Allergy - 2022 - Farraia - Allergen Immunotherapy For Asthma Prevention A Systematic Review and Meta Analysis of

13989995, 2022, 6, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/all.15295 by Cochrane Portugal, Wiley Online Library on [16/03/2023].
See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
| |
Received: 4 February 2022 Revised: 3 March 2022 Accepted: 8 March 2022
DOI: 10.1111/all.15295
REVIEW ARTICLE
Allergen immunotherapy for asthma prevention: A systematic

review and meta-analysis of randomized and non-randomized
controlled studies
Mariana Farraia1,2,3 | Inês Paciência1,2 | Francisca Castro Mendes1,2,3 |

João Cavaleiro Rufo1,2 | Mohamed Shamji4,5 | Ioana Agache6 |
André Moreira1,2,3,7
1
EPIUnit -Institute of Public Health,
University of Porto, Porto, Portugal Abstract
Background: Allergen immunotherapy (AIT) is a disease-modifying treatment for IgE-
2
Laboratory for Integrative and
Translational Research in Population
Health (ITR), Porto, Portugal
mediated diseases. Randomized controlled trials (RCTs) support AIT's potential role in
3
Basic and Clinical Immunology Unit, asthma prevention but evidence from non-randomized studies of interventions (NRSI)
Department of Pathology, Faculty of and longitudinal observational studies has been poorly addressed. Therefore, we
Medicine, University of Porto, Porto,
Portugal aimed to conduct a systematic review and meta-analysis to assess clinical data from
4
National Heart and Lung Institute, all study types to evaluate quantitatively the preventive role of AIT in asthma onset.
Imperial College London, London, UK
5
Methods: We search three databases. Studies were screened, selected and evalu-
NIHR Imperial Biomedical Research
Centre, London, UK ated for quality using risk-of-bias (ROB) tools. Data were descriptively summarized
6
Faculty of Medicine, Transylvania and meta-analysed using random effects. We performed a sensitivity, influence and
University, Brasov, Romania
subgroup analyses. Publication bias and heterogeneity were assessed.
7
São João University Hospital Center,
Porto, Portugal
Results: From the 4549 identified studies, 24 (12 RCTs and 12 NRSI) were included
in the qualitative synthesis and 18 underwent meta-analysis. One study was at low
Correspondence
Mariana Farraia, Institute of Public Health,
ROB, seven had moderate ROB, and 15 were proven of high ROB. Random-effects
University of Porto, Rua das Taipas 135, analysis showed a significant decrease in the risk of developing asthma following AIT
4050-091 Porto, Portugal.
Email: mariana.farraia@gmail.com
by 25% (RR, 95% CI: 0.75, 0.64–0.88). This effect was not significant in the sensitivity
analysis. Publication bias raised concerns, together with the moderate heterogene-
Funding information
Mariana Farraia is funded by Fundação
ity between studies (I2 = 58%). Subgroup analysis showed a remarkable preventive
para a Ciência e Tecnologia through effect of AIT in children (RR, 95% CI: 0.71, 0.53–0.96), when completing 3 years of
the PhD Grant number SFRH/
BD/145168/2019. João Cavaleiro Rufo
therapy (RR, 95% CI: 0.64, 0.47–0.88), and in mono-sensitized patients (RR, 95% CI:
is funded by Fundação para a Ciência 0.49, 0.39–0.61).
e Tecnologia through the Stimulus for
Scientific Employment Individual Support
Conclusions: Our findings support a possible preventive effect of AIT in asthma onset
(2020.01350.CEECIND). and suggest an enhanced effect when administered in children, mono-sensitized, and
for at least 3 years, independently of allergen type.
KEYWORDS
allergen immunotherapy, allergy treament, asthma, prevention, rhinitis
© 2022 European Academy of Allergy and Clinical Immunology and John Wiley & Sons Ltd.
Allergy. 2022;77:1719–1735. |
wileyonlinelibrary.com/journal/all 1719
|
13989995, 2022, 6, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/all.15295 by Cochrane Portugal, Wiley Online Library on [16/03/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
1720 FARRAIA et al.
1 | I NTRO D U C TI O N asthma* AND (prevent* OR risk). During the search, we selected

