Introduction

Polysensitization can be defined, in line with different published descriptions, as sensitization, according to standardized skin prick tests (SPTs) or specific immunoglobulin E (sIgE) assays, to two or more allergens from nonrelated sources.¹ Polysensitization is more prevalent than monosensitization as defined in several surveys using standard panels of allergens in the general population of Europe (12.8–25.3%) and the USA.^2,3 These data are transferred from the general population to the specialized practice, where a majority of patients (up to 80%) are polysensitized.^4,5 In Spain, polysensitization is also frequent. According to epidemiological data, 23.3% of patients are sensitized to two types of allergens and 10.1% to three or more.⁶ The mean ± standard deviation (SD) number of positive SPT per patient was 6.5 ± 2.⁷

Not all sensitizations induce allergic symptoms, while all polyallergic patients are polysensitized.^1,4,8 Pollen is the most common allergen related to respiratory diseases, followed by house dust mites, animal dander, and moulds.⁶ In patients polysensitized to pollens, it is difficult to distinguish whether the symptoms are due to pollen or other allergen sources. Moreover, it is difficult to establish which pollen source is responsible for the respiratory symptoms. The situation is more complicated when the patient is sensitized to pollens with similar pollination periods.^8,9

The severity of the allergic disease is directly proportional to the number of sensitizations.^10,11 In addition to classical diagnostic methods, molecular diagnosis and component-resolved diagnosis (CRD) have become widespread in clinical practices.¹² This has led to improved accuracy of the diagnostic process and the specific AIT prescription.^12,13 The therapeutic approach to respiratory allergy is based on allergen avoidance when possible, treatment with symptomatic drugs (including antihistamines, inhaled, intranasal and systemic corticosteroids, bronchodilators, and leukotriene receptor antagonists), and allergen immunotherapy (AIT).^11,14,15 AIT represents the only currently available treatment targeting the underlying pathophysiology of respiratory allergy and has shown a disease-modifying effect.¹⁶ AIT is recommended for patients with allergic rhinitis (AR) who have moderate-to-severe symptoms and controlled bronchial asthma despite regular and/or avoidance strategies, and evidence of IgE sensitization to one or more clinically relevant allergens.¹⁶ AIT may also be considered in less severe AR where a patient wishes to take advantage of its long-term effect on rhinitis and the potential to prevent asthma with Grass pollen AIT.¹⁶ However, ways to modulate AIT as a function of the number or nature of the patients’ sensitizations are not clearly stated. In Europe, polysensitized patients are typically treated for one or several allergens that are considered clinically relevant.^11,17,18 In the CONDOR Delphi consensus carried out in Spain in recent years, it was concluded that the choice of allergens for AIT in polysensitized patients is an enormously difficult task.⁹ The conclusions of this work recommended that no more than three allergenic sources should be mixed in the same AIT vaccine.⁹ Furthermore, the results recommended that the AIT prescription should be based on solid scientific evidence and that AIT treatments should have their own safety and efficacy studies.⁹

The debate on whether to prescribe single or multiallergen formulations of AITs and the timing of AIT initiation are currently ongoing. Hence, there is a need to carry out and analyze new consensus works in polysensitized patients and, more specifically, in the case of pediatric patients.

The CAPP project was a nationwide, multicenter, two-round Delphi study to determine the expert opinion on the clinical management of polysensitized children in Spain in order to provide guidance on the AIT indication and prescription.

Materials and Methods

Study design

The study started as an initiative of the Spanish Society of Pediatric Allergy, Asthma, and Clinical Immunology (SEICAP).

The approval of the Institutional Review Board (IRB) or by the equivalent ethics committee(s) was not required as this Delphi study does not involve research with human subjects. No patient data were collected for this study, based on feedback and opinions from experts.

The present study followed a modified two-round Delphi methodology, which is an established approach in consensus-building studies.^19,20 The Delphi process is a widely accepted scientific method of systematic collection of information from a group of experts (termed as Delphi expert panel) on controversial or complex topics.²¹ Each panel expert provides opinions individually and anonymously without the biasing effect of dominant individuals or group pressure.^22–24 The Delphi process ends when an agreement has been reached on the discussed topics.

