Efficacy and safety of allergen-specific immunotherapy for allergic asthma: a meta-analysis comparing sublingual and subcutaneous routes across allergen types and age groups

Abstract

Purpose

This meta-analysis aims to evaluate the efficacy and safety of allergen-specific immunotherapy (AIT) in treating allergic asthma.

Methods

A comprehensive search across PubMed, Cochrane Library, Embase and Web of Science was conducted. Randomized controlled trials (RCTs) evaluating AIT for allergic asthma were included. Data analysis was performed using RevMan 5.3, with evaluations for heterogeneity and publication bias. Subgroup analyses were stratified based on different treatment type, duration, age and allergen type (house dust mites vs. grass pollen).

Results

A total of 28 RCTs were included in the meta-analysis. AIT significantly improved asthma symptoms (p < 0.001), with SCIT showing greater efficacy than SLIT. Treatment duration and age did not significantly impact the outcomes. AIT was effective against both house dust mite and grass pollen allergies. It also notably improved medication scores (p < 0.001) and positively impacted FEV1 (p < 0.001). Regarding safety, AIT did not increase the total number of local side effects but significantly increased systemic reactions.

Conclusion

AIT is effective in improving asthma symptoms and reducing medication use, with subcutaneous immunotherapy being more efficacious than sublingual immunotherapy. The effectiveness varies by allergen type, with no substantial differences across age groups and treatment durations.

Introduction

Allergen-specific immunotherapy (AIT) has emerged as a potential disease-modifying treatment for allergic asthma. AIT involves the subcutaneous and sublingual administration of gradually increasing quantities of an allergen extract to modify the immune system’s response to that allergen, thereby reducing the severity of allergic reactions and asthma symptoms [1]. Numerous studies demonstrated its potential in altering the natural course of allergic diseases, with benefits persisting for many years post-treatment [2]. In allergic asthma, AIT has been shown to decrease airway hyperresponsiveness, improve asthma control and reduce the need for pharmacological treatment [2,3]. Despite these promising findings, the acceptance of AIT, particularly subcutaneous immunotherapy (SCIT), is not universal. Concerns over safety and variability in treatment outcomes have limited its widespread adoption [4]. While SCIT has been extensively studied, concerns about safety and patient compliance due to its invasive nature have led to increased interest in sublingual immunotherapy (SLIT) [5,6].

Several meta-analyses have investigated the efficacy and safety of AIT in allergic asthma. Mösges et al. [7] focused on SCIT using depigmented-polymerized allergen extracts but limited their analysis to a specific extract and included patients with allergic rhinitis with or without asthma. Lu et al. [8] examined SCIT in mite-sensitized asthmatic subjects but did not consider other allergens or compare different administration routes. Zheng et al. [9] evaluated SCIT in asthmatic children allergic to house dust mites (HDM), thereby restricting their findings to a paediatric population and a single allergen type. Despite these valuable contributions, there remains a need for a comprehensive analysis that encompasses both SCIT and SLIT, includes multiple allergen types, and considers a broader age range. Furthermore, direct comparisons between SCIT and SLIT in allergic asthma are scarce, and the impact of factors such as treatment duration and patient age on AIT efficacy and safety is not well-established. To address these gaps, we conducted a comprehensive meta-analysis of randomized controlled trials (RCTs) to evaluate the efficacy and safety of AIT in patients with allergic asthma. Our study includes both SCIT and SLIT modalities, various allergen sources (HDM and grass pollen), and encompasses both paediatric and adult populations. We performed subgroup analyses to compare the effects of different administration routes, treatment durations, allergen types and age groups on clinical outcomes.

This systematic review and meta-analysis aim to offer clinicians and researchers a deeper understanding of its role in managing allergic asthma. This study not only updates the existing evidence but also introduces novel insights by directly comparing SCIT and SLIT, evaluating multiple allergens and analysing a wider patient demographic. Our findings have the potential to inform clinical practice and guide personalized treatment strategies for patients with allergic asthma.