studies conducted in humans and published in English, Spanish
Allergen immunotherapy (AIT), administered either by the sublingual or Portuguese. No data restrictions were employed, and the
(SLIT) or the subcutaneous (SCIT) route, has been used to treat some search algorithm was repeated in the final stage of the review
IgE-mediated allergic diseases for many decades.1,2 Recent literature (date: 30.12.2021) to confirm whether new papers were avail-
has highlighted the importance of studying allergic conditions using able. Duplicates were removed, and two independent reviewers
a multimorbidity framework instead of the classic transition from screened the search results based on the title and abstract. Any
one disease to another, allowing for a mechanism-based treatment original study that evaluated the effect of AIT on the develop-
approach, such as AIT.3–5 However, it is not fully understood how IgE ment of asthma in healthy subjects or subjects with other allergic
profiles and other underlying allergic (and non-allergic) mechanisms diseases was included. Two independent reviewers read the full
in multimorbidities relate to better treatment response and prog- text of the previous selected studies and those accomplishing
nosis, including the risk of developing other allergic conditions.3,5,6 the inclusion criteria were selected for data extraction. In case
Atopic sensitization is a risk factor for the development of allergic of disagreement, a third reviewer was consulted. References
4,6,7
multimorbidities, and the presence of allergic diseases, such as were managed with EndNote X9 Software (Clarivate). A manual
food allergy, atopic dermatitis, rhinitis and rhinoconjunctivitis, is a search of the reference lists from all the relevant original stud-
significant risk factor for asthma development.8 Children with aller- ies and review articles to identify additional eligible studies was
gic rhinitis have a higher risk of developing asthma later in life, and performed.
the co-existence of these diseases is very frequent since both share
common pathophysiological features.7,9
Allergen immunotherapy induces allergen-specific immune tol- 2.2 | Inclusion criteria
erance and suppresses allergic inflammation by inducing changes
in the profile of T- and B-cell responses to allergens, by decreasing Inclusion criteria were defined following the PICOS framework. We
activation and degranulation of effector cells, and by reducing effec- were interested in studies conducted on subjects of any age, either
tor cells pro-inflammatory mediators in mucosal tissues.10,11 Studies healthy or with allergic conditions such as rhinitis, atopic dermatitis
evaluating the effectiveness of AIT in patients with allergic rhinitis or rhinoconjunctivitis. AIT studies including subjects with comorbid
12,13
suggest a reduction in the risk of developing asthma. asthma were included if data were available for the outcome of in-
Asthma affects more than 300 million people worldwide and it terest in those without asthma. The intervention was AIT with any
is expected to increase in the upcoming years, being one of the most allergen administered through any route, SLIT or SCIT. The com-
common non-communicable diseases with a high impact on patient's parator group included placebo or no intervention (control group).
14
lives and healthcare systems. Thus, there is interest in gathering in- The outcome of interest was the development of asthma. Different
formation from all studies investigating this potential preventive ef- asthma definitions were considered for the inclusion of studies to
fect of AIT in asthma onset. In 2017, the European Academy of Allergy guarantee a comprehensive analysis: asthma symptoms and/or
and Clinical Immunology (EAACI) conducted a systematic review (SR) asthma medication, and GINA-based or medical diagnosis. RCTs and
to assess the effectiveness of AIT in the prevention of asthma and the NRSI (cross-sectional, retrospective or prospective) were consid-
development of new sensitizations using randomized controlled trials ered for selection and data extraction.
(RCTs).12 This study showed a significant reduction in the short-term
risk of patients with allergic rhinitis developing asthma (RR: 0.40; 95%
CI: 0.29–0.54) when data from six RCTs were combined. 2.3 | Data extraction and synthesis
In this SR, we aimed to assess and update the current knowl-
edge in this field by assembling and presenting new evidence, not One reviewer extracted the main characteristics of eligible studies
limited to RCTs, but including non-randomized studies of interven- (study design, population characteristics, intervention, compara-
tions (NRSI) and longitudinal observational studies. The outcome of tor, follow-up, baseline allergic conditions, asthma definition and
interest in this SR is the development of asthma. Different outcome outcome of interest) onto a customized data extraction sheet in
definitions were considered to achieve a comprehensive analysis. Microsoft Excel. A descriptive summary and data tables were pro-
duced to summarize the literature. The outcome of interest was
extracted, when possible, as the absolute number of subjects devel-
2 | M E TH O D S oping asthma (event rate data). When not reported, this number was
estimated if the percentage of subjects developing the outcome was
2.1 | Search strategy and study selection available or by contacting the authors of the study. If none of the
previous data was reported, the study was only considered for the
A search was performed in Pubmed, Scopus and Cochrane data- qualitative synthesis. The data available for the intention-to-treat
bases for articles published until 6 April 2021 using the following population were preferable and, if not reported, the per-protocol
search terms: (immunotherapy OR vaccine OR allergy shot) AND population was extracted.
|
FARRAIA et al. 1721
2.4 | Quality assessment checklist 2020 and the Cochrane Handbook for Systematic Reviews
of Interventions Version 6.1.17,21 This review is registered with
The revised Cochrane Risk of Bias tool (RoB2) for randomized tri- PROSPERO registration number: CRD42021251166.
als15 was used to assess the risk of bias (ROB) of the included
RCTs. The ROB was judged as low, moderate, high or unclear for
each domain: randomization process, deviations from intended in- 3 | R E S U LT S
terventions, missing outcome data, measurement of the outcome,
selection of the reported result and overall bias. This assessment 3.1 | Study selection and characterization
was performed independently by the two reviewers and discussed
if results differed across the evaluations. The Risk of Bias tool for In this systematic review, we identified 4549 potentially eligible
non-randomized studies of interventions (ROBINS-I) was used to as- studies after the removal of duplicates. During the screening of titles
sess the ROB of the included observational studies.16 The domains and abstracts using the prespecified inclusion criteria, 4482 stud-
included in this tool are divided according to the stage of the inter- ies were rejected, yielding 67 studies for a complete revision. Each
vention: pre-intervention (bias due to confounding, bias in selection of those studies was then reviewed. Forty-three studies were later
of participants into the study), at intervention (bias in classification excluded. Thus, 24 studies were included for the qualitative analysis
of interventions) and post-intervention (bias due to deviations from and 18 for the quantitative analysis. Figure 1 shows the flow diagram
intended interventions, bias due to missing data, bias in measuring of the search and selection process.
outcomes and bias in a selection of the reported result). The domains Regarding study design, 12 RCTs were included. 22–33 Of these,
and the overall ROB were judged as low, moderate, high (included three publications are from the PAT study (the Preventive Allergy
serious and critical judgements) or unclear. Treatment study) but with different updates according to the fol-
low-up duration; after 3 years of therapy, 28 and after 2 and 7 years
of therapy discontinuation. 26,27 Twelve studies were NRSI.34–45 Of
2.5 | Data analysis these, six studies used secondary data from healthcare databases;
one analysed real-world data from a population-based administra-
Meta-analysis using the random-effects model approach was per- tive healthcare database,36 while five used real-world data from a
formed in the available data provided by 18 studies and using RStudio health insurance prescription database.35,42–4 4,46 Three studies were
software v3.4.2 (R Foundation). For dichotomous variables, the event not eligible for meta-analysis; one due to the single-arm design (no
rate data were pooled as risk ratio (RR) with a 95% confidence in- comparator group) 40 and two studies used a mixed population (with
terval using the Der Simonian-L aird estimator.17,18 The magnitude of asthma and/or rhinitis) but did not present the data for those only
heterogeneity between the included studies was estimated using the with rhinitis at baseline. 25,32 Additionally, due to different follow-up
Higgins I² statistic, Cochran's Q, tau-squared and prediction interval. times of the same population, three studies were also not included
Outlier and influential analyses were performed to improve the ro- in the main meta-analysis. 26,27,41
bustness of the results. Publication bias was assessed through the The population in each study varied significantly, ranging from
19
creation of a funnel plot and using Egger's test. In case of publica- 28 subjects (14 in the intervention group vs. 14 in the comparator
tion bias, Thuval & Tweedie's trim-and-fill procedure was applied to group) up to 118,754 subjects (2431 in the intervention group vs.
estimate the effect size if missing small studies have been published.20 116,323 in the comparator group). Eleven studies were performed
Sensitivity analysis was performed according to the risk of bias/ only in children, one in adolescents, one in adults and eleven in a
quality of evidence by comparing the summary estimates obtained mixed population (children and adults). Most studies were per-
by excluding studies to be at high risk of bias with those considered formed in a population with allergic rhinitis (n = 12), followed by
at low and moderate risk. Subgroup analyses were undertaken ac- rhinoconjunctivitis (n = 9), atopic dermatitis (n = 2) and healthy indi-
cording to the type of study (RCT vs. NRSI), AIT administration route viduals (n = 1). Eleven studies also included participants with aller-
(SLIT tablets vs. SLIT drops; any SLIT vs. SCIT), population (children gic asthma; nine presented the data for the subpopulation without
vs. adults), asthma definition, follow-up and therapy duration, type asthma at baseline. Eight only included participants mono-sensitized
of outcome analysis (primary or secondary outcomes/endpoints), to the specific allergen under investigation. Asthma onset was the
allergen (grass pollen vs. house-dust mites), history of sensitization primary outcome/endpoint only in seven publications.
(mono-vs. poly-sensitized) and funding (pharmaceutical industry vs. SLIT was evaluated in thirteen studies, SCIT in eight, and both
academic/other). in three studies of which two studies presented data separately ac-
cording to the administration route. The allergens used for AIT were
grass pollen (7 studies), house dust mites (5 studies), birch pollen
2.6 | Registration and reporting (2 studies) and Parietaria judaica (1 study). Nine studies evaluated
more than one allergen but results were not presented separately
This systematic review was reported per the Preferred Reporting according to the type of allergen used for AIT. The follow-up after
Items for Systematic Reviews and Meta-analyses (PRISMA) treatment discontinuation varied from 9 months to 12 years, while
|
1722 FARRAIA et al.
some studies did not conduct a follow-up of participants after does not allow to overly confidence in the preventive effect of
discontinuation. the intervention, meaning that AIT may not be protective in every
The definition of the outcome (asthma) varied between studies, context. In addition, after excluding studies with high risk of bias
and it was grouped as follows: a) asthma symptoms, presence of one the result did not remain significant (RR, 95% CI: 0.76, 0.45–1 .27;
or more symptoms of asthma (nine studies); b) asthma medication, N = 7; I2 = 27%). When including the study considered an out-
prescription and use of asthma medication (5 studies); c) asthma lier (Friztching et al, 2021) and when excluding influential studies
symptoms and medication use; d) the combination of the previous from the analysis (Friztching et al, 2021 and Wahn et al, 2018),
definitions and the presence of both to define asthma (4 studies); e) the results remained significant and similar (Figures S3, S4, S5, S6)
and mixed criteria, according to symptoms, lung function and medi- showing the robustness of these findings. The evidence of publi-
cation use or following the GINA guidelines (5 studies). cation bias needs to be considered since the funnel plot (Figure 3)
The key characteristics and main findings are presented in shows substantial asymmetry and the Egger's test is statistically
Table 1. Risk-of-bias results are detailed in the Appendix S1. significant (p < .01), denoting that small studies without a signifi-
cant effect might be missing. When performing the trim-a nd-f ill
analysis to impute data from missing studies, the result remained
3.2 | Random-effects meta-analysis and marginally significant (RR, 95% CI: 0.82, 0.68–0 .98) and the het-
sensitivity analysis erogeneity remained moderate-to-h igh (I2 = 59.7%). The analysis
by study design evidenced the protective effect of AIT in prevent-
Random-effects meta-a nalysis of studies demonstrated a signifi- ing asthma onset in RCTs (RR, 95% CI: 0.59, 0.40–0 .88; I2 = 17%),
cant decrease in the risk of developing asthma following AIT by while in NRSI studies, the effect was not significant (RR, 95% CI:
25% (RR, 95% CI: 0.75, 0.64–0 .88; N = 17) (Figure 2). The three in- 0.82, 0.66–1 .02; I2: 86%). The between-s tudy heterogeneity in
dicators of between-s tudy heterogeneity show moderate-to-h igh RCTs was low (I2 = 17%), and the prediction interval remained
heterogeneity between the pooled studies (Figure 2). The broad wide (0.31–1 .12); thus, these results need to be interpreted with
prediction interval (0.53–1 .08), which surpasses the value one, caution.
F I G U R E 1 PRISMA flow diagram of the