Delphi process

Selection of Delphi participants

The scientific expert committee comprised 10 specialists experienced in managing polysensitized patients and AIT and recognized experts in the field, who are also members of the SEICAP AIT board. The scientific expert committee was responsible for reviewing the existing literature on the subject, drawing up the list of dimensions and items to be evaluated, and establishing the criteria to be met by the panelists who participated in rounds 1 and 2 in the development of this Delphi project.

A total of 46 specialists from 40 hospitals across Spain were invited to participate in the project as Delphi panel experts (panelists) in both rounds of the Delphi process. The expert panel consisted of allergists and pediatric allergists with at least 10 years of full-time dedication to allergology and extensive experience in prescribing AIT. Participants were selected, with representation from all Spanish regions, from members of SEICAP and the Spanish Society of Allergology and Clinical Immunology (SEAIC), the two most influential allergology societies in Spain. Most panelists worked in the Public Health Service. They had participated in clinical trials and studies, with publications and/or communications to conferences related to AIT. This selection process aimed to ensure a diverse group of experts with varied backgrounds and perspectives while maintaining a high level of expertise in managing polysensitized pediatric patients and AIT.

The expert panel members were provided with an informative leaflet outlining the aims and the study procedure, including an electronic link to the online survey. The purpose of the expert panel was to reach a consensus based on the current clinical evidence and their daily practice in and knowledge of the AIT management of polysensitized children.

Selection of Delphi questionnaire dimensions and items

The scientific expert committee carried out a systematic literature review on the main topics, focusing on current controversial and unresolved topics. After a careful and critical review of the selected literature and based on their knowledge of the clinical management of the pathology, the scientific expert committee developed the first set of domains and items for the Delphi questionnaire in a meeting.

Round 1

The members of the Delphi expert panel were asked during March 2022 to rate their level of agreement with each questionnaire item on a 9-point Likert scale from 1 (completely disagree) to 9 (completely agree). Each item was categorized according to the scores as rejected (scores 1–3), undetermined (scores 4–6), or accepted (scores 7–9). Panelists were also encouraged to provide comments after scoring each item using open-text comment fields included in the online survey.

After analyzing the data from the first Delphi round, the scientific committee experts participated in a meeting where the Delphi survey results were presented and discussed. Item selection was based on the acceptance of questionnaire items by ≥66.7% of the expert panel and the agreement of the scientific committee. Statements not achieving 66.7% agreement were removed or modified according to the feedback provided by the expert panel. The updated questionnaire was redistributed to the panelists for round 2.

Round 2

In round 2, the same panel members were asked between July and September 2022 to evaluate the list of items that did not meet the consensus from round 1. For this evaluation, the panel members were provided with a summary of the opinions issued anonymously by the participants in the first round, in addition to any other information that the scientific committee deemed appropriate to make available to the panelists to achieve consensus, without conditioning their decision.

After analysis of the responses described in round 1, the statements that did not meet the expert agreement were retained for discussion.

Concluding round

The concluding round comprised a teleconference meeting among the scientific committee experts to assess the nonconsensus items in round 2 until an agreement was reached to retain or eliminate them from the final consensus guidelines.

Statistical analysis

A descriptive statistical analysis of the data obtained from the assessment of the Delphi questionnaire items in rounds 1 and 2 was conducted. The distribution of frequencies of panel responses on the 9-point scale was calculated to establish the level of consensus for each questionnaire item.

A descriptive statistical analysis of the characteristics of the Delphi expert panel was also performed, including calculation of central tendency and dispersion (mean ± standard deviation, median and interquartile range) for quantitative variables and frequencies and valid percentages for qualitative variables.

The online survey was designed to prevent missing data and inconsistencies. Only complete responses were included in the analysis; partial responses were automatically excluded from the final dataset. This approach ensured data integrity and consistency across all participants’ responses.

The statistical analysis was performed using the Statistical Package for the Social Sciences (SPSS) version 18.0 (SPSS Inc., Chicago, IL, USA).

Results

Panel experts

All 46 panel experts participated in round 1, whereas 44 completed the round 2 survey. The characteristics of the Delphi experts are summarized in Table 1. Briefly, the median (range) professional experience in pediatric allergology was 20.5 (14.0–28.5) years. Most participants were specialists in pediatrics (n = 21; 45.65%), followed by allergology, and double speciality in pediatrics and allergology (n = 11; 23.9%).