Methods

Literature search and study selection

We conducted a comprehensive literature search following the PICOS (Population, Intervention, Comparison, Outcomes, and Study Design) framework and adhering to PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) guidelines. The study protocol was registered online in the International Prospective Register Systematic Reviews (PROSPERO number: CRD42023477577). Ethical Committee acceptance was not requested because only published data were used.

The search spanned several databases, including PubMed, the Cochrane Library (comprising the Cochrane Database of Systematic Reviews and the Cochrane Central Register of Controlled Trials), Embase and Web of Science. The time frame for the database search covered the period from the inception of each database to 31 July 2023. The detailed search strategy was provided in Supplemental Table S1. Two independent reviewers screened for eligible studies meeting the inclusion criteria outlined below.

Inclusion criteria: 1) patients diagnosed with allergic asthma based on objective biomarkers (e.g. skin prick test or specific IgE assessments); 2) patients treated with AIT; 3) studies that provide the assessment of efficacy through symptom and/or medication scores; 4) outcomes related to patient lung function assessment and safety (including local and systemic reactions); 5) RCTs.

Exclusion criteria: 1) duplicate publications; 2) non-clinical studies, such as commentaries, expert opinions, case reports, conference abstracts, reviews, animal studies; 3) studies with inaccessible full-texts or those with incomplete/unextractable data; 4) non-English publications.

Data extraction and quality assessment

Information was independently extracted from the original publications by two reviewers utilizing a pre-designed data extraction form. In parallel, the risk of bias for each RCT was assessed using the tool developed by the Cochrane Collaboration. Discrepancies were jointly deliberated, and if unresolved, a third reviewer was consulted for arbitration.

Data synthesis

Our primary focus was on the efficacy and safety of AIT in patients with asthma. For efficacy outcomes, which were primarily continuous variables, we calculated the standardized mean difference (SMD) along with 95% confidence intervals (CIs). Safety outcomes, categorized as dichotomous variables, were evaluated using odds ratios (ORs) for local and systemic adverse events. Data synthesis was performed using Review Manager 5.3. Heterogeneity was quantified using the I² statistic. A random-effects model was employed in the presence of significant heterogeneity (I²≥ 50%). Publication bias was assessed using funnel plots, and for analyses including 10 or more studies, Begg’s and Egger’s tests were applied.

Subgroup analysis

Subgroup analyses were conducted to further investigate the statistically significant factors and to reduce heterogeneity. These analyses focused on comparing the impact of age, forms of AIT administration, types of allergens and duration of treatment on the efficacy of AIT in patients with asthma.

Results

Following our search strategy, a total of 1,439 relevant publications were identified. After careful screening and assessment, 29 RCTs [10–38] met our inclusion criteria (Figure S1). The patient population in the included studies was predominantly younger and underwent longer treatment periods. According to the Global Initiative for Asthma (GINA) guidelines, most participants suffered from mild to moderate asthma. The allergens primarily involved in these studies were categorized into two types: house dust mites (HDM) and grass pollen, with the majority of patients being allergic to a single allergen source. The characteristics of the included studies were detailed in Table S2. The quality assessment of the literature was depicted in Figure S2.

Primary outcomes

Symptom scores

Twelve RCTs reported on the impact of AIT on asthma symptoms, one of the articles compared SLIT and SCIT with a control group. The meta-analysis demonstrated that AIT significantly improved the symptoms in asthma patients (SMD −2.19 [95% CI −3.19, −1.19], I² = 96%, p < 0.001) (Figure 1(A)). A funnel plot (Figure S3A) and Egger’s test (p = 0.019) suggested the presence of publication bias, although this was not corroborated by Begg’s test (p = 0.855). Additionally, six studies focused on daytime asthma scores (Figure 1(B)) and five studies on night-time asthma scores (Figure 1(C)). However, the results indicated that AIT did not have a significant impact on asthma scores during these specific times of day or night. Notably, one study [31] showed that, in adolescents with moderate asthma, there was an improvement in asthma symptom scores in the AIT-treated group compared to the control group, but the difference was not statistically significant, neither after 1 year nor 2 years of treatment.

