EAACI guidelines on allergen immunotherapy: Hymenoptera venom allergy

2.1 Clarifying the scope and purpose of the guideline

The scope of this EAACI guideline is multifaceted, providing statements that assist clinicians in the optimal use of use of VIT in the management of patients with hymenoptera venom allergy and identifying gaps for further research.

2.2 Ensuring appropriate stakeholder involvement

Participants in the EAACI Taskforce on VIT represented a range of 16 European countries and disciplinary and clinical backgrounds, including allergists, pediatricians, primary care practitioners, ophthalmologists, ear, nose, and throat (ENT) specialists, pharmacists, immunologists, nurses, and patient representatives. Representatives of immunotherapy product manufactures were given the opportunity to review and comment on the draft guideline as part of the peer review and public comment process. These comments were considered by the taskforce, and, where appropriate, revisions were made.

2.3 Systematic reviews of the evidence

The initial full range of clinical questions that were considered important were rationalized through several rounds of iteration to agree on one key question: What is the effectiveness, cost-effectiveness, and safety of VIT in patients. This was then pursued through a formal systematic review and meta-analysis of the evidence.1 We continued to track evidence published after our systematic review and meta-analysis with a cutoff date of July 1, 2017, and, where relevant, studies were considered by the taskforce chairs. This evidence will formally be considered in the systematic review update that will precede the update of this guideline, which is scheduled for publication in 2022.

2.4 Formulating recommendations

We graded the strength and consistency of key findings from these systematic reviews1 to formulate evidence-based recommendations for clinical care by applying the GRADE process13 (Box 2). This involved formulating clear recommendations with the strength of evidence underpinning each recommendation. Where the systematic review did not cover the clinical area, we took a hierarchical approach reviewing other evidence until we could formulate a recommendation, that is: (i) other systematic reviews on the subject to see whether these provided any clarity on the topic; (ii) randomized controlled trials (RCTs) within these systematic reviews; (iii) other RCTs known to Taskforce members; and (iv) a consensus-based approach using an expert panel. Recommendations apply to all ages unless otherwise indicated in the tables. Experts identified the resource implications of implementing the recommendations, barriers, and facilitators to the implementation of each recommendation, advice on approaches to implementing the recommendations and suggested audit criteria that can help with assessing organizational compliance with each recommendation.

2.5 Peer review and public comment

A draft of this guideline was externally peer-reviewed by invited experts from a range of organizations, countries, and professional backgrounds. Additionally, the draft guideline was made available on the EAACI Web site for a 3-week period in May 2017 to allow a broader array of stakeholders to comment. All feedback was considered by the taskforce and, where appropriate, final revisions were made in light of the feedback received. We will be pleased to continue to receive feedback on this guideline, which should be addressed to the corresponding author.

2.6 Identification of evidence gaps

The process of developing this guideline has identified a number of evidence gaps which are prioritized.

2.7 Editorial independence and managing conflict of interests

The production of this guideline was funded and supported by EAACI. The funder did not have any influence on the guideline production process, on its contents or on the decision to publish. Taskforce members’ conflict of interests were declared at the start of the process and taken into account by the taskforce chairs as recommendations were formulated. Final decisions about the strength of evidence for recommendations were checked by the methodologists who had no conflict of interests in this area.

2.8 Updating the guideline

EAACI plans to update this guideline in 2022 unless there are important advances before then.

2.9 General considerations before initiating venom immunotherapy

2.10 Absolute and relative contraindications and VIT in patients with special conditions

An European position paper on clinical contraindications has been published in 2015 tackling all relevant contraindications in detail.20 In a recently published survey among 520 mainly European allergists, up to 47% had experience with administration of AIT in patients with risk conditions such as cardiovascular disease, taking ACEI or beta-blockers, malignant disease in remission, and autoimmune disease which previously had been considered as contraindications.21 Problems were uncommon and mostly minor so we have reconsidered contraindications in VIT. Below contraindications are briefly described, and recommendations are given in Table 2.