studies included
|
FARRAIA et al. 1723
3.3 | Subgroup analysis overall heterogeneity between studies was moderate, supporting
the meta-analysis using a random-effects model methodology. The
Given the number of studies needed to perform the subgroup analy- heterogeneity may be explained by differences in study design, ther-
sis to evaluate AIT in different contexts, we decided to perform the apy duration, time of follow-up and population considered across
analysis combining all types of studies independently of the design. studies. Third, the detailed subgroup analysis allowed the scrutiny of
This analysis showed that AIT is beneficial in specific circumstances data, unravelling important clinical implications and supporting the
(Table 2): a) studies conducted only in children (RR 95% CI: 0.71, consistency of the main results. For the first time, a meta-analysis
0.53–0.96; N studies = 8); b) mono-sensitized patients (RR 95% CI: pooled separate effect estimates for SLIT tablets and SLIT drops.
0.49, 0.39–0.61; N = 6); c) AIT administration for at least 3 years (RR Finally, the influence analysis was important to identify high impact
95% CI: 0.64, 0.47–0.88; N = 8); d) follow-up after AIT discontinu- studies driving the results and to increase the robustness of findings.
ation for up to 3 years (RR 95% CI: 0.74, 0.62–0.88; N = 14); and The main limitation is the low number of high-quality studies;
e) evidence from academic/or studies (RR 95% CI: 0.68, 0.53–0.88; specifically, only one was considered of low risk of bias. The sen-
N = 10). When asthma was defined by symptoms, medication and sitivity analysis supports the importance of this gap since it led to
lung function (mixed criteria) or only by medication, the results did imprecise results. However, the biological plausibility of AIT effects
not remain significant. The AIT administration route did not influ- in allergic diseases and the absence of studies pointing for a statisti-
ence the results since results were significant for both SCIT and cally significant increased risk of using AIT as a prevention strategy
SLIT. We were able to pool the results for grass pollen and HDM support the results presented. Most RCTs were conducted using
studies and results were significant for both. Concerning SLIT ad- small samples contributing to imprecision (wider confidence inter-
ministration, both drops and tablets were statistically significant, vals) of individual studies. The effect estimate generated in NRSI
with a larger effect observed when administering SLIT in drops. The was based on the raw results presented in each study. However, four
subgroup analysis was further stratified according to SLIT and SCIT studies presented the adjusted relative risk ratio which was system-
administration, but there were some limitations due to the low num- atically lower than the relative risk obtained in this meta-analysis;
ber of studies considered in each group, and results need to be inter- thus, the results provided by observational studies on the pooled
preted cautiously (Table 2). meta-analysis might be underestimated. Nevertheless, NRSI results
should be analysed cautiously having in mind the risk-of-bias and
confounder factors. Another limitation is the fact that in very few
4 | D I S C U S S I O N studies asthma onset was a primary or a secondary endpoint, high-
lighting the need for studies specifically designed for the assessment
This meta-analysis suggests a preventive effect of allergen immuno- of asthma onset. Additionally, asthma definition should follow GINA
therapy in asthma development, confirming the association found guidelines whenever possible to guarantee precise results and allow
in previous studies. The preventive effect of AIT was noteworthy better comparisons across studies; specifically, when using a more
in children, mono-sensitized patients, and when administered for at complete definition of asthma (including symptoms, medication and
least 3 years, independently of the allergen (grass pollen or HDM) lung function assessments), in a subgroup analysis, the results on
and administration route (SCIT, SLIT tablets or SLIT drops), indicating asthma prevention were not significant, although the results sug-
an overall effect of this therapy on the outcome of interest (Figure 4). gest publication bias further evaluation using a trim-and-fill method
The results were robust for the short-term (3 years of follow-up). This to compute and estimate the impact of small studies missing in the
is the first meta-analysis combining all the evidence, independently analysis did not change the estimate, supporting their consistency.
of study design, since both RCTs and NRSI were reviewed, analysed Finally, the subgroup analysis according to SLIT administration routes
and included in the analysis whenever possible. Thus, it was possible (drops vs. tablets) was hampered by the small number of studies.
to pool the data from eighteen studies, achieving a relevant number The evidence provided by RCTs is more robust when compared
of participants while having moderate heterogeneity. to NRSI studies because of the randomization process which make
There are several strengths of this study. First, the compre- results less prone to bias compared to NRSI. Furthermore, ther-
hensive search strategy and eligibility criteria with the inclusion of apy compliance and treatment duration according to recommen-
evidence from NRSI provide new insights on the preventive effect dations tended to be better accomplished and evaluated in RCTs,
of AIT supported by real-world evidence (RWE) data from a larger which could contribute to the observed positive effects of asthma
number of studies and, consequently, participants. The inclusion prevention in these studies. However, some RCTs included in this
of non-randomized studies, specifically retrospective cohorts of review raise some concerns due to the unblinding of participants
prescriptions and/or routine healthcare databases, was important and outcome assessors, contributing to a lower quality of the evi-
to counterpart evidence from RCTs because these studies reflect dence. 26–28,30,39,41 Authors justified the lack of use of controls due
the complexity of clinical settings.47 AIT results in clinical practice to ethical issues concerning the use of placebo in children for a
depend on the population's characteristics such as patients' prefer- long time and the difficulties in justifying its use to parents when
ences, comorbidities, lifestyle, personal expectations and priorities, asthma is a possible outcome.30,41 However, more studies in this age
and the accessibility to correct diagnosis and treatment. Second, the group are needed despite the anticipated limitations and challenges
|
1724 FARRAIA et al.
TA B L E 1 Characteristics and main results of the original studies including in the systematics review
Participants, allergic disease(s),

Author, year, country Design sensitization Intervention group Comparator group
Zhou et al, 2021, NRSI Mixed (adults and children, N = 91. SCIT (HDM, Allergovit, N = 116. control
China45 5–45 years old); Atopic maintenance dose: 5000 TU/ml)
dermatitis and Rhinitis; mono-
or poly-sensitized
Fritzsching NRSI Mixed (children and adults); N = 27,039. SCIT, SLIT drops, or SLI N = 27,039. Controls,
et al (REACT Rhinitis; mono-or tablets (against HDM, grass, tree, selected using
study), 2021, poly-sensitized or other). propensity
Germany44 matching scores
a a
Jutel et al, 2020, NRSI Mixed (adults and children, N = 1492. SCIT (HDM extract, N = 38,219. Control
Germany43 5–50 years old); Rhinitis; Acaroid®, 1000 or 10 000 TU/ml)
mono-or poly-sensitized
Alviani et al, 2020, RCT Children (aged 6–12 months); N = 54. SLIT drops (HDM extract, ALK N = 53. Placebo
UK 22 Healthy; No sensitization Abelló, 2000 standard treatment
units per day)
a a
Devillier et al, 2019, NRSI Mixed (adults and children over N = 686. SLIT tablets (grass pollen; N = 16,699. Control
France34 5 years old); Rhinitis, asthma; oralair, Stallergenes Greer; or
mono-or poly-sensitized grazax, ALK Abelló)
Wahn et al, 2018 42 NRSI Mixed (adults and children over a

N = 6349. SLIT drops (N = 634) or SCIT a
N = 31,683. Control
5 years old); Rhinitis, asthma; (N = 5715) (birch pollen; natural
mono-or poly-sensitized SLIT drops, natural SCIT or allergoid
SCIT preparations)
a a
Zielen et al, 2018, NRSI Mixed (adults and children over N = 2191. SLIT tablets (grass pollen, N = 53,718. control
Germany35 5 years old); Rhinitis, asthma; grazax or oralair, ALK Abelló or
mono-or poly-sensitized Stallergenes Greer)
|
FARRAIA et al. 1725
Therapy duration; Asthma, N of Risk of