Figure 1.

Efficacy of AIT on Asthma symptoms.

(A) Forest plot in asthma symptom scores comparing AIT with placebo. (B) Daytime asthma scores. (C) Nighttime asthma scores.

SLIT vs. SCIT

In the included studies on AIT, there were two studies on SLIT and five studies on SCIT. The results revealed no significant difference in the efficacy of these two treatment modalities for asthma, which may be attributed to the high heterogeneity in effect sizes across the studies (Figure 2(A)). Notably, two studies directly compared the different therapeutic effects of SCIT and SLIT in asthma patients [34,35]. Yukselen et al. [34] did not compare symptom scores, while Li H et al. [35] reported that SLIT appeared to be more effective than SCIT.

Figure 2.

Subgroup analysis of the efficacy of AIT on asthma symptoms.

(A) Comparative efficacy of SLIT and SCIT on asthma symptoms. (B) Analysis of AIT efficacy based on treatment duration (<6 months, 6–12 months, >12 months). (C) Comparison of AIT efficacy in different age groups (<18 years, ≥18 years). (D) Comparison of AIT efficacy for different allergen types.

< 6 months vs. 6 months–12 months vs. >12 months

Subgroup analysis based on treatment duration showed that AIT was effective in < 6 months or > 12 months, but there was no significant difference when the treatment was 6–12 months (Figure 2(B)). This suggested that early treatment or long-term treatment of asthma was more effective for patients. However, one study could not be included in the analysis due to unconvertible data. In this study [33], patients treated with AIT showed a significant increase in asthma scores compared to the control group. Yet, no further improvement in treatment efficacy was observed at 1-, 2- and 3- year post-treatment, indicating no significant differences in asthma scores over these time points.

<18 years vs. ≥18 years

The subgroup analysis based on patient age indicated that AIT was not effective in different age groups and age did not appear to be a significant factor influencing symptom scores (p = 0.66) (Figure 2(C)).

House dust mites vs. grass pollen

The subgroup analysis based on allergen types showed that AIT was effective for both HDM (p = 0.001) and grass pollen (p = 0.02) allergies. However, there seemed to be no difference in the therapeutic effect of AIT on asthma caused by HDM or grass pollen (p = 0.11) (Figure 2(D)). There was no publication bias in this subgroup. This suggested that the publication bias observed in the symptom scores was not attributable to differences in allergen types.

Medication score

Eleven RCTs reported on medication scores in AIT treatment, compared with placebo, one of the articles compared SLIT and SCIT with a placebo group. The meta-analysis revealed that AIT was effective in improving medication scores, signifying a reduction in medication use (SMD −1.10 [95% CI −1.76, −0.44], p = 0.001), which corresponds to a substantial effect (Figure 3(A)). The funnel plot (Figure S3B), Begg’s test (p = 0.193) and Egger’s test (p = 0.120) all indicated no evidence of publication bias.

Figure 3.

Effect of AIT on medication scores.

(A) Meta-analysis of medication score changes with AIT treatment. (B) Subgroup analysis comparing the efficacy of SLIT and SCIT on medication scores. (C) Analysis based on AIT treatment duration for medication scores.

SLIT vs. SCIT

In the included studies on AIT, there were four studies investigating SLIT and six studies focusing on SCIT. Results suggested that SCIT had a better therapeutic effect on asthma compared to SLIT, which differs from the findings reported by Li H et al. [35] (Figure 3(B)).

< 6 months vs. 6 months–12 months vs. >12 months

Subgroup analysis according to AIT treatment duration showed that treatment duration had no significant effect on medication use scores. However, at >12 months, AIT treatment had a significant effect on medication use scores (p = 0.009) (Figure 3(C)).

Secondary outcomes

Pulmonary function-Forced Expiratory Volume in the first second (FEV1)]

Ten studies reported the effects of AIT on FEV1. The meta-analysis showed that AIT improved FEV1 compared to placebo, though with high heterogeneity (SMD 0.88 [95% CI 0.39, 1.37], I² = 89%, p < 0.001) (Figure 4(A)). The funnel plot did not indicate publication bias (Figure S3C), but Egger’s test suggested potential publication bias (p = 0.013), while Begg’s test showed no publication bias (p = 0.951).