2.11 Quality of life

For most patients, and their families, any allergic reaction (regardless of severity) is a frightening experience. Given the effort required to avoid accidental exposures and the inherent uncertainty of success, living with HVA negatively influences QoL. This is particularly due to emotional distress of being alert during activities of daily living.49 VIT improves QoL in vespid venom-allergic patients even when they do not experience a resting.50 In a study where patients were offered a sting challenge after VIT, 80% of patients reported a significantly increased QoL after tolerating a sting challenge.51 In contrast, therapy with the AAI alone was shown to negatively impact on health-related QoL,50, 52 a significantly increased burden for patients53 and a higher level of anxiety and depression.54 In contrast, more than 90% of patients perceived VIT as (extremely) positive,53 with health and allergy-related QoL improving significantly during treatment,50, 52, 55 dysfunctional beliefs decreasing,55 and anxiety and depression levels, were the lowest among VIT-treated subjects.54 In a randomized study evaluating dermal reactors, QoL was also impaired in these systemic reactors and VIT was also able to improve their QoL in contrast to the AAIs.52

2.12 Venom immunotherapy: evidence-based clinical recommendations

2.13 Preventive pretreatment

In several double-blind, placebo-controlled trials, it has been shown that pretreatment with H₁ antihistamines improves the tolerability of VIT.65-68 In detail, it was reported that levocetirizine decreased the rate of SSR68 and fexofenadine decreased the rate of LLR and cutaneous SSR67 (Table 3). Importantly, effectiveness of VIT was not negatively influenced.68, 69 Antihistamines were usually administered 1-2 h before the injections or sometimes twice daily. In case of repeated adverse events during updosing, pretreatment with omalizumab may be recommended.70-72

2.14 Treatment protocols

VIT is performed by subcutaneous injections. VIT consists of an updosing phase and a maintenance phase, which is necessary to ensure a sustained effect of VIT. Conventional protocols, where the maintenance dose is reached in several weeks to months, can be administered in outpatient clinics.73 In an effort to reach the maintenance dose faster, rush73-77 and ultra-rush protocols78-81 with several injections per day on consecutive days are performed in hospitals. Maintenance dose is reached either within a few hours or within a few days, respectively. Cluster protocols, with several injections per day usually 1-2 weeks apart, are also a quick alternative to conventional protocols.82, 83 Importantly, the risk of adverse events is not associated with the severity of initial reactions,24, 25, 84 high venom-specific IgE levels, or skin test reactivity at low venom concentrations.84, 85 Conventional regimes appear to be best tolerated, while rush and ultra-rush protocols are more frequently associated with adverse events.24

2.15 Updosing

The recommended starting dose in updosing protocols lies between 0.001 and 0.1 μg, but it has also been shown that a starting dose of 1 μg is usually safe and not associated with a higher rate of side-effects in adults or in children.86 A maximum dose of 100 μg venom allergen dose usually offers adequate protection against systemic-allergic sting reactions in the majority of venom-allergic individuals.87-89

2.16 Maintenance dosing

A maintenance dose of 100 μg venom is significantly more effective than 50 μg.88 This dose is equivalent to the dry weight of approximately two honeybee stings or five wasp stings90 and has been adhered to as the recommended maintenance dose since the first controlled trial.87 A further increased dose gives a better protection when needed.91 A dose of 200 μg is recommended in patients who develop systemic-allergic reactions following a field sting or sting challenge while on 100 μg maintenance VIT.91 An increased maintenance dose should also be considered in allergic populations at high risk of multiple stings, such as beekeepers92 and in exceptional cases where patients have accumulated risk factors for treatment failure.

Although the European Medicines Agency (EMA) had no safety concerns regarding aluminum toxicity from their pharmacovigilance review of aluminum hydroxide in standard AIT, high dose VIT and lifelong therapy have not been specifically evaluated. As a precaution, where lifelong therapy is planned is can be undertaken with aqueous preparations. If a 200 μg dose is required for maintenance, half can be given as an aqueous preparation.

The interval for maintenance VIT with 100μg venom recommended by the manufacturers has been 4-6 weeks for aqueous preparations and 6-8 weeks for purified aluminum hydroxide adsorbed preparations (depot preparations). According to expert consensus, injections are usually given every four weeks in the first year of treatment, every six weeks in the second year, and in case of a five-year treatment every eight weeks from year 3-5.93 Extending the maintenance interval to three months does not seem to reduce effectiveness or increase adverse events,94-96 which could be relevant in terms of convenience and economic savings if lifelong treatment is necessary. As there is no specific study available for mastocytosis patients with severe initial SSR, caution should be used in extending the intervals to three months in those patients. A dose interval of six months did not provide suitable protection in honeybee venom-allergic patients97 and is therefore not recommended for standard practice (Table 3).