Overview of the Delphi study

Figure 1 illustrates the results of the Delphi study. Due to the scope of the project and the volume of results, the scientific expert committee decided to split the work into two publications. The first part (present article) includes the results related to diagnosis, therapeutic management, and allergy to pollens (dimensions 1 to 3); while the second part includes the results related to allergy to mites, moulds, animals, Hymenoptera venom, and allergen mixtures (dimensions 4–8).

Figure 1 Summary of results of the Delphi study.

Results from dimensions 1–3

Tables S1–S3 summarize the results from the Delphi process and the level of agreement after the 2 rounds for the statements related to the diagnosis (Table S1), therapeutic management with AIT (Table S2), and pollens (Table S3).

Dimension 1: Diagnosis

Consensus was reached for 24 items (88.9%) (Figure 2). A total of 23 and 1 statements reached an agreement in rounds 1 and 2, respectively. Three statements did not achieve consensus in round 2.

Figure 2 Consensus achievement and key recommendations emerging from the consensus in Dimensions 1 (Diagnosis) and 2 (Therapeutic management).

AIT, allergen immunotherapy; IgE, immunoglobulin E; sIgG4, specific immunoglobulin G4; sIgE, specific immunoglobulin E.

With an agreement of 79.5%, polysensitization is defined as “Sensitisation to two or more allergens of different homologous groups.” As for the diagnosis of sensitized children, the agreements vary between 73.9 and 100% in which SPT and total IgE, specific to the whole source and its components, are fundamental. No agreement was achieved (27.3%) on the size of the prick papule and its relationship with the symptomatology. With 91.3–100% agreement, the importance of molecular diagnosis and the distinction between co-sensitisation and cross-reactivity is emphasized.

Dimension 2: Therapeutic management

Consensus was reached for 21 items (63.6%) (Figure 2). A total of 19 and 1 statements reached an agreement in rounds 1 and 2, respectively. Fourteen statements did not achieve consensus in round 2, of which the scientific experts committee decided to accept two statements.

The refusal to set a minimum age for AIT prescription (items 1 and 2) stands out among different agreed items. Regarding the preference for prescribing SCIT mixtures of two or more allergens, rather than SLIT, the agreement achieved 78.3%.

Regarding the evaluation of the effectiveness of the treatment, the agreement was broad among the participants on the use of clinical practice tools (including symptom scales, medication savings, and quality of life), but not on the use of in vitro methods (total IgE, sIgE, and sIgG4 for whole allergen sources, and components and/or ratios among them), or the use of nasal provocation test.

Importantly, no agreement was achieved on whether to discontinue, continue for another year, or change the route of administration in the case of no clinical improvement in the first year of treatment with AIT.

Dimension 3: Pollens

The consensus was reached for 27 items (77.1%) (Figure 3). A total of 24 and 1 statements reached agreements in rounds 1 and 2, respectively. Ten statements did not achieve consensus in round 2, of which the scientific experts committee decided to accept two statements.

Figure 3 Consensus achievement and key recommendations emerging from the consensus in Dimension 3 (Pollens).

AIT, allergen immunotherapy; IgE, immunoglobulin E; SCIT, subcutaneous immunotherapy; sIgE, specific immunoglobulin E; SLIT, sublingual immunotherapy.

With 79.5% agreement, the participants considered that pollens are the most representative allergens in AR. In addition, the consideration that allergens such as Php p 1 are initiator molecules (molecular spreading) achieved an agreement of 84.8%. Likewise, it is accepted that pollen AIT is more effective in the treatment of bronchial asthma than AR. In the case of routes of administration, the experts believed that SCIT does not show more effectiveness than SLIT for the AR treatment.

Among other issues, the question of whether the levels of IgE to the whole allergen source and its components are positively related to the intensity of the symptomatology stands out with an agreement of 65.9%. Similarly, the statement related to the level of Phl p 1/Phl p5 pretreatment and the efficacy of Gramineae AIT achieved an agreement of 73.9%.