Figure 4.

Impact of AIT on FEV1.

(A) Forest plot showing the effect of AIT on FEV1. (B) Subgroup analysis based on treatment duration (≤2 years, >2 years). (C) Comparison of AIT efficacy in patients with house dust mite vs. grass pollen allergies.

≤2 years vs. > 2 years

SMD was 0.80 (95% CI 0.27, 1.33) for treatments lasting ≤2 years and 1.47 (95% CI 0.20, 2.74) for treatments lasting >2 years. Both treatment durations were found to be effective. However, treatment duration did not significantly affect FEV1, with low heterogeneity (p = 0.34, I² = 0%) (Figure 4(B)).

House dust mites vs. pollen

For allergen type, seven RCTs about HDM showed a significant improvement in FEV1 with AIT versus control (SMD 0.47 [95% CI 0.10, 0.84], I² = 81%, p < 0.001) (Figure 4(C)). Among studies evaluating pollen allergens, one RCT was available for each of grass, birch and Parietaria pollen. Both the grass pollen and birch pollen studies reported a statistically significant improvement in FEV1 in the AIT group compared with the control group. In contrast, the Parietaria pollen study found no significant difference in FEV1 between the two groups.

Safety

Total local side effects

Four RCTs reported on the total number of local side effects associated with AIT treatment. The meta-analysis revealed that the incidence of local side effects in the AIT group was comparable to that in the control group, with no statistically significant difference between the two groups (Figure 5(A)). The funnel plot suggested no publication bias (Figure S3D). Furthermore, a study indicated that patients receiving SCIT experienced a higher total number of local side effects compared to those undergoing SLIT [35].

Figure 5.

Safety of AIT in patients.

(A) Analysis of total local side effects associated with AIT. (B) Meta-analysis of total systemic reactions with AIT treatment. (C) Subgroup analysis of systemic reactions based on treatment duration (≤6 months, >6 months).

Total systemic reactions

Five RCTs assessed the impact of AIT on the total number of systemic reactions. The meta-analysis found that the incidence of systemic reactions was significantly higher in the AIT group compared with the control group, with low heterogeneity (p = 0.006, I²=15%) (Figure 5(B)). The funnel plot indicated no publication bias (Figure S3E).

≤6 months vs. >6 months

Two studies had a treatment duration of ≤6 months, and three studies had a treatment duration of >6 months. Studies with treatment duration >6 months showed a significant increase in systemic reactions in the AIT group (OR 2.28, [95% CI 1.40, 3.72], I² = 0%, p < 0.001), while no significant difference was observed in studies with treatment duration ≤6 months (OR 0.95, [95% CI 0.43, 2.06], I² = 0%, p = 0.89). Treatment duration did not significantly influence the total number of systemic reactions, but with high heterogeneity (p = 0.06, I²=71.6%) (Figure 5(C)).

Discussion

This comprehensive meta-analysis evaluated the efficacy and safety of AIT in patients with allergic asthma demonstrate that AIT significantly improves asthma symptoms, reduces medication use and enhances pulmonary function as measured by FEV1. Notably, SCIT showed greater efficacy than SLIT in improving medication scores. Additionally, AIT was effective across different allergen types—specifically house dust mites and grass pollen—and age groups, with treatment duration showing variable effects. However, AIT did not increase the total number of local side effects, but systemic reactions were significantly increased.