follow-up events Asthma definition bias Main findings
Therapy: 3 years; Int: 19; Cp: 45 Allergic asthma (no more High New asthma onset was a secondary endpoint.
Follow-up: 2 years specifications) Among patients with atopic dermatitis and rhinitis at baseline
(but not asthma), the RR of asthma diagnosis for the
control group was 1.86 (95% CI: 1.17–2.95), considering
an interval of 5 years (3 of intervention +2 follow-up).
Therapy: at least Int: 1148; Cp: At least one asthma High New asthma onset was a secondary endpoint.
1 year; Follow-up: 950 diagnosis (coded at Time-to-first onset of asthma showed an increased risk
at least 1 year (and database) and two for the AIT group (HR: 1.22; 95% CI: 1.12–1.32). In the
up to 9 years) prescriptions of non-asthma cohort, the number of prescriptions for
SABA/ICS asthma medication over time was low and remained
stable for the AIT group across the 9 follow-up years,
but prescriptions tended to increase in the control group
(indicating reporting bias). Data were not presented
separately according to administration route, allergen or
population (children vs. adults).
Therapy: at least Int: 208; Cp: Prescription data of High New asthma onset was a secondary endpoint. After
1 year; Follow-up: 6829 asthma medication covariate adjustment in the regression analysis, the
at least 2 years probability of asthma development was significantly
lower in the patients receiving HDM allergoid (OR = 0.81;
95% CI: 0.70–0.95). When considering only children, the
adjusted OR was 0.83 (95% CI: 0.69–1.00). Number of
asthma events presented separately for children (used in
meta-analysis, for the sensitivity analysis).
Therapy: 1 year; Int: 1; Cp: 5 Based on symptoms, Mod. Asthma development was the primary outcome. Early-life
Follow-up: 5 years medication, and lung administration of HDM SLIT may reduce childhood
function asthma. Only RCT investigating primary prevention of
asthma and atopy using AIT.
Therapy: at least Int: 94; Cp: Prescription data of High New asthma onset was a secondary endpoint. The number
2 years; Follow-up: 2836 asthma medication of patients with a dispensed prescription for asthma
at least 1 year medication (those who had not been treated for asthma
at index date) was lower in the SLIT group (13.7%) than
in the control group (17.0%). The data presented is
for the primary analysis reported by the authors (No
age-matching). For the secondary analysis, with age-
matching of participants, adjusted OR was 0.549 (95% CI:
0.431–0.700).
SLIT tablets on AR may have a potential effect in reducing
asthma onset and asthma progression in routine clinical
practice, using medication dispensing as a proxy.
Therapy: at least Int: 793 Prescription data of High New asthma onset was a secondary endpoint. During
2 years; Follow-up: (SLIT = 61; asthma medication treatment, new-onset asthma risk was significantly
at least 2 years SCIT: 732); reduced in the AIT vs. non-AIT group (OR: 0.83; p = .001).
Cp: 4159 During the follow-up/post-treatment period, the OR for
the AIT group was close to equality and therefore not
significant (OR: 1.02, p = .77). Number of asthma events
were presented separately for SCIT and SLIT drops (used
in meta-analysis, for the sensitivity analysis). Birch pollen
AIT demonstrated real-world benefits up to 6 years post-
treatment cessation through significantly decreased risk
of new-onset asthma medication use on-treatment.
Therapy: at least 2 Tx Int: 208; Cp: Prescription of asthma High New asthma onset was a secondary endpoint. After
cycles; Follow-up: 6222 medication adjustment for covariates, the odds ratio for new asthma
at least 2 years onset evidenced a reduction in the risk of asthma onset
in the SLIT tablet group in all analytical periods (OR:
0.696, 95% CI: 0.552–0.877). Real-world treatment of
AR patients with grass pollen SLIT tablets was associated
with less frequent asthma onset.
(Continues)
|
1726 FARRAIA et al.
TA B L E 1 (Continued)

Valovirta et al (GAP RCT Children (5–12 years old); RC; N = 398. SLIT tablets (grass pollen, N = 414. Placebo
study), 2018, mono-or poly-sensitized grazax, ALK Abelló)
Europe (11
countries)23b
Schmitt et al, 2015, NRSI Mixed; Rhinitis; mono-or N = 2431. Any SLIT, SLIT drops N = 11,6323. Control
Germany 36 poly-sensitized (N = 248) or SCIT (N = 2030) (any
allergen type was included in the
analysis)
a a
Peng et al, 2013, NRSI Mixed; Rhinitis, asthma; only N = 55. SCIT (standardized mite depot- N = 23. Control
China37 HDM sensitized allergen extract, Allergopharma)
Holt et al, 2013, RCT Children (12 months old); Atopic N = 25. SLIT drops (mixture of soluble N = 25. Placebo
Australia and dermatitis; sensitization to ≥1 allergens of HDM, cat, and timothy
USA 24 food allergen grass)
Milani et al (EFESO NRSI Adolescents (mean aged 22 years N = 154. SLIT drops (SLITone, any N = 151. control
trial), 2008, Italy38 old); Rhinitis; mono-or allergen type, ALK Abelló)
poly-sensitized
Marogna et al, 2008, RCT Children (5–17 years old); Rhinitis, N = 144. SLIT (Any allergen type, N = 72. Control
Italy25c intermittent asthma; mono-or Anallergo)
poly-sensitized
a a
Jacobsen et al (PAT RCT Children (6–14 years old); RC, N = 79. SCIT (grass and/or birch pollen, N = 72. Control.
study, 10 years), mild seasonal asthma; only Alutard or Aquagen, ALK Abelló). Sample size after
2007, Europe26d sensitized to grass or birch Sample size after follow-up: 64 follow-up: 53
a a
Niggermann et al (PAT RCT Children (6–14 years old); RC, N = 79. SCIT (grass and/or birch pollen, N = 72. Control.
study, 5 years), mild seasonal asthma; only Alutard or Aquagen, ALK Abelló). Sample size after
2006, Europe27d sensitized to grass or birch Sample size after follow-up: 95. follow-up: 88.
a a
Moller et al (PAT RCT Children (6–14 years old); RC, N = 79. SCIT (grass and/or birch pollen, N = 72. Control
study, 3 years), asthma; only sensitized to Alutard or Aquagen, ALK Abelló)
2002, Europe28d grass or birch
|
FARRAIA et al. 1727

Therapy: 3 years; Int: 34; Cp: 39 Based on symptoms, Low Asthma onset was the primary outcome. Asthma according
Follow-up: 2 years medication, and lung to symptoms or medication intake was a secondary
function outcome. Treatment reduced the risk of experiencing
asthma symptoms and using asthma medication. HR
(95% CI) for asthma onset: 0.90 (0.57–1.43). OR (95%
CI), and for asthma symptoms and medication used: 0.67
(0.46–0.97).
Treatment with SLIT tablets reduced the risk of experiencing
asthma symptoms and using asthma medication.
Therapy: at least Int: 33 (SLIT Diagnoses according Mod. Asthma onset was de primary outcome. AIT demonstrated
1 prescription; drops: 3; to the ICD−10 an RR of 0.60 (95% CI: 0.42–0.84), indicating a 40% risk
Follow-up: up to SCIT: 25); code and at least 2 reduction for patients with AR for the development of
6 years Cp: 1613 prescriptions of ICS asthma. Our data suggest that Tx duration of 3 years
or more have slightly better preventive effects on the
development of asthma.
Number of asthma events presented separately for SCIT
and SLIT drops (used in meta-analysis, for the sensitivity
analysis).
Therapy: 3 years; Int: 6; Cp: 7 Based on GINA High The outcome of asthma development showed an OR of 3.57
Follow-up: 2 years. guidelines (1.05–12.19; p < .05) in favour of the hypothesis that
SCIT can prevent the development of asthma meaning
that the risk of developing asthma was 3.57 times higher
in the control group.
Therapy: 1 year; Int: 4; Cp: 4 Asthma symptoms. Mod. Asthma onset was a secondary endpoint. It is not possible to
Follow-up: 3 years draw firm conclusions concerning the potential efficacy
of this overall approach for allergy prophylaxis. OR (95%
CI): 1.00 (0.16–6.1).
Therapy: 2 years; Int: 13; Cp: 30 Asthma symptoms High Asthma onset was a secondary outcome. SLIT was associated
Follow-up: 0 (ISAAC questionnaire: with a lower incidence of asthma-related symptoms, such
‘Have you had as wheezing, and new sensitizations at the end of the
wheezing or whistling observational period in comparison with allergic patients
in the chest in the last not treated with SLIT. Age, sex distribution, duration of
12 months?’) allergy symptoms and type of AR and poly-sensitized
patients were equally distributed in the two groups.
Therapy: 3 years; No data for Based on GINA High Asthma onset was the primary outcome. This study shows a
Follow-up: 0 subjects guidelines preventive effect on the onset of mild persistent asthma
without (intermittent or mild and remained statistically significant even after the
asthma at asthma). worst-c ase analysis.
baseline
Therapy: 3 years; Int: 16; Cp: 24. Asthma symptoms High Asthma development was the primary outcome. This
Follow-up: 7 years and response to 10-year follow-up study demonstrates that SCIT for
B2-agonists. 3 years shows the persistent long-term effect on
clinical symptoms and a preventive effect on the later
development of asthma in children with seasonal RC.
Asthma development showed an OR of 2.5 (95% CI: 1.1–
5.9) in favour of the hypothesis that SCIT could prevent
the long-term development of asthma.
Therapy: 3 years; Int: 15; Cp: 29. Asthma symptoms High Asthma development was the primary outcome. This
Follow-up: 2 years and response to follow-up study indicates that SIT for 3 years had a long-
B2-agonists. term effect on symptoms and a reduction of the risk of
developing asthma in children. OR (95% CI) of 2.68 (1.3–
5.7, p < .05) in favour of the hypothesis that SCIT can
prevent the long-term development of asthma.
Therapy: 3 years; Int: 19; Cp: 32. Asthma symptoms and High Asthma development was the primary outcome. A 3-year
Follow-up: 0 use of B2-agonists. course of SIT in children with allergic RC significantly
reduces the risk of developing clinical asthma and
improves BHR.
(Continues)
|
1728 FARRAIA et al.

