A 13-year real-life study on efficacy, safety and biological effects of Vespula venom immunotherapy

Insect sting allergy is responsible for severe and, sometimes, life-threatening reactions. Venom immunotherapy (VIT) was proven to be effective and safe in patients with venom allergy-induced anaphylaxis [1, 2]. The clinical efficacy of VIT is commonly defined by a reduction of the severity of the allergic reactions after Hymenoptera stings. In clinical practice, the reduction of both skin reactivity to insect venom and specific IgE levels in serum helps corroborate the assessment of the IT clinical efficacy [3]. Indeed, taken together, these two parameters define the global levels of allergen-specific IgE, as skin reactivity is quantitatively proportional to mast cell bound IgE. Of note, the vast majority of IgE are mast cell bound, whereas serum IgE reflect the minor pool of unbound/circulating IgE.

Only some reports exist on the long-term clinical efficacy of VIT and the kinetics of bound and unbound IgE after IT discontinuation (Additional file 1: Table S1) [4‐10].

In the present study, we investigated, retrospectively, the real-life long-term efficacy of Vespula (Vespula spp.) VIT and its effects of specific IgE and IgG₄. Thus, 23 patients (18 men and 5 women), with a history of severe allergic reaction to Vespula sting, underwent Vespula VIT for 5 years, followed by an 8-year follow-up. During the study period, we monitored: (i) allergic reactions to Vespula stings; (ii) performed rigorously standardized quantitative skin testing; (iii) evaluated Vespula-specific IgE levels.

Several mechanisms have been proposed to account for VIT-induced clinical efficacy, including drop of the allergen specific IgE levels and induction of protective IgG₄ antibodies [11‐14]. In fact, IgG₄ are considered protective antibodies for several reasons: (i) among all the IgG subtypes, they have a weak capacity to bind Fcγ Receptors and thereby a reduced ability to activate immune cells [15]; (ii) the Fc portion of IgG₄ molecule does not fix complement, due to the low affinity for complement factor C1q [16]; (iii) IgG₄ are functionally monovalent and unable to form immune complexes. Indeed, IgG₄ are dynamic molecules that exchange Fab arms by swapping heavy-light chain pairs with IgG₄ molecules of different specificities [17, 18]. This results in the production of bispecific antibodies with a substantially decreased capacity for antigen cross-linking [15, 17, 18].

Thus, we also investigated the kinetics of IgG₄ in the 23 patients during the 5-year VIT course and the follow-up.

It is widely accepted that Vespula VIT induces protection from stings in allergic patients during VIT and after discontinuation. In 2004, Golden and Colleagues monitored children with allergy to insect stings that had undergone VIT [8]. Importantly, they demonstrated that VIT in children leads to a significantly lower risk of systemic reactions to stings, up to 20 years after the treatment is stopped, suggesting that clinical efficacy is long lasting [8]. In 2008, Hafner et al. performed a long-term survey in adult patients who had previously discontinued VIT. The majority of the patients reported that the symptoms experienced with stinging events after VIT discontinuation were milder than symptoms before VIT, suggesting again long-term efficacy [9]. Moreover, Pravettoni and co-workers, analysing a series of 232 patients, found that 35.2% of the patients who could be monitored (159) reported at least one field sting up to 10 years after VIT discontinuation. None of them suffered from any systemic reactions [28].

However, few reports exists on VIT clinical efficacy long after discontinuation in adult patients (Additional file 1: Table S1) and most of them rely on data obtained through mail questionnaires sent to the patients after VIT discontinuation or telephone interviews.

Moreover, near-all these studies present some points of relative weakness, mostly because of considerable heterogeneity at different levels (Additional file 1: Table S1): (i) age of the patients, (ii) duration of the IT, (iii) type(s) of venom(s) administered during VIT (e.g. honeybee, Vespula) and (iv) VIT supplier used.

As for the age of patients, the outcome of the immune response is age-related [29]. Therefore, the final outcome of VIT and its long-term efficacy may vary accordingly. Moreover for duration of VIT, some of the studies on long-term efficacy included patients that underwent variable periods of VIT (from 3 to 5 years). Nonetheless, such difference might influence the outcome of the long-term immunological memory and therefore the extent of the long-term protection. Moreover, Hymenoptera venoms are, to some extent, cross-reactive and this could introduce biases in the assessment of VIT outcomes, when two therapies are given together. As an example, Vespula antigen 5 (Ves v 5), one of the major Vespula allergen, contains 204 amino acids residues and shares 60% of sequence identity with Polistes dominulus antigen 5 (Pol d 5) [30]. Therefore, patients undergoing VIT for these two different venoms could possibly develop a longer and more robust protection compared to patients receiving VIT for a single venom.