Firstly, our results demonstrate a significant improvement in asthma symptom scores with AIT compared to placebo. This finding is consistent with previous studies, which have established the efficacy of AIT in treating asthma, showing significant improvements in symptom scores versus placebo [39,40]. This improvement was consistent across different time points, with no significant differences noted between daytime and night-time scores. This could imply that AIT’s effectiveness is not confined to any specific time of the day, thereby offering a consistent therapeutic benefit throughout. Regarding the consistency of improvements across different times of the day, existing evidence supports AIT’s broad therapeutic efficacy but does not explicitly differentiate between daytime and night-time symptom control [41]. Regarding the impact of treatment duration on symptom scores, our results indicate that AIT was effective in both durations shorter than 6 months and longer than 12 months, but no significant differences were observed when the treatment lasted between 6 and 12 months. This finding is intriguing, as it challenges conventional understanding. Stelmach et al. [28] reported no further improvement in symptom scores after the first year of treatment. However, other research highlights continued benefits beyond 1 year, which may indicate variability based on individual patient characteristics or specific allergens [42]. This apparent contradiction in findings might suggest a potential plateau effect in AIT efficacy, which is an important consideration for clinical practice. Notably, the European Academy of Allergy and Clinical Immunology (EAACI) guidelines recommend at least 3 years of continuous AIT to achieve sustained clinical efficacy [43]. Similarly, this emphasizes that HDM SLIT can serve as an add-on treatment for allergic asthma patients whose symptoms remain uncontrolled despite inhaled corticosteroid therapy and that 3–5 years of effective treatment are generally required before determining whether to continue therapy [44]. Therefore, although our findings confirm that shorter treatment courses can produce measurable symptom improvement, they do not yet provide evidence of long-term efficacy or disease-modifying effects. Further randomized controlled trials with extended follow-up durations (≥ 3 years) are warranted to validate the persistence of clinical benefit and to determine the optimal treatment duration across different allergen types and patient subgroups. Our analysis also highlighted that symptom improvement was consistent regardless of patient age or allergen type, indicating the broad applicability of AIT in managing allergic asthma.

A notable outcome of AIT treatment was the improvement in pulmonary function, as evidenced by increased FEV1 values. This observation is critical as it directly correlates with the primary pathophysiology of asthma-airway obstruction and hyperresponsiveness. The improvement in FEV1 signifies not just symptomatic relief but possibly an alteration in the underlying disease process, echoing the findings of other researchers in the field. Studies have confirmed that AIT can reduce symptoms and induce allergen-specific immune tolerance, potentially improving pulmonary function in asthma patients [45,46].

Safety is a paramount concern with AIT, particularly with SCIT, due to the potential risk of systemic reactions. Our analysis revealed that AIT does not cause more local reactions than control injections. On the other hand, the incidence of systemic reactions was significantly higher in the AIT group than in the control group, indicating that although such reactions are relatively uncommon, they represent an inherent risk of AIT. Subgroup analysis further revealed that when the treatment duration exceeded 6 months, the risk of systemic reactions increased significantly, whereas no significant difference was observed with shorter (≤6-month) treatment courses. This finding suggests that the risk of systemic reactions may rise with cumulative allergen exposure. Nevertheless, these reactions are typically mild to moderate and can be effectively managed under appropriate medical supervision, the incidence of serious adverse events associated with both SCIT and SLIT is extremely low when standardized safety protocols are followed [47]. Importantly, the observed increase in systemic reaction rates should not overshadow the well-established immunological benefits of AIT, including the induction of immune tolerance and the modulation of IgE- and T-cell-mediated pathways [48]. These mechanisms underpin the disease-modifying potential of AIT and support its continued use under guideline-based safety management strategies. Taken together, although systemic reactions occurred more frequently in the AIT group compared with controls, their predictable and manageable nature indicates that, when appropriately administered and monitored, AIT remains a clinically safe and effective therapeutic option for allergic asthma.