Eng et al, 2006, NRSI Children (5–16 years old); RC; N = 14. SCIT (grass pollen, Allergovit, N = 14. Control
Switzerland39e only sensitized to grass or tree Allergopharma)
Eng et al, 2002, NRSI Children (5–6 years old); RC; only N = 14. SCIT (grass pollen, Allergovit, N = 14. Control
Switzerland41e sensitized to grass or tree Allergopharma)
Crimi et al, 2004, RCT Adults (20–54 years old); Rhinitis; N = 15. SCIT (Parietaria judaica, N = 15. Placebo
Italy29 mono-sensitized only Alutard, ALK Abelló)
Novembre et al, 2004, RCT Children (5–14 years old); RC; only N = 54. SLIT drops (grass pollen, ALK N = 59. Control. After
Italy30 mono-sensitized Abelló). After follow-up: 47 follow-up: 50
Madonini et al, 2003, NRSI Mixed (2–68 years old); Rhinitis, N = 281. SLIT (Any allergen type, No comparator
Italy40b asthma Allergopharma)
Grembiale et al, 2000, RCT Mixed (10–38 years old); Rhinitis; N = 22. SCIT (HDM, Conjuvac, Bayer) N = 22. Placebo
UK31 mono-sensitized only
Pradalier et al, 1999, RCT Mixed (>6 years old); Seasonal N = 62. SLIT (5-grass pollen, N = 61. Placebo
France32b rhinitis, RC, mild asthma; Stallergenes Greer)
sensitized to grass pollen only
Moller et al, 1986, RCT Children (8–16 years old); RC; N = 16. SLIT tablets (birch pollen N = 16. Placebo
Sweden33 mono or poly-sensitized capsules)
Note: Studies presenting quantitative results regarding the effects of AIT on asthma were included in the meta-analysis (if data available on the
column ‘Asthma, N of events’).
Abbreviations: AIT, Allergen immunotherapy; AR, allergic rhinitis; CI, confidence interval; Cp, Comparator group; HDM, house-dust mite; ICD-10,
International Statistical Classification of Diseases; ICS, inhaled corticosteroids; Int, Intervention group; Mod., Moderate; N, number of cases;
NA, not applicable; NRSI, non-randomized studies of interventions; OR, odds ratio; RC, rhinoconjunctivitis; RCT, randomized controlled trial;
SABA, short-acting beta agonists; SCIT, subcutaneous immunotherapy; SLIT, sublingual Immunotherapy; Tx, treatment.
a
Data were presented (sample size) for individuals without asthma at baseline.
b
The primary outcome data were used in the meta-analysis.
c
Not included in the meta-analysis.
d
PAT study articles. For the meta-analysis, we used Moller et al, 2002. When considering the subgroup analysis by time of follow-up, we considered
also Jacobsen 2007 due to the longer follow-up time.
e
Studies with same population but different follow-up times (both, long-term follow-up, >6 years). For the meta-analysis, we used Eng et al, 2006.
|
FARRAIA et al. 1729

Therapy: 3 years; Int: 4; Cp: 7 Asthma symptoms. High Asthma onset was a secondary endpoint. This prospective
Follow-up: controlled long-term follow-up study demonstrates that
12 years both the clinical efficacy and the preventive capacity
of grass pollen SCIT are evident 12 years after the end
of SIT when compared with seasonal pharmacotherapy
alone. This study is the continuation of Eng et al, 2002.
Therapy: 3 years; Int: 3; Cp: 7 Asthma symptoms. High Asthma onset was a secondary endpoint. It demonstrates
Follow-up: 6 years that SIT in children with monovalent pollen allergy
prevents the development of new sensitization, reduces
the prevalence of subsequent asthma and therefore has
the potential to modify the natural course of the allergic
disease.
Therapy: 3 years; Int: 2; Cp: 7 Asthma symptoms. Mod. Asthma onset was a secondary outcome. The preventive
Follow-up: 0. effect of SIT did not reach significance (p = .056) but a
trend was apparent; 47% of subjects in the placebo group
developed asthma symptoms by the end of the study,
compared to 14% in the SIT group.
Therapy: 3 years; Int: 8; Cp: 18 Asthma symptoms and High Asthma development was not defined as the primary or
Follow-up: 0. medication use. secondary outcome. The results of this study show that it
can prevent the onset of asthma in children with RC.
Therapy: 26 (mean) Int: 2 Asthma symptoms. NA Not specified if asthma is a primary or secondary outcome.
months; Follow-up: Despite the limitations of a retrospective, uncontrolled
9 months (mean) study, the results point to the long-term and preventive
effects of SLIT benefits.
Therapy: 2 years; Int: 0; Cp: 2 Asthma symptoms. Mod. Asthma was a secondary outcome. This pilot study, although
follow-up: 0. not designed to detect differences in the incidence of
asthma between the SCIT and placebo groups, suggests
that individuals with allergic rhinitis should undergo
testing of their bronchial reactivity to identify those
with BHR and to provide a window of opportunity in
which SCIT may be effective at preventing progression
to asthma.
Therapy: 6 months. No data for Asthma symptoms. Mod. Asthma onset was a secondary outcome. The proportion
Follow-up: 0. subjects of patients who experienced asthma symptoms at least
without once during the pollen season was significantly smaller
asthma at (p = .005) in the active treatment group (15%) than in the
baseline placebo group (38%). Adjusting for a personal history of
asthma did not abolish the significant between-group
differences.
Therapy: 10 months; Int: 0; Cp: 5 Lung symptoms. Mod. Asthma onset was a secondary outcome. AIT would appear
Follow-up: 0. to have protected against the development of asthma
during this period.
|
1730 FARRAIA et al.
F I G U R E 2 Random-effects meta-analysis of effectiveness of AIT in the prevention of asthma (N of studies =17). Diamond: pooled effect;
grey square: represents study weight; black horizontal line: study confidence interval
F I G U R E 3 Publication bias assessed by

funnel plot
regarding blinding. Still, the recent high-quality RCT conducted by Non-randomized studies of interventions reflect better the real-
23
Valovirta et al. in children used a large placebo group. The study world effectiveness of AIT but raises serious concerns regarding the
was primarily designed to detect differences in asthma onset ac- selection of participants, reporting bias and confounders. Allergic
cording to different criteria. Despite not being significant for the patients receiving AIT are generally from higher socioeconomic
time of asthma onset strictly defined by symptoms, medication and status, with facilitated access to healthcare specialists, overesti-
lung function, the adjusted odds ratio was statistically significant mating the effect. However, better access to the healthcare system
when considering asthma defined by symptoms and medication use, reduces the likelihood of undiagnosed asthma in the intervention
indicating a reduction of 33% in the risk of developing asthma, which group, which may be the opposite in patients with less access or in
corroborates with our findings. 23 the control group, underestimating the results. Another limitation is
|
FARRAIA et al. 1731
TA B L E 2 Meta-analysis when combining all studies, all studies except the outlier (Fritzsching et al, 2021), and by AIT administration route
(SLIT or SCIT)
All studies except the