Regarding pollen mixtures, it was accepted that AIT with mixtures of more than three pollen sources is not as effective as a monocomponent AIT with a single pollen (63.6% agreement, agreed after scientific committee evaluation). However, it was agreed that mixtures, including Gramineae and Olea, Gramineae and Cupressaceae, Cupressaceae and Olive, are suitable, establishing a maximum mixture of three allergens. On the other hand, an agreement was not achieved about mixing pollen from Parietaria with any other allergen.

Discussion

Polysensitization is common in Spain, as in other surrounding countries.^1,4,6,11 This fact is even present in the management of children with allergic conditions. However, there is a lack of scientific evidence concerning the diagnosis and therapeutic management of polysensitized children.¹⁷ All these make decision-making difficult for the specialist. In this context, the Delphi method is important in medical science as it makes it possible to collect experts’ opinions, promote reflection and opinion sharing, reach a consensus, and ensure the reliability of the results.^20,25,26

Overall, in the present study, a total of 148 of 204 statements (72.5%) were agreed upon. However, the consensus reached was different depending on the area of interest evaluated. Regarding the data analyzed in the present article, consensus was reached in 24 (88.9%), 21 (63.6%), and 27 (77.1%) of the statements for the areas of diagnosis, AIT management, and pollen allergy, respectively. To simplify the interpretation of the results, a set of recommendations was developed with the most relevant items that reached expert consensus in the Delphi process (Table 2).

Diagnosis

One of the first important aspects when analyzing the results of the group of statements related to diagnosis in polysensitized children is linked to the definition of polysensitization. The two definitions that reach consensus (although <80%) are those that introduce the concept of sensitization to ≥2 allergens without cross-reactivity and of different homologous groups. In this sense, European Medicines Agency (EMA) guidelines on allergen products indicate that allergen extracts may be grouped into homologous groups according to several criteria, including comparable physicochemical and biological properties of the source material; cross-reactivity or structural homology; and identical formulation and production process of the finished product.²⁷

Since diagnosis is the first step in determining appropriate and personalized management for each patient, it is important to establish a definition of polysensitization, especially from a clinical point of view. Recommendations from the European Academy of Allergy and Clinical Immunology (EAACI) and Allergic Rhinitis and its Impact on Asthma (ARIA) clearly state that the number of sensitisations is less important than the clinical relevance of each allergen.^28–30 For that reason, the identification of the clinically relevant allergen should be the basis of the diagnosis, in terms of type and severity of symptoms, duration of these symptoms over the year, the impact on quality of life (QoL) and the possibility of avoiding the allergen.^28–30

Furthermore, emphasis is placed on the fact that sensitization must be confirmed by SPT and/or IgE determination. In addition, the results reflect a broad consensus on the limitations of SPT itself, especially in terms of the AIT treatment decision. A high percentage of experts (97.8%) recognize the impossibility of distinguishing between cosensitization and cross-reactivity, except for single component prick tests (Profilin, Alt a 1 and Cup a 1). In addition, experts also agreed that, once the usefulness of SPT and total and sIgE testing has been established, it would be advisable to perform CRD (defined as the detection in the blood serum of sIgE against single allergen components or molecules from a specific allergen source) to accurately determine the relevant allergens. The experts also agreed on the potential of molecular diagnostics, even considering the current limitations in relation to the number of allergens available, mainly to establish the most accurate indication for AIT. In particular, molecular diagnosis-related items reached agreement levels from 95.7 to 100%, indicating the essential use of these tools to distinguish between cosensitization and cross-reactivity, the specific allergens of each source, and to clarify the composition of the AIT. Several studies in real clinical practice have shown that the use of CRD molecular diagnostics was an essential tool for the management of AIT and the follow-up of the disease.^8,31 In addition, current EAACI guidelines establish to use the adequate diagnosis algorithms, including molecular diagnosis.³²

Regarding the nasal challenge test, experts show a large degree of agreement regarding its use and possible limitations, especially concerning the methodology. These unmet needs are mainly allergen dose and quality, allergen application technique, the need for titration process, or the methodology to assess objective outcomes.^33,34 However, a position paper of the EAACI established nasal allergen challenge as an important tool.³⁵ In addition, a recent work indicated that the nasal allergen challenge tests are safe and highly reproducible diagnostic tests in clinical practice, both in children and adults.³⁶