In comparing SLIT and SCIT in symptom scores, meta-analyses have shown no significant differences in the efficacy of these two treatments for asthma. However, one direct comparison of SCIT and SLIT suggested that SLIT may achieve better symptom score improvements than SCIT [35]. Although meta-analyses found no significant differences, the high heterogeneity among included studies may limit the conclusions, and further research is needed to explore their relative efficacy. Regarding medication scores, our findings showed a significant reduction in pharmacotherapy use with AIT, particularly with SCIT, which has been shown to offer greater benefits in reducing medication dependency. This is an important aspect, considering the chronic nature of asthma and the potential side effects associated with long-term pharmacotherapy. The reduced medication requirement not only points towards the efficacy of AIT in symptom control but also indicates its role in potentially altering the underlying allergic response. A review has demonstrated that allergen immunotherapy, especially subcutaneous immunotherapy, may help reduce pharmacotherapy use for allergic asthma [49]. Additionally, SCIT has been shown to transform extracellular vesicles into a suppressive phenotype of type 2 allergic inflammation, potentially reducing pharmacotherapy use for asthma [50]. However, conflicting views exist in the literature. For example, a study by Li et al. comparing SCIT and SLIT in Chinese paediatric asthma patients found that SLIT may be more favourable due to fewer adverse effects, although both treatments were effective in reducing medication scores [35]. This discrepancy may be attributable to differences in the non-invasive nature of SLIT and varying tolerability profiles.

Although this meta-analysis included only RCTs to ensure internal validity, an increasing body of real-world evidence (RWE) provides important complementary insights into the long-term effectiveness and safety of AIT in routine clinical practice. Several large-scale observational cohort and registry studies have demonstrated that AIT can significantly reduce asthma exacerbation risk, decrease medication use and maintain sustained clinical benefits during long-term follow-up. For example, Fritzsching et al. reported that AIT led to persistent improvement in asthma control and fewer emergency visits under real-world conditions [51]. Similarly, Buta et al. found that AIT may delay or even interrupt the allergic march in children and adolescents [52]. The EAACI position paper emphasized that RWE is an essential complement to RCTs, particularly for evaluating diverse populations, long-term disease management and treatment adherence [53]. Together, these findings support the effectiveness and safety of AIT from a broader clinical perspective and highlight the need for integrated analyses combining RCT and RWE data to provide a more comprehensive assessment of AIT’s long-term efficacy and real-world applicability.

Despite positive findings, our study has several limitations. First, the high heterogeneity observed in some analyses, particularly in symptom scores and FEV1, may limit the generalizability of the results. The heterogeneity could be due to variations in study designs, patient populations, allergen extracts, dosing regimens and outcome measurements. Second, although major allergen categories such as HDM and grass pollen were well represented, evidence on AIT for tree pollen-induced allergic asthma was limited. Only one included study involved birch pollen, while most existing trials on tree pollen allergens primarily targeted allergic rhinitis rather than asthma. This evidence gap may affect the comprehensiveness of allergen-type comparisons. Third, potential publication bias detected in some analyses may have influenced the results, although Begg’s test did not confirm this bias. Moreover, our study design did not evaluate outcomes at the >3-year recommended by guidelines, precluding firm conclusions about long-term durability or disease-modifying effects. Besides, the majority of included studies involved patients with mild-to-moderate asthma, which may not represent the efficacy and safety of AIT in severe asthma cases. Lastly, the limited number of studies directly comparing SCIT and SLIT restricts our ability to definitively conclude the superiority of one method over the other.

Conclusion

Our meta-analysis supports the efficacy and safety of AIT in the management of allergic asthma. AIT significantly improves asthma symptoms, reduces medication use and enhances pulmonary function, with SCIT showing greater benefits in medication reduction compared to SLIT. AIT is effective across different allergen types and age groups, with early initiation and prolonged treatment potentially offering greater benefits. Future studies should aim to address the heterogeneity among trials, include patients with varying asthma severities, and explore the long-term benefits and safety of AIT in diverse patient populations. Additionally, more direct comparisons between SCIT and SLIT are warranted to guide clinical decision-making.

Funding Statement

This study was supported by Liuzhou Science Research and Technology Development Project (2020NBAB0816), National Natural Science Foundation of China (81460250) and The Scientific Research Start-Up Fund for High Level Talents of Liuzhou People’s Hospital (LRYGCC202120).

Ethics approval statement and patient consent statement

All analyses in this study were based on previously published results and did not require ethical approval or patient consent.

Disclosure statement

All authors declare that they have no conflicts of interest.

Data availability statement

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