should not be interpreted as evidence of a direct regulatory relationship. Further mechanistic studies are needed to determine whether miR-223 influences the IL-33/ST2 axis or whether both biomarkers are independently regulated by the allergic inflammatory environment.

The current study found no statistically significant differences in age, sex distribution, or BMI between the MSAR group and HC group. These findings indicate that the two groups were demographically comparable at baseline, thereby reducing the likelihood of confounding influences related to age, sex, or anthropometric parameters.

In the current study, MSAR patients exhibited significantly higher serum sST2 levels at baseline compared with healthy controls, along with higher symptom severity scores (TNSS and VAS). These findings are consistent with the established involvement of the IL-33/ST2 axis in type 2 inflammatory responses, particularly in allergic airway diseases. sST2 functions as a soluble decoy receptor that binds IL-33 and can limit IL-33-mediated signaling, thereby contributing to the regulation of mast cell, eosinophil, and T-helper type 2 (Th2) cell responses. Cayrol and Girard described IL-33 as an alarmin that activates mast cells, eosinophils, dendritic cells, and group 2 innate lymphoid cells, thereby promoting Th2-skewed immune responses²³. The positive correlations observed between sST2 levels, HDM-specific IgE, and peripheral blood eosinophil counts further support an association between elevated sST2 and the allergic inflammatory response.

Comparable patterns have been reported in other Th2-mediated conditions, including asthma, atopic dermatitis, and food allergy, in which the IL-33/ST2 pathway contributes to the regulation of type 2 inflammation. In asthma, elevated sST2 levels have been reported during disease exacerbations and have been associated with clinical severity, potentially reflecting a compensatory response aimed at limiting IL-33 signaling and modulating airway inflammation and hyperresponsiveness²⁴. In food allergy, ex vivo studies have shown that soluble ST2–Fc fusion proteins can suppress IL-33-induced type 2 recall responses by up to 80%, highlighting the potential regulatory capacity of soluble ST2 in IgE-mediated hypersensitivity²⁵.

Our findings are partially consistent with those of Zhu et al.¹³, who reported significantly elevated serum sST2 levels in patients with HDM-induced AR and positive correlations with TNSS, VAS, and HDM-specific IgE. In the present study, sST2 levels were also significantly elevated in patients with HDM-induced MSAR and showed a positive correlation with HDM-specific IgE. However, no significant correlations were observed between pre-SLIT sST2 levels and either TNSS or VAS. This difference may be related to variations in disease severity, study populations, sample size, or other methodological characteristics. Nevertheless, the observed association between sST2 and HDM-specific IgE supports the potential utility of sST2 as a biomarker reflecting allergic inflammatory activity.

However, contrasting evidence also exists. A Chinese pediatric study comparing intermittent AR, allergic asthma, and HC reported no significant differences in serum ST2 concentrations among groups, despite elevated IL-33 and IL-31 in AR patients²⁶. This suggests age-related or assay-specific factors may influence measurable sST2 levels in children, contrasting with our adult cohort findings. Ketelaar et al. found that higher sST2 levels in preschool wheezers predicted later eosinophilic asthma but were inversely correlated with fractional exhaled nitric oxide (FeNO), suggesting a modulatory rather than purely proinflammatory role for sST2 in asthma pathogenesis²⁷. Similarly, Quoc et al. found that increased sST2 is linked to neutrophilic inflammation and decreased FeNO in uncontrolled adult asthma, suggesting context-dependent roles for sST2 beyond Th2 inhibition²⁸. These contrasting asthma data indicate that sST2 dynamics may differ across airway diseases and with disease chronicity.

SLIT responders had numerically lower baseline sST2 levels than non-responders, although the difference was not statistically significant. In the present study, treatment response was defined clinically as a ≥ 30% reduction in TNSS from baseline after six months of treatment, without incorporating sST2 or miR-223 into the response definition. Therefore, baseline sST2 levels did not demonstrate a significant association with clinical treatment response. In contrast, the observed reduction in sST2 levels following SLIT may reflect treatment-associated changes in allergic inflammation. As proposed by Palmer et al.²⁹, sST2 may play a compensatory role in limiting type 2 inflammation during allergen immunotherapy.