Therapeutic management

In the present study, a broad consensus was reached that AIT prescription should only be made once the relevant allergens have been identified and the factors to be considered for initiating treatment (including clinical factors, the widely described recommendations previous to AIT treatment, the use of extracts with proven efficacy, and the assessment of possible patient adherence, among others). EAACI guidelines established that if single AIT extracts cannot cover clinically relevant allergens, consider a maximum of three allergens.³² The experts also agreed that the evaluation of the AIT effectiveness should include mainly clinical assessments, quality of life evaluations, and treatment adherence. Long-term effects of AIT include reductions in symptoms and rescue medication use, as well as improvement in QoL.³⁷ AIT benefits for asthma control are reflected in an improvement in airway hyperreactivity, the reduction of regular pharmacologic treatment, and a steroid-sparing effect.^38,39 The time after inhaled corticosteroid withdrawal to first exacerbation was recently considered for the EAACI as the primary outcome measure of asthma endpoints in AIT trials.⁴⁰ However, no consensus was achieved on the use of in vitro methods (total IgE, sIgE, sIgG4, and ratios among them) for the evaluation of the efficacy of AIT treatments. This is probably due to the fact that biomarkers of efficacy are not yet developed for clinical practice. Follow-up should be performed 6 months after treatment initiation, and a switch to another type of AIT should be advisable in patients without clinical improvement after 1 year of treatment. Switching between different types of AIT may be necessary if the initial treatment is not effective or well-tolerated. The decision to switch AIT should be made individually considering the patient’s clinical history and symptoms.⁴¹

Several studies showed that for a specific allergen extract, the benefit received for polysensitized patients should be similar to monosensitized patients treated with a single-allergen extract.^42–45 In this respect, in the present consensus, the statement “AIT administered subcutaneously is preferable to sublingual administration in polysensitized patients where the use of AIT with mixtures of more than two allergen extracts is considered” reached an agreement of 78.3%. In this sense, this perception, evidenced by the degree of agreement reached, would have no scientific evidence to support it and seems to be rather based on clinical experience and routine clinical practice.^18,46,47

AIT efficacy and safety are dose- and time-dependent.^48,49 According to several studies, the treatment effect plateaus after 16–26 weeks of treatment.⁵⁰ Switching between different types of AIT may be necessary if the initial treatment is not effective or well-tolerated. The decision to switch AIT should be made on a case-by-case basis considering the patient’s clinical history and symptoms.⁴¹

Pollens

The results in the block of statements related to pollen polysensitization and therapeutic approach with pollen extract AIT indicated a high degree of agreement on the use of the diagnostic tools discussed above (medical history, SPT). In general, there is a high degree of agreement in considering molecular diagnostics analysis to make an appropriate prescription of AIT and to identify possible good effectiveness or even the possible risk of adverse reactions. In this sense, panelists agreed that, in the case of Gramineae pollens, sIgE to rPhl p 1, rPhl p 5b, rPhl p 7, and rPhl p 12 should be identified to assess the AIT indication, and sensitization to Phl p 12 indicates a worse AIT response. According to the experts’ opinion, when assessing Olive pollen allergy, Ole e 7, and Ole e 9 determinations are related to a higher risk of more severe bronchial asthma symptoms and worst AIT safety.

Previous consensus studies have already indicated a high degree of agreement in considering Phl p 1, Phl p 5, and Ole e 1 sIgE measurements as phenotypic markers of Grass and Olive pollen sensitization and their usefulness for the composition of AIT prescription.⁹ Likewise, other allergens such as Ole e 7 and Ole e 9 would also be of importance in regions with high levels of Olive pollen exposure.⁹

The role of Phl p 12 (profilin), as one of the main components related to cross-reactivity, makes the determination of sIgE levels to this allergen a particularly important tool.^51,52 However, sensitization to this allergen may be influenced by specific patient characteristics (men and children seem to have higher values) and by geographical location.^53–55 In line with what the experts of the present study agree, different works showed that the detection of IgE to Phl p 12 could predict oral allergy syndrome and an increased risk of adverse reactions during AIT with Grass extract.^56,57