2.17 Duration of VIT

Termination after approximately one or two years leads to a relapse rate of 22%-27%.41, 42 Some studies have concluded that VIT for three years may be sufficient,98 particularly in patients with only mild-to-moderate initial sting reactions.98 Nevertheless, most of the studies concluded that a minimum of a five-year treatment is superior for long-term effectiveness.99-102 Lifelong therapy should be considered in patients with severe initial SSR, systemic adverse events during VIT, and honeybee venom-allergic patients with high risk of future honeybee stings (Tables 3 and 4).

2.18 Adherence

Adherence to VIT is high, possibly because of patients’ perception of an unpredictable risk of life-threatening sting reactions. In a recent study, 95% and 84% of patients still continued VIT after three and five years, respectively.103

2.19 Effectiveness

Treatment with ant venom is very effective as 97 to 98% are protected after VIT.9, 10 The effectiveness of honeybee and vespid VIT is different and ranges from 77 to 84% for honeybee venom compared to 91 to 96% for vespid venom.7, 8 The underlying reasons are still unclear. It has been speculated that the amount of venom delivered by a honeybee sting is much larger and more consistent.90 This may also explain the difference in the reaction rate to sting challenges, which has also been observed in untreated patients.104-106 It also appears that the broad sensitization pattern in honeybee venom-allergic patients may play a role in the lower effectiveness of honeybee VIT.107 For example, some patients are predominantly sensitized to Api m 10, which may be underrepresented in certain available honeybee venom preparations.108, 109 However, none of these studies included a patient analysis of molecular sIgE binding patterns to honeybee venom allergens before the start of VIT. Without such a specific IgE stratification aligned with the clinical outcome, the conclusions are of limited value. The specific preparation does not seem to have an impact on the effectiveness. The effectiveness of aqueous and purified aluminum hydroxide adsorbed preparations has been shown to be similar.60, 110

2.20 Effectiveness of VIT after updosing phase

Only one recent study has looked at how rapidly protection occurs. In honeybee VIT, 89% tolerated the sting challenge one week after reaching the maintenance dose in a 3- to 5-day rush protocol or a 3- to 4-month conventional protocol. Those patients who were not protected with 100 μg venom tolerated the sting challenge immediately after reaching the dose of 200μg.89

2.21 Effectiveness during/after maintenance VIT

Most effectiveness data are obtained during VIT. Resting reaction rates of 0%-10% 1-5 years after discontinuation of vespid VIT have been reported.100, 101, 111 Relapses after honeybee VIT are more frequent as 17% are reported to relapse one year after stopping VIT.112 There are only few reports on the outcome following VIT withdrawal for more than five years, and there are no data for more than 10 years after discontinuing VIT. In two studies, 7%-7.5% of patients treated with vespid venom relapsed after 7 to 10 years,98, 99 while 15.8% after stopping honeybee VIT had resting reactions.99 Another study compared relapse rates after four and approximately 10 years and reported relapse rates of 10.2% and 16.2%, respectively.113 In children, the long-term outcome is superior compared to adults as only 5% with moderate-to-severe reactions relapsed after up to 20 years after stopping VIT.15

2.22 Carriage of adrenaline autoinjectors during and after VIT

It is still a debated issue whether AAI should be carried during and after VIT, and it has also been difficult to reach a consensus on this topic. Most patients are protected after reaching the maintenance dose.89 Therefore, patients usually do not need to carry AAIs at this point, particularly if their sting reaction had been mild or they had tolerated a sting challenge or field sting during VIT. It should also be considered that carrying an AAI can negatively impact on health-related QoL50, 52 (Table 3). According to the EAACI position paper “Self-medication of anaphylactic reactions due to Hymenoptera stings”, 13% of experts/authors would still prescribe an AAI to patients who initially only had generalized skin symptoms after discontinuation of VIT, and 100% considered recommending carrying an AAI in patients who initially suffered from moderate-to-severe reactions after terminating VIT if risk factors for treatment failure were present.114

2.23 Risk factors for systemic adverse events with VIT and relapse of SSR

2.24 Risk factors for relapse of SSR

2.25 Procedures to monitor VIT

Many attempts have been made to identify biomarkers to monitor AIT. In peripheral venous blood samples of treated patients, there are significant changes of venom-specific T-cell populations, secreted cytokine patterns, and immunoglobulin levels but these are not appropriate to estimate the individual risk for relapse of SSR. The sting challenge remains the gold standard in identifying unprotected patients (Table 5).