All studies outlier SLIT SCIT
Meta-analysis, RR (95% CI) 0.77 (0.64; 0.92) 0.75 (0.64; 0.88) 0.77 (0.66; 0.89) 0.72 (0.56; 0.93)
Heterogeneity: I2; Q 79.5%; 20.63 58%; 38.01 17%; 10.79 70%; 26.37
N of studies analysed 18 17 10a 9a
Sensitivity, RR (95% CI) 0.76 (0.45; 1.27) 0.76 (0.45; 1.27) 0.77 (0.39; 1.54) 0.62 (0.13; 2.94)
Heterogeneity: I2; Q 27%; 8.25 27%; 8.25 12%; 4.56 33%; 2.97
N of studies analysed 7 7 5 3
Subgroup analysis
Study design
RCT, RR (95% CI) 0.59 (0.40; 0.88) 0.59 (0.40;0.88) 0.63 (0.29; 1.35) 0.50 (0.27; 0.93)
Heterogeneity: I2; Q 17%; 8.44; 17%; 8.44 33%; 5.93 0%; 0.87
NRSI, RR (95% CI) 0.82 (0.66–1.02) 0.79 (0.66; 0.94) 0.78 (0.66; 0.93) 0.78 (0.59; 1.04)
Heterogeneity: I2; Q 86%; 65.41; 67%; 24.08 15%; 4.71 73%; 18.41
Outcome
Primary analysis, RR (95% CI) 0.84 (0.67; 1.06) 0.84 (0.67; 1.06) 0.76 (0.57; 1.03) 0.83 (0.59; 1.17)
Secondary analysis, RR (95% CI) 0.67 (0.50; 0.89) 0.65 (0.52; 0.82) 0.73 (0.54; 1.00) 0.48 (0.36; 0.65)
Population
Children, RR (95% CI) 0.71 (0.53; 0.96) 0.71 (0.53; 0.96) 0.63 (0.29; 1.35) 0.71 (0.37; 1.38)
Adults, RR (95% CI) 0.75 (0.56; 1.01) 0.75 (0.56; 1.01) 0.78 (0.58; 1.05) 0.54 (0.21; 1.35)
Therapy duration
<3 years, RR (95% CI) 0.85 (0.67; 1.07) 0.81 (0.68; 0.96) 0.76 (0.61; 0.94) 0.87 (0.59; 1.36)
≥3 years, OR (95% CI) 0.64 (0.47; 0.88) 0.64 (0.47; 0.88) 0.78 (0.35; 1.74) 0.60 (0.42; 0.85)
Follow-up after treatment discontinuation
Short-term (≤3 years), RR (95% CI) 0.74 (0.62; 0.88) 0.74 (0.62; 0.88) 0.77 (0.65; 0.91) 0.70 (0.50; 0.97)
Long-term (>3 years), RR (95% CI) 0.88 (0.51; 1.49) 0.73 (0.38; 1.40) Not enough data 0.76 (0.42; 1.37)
Heterogeneity: I2; Q 67%; 11.93 33%; 4.47 0%; 1.62
N of studies analysed 5 4 3
Funding
Pharma, RR (95% CI) 0.84 (0.64; 1.11) 0.79 (0.62; 1.02) 0.77 (0.63; 0.95) Not enough data
Academic/others, RR (95% CI) 0.68 (0.53; 0.88) 0.68 (0.53; 0.88) 0.59 (0.27; 1.31) 0.69 (0.53; 0.91)
Asthma definition
b
Mixed criteria, RR (95% CI) 0.77 (0.34; 1.73) 0.77 (0.34; 1.73) 0.85 (0.36; 2.03) Not enough data
(Continues)
|
1732 FARRAIA et al.
All studies except the