In the current study, immunological markers, including total IgE, HDM-specific IgE, and peripheral blood eosinophil counts, were significantly elevated in patients with MSAR compared with healthy controls. Among patients with MSAR, correlation analysis at baseline demonstrated a relatively strong positive correlation between serum sST2 levels and HDM-specific IgE (ρ = 0.671, p < 0.001), as well as a moderate positive correlation with peripheral blood eosinophil counts (ρ = 0.398, p = 0.003). Following six months of SLIT, the positive correlation between sST2 and HDM-specific IgE remained significant, although slightly attenuated (ρ = 0.620, p < 0.001), while the correlation with eosinophil counts remained significant and increased in strength (ρ = 0.502, p < 0.001). In addition, post-SLIT sST2 levels showed weak positive correlations with TNSS (ρ = 0.305, p = 0.025) and VAS (ρ = 0.269, p = 0.049). These correlations likely reflect that sST2 correlates with disease burden following treatment of SLIT, which reduces the allergic inflammatory burden and alters the cytokine environment, leading to disease improvement. Similarly, GA²LEN’s 2009 meta-analysis shows clear reductions in symptom scores and medication use for HDM-SLIT in AR and asthma³⁰. This was echoed by the 2023 Malaysian Society Consensus, highlighting HDM-SLIT’s safety and sustained benefit³¹.

Serum sST2 and miR-223 may reflect systemic allergic inflammation but may not fully represent local nasal inflammation. Future studies comparing serum and nasal biomarkers are needed to clarify their relationship with local disease activity.

The comparison between SLIT responders and non-responders provided additional insights into the potential clinical relevance of sST2. Responders exhibited significantly lower total IgE and HDM-specific IgE levels, peripheral blood eosinophil counts, and clinical symptom scores compared with non-responders. Although both pre- and post-SLIT sST2 levels were numerically lower in responders, the difference was statistically significant only post-SLIT. These findings suggest that sST2 may reflect treatment-associated changes in allergic inflammation rather than serve as a reliable baseline predictor of clinical response. Similarly, Zhu et al.¹³ demonstrated that, in patients with HDM-induced AR, sST2 levels correlated significantly with TNSS and VAS at baseline. After three years of SLIT, symptom severity decreased and these correlations were significantly attenuated in responders, supporting the potential role of sST2 as a dynamic marker of allergic inflammatory activity.

Importantly, baseline sST2 levels did not differ significantly between SLIT responders and non-responders, further indicating that baseline sST2 may not serve as a reliable predictor of treatment response. In contrast, post-SLIT sST2 levels were significantly lower in responders than in non-responders, and sST2 levels decreased significantly following treatment. These findings suggest that sST2 may be more useful as a marker of treatment-associated changes and residual allergic inflammatory burden than as a pretreatment predictor of SLIT efficacy.

However, contrasting findings have also been reported. Yang et al.³² reported persistent correlations between sST2 and IgE after SLIT, suggesting that the relationship between sST2 and allergic inflammation may vary among individuals and may be influenced by differences in treatment duration, treatment protocols, or patient characteristics. These findings highlight the context-dependent nature of sST2 responses following allergen immunotherapy and indicate that further studies are needed to clarify its role as a marker of treatment-associated immunological changes.

In the current study, ROC curve analysis demonstrated that pre-SLIT serum sST2 levels effectively discriminated patients with MSAR from healthy controls, with excellent discriminative performance (AUC = 0.933). These findings suggest that serum sST2 may have potential as a biomarker for distinguishing patients with MSAR from healthy individuals. In addition, the significant correlations observed between sST2 levels and immunological parameters, particularly HDM-specific IgE and peripheral blood eosinophil counts, further support its association with the allergic inflammatory status of HDM-induced MSAR.

Similarly, Zhu et al.¹³ reported significantly elevated baseline serum sST2 levels in patients with moderate-to-severe allergic rhinitis compared with healthy controls (p < 0.001). Their ROC analysis, which primarily evaluated the ability of sST2 to predict the clinical efficacy of SLIT, yielded an AUC of 0.839 (p < 0.001). Although the AUC in the current study was higher (0.933), direct comparison between the two AUC values should be interpreted with caution because the ROC analyses evaluated different clinical outcomes. The observed difference may also be attributable to variations in study populations, sample sizes, assay methods, and study designs.

In the current study, ROC curve analysis demonstrated that pre-SLIT serum sST2 levels effectively discriminated patients with MSAR from healthy controls, with excellent discriminative performance (AUC = 0.933). These findings suggest that serum sST2 may have potential as a biomarker for distinguishing patients with MSAR from healthy individuals. In addition, the significant correlations observed between sST2 levels and immunological parameters, particularly HDM-specific IgE and peripheral blood eosinophil counts, further support its association with the allergic inflammatory status of HDM-induced MSAR.