2.26 Sting challenges/field stings

Performing sting challenges is still the most reliable method and gold standard to monitor the effectiveness of VIT. VIT is effective immediately after reaching the first maintenance dose.89 Therefore, if feasible, sting challenges should be performed as early as possible to identify those who are not protected with the maintenance dose of 100 μg. If sting challenges cannot be performed, information about field stings may be helpful. However, the risk of misidentification of the stinging insect and the nonstandardized sting procedure reduce reliability.112

The reproducibility of sting challenges, at least for diagnostic purposes, is a debated issue. A study on 129 patients revealed that in 95% of patients, a diagnostic sting challenge provided a good prediction of tolerance for subsequent field stings.119 On the other hand, it has been shown that 21% of patients not treated with VIT, who initially tolerated a sting challenge, had systemic symptoms after a second challenge.120 The reliability of early sting challenges to monitor effectiveness of VIT appears to be high,121 although repeated sting challenges during three to five years after treatment identified 8%-10% of patients who relapsed.101, 117 Importantly, tolerated sting challenges can improve health-related QoL, especially in patients reporting high impairment of health-related QoL before the sting challenge.51 Thus, sting challenges should not only be seen in the context of evaluating effectiveness but also in terms of fostering individual belief in disease-specific safety.

2.27 Specific IgE and IgG4 levels

It has been repeatedly shown that specific IgE levels to the respective venom decrease during VIT after an initial rise during the first months of treatment60, 121; they usually remain low even after stopping VIT.117 VIT is associated with a significant increase in specific IgG antibodies that has initially been suggested as a marker of effectiveness122; these immunological changes induced by VIT were also reported in honeybee venom-allergic children.123 The subclass of IgG antibodies is usually restricted to IgG1 and IgG4.121 However, after stopping VIT, specific IgG starts to decrease99, 124, 125 and patients appear to be protected by a mechanism independent from venom-specific IgG.122 Taken together, available data do not support the use of specific IgE, specific IgG, or specific IgG subclasses or even ratios can be used as predictors for protection during and after VIT in the individual patient.

2.28 Intradermal testing

Similar to the decline of specific IgE levels during VIT, intradermal test end point concentrations usually decrease from before to after VIT.99, 101 No study has been able to identify a relevant difference in skin test reactivity between tolerant subjects and patients with relapses.99, 100, 112 Moreover, patients with negative intradermal tests have been reported to have significant relapse, a few with near-fatal reactions.102, 113

2.29 Basophil activation test (BAT)

Allergen-specific basophil response remains positive126 or even unchanged125 during VIT. However, basophil responses at submaximal allergen concentrations are markedly decreased after VIT in tolerant subjects and this decline seemed to be associated with the induction of tolerance.125, 127 Also the measurement of basophil threshold sensitivity to anti-FcεRI stimulation has been proposed to monitor an early protective effect of VIT.128 BAT inhibition with sera of treated subjects correlated well with effectiveness of AIT in grass pollen-allergic patients129 but this has not yet been shown in patients with HVA.

2.30 Enzyme-linked immunosorbent facilitated antigen binding (ELIFAB)

The ELIFAB is a cell-free assay which is used to demonstrate inhibition of allergen-specific IgE binding by blocking antibodies.130 One study measured the serum inhibitory activity of VIT-treated vespid venom patients.124 During VIT, patients displayed an increased ability to inhibit Ves v 5 binding by IgE antibodies. This allergen-blocking capacity correlated with serum concentrations of Ves v 5-specific IgG4. However, both the inhibitory activity and specific IgG4 levels were again reduced in patients who stopped VIT several years ago.124

Despite of the availability of new methods such as the BAT and the ELIFAB, most of the parameters cannot precisely distinguish between patients who are protected from future SSR and those who are at risk. Currently, it is not possible to estimate the individual risk for relapse of SSR with any of the currently available parameters (Table 5).