Regarding the statement related to the level of sIgE to Phl p 1 and/or 5 and the correlation with the effectiveness of AIT to Grass pollen, this item reached an agreement of 73.9%. The literature does not appear to be conclusive in this regard, with some published work claiming that the sIgE/total IgE ratio could be used for predicting the clinical response for specific allergens.^58,59 On the contrary, some recent work suggests that the relationship between pretreatment allergen-sIgE concentration, or demographic characteristics, and the effect of AIT is unclear.⁶⁰

Experts agreed that AIT with pollen extracts is more effective for the treatment of bronchial asthma than rhinoconjunctivitis. No agreement was reached, however, on possible differences in the use of SCIT or SLIT in patients polysensitized to pollens of different origins. The efficacy of SCIT in allergic rhinoconjunctivitis induced by pollen allergy presents strong evidence for adult patients.⁶¹ However, only a few studies showed it for children and adolescents.⁴¹ Treatment with SCIT indicated the control of bronchial asthma, as reported by the Global Initiative for Asthma (GINA) if a clear causal link between respiratory symptoms and the relevant allergen is demonstrated.^39,41,62 However, more evidence would be needed for SCIT. In the case of SLIT, efficacy in Grass pollen-induced allergic rhinoconjunctivitis was demonstrated to a greater extent than in tree pollen allergy.⁶³ In contrast, only a few studies examined the efficacy of SLIT with pollen extracts in allergic asthma.⁴¹

Regarding the use of pollen mixtures in AIT, it is important to note the degree of agreement reached on the types of mixtures recommended. The utilization of Grass, Olive and Cupressus (three pollens) in the same preparation was accepted by 95.7% of panelists. Experts agreed that AIT with a single Grass species (e.g., Phleum pratense) is effective for the treatment of sensitisation to the whole group (87%). Several works reported the use of primarily single-allergen AIT in polysensitized patients and showed that symptoms and medication scores decreased and the quality of life improved.^43,64,65 The statements indicating that a mixture of Grasses and Cupressaceae is acceptable in polysensitized patients and can prevent AR progression to asthma achieved an agreement of 93.5% and 80.4%, respectively. Finally, after identification of the relevant allergens, experts agreed that AIT could be indicated by including a maximum of three allergens in the AIT composition (87%). The EAACI guidelines recommend limiting allergen formulations to a maximum of two allergens for pollens when possible.³² However, some working groups propose using up to three allergens, despite the lack of clinical evidence supporting this practice.³² The Delphi study conducted in Spain exploring consensus on treatment strategies for polysensitized patients reached a high agreement for the two primary pollen sources in the country (Grasses and Olive).⁹ However, no consensus was achieved for patients sensitized to more than two allergens, highlighting the ongoing debate and need for further research in managing polysensitized patients.

Limitations and strengths

The Delphi method is a valuable tool for gathering insights and making informed decisions in healthcare practice. However, one of its limitations is that the anonymity of the Delphi method can help mitigate biases, but it is still possible for participants to exhibit response bias. In addition, it should be considered that the Delphi methodology itself means that the fact that experts express a high degree of agreement does not directly imply that a recommendation is necessarily effective. The results of the study represent the starting point to design appropriate studies related to the field, and for the development of recommendation documents and management guidelines.

The variability in consensus observed in this Delphi study can be attributed to several factors specific to clinical practice and healthcare in Spain. Most panellists work within the Spanish Public Health System, where diagnosis and treatment are publicly funded, influencing appointment scheduling and diagnostic approaches for polysensitized patients. Molecular diagnostics are widely used, but their availability and experience vary depending on the allergens involved.^8,13,66 Spanish clinical practice predominantly favours SCIT over SLIT, and chemically modified allergen extracts (allergoids) are frequently used in AIT, though their standardization and quality control are more complex than native extracts.^67,68 Additionally, the regulatory process for AIT in Spain is still evolving, suggesting that future consensus analyses may be needed as the regulatory framework becomes more established.

In summary, the results of this Delphi study suggested that, although the general lines of recommendations and suggestions are in line with the management recommendations established by, among others, the EAACI, an adaptation to the specific characteristics of the polysensitized children and the Spanish population is necessary.

Conclusion

The expert consensus recommendations derived from this Delphi panel study may provide support and guidance on clinical decision-making regarding the management of polysensitized children in real-world clinical practice. Additionally, this consensus analysis may encourage discussion on controversial issues addressed in the consensus statements.