All studies outlier SLIT SCIT
Symptoms + medication, RR (95% CI) 0.52 (0.45; 0.60) 0.52 (0.45; 0.60) 0.50 (0.40;0.61) Not enough data
Heterogeneity: I2; Q 0%; 0.07 0%; 0.07 0%; 0
Medication, RR (95% CI) 0.91 (0.72; 1.14) 0.85 (0.73; 0.99) 0.80 (0.71; 0.91) Not enough data
Symptoms, RR (95% CI) 0.45 (0.29; 0.72) 0.45 (0.29; 0.72) 0.47 (0.10; 2.24) 0.44 (0.16; 1.16)
SLIT administration route
Drops, RR (95% CI) - - 0.65 (0.48; 0.89) -
Heterogeneity: I2; Q 5%; 5.28
N of studies analysed 6
Tablets, RR (95% CI) - - 0.82 (0.70; 0.96) -
Heterogeneity: I2; Q 0%; 2.65
N of studies analysed 4
Previous sensitization
Only mono-sensitized, RR (95% CI) 0.49 (0.39; 0.61) 0.49 (0.39; 0.61) Not enough data 0.49 (0.36; 0.66)
Mono-or poly-sensitized, RR (95% CI) 0.85 (0.70; 1.03) 0.81 (0.70; 0.95) 0.79 (0.69; 0.90) 0.83 (0.58; 1.17)
Heterogeneity: I2; Q 84%; 63.43 61%; 22.85 15%;7.53 79%; 14.04
Allergen
Grass pollen, RR (95% CI) 0.79 (0.68; 0.92) 0.79 (0.68; 0.92) 0.81 (0.71; 0.94) 0.55 (0.42; 0.72)
House dust mite, RR (95% CI) 0.60 (0.38; 0.96) 0.60 (0.38; 0.96) Not enough data 0.63 (0.38; 1.05)
Note: For each output, we report the corresponding RR (95% CI), the heterogeneity parameters, and the number of studies included. Results for
sensitivity analysis and subgroup analysis are also included.
Abbreviations: CI, confidence interval; I2, Higgins & Thompson's I2 Statistic; N, number of studies; NRSI, non-randomized studies of interventions;
Q, Cochran's Q; RCT, randomized controlled trial; RR, risk ratio; SCIT, subcutaneous immunotherapy; SLIT, sublingual immunotherapy.
a
Two studies presented data for both SLIT and SCIT administration routes. The study by Fritzching et al, 2021. presented mixed data for both SLIT and
SCIT but not separated by each administration route and, thus, was not included in these analyses.
b
Mixed criteria were defined as a combination of symptoms, medication and lung function assessments to diagnose asthma; thus, in these studies,
asthma diagnosis was more accurate because it followed a more complete and strict criteria. One of these studies reported an asthma diagnosis
based on the GINA guidelines.
Bold values denote statistical significance at the p < 0.05 level.
the incomplete or absence of data on relevant prognostic variables, according to reporting of the ICD-10 code on the administrative da-
especially in retrospective studies, such as multiple sensitization sta- tabase, at least two times, and prescription of asthma medication. At
tus, severity and duration of allergic disease for which AIT is being baseline, AIT patients were more likely to have higher prescriptions
prescribed, and previous history of asthma. The previous variables of antihistamines and healthcare use because of allergic rhinitis,
may underestimate the results because patients requesting AIT usu- being more likely to be diagnosed with asthma, which may under-
ally have a longer and more severe disease, a previous or familiar his- estimate the results.36 Some studies might have reported overesti-
tory of asthma, and are multiple sensitized. The retrospective study mated results since control patients had more severe disease (higher
conducted by Schmitt et al. was considered of moderate quality due number of allergic rhinitis prescriptions).34,35 The study by Schmitt
to efforts in minimizing bias and confounding, such as the criteria et al. reported a markedly decrease in the adjusted relative risk of
to select controls and adjustment of the effect measure for a priori- 40% in developing asthma compared with the conservative result
defined confounder (age, sex, healthcare use, antihistamines use), calculated by the random-effects meta-analysis that was based on
as well as the outcome definition reliably.36 Asthma was defined raw event counts without any adjustments for covariates. Still, the
|
FARRAIA et al. 1733
F I G U R E 4 This systematic review and meta-analysis included 18 studies (RCT and NRSI) in the quantitative synthesis. Our findings
support a possible preventive effect of AIT in asthma onset. This effect might be enhanced when administered in children, mono-sensitised,
and for at least three years, independently of allergen type. NRSI, non-randomised studies of intervention; RCT, randomised controlled trial;
SCIT, subcutaneous immunotherapy, SLIT, sublingual immunotherapy
adjusted association measure highlighted a higher preventive effect initiate AIT and to be more compliant, although future expectations
for SCIT, independently of allergen (seasonal or perennial) and du- must be managed carefully as it is not fully understood how to select
36
ration of AIT. Finally, we cannot rule out indication bias, which is patients that will benefit the most from the preventive effect of AIT.
inherent to these studies designs, as well as reporting bias since pa- Concerning research implications, our study reveals the lack of high-
tients adhering to AIT are more carefully accompanied by physicians quality well-powered studies, independent of study design, espe-
over time and asthma diagnosis can be reported more frequently. cially for the long-term effects after AIT discontinuation. Evidence
The study considered as an outlier showed different effects on for implementation is also lacking. RWE provided by health-related
asthma prevention.44 The results were explained by the authors to nationwide and representative databases may overcome such limita-
be due to reporting bias since patients from the intervention group tions; however, researchers must ensure a suitable and comparable
visit more frequently specialists and asthma definition required a control group of patients and minimize, whenever possible and em-
medical diagnosis coded onto the database. This reasoning was sup- ploy appropriate methodologies, confounding effects.47 This study
ported by the major conclusions of the study that pointed a preven- also provides data and highlights the importance of including asthma
tive effect of AIT in asthma progression from mild to severe.44 prevention outcomes in cost-effectiveness models.50
Treating allergic multimorbidities may prevent the onset of asthma
because of induced allergen-specific immune tolerance, impacting
underlying mechanisms of asthma development.11 However, it is im- 5 | CO N C LU S I O N S
portant to unravel the underlying mechanisms conferring higher risk
of developing asthma and who will better respond to AIT, highlighting This systematic review with meta-analysis suggests a possible preven-
the importance of discovering treatable traits using the multimorbid- tive role of AIT in the short-term reinforcing the body of evidence and
ity framework. Furthermore, it should be noted that AIT is not cur- highlights a more pronounced effect in children, mono-sensitized, and
rently recommended in healthy individuals nor with atopic dermatitis for the patients completing 3 years of therapy, independently of al-
because the link between atopic dermatitis and asthma development lergen used for AIT. However, we should be aware that this effect was
48
is not clear and studies are generally inconclusive. Still, this work in- observed when considering the diagnosis of asthma based on symp-
cluded evidence from studies conducted in the aforementioned pop- toms or medication, which may indicate that AIT can prevent the onset
ulation to ensure a comprehensive analysis of the literature. of asthma-like symptoms and the use of asthma medication. It may
Our data have important clinical and research implications. In be argued that secondary prevention of asthma using AIT has its best
terms of clinical practice, they may contribute to the further im- chance in stopping the progression of allergic multimorbidities when
provement of guidelines by extending the efficacy results of AIT administered early in life and at the beginning of the allergic disease
in treating allergic diseases to its secondary preventive role in and sensitization. Results were more noticeable for SCIT and SLIT
asthma.49 Moreover, it might be an additional reason for patients to drops, but SLIT tablets also demonstrated protective effects.
|
1734 FARRAIA et al.
This work highlights the importance of gathering data from dif- 7. Siroux V, Ballardini N, Soler M, et al. The asthma-rhinitis multimor-
bidity is associated with IgE polysensitization in adolescents and
ferent sources, both RCTs and NRSI, allowing for a comprehensive
adults. Allergy. 2018;73(7):1447-1458.
analysis of different perspectives and reflecting the real-world com- 8. Toppila-Salmi S, Chanoine S, Karjalainen J, Pekkanen J, Bousquet
plexity and preferences of populations. Future implications of using J, Siroux V. Risk of adult-onset asthma increases with the num-
AIT as a secondary allergy prevention strategy while treating aller- ber of allergic multimorbidities and decreases with age. Allergy.
2019;74(12):2406-2416.
gic rhinitis, atopic dermatitis, or rhinoconjunctivitis, depends on the
9. Khan DA. Allergic rhinitis and asthma: epidemiology and common
availability of high-quality evidence, compliance and adherence of
pathophysiology. Allergy Asthma Proc. 2014;35(5):357-361.
participants, and costs. 10. Jutel M, Van de Veen W, Agache I, Azkur KA, Akdis M, Akdis CA.
Mechanisms of allergen-specific immunotherapy and novel ways
AC K N OW L E D G E M E N T S for vaccine development. Allergol Int. 2013;62(4):425-433.
11. Jensen-Jarolim E, Roth-Walter F, Jordakieva G, Pali-Scholl I.
None.
Allergens and adjuvants in allergen immunotherapy for immune
activation, tolerance, and resilience. J Allergy Clin Immunol Pract.
C O N FL I C T O F I N T E R E S T 2021;9(5):1780-1789.
None of the other authors report any conflict of interest in relation 12. Kristiansen M, Dhami S, Netuveli G, et al. Allergen immunother-
apy for the prevention of allergy: a systematic review and meta-
to this work.
analysis. Pediatr Allergy Immunol. 2017;28(1):18-29.
13. Gradman J, Halken S. Preventive effect of allergen immunother-
AU T H O R C O N T R I B U T I O N S apy on asthma and new sensitizations. J Allergy Clin Immunol Pract.
AM, JCR and MF conceptualized the work and performed the lit- 2021;9(5):1813-1817.
14. To T, Stanojevic S, Moores G, et al. Global asthma prevalence in
erature search. AM, IA and MS discussed and approved the clinical
adults: findings from the cross-sectional world health survey. BMC
aspects and conclusions as medical experts in this field. MF and JCR Public Health. 2012;12:204.
selected, extracted the data and evaluated the risk of bias of the 15. Sterne JAC, Savović J, Page MJ, et al. RoB 2: a revised tool for as-
studies. MF summarized data and conducted the meta-analysis. All sessing risk of bias in randomised trials. BMJ. 2019;366:l4898.
16. Sterne JA, Hernan MA, Reeves BC, et al. ROBINS-I: a tool for as-
authors discussed the results and were involved in the manuscript
sessing risk of bias in non-randomised studies of interventions.
drafting and writing. All authors revised critically the manuscript. BMJ. 2016;355:i4919.
17. Higgins JPT, Thomas J, Chandler J, et al. Cochrane handbook
ORCID for systematic reviews of interventions version 6.1 (updated
September 2020). Cochrane, 2020. www.training.cochrane.org/
Mariana Farraia https://orcid.org/0000-0002-4959-0975
handbook. Accessed May 30, 2021.
Inês Paciência https://orcid.org/0000-0001-5762-7814
18. Veroniki AA, Jackson D, Viechtbauer W, et al. Methods to estimate
Francisca Castro Mendes https://orcid. the between-study variance and its uncertainty in meta-analysis.
org/0000-0002-2400-8408 Res Synth Methods. 2015;7(1):55-79.
João Cavaleiro Rufo https://orcid.org/0000-0003-1175-242X 19. Rucker G, Carpenter JR, Schwarzer G. Detecting and adjusting for
small-study effects in meta-analysis. Biom J. 2011;53(2):351-368.
Mohamed Shamji https://orcid.org/0000-0003-3425-3463
20. Duval S, Tweedie R. “Trim and fill: a simple funnel-plot–based
Ioana Agache https://orcid.org/0000-0001-7994-364X method of testing and adjusting for publication bias in meta-
André Moreira https://orcid.org/0000-0002-7294-9296 analysis”. Biometrics. 2000;56(2):455-463. Wiley Online Library.
21. Page MJ, McKenzie JE, Bossuyt PM, et al. statement: an updated
guideline for reporting systematic reviews. BMJ. 2020;2021:372.
REFERENCES
22. Alviani C, Roberts G, Mitchell F, et al. Primary prevention of asthma
1. Agache I, Lau S, Akdis CA, et al. EAACI guidelines on allergen im- in high-risk children using HDM SLIT: assessment at age 6 years. J
munotherapy: house dust mite-driven allergic asthma. Allergy. Allergy Clin Immunol. 2020;145(6):1711-1713.
2019;74(5):855-873. 23. Valovirta E, Petersen TH, Piotrowska T, et al. Results from the 5-
2. Roberts G, Pfaar O, Akdis CA, et al. EAACI guidelines on al- year SQ grass sublingual immunotherapy tablet asthma preven-
lergen immunotherapy: allergic rhinoconjunctivitis. Allergy. tion (GAP) trial in children with grass pollen allergy. J Allergy Clin
2018;73(4):765-798. Immunol. 2018;141(2):529-538.e13.
3. Custovic A, Custovic D, Kljaic Bukvic B, Fontanella S, Haider S. 24. Holt PG, Sly PD, Sampson HA, et al. Prophylactic use of sublingual
Atopic phenotypes and their implication in the atopic march. Expert allergen immunotherapy in high-risk children: a pilot study. J Allergy
Rev Clin Immunol. 2020;16(9):873-881. Clin Immunol. 2013;132(4):991-993.e1.
4. Bousquet J, Anto JM, Wickman M, et al. Are allergic multimorbid- 25. Marogna M, Tomassetti D, Bernasconi A, et al. Preventive ef-
ities and IgE polysensitization associated with the persistence or fects of sublingual immunotherapy in childhood: an open
re-occurrence of foetal type 2 signalling? The MeDALL hypothesis. randomized controlled study. Ann Allergy Asthma Immunol.
Allergy. 2015;70(9):1062-1078. 2008;101(2):206-211.
5. Blondal V, Sundbom F, Borres MP, et al. Study of atopic multimor- 26. Jacobsen L, Niggemann B, Dreborg S, et al. Specific immunother-
bidity in subjects with rhinitis using multiplex allergen component apy has long-term preventive effect of seasonal and perennial
analysis. Clin Transl Allergy. 2020;10:6. asthma: 10-year follow-up on the PAT study. Allergy. 2007;62(8):
6. Custovic A, Sonntag HJ, Buchan IE, Belgrave D, Simpson A, 943-948.
Prosperi MCF. Evolution pathways of IgE responses to grass 27. Niggemann B, Jacobsen L, Dreborg S, et al. Five-year follow-up on
and mite allergens throughout childhood. J Allergy Clin Immunol. the PAT study: specific immunotherapy and long-term prevention
2015;136(6):1645-1652 e8. of asthma in children. Allergy. 2006;61(7):855-859.
|
FARRAIA et al. 1735
28. Möller C, Dreborg S, Ferdousi HA, et al. Pollen immunother- 41. Eng C, Reinhold M, Gnehm HPE. Long-term efficacy of preseasonal
apy reduces the development of asthma in children with sea- grass pollen immunotherapy in children. Allergy. 2002;57:306-312.
sonal rhinoconjunctivitis (the PAT-study). J Allergy Clin Immunol. 42. Wahn U, Bachert C, Heinrich J, Richter H, Zielen S. Real-world ben-
2002;109(2):251-256. efits of allergen immunotherapy for birch pollen-associated allergic
29. Crimi N, Li Gotti F, Mangano G, et al. A randomized, controlled rhinitis and asthma. Allergy. 2018;74(3):594-604.
study of specific immunotherapy in monosensitized subjects with 43. Jutel M, Bruggenjurgen B, Richter H, Vogelberg C. Real-world evi-
seasonal rhinitis: effect on bronchial hyperresponsiveness, sputum dence of subcutaneous allergoid immunotherapy in house dust mite-
inflammatory markers and development of asthma symptoms. Ann induced allergic rhinitis and asthma. Allergy. 2020;75(8):2046-2054.
Ital Med Int. 2004;19(2):98-108. 4 4. Fritzsching B, Contoli M, Porsbjerg C, et al. Long-term real-world
3 0. Novembre E, Galli E, Landi F, et al. Coseasonal sublingual immu- effectiveness of allergy immunotherapy in patients with allergic
notherapy reduces the development of asthma in children with rhinitis and asthma: results from the REACT study, a retrospective
allergic rhinoconjunctivitis. J Allergy Clin Immunol. 2004;114(4): cohort study. Lancet Reg Health Eur. 2021;13:100275.
851-857. 45. Zhou J, Chen S, Song Z. Analysis of the long-term efficacy and
31. Grembiale RD, Camporota L, Naty S, Tranfa CME, Djukanovic R, safety of subcutaneous immunotherapy for atopic dermatitis.
Marsico SA. Effects of specific immunotherapy in allergic rhinitic Allergy Asthma Proc. 2021;42(2):e47-e54.
individuals with bronchial hyperresponsiveness. Am J Respir Crit 46. Devillier P, Wahn U, Zielen S, Heinrich J. Grass pollen sublingual
Care Med. 2000;162(6):2048-2052. immunotherapy tablets provide long-term relief of grass pollen-
32. Pradalier A, Basset D, Claudel A, et al. Sublingual-swallow immuno- associated allergic rhinitis and reduce the risk of asthma: findings
therapy (SLIT) with a standardized five-grass-pollen extract (drops from a retrospective, real-world database subanalysis. Expert Rev
and sublingual tablets) versus placebo in seasonal rhinitis. Allergy. Clin Immunol. 2017;13(12):1199-1206.
1999;54:819-828. 47. Devillier P, Demoly P, Molimard M. Allergen immunotherapy: what
33. Moller C, Dreborg S, Lanner A, Bjorksten B. Oral immunotherapy of is the added value of real-world evidence from retrospective claims
children with rhinoconjunctivitis due to birch pollen allergy. Allergy. database studies? Expert Rev Respir Med. 2020;14(5):445-452.
1986;41:271-279. 48. Tam HH, Calderon MA, Manikam L, et al. Specific allergen immuno-
3 4. Devillier P, Molimard M, Ansolabehere X, et al. Immunotherapy therapy for the treatment of atopic eczema: a cochrane systematic
with grass pollen tablets reduces medication dispensing for aller- review. Allergy. 2016;71(9):1345-1356.
gic rhinitis and asthma: a retrospective database study in France. 49. Halken S, Larenas-Linnemann D, Roberts G, et al. EAACI guidelines
Allergy. 2019;74(7):1317-1326. on allergen immunotherapy: prevention of allergy. Pediatr Allergy
35. Zielen S, Devillier P, Heinrich J, Richter H, Wahn U. Sublingual Immunol. 2017;28(8):728-745.
immunotherapy provides long-term relief in allergic rhinitis and 50. Di Bona D, Bilancia M, Albanesi M, Caiaffa MF, Macchia L. Cost-
reduces the risk of asthma: a retrospective, real-world database effectiveness of grass pollen allergen immunotherapy in adults.
analysis. Allergy. 2018;73(1):165-177. Allergy. 2020;75(9):2319-2329.
36. Schmitt J, Schwarz K, Stadler E, Wüstenberg EG. Allergy immu-
notherapy for allergic rhinitis effectively prevents asthma: results
from a large retrospective cohort study. J Allergy Clin Immunol. S U P P O R T I N G I N FO R M AT I O N
2015;136(6):1511-1516. Additional supporting information may be found in the online
37. Peng H, Wei Li C, Lin ZB, Li TY. Long-term efficacy of specific im-
version of the article at the publisher’s website.
munotherapy on house dust mite-induced allergic rhinitis in China.
Otolaryngol Head Neck Surg. 2013;149(1):40-46.
38. Milani M, Pecora S, Burastero S. Observational study of sublingual
specific immunotherapy persistent and intermittent allergic rhini- How to cite this article: Farraia M, Paciência I, Castro
tis: the EFESO trial. Curr Med Res Opin. 2008;24(9):2719-2724. Mendes F, et al. Allergen immunotherapy for asthma
39. Eng PA, Borer-Reinhold M, Heijnen IA, Gnehm HP. Twelve-year fol-
prevention: A systematic review and meta-analysis of
low-up after discontinuation of preseasonal grass pollen immuno-
therapy in childhood. Allergy. 2006;61(2):198-201. randomized and non-randomized controlled studies. Allergy.
4 0. Madonini E, Agostinis F, Barra R, et al. Long-term and preven- 2022;77:1719–1735. doi:10.1111/all.15295
tive effects of sublingual allergen-specific immunotherapy: a
retrospective, multicentric study. Int J Immunopathol Pharmacol.
2003;16(1):73-79.