Finally, in near-all the reports present in the literature the VIT supplier is not declared. This latter point appears to be particularly relevant as Hymenoptera venoms comprise a complex mixture of different proteins that might all contribute to sensitization. Indeed, in vespid venom three main allergens have been described: Phospholipase A1, Hyaluronidase and Antigen 5 [31]. Importantly, the purification process of the venom extracts may vary between manufacturers [32, 33] and, therefore, the final quality and composition of the venom extracts might not be fully comparable. Thus, a different outcome in the final immune response might be obtained using different venom preparations.

In our work, we analysed a homogenous group of adult patients (mean age of 42.8 years ± 11.8) that underwent Vespula VIT only, for almost exactly 5 years (on average 61.2 months ± 4.0). After VIT discontinuation, we monitored all the patients over a prolonged follow-up (Table 1). Moreover, in our study, 11 patients were treated with DHS and 12 with ALK-Abellò ITs. It has to be noticed that in all the patients nearly the same induction/maintenance protocol was used (Additional file 2: Table S2 and Additional file 3: Table S3) and the final dose of venom received was fully comparable, regardless of the suppliers [20]. Interestingly, no differences were observed in terms of clinical protection in the patient treated with VIT of either of the two suppliers (data not shown). Moreover, both DHS and ALK-Abellò ITs induced comparable effects on circulating Vespula-specific IgE and IgG₄ (Additional file 4: Figure S1A, B). This latter observation indirectly confirms that the immunogenicity of the VIT used was also similar.

We demonstrated that Vespula VIT confers a robust protection not only during the VIT course, but also long after its discontinuation. Indeed, during the follow-up, 60% of the patients were re-stung, of whom 93% were minimally symptomatic or asymptomatic and none of the patients needed to resume Vespula VIT, which suggests that a 5-year IT course is possibly protective and can be recommended, at least in Vespula allergic patients. In order to evaluate clinical efficacy of Vespula VIT the patients in our study were interviewed for stinging events at each single administration during VIT and at two time (at least) points during the follow-up. In the case of a sting, they were asked to recognize the culprit insect at every interview, using an entomologic display case. It has to be noticed that Vespula appears to be correctly recognized in 72.3% of the cases from Vespula allergic patients [34]. This rather stringent approach represents a novelty.

Moreover, aside clinical efficacy, we also evaluated the biological effects of Vespula VIT. Indeed, as shown in Additional file 1: Table S1, few reports assessed the long-term biological effects of venom IT. Even though a drop of circulating specific IgE titres was sometimes observed in up to 70% of the patients undergoing VIT [35], only little information is present in the literature on the kinetics of venom-specific IgE long after treatment discontinuation.

As it is known, IgE molecules have a unique immunological behaviour. Indeed, at the steady state, the vast majority of IgE is bound on the surface of the mast cells via the interaction with FcεRI, a high affinity receptor, able to bind a single IgE molecule with K_d of 10⁻⁹ [23]. The remaining smaller pool of IgE circulates in plasma in unbound form. Therefore, in our study, we monitored both bound and unbound pools of Vespula-specific IgE.

On the one hand, changes in the bound pool of IgE were assessed with skin testing, performed in a strictly quantitative fashion. Data of this kind are particularly scarce in the context of long-term studies. Indeed, skin test results might be influenced by several factors such as circadian rhythm and technical expertise of the operator performing skin testing, thus demanding methodological rigour. Moreover, in order to render our data comparable with each other, we put emphasis on achieving normalization of the results. To this aim, we expressed this parameter by a Skin Index, previously defined as the ratio between the area of the wheal generated by the Vespula allergen and the area of the wheal generated by exogenous histamine [21]. Remarkably, we observed a robust reduction of skin reactivity (62.5% reduction after 5 years of VIT) after VIT that was maintained throughout the follow-up (Fig. 2a). To our knowledge, few works have so far analysed skin reactivity/variation in mast cell-bound specific IgE in a strictly quantitative fashion at different time points. In particular, before VIT, during the VIT course and long after VIT discontinuation.

On the other hand, the unbound pool of Vespula specific IgE was assessed by RAST. In line with other reports [36, 37], we observed a transient increase of circulating Vespula-specific IgE early after VIT initiation, peaking at 3 months (Fig. 2b). This time-point corresponds to the end of the IT induction phase, during which the allergen is administered weekly, at increasing doses. Thus, the increase in Vespula-specific IgE titres might be explained by the presence in allergic subjects of memory IgE B cells. These cells are able to differentiate into IgE-producing plasma cells upon allergen encounter [38, 39]. Interestingly, during this initial phase of VIT, the patients are already protected towards re-stinging events. Indeed, during the first year of VIT course, we observed 13 re-stinging events (in 8 patients) that were either asymptomatic or followed by a reaction of grade I of the Müller scale. This observation suggests that the clinical protection is already present early during the VIT course, despite the high Vespula-specific IgE levels. This initial increase was followed by a progressive and steady decline during the maintenance period. Surprisingly, we observed a persistent and further reduction of circulating IgE levels long after VIT discontinuation. This latter data suggest the induction of immunological changes, possibly inducing tolerance, resulting in the decline of specific IgE levels (Even though mast-cell bound Vespula-specific IgE levels seemed not to decline further during the follow-up) (Fig. 2a, b).