In light of these findings, sST2 and miR-223 may represent promising candidate biomarkers associated with allergic inflammatory activity in HDM-induced MSAR, while sST2 may also reflect treatment-associated changes following SLIT. Their associations with immunological parameters, together with their ability to discriminate between patients with MSAR and healthy controls, support further investigation in larger and independent cohorts. These findings may contribute to the future development of biomarker panels to support more personalized management of patients undergoing SLIT.

Limitation

The study has several limitations. First, the modest sample size and single-center design may limit generalizability and the precision of the ROC estimates; therefore, the AUC of 0.933 should be considered preliminary and requires validation in larger multicenter cohorts. Second, the six-month SLIT duration may not capture long-term clinical and immunological changes. Third, the absence of IL-33 and other cytokine measurements limits mechanistic interpretation of the IL-33/ST2 axis. Finally, serum sST2 and miR-223 may not fully reflect local nasal inflammation. Future studies should include longer follow-up, additional immunological markers, and comparisons with nasal biomarkers.

Conclusion

In conclusion, serum sST2 and miR-223 levels were elevated in patients with HDM-induced MSAR and were associated with selected clinical and immunological markers. Both biomarkers showed excellent discrimination between patients with MSAR and healthy controls. Serum sST2 also decreased after six months of SLIT and was associated with clinical improvement. These exploratory findings suggest that sST2 and miR-223 may represent candidate biomarkers of allergic inflammatory activity, while sST2 may also reflect treatment-associated changes. Further validation in larger, prospective, multicenter studies is warranted.

Recommendations

Given these findings, future multicenter studies with larger sample sizes are needed to validate the clinical utility of serum sST2 in HDM-induced AR. Long-term follow-up is also warranted to assess the durability of sST2 changes following SLIT and to determine whether serial sST2 measurements may help monitor treatment-associated changes over time. Incorporating additional immunological markers, such as IL-33, thymic stromal lymphopoietin (TSLP), and Th2-related cytokines, may provide further mechanistic insights and improve biomarker-based patient stratification. Although sST2 is biologically relevant to the immunopathology of AR, its specificity and clinical utility remain to be established. Therefore, sST2 should currently be considered an investigational biomarker pending validation in larger and diverse patient populations.

Abbreviations

sST2

Soluble suppression of tumorigenicity 2

IL-33

Interleukin 33

SLIT

Sublingual immunotherapy

HDM

House dust mite

TSLP

Thymic stromal lymphopoietin

MSAR

Moderate-to-severe allergic rhinitis

TNSS

Total nasal symptom score

VAS

Visual analogue scale

Author contributions

Conceptualization, G.A.M., S.A.M.; Data curation, H.A.N., M.G.G., H.M.M,; Formal analysis, H.A.N.,M.G.G.,H.M.M; Investigation, M.S., A.N.H., S.W.R, A.M.A., A.A.A., F.H.O., S.A.M., Methodology, G.A.M., S.A.M., M.S., M.G.G, H.M.M., A.N.H., S.W.R, A.M.A., A.A.A., F.H.O., S.A.M., N.M.N.; Resources, M.G.G., H.M.M,; Supervision, H.A.N., N.M.M.; Validation, H.A.N.,N.M.N, G.A.M, M.G.G.; funding acquisition: N.M.N Writing – original draft, All authors; Writing – review & editing, All authors.

Funding

The authors declare that financial support was sponsored by Princess Nourah bint Abdulrahman University, researchers supporting project number (PNURSP2026R768), Princess Nourah bint Abdulrahman University Riyadh, Saudi Arabia.

Data availability

The datasets during and/or analyzed during the current study are available from the corresponding author on reasonable request.

Declarations

Competing interests

The authors declare no competing interests.

Consent for publication

Not applicable.

Ethical Approval

The study was approved by the Institutional Review Board (IRB) of the Faculty of Medicine, Zagazig University (IRB# 1152). Written informed consent was obtained from the parents and legal guardians of all minor participants prior to study enrollment; The study was conducted in accordance with the principles of the Declaration of Helsinki. Confidentiality of patient information was strictly maintained, and all participants retained the right to withdraw from the study at any time without consequence. Trial registration: Retrospectively registered on ClinicalTrials.gov, registered at 25/2/2026, Identifier (NCT07436208).