Allergy - 2022 - Farraia - Allergen Immunotherapy For Asthma Prevention A Systematic Review and Meta Analysis of

Uploaded by

Copyright:

Available Formats

Allergy - 2022 - Farraia - Allergen Immunotherapy For Asthma Prevention A Systematic Review and Meta Analysis of

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Allergy - 2022 - Farraia - Allergen Immunotherapy For Asthma Prevention A Systematic Review and Meta Analysis of

Uploaded by

Copyright:

Available Formats

13989995, 2022, 6, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/all.15295 by Cochrane Portugal, Wiley Online Library on [16/03/2023].

Allergen immunotherapy for asthma prevention: A systematic

Mariana Farraia1,2,3 | Inês Paciência1,2 | Francisca Castro Mendes1,2,3 |

1 | I NTRO D U C TI O N asthma* AND (prevent* OR risk). During the search, we selected

F I G U R E 1 PRISMA flow diagram of the

Participants, allergic disease(s),

Wahn et al, 2018 42 NRSI Mixed (adults and children over a

Therapy duration; Asthma, N of Risk of

Participants, allergic disease(s),

Therapy duration; Asthma, N of Risk of

Participants, allergic disease(s),

Therapy duration; Asthma, N of Risk of

F I G U R E 3 Publication bias assessed by

All studies except the

All studies except the

You might also like