The cross-comparison between the clinical data, the quantitative skin test analysis and the data on circulating Vespula-specific IgE seems to suggest that the drop of Vespula-specific IgE levels might be mechanistically related to the long-term efficacy of VIT. As a consequence, based on our data, one can propose to use the venom-specific IgE levels (assessed after discontinuation) as a biomarker for long-term VIT clinical efficacy, at least in Vespula allergy.

However, the precise immunological mechanisms underlying VIT efficacy are probably more diverse and complex and still unclear [40‐43]. Particularly, VIT has been shown to induce a rise in specific IgG₄ levels. It is well known that a regular and persistent exposure to an antigen is able to induce IgG₄ antibodies [44]. As mentioned above, IgG₄ should be considered anti-inflammatory antibodies. Indeed, IgG₄ are dynamic molecule that behave as monovalent antibodies and therefore cannot form immune complexes [17]. Moreover IgG₄, due to their poor binding capacity to both complement and FcγRs, activate only weakly Fc-dependent immune mechanisms, such as antibody dependent cytotoxicity or the complement cascade [15, 16]. However, even though IgG₄ have been previously proposed as a contributing factor to the clinical efficacy of VIT [45, 46], no reports have analysed the changes in the IgG₄ levels long-after VIT discontinuation, so far (Additional file 1: Table S1).

Thus, we evaluated changes in the levels of Vespula-specific IgG₄ throughout VIT and follow-up in the 23 patients studied. Remarkably, we found that Vespula-specific IgG₄ rose and then reached a plateau during the VIT course but declined substantially during the follow-up, since the IgG₄ titres at 3 and 8 years follow-up time-points are comparable to the ones observed before VIT (Fig. 3a).

This latter result suggests that long-term protection induced by VIT could be IgG₄-independent. Interestingly, a previous report by Varga et al. [47] analysed the role of honeybee specific IgG₄ in a group of 10 children that had undergone honeybee VIT. In line with our results, the authors of this study after a 2-year follow-up found no correlation between honeybee specific IgG₄ levels and long-term efficacy of VIT.

In our study we evaluated, for the first time, Vespula-specific IgG₄ using two different technical approaches. In particular, the commercial kit for Vespula-specific IgG₄ was based on the use of Vespula allergen pre-coated strips. In this assay, after the incubation with the patient serum, the presence of IgG₄ is revealed using of anti-human IgG₄-antibody conjugated with horseradish peroxidase. Nonetheless, in this experimental setting, the total amount of IgG₄ revealed might be underestimated. Indeed, during the incubation, other Vespula-specific antibodies with different isotypes (e.g. IgG₂, IgE), present in the patient serum, will most likely compete with IgG₄ for the binding to the cognate antigen. In order to overcome this possible competition and to ascertain the results obtained with the commercial kit for Vespula-specific IgG₄, we developed an in house-technique that we applied to six patients and nine normal controls. This assay relies on the use of pre-coated anti-IgG₄ wells. After incubation with the patients’ sera, Vespula-specific IgG₄ are revealed using biotinylated Vespula allergen. Remarkably, the results obtained with the two distinct technique were comparable, thus validating our findings on IgG₄ (Fig. 3b).

It has to be noticed that IgG₄ titres at the beginning of the VIT were already higher compared to the 20 healthy controls, suggesting that in Hymenoptera venom allergic subjects an IgG-driven immune response already occurs independently from VIT. To our knowledge, this is the only study in which pre-VIT values of IgG₄ in Vespula allergic patients are compared to those found in a group of normal adult individuals. Finally, we analysed the trend of the ratio between the Vespula-specific IgE and IgG₄ over all the study period. Interestingly, this ratio progressively declines (from 70.1 to 27.2, at the end of the VIT course). This decline appears to be even more pronounced at the end of the follow-up (14.2) (Additional file 4: Figure S1C). This latter result seems to corroborate the hypothesis that indeed Vespula venom VIT induces long-lasting immunological changes.

In conclusion, our results obtained in a relatively small but well-controlled cohort of patients (out of a series of 686 patients) show that Vespula VIT induces a robust and long-lasting protection towards re-stinging events. The results obtained on Vespula-specific IgE levels show that IgE levels (bound and unbound) closely match the VIT clinical efficacy. Particularly, IgE levels in the follow-up, are perhaps the best predictor of clinical efficacy of Vespula VIT. Furthermore, IgG₄ levels were sustained during VIT course but declined substantially after the end of VIT and, therefore, are less useful in assessing clinical efficacy of VIT in the long term. Rather, specific IgG₄ could be used as a signature of valid immune stimulation induced by VIT (Fig. 4).