Introduction

Allergic rhinitis (AR) is a type-I hypersensitivity (allergic) disease resulting in biphasic clinical response. It is characterised by chronic inflammation of the lining of the nasal mucosa resulting from early-and late-phase responses upon exposure to prevailing common aeroallergens such as tree, grass, or weed pollens in the spring and summer months [1, 2]. Typical symptoms include sneezing, rhinorrhoea, nasal congestion, and nasal itching. AR is often associated with ocular (eye) symptoms of allergic conjunctivitis [3]. Recent classification of allergic rhinitis has taken into consideration the impact of symptoms on quality of life. The severity of symptoms is divided into mild or moderate/severe [4, 5]. A substantial increase in the prevalence of seasonal allergic rhinitis has been reported in industrialised countries, particularly in Western Europe and the United States, where it affects one-quarter of the population [68]. Moreover, it is a risk factor for developing allergic asthma in adults, with up to 30 % of patients reporting asthma symptoms [911]. AR impacts on a patient’s social life. It affects learning performance in school children [12] and work productivity in adults [13]. It has a major impact on quality of life and is a proven substantial socioeconomic burden [14].

The current management of allergic rhinitis consists of patient education, avoidance of allergen, pharmacotherapy, and allergen immunotherapy [15]. Antihistamines can be used for both intermittent and persistent disease, reducing symptoms such as nasal itchiness, sneezing, and rhinorrhea, and they have been shown to improve work and school performance in pollen-induced rhinitis [16]. Oral and topical decongestants have been shown to relieve nasal congestion effectively in the short term but are not effective for other symptoms such as sneezing and rhinorrhoea, and their long-term use is to be avoided due to the risk of developing rebound congestion and rhinitis medicamentosa. Chromones, which stabilise mast cells, are recommended for topical intraocular treatment. In placebo-controlled studies the leukotriene receptor antagonist montelukast has been shown to reduce symptoms of nasal congestion as well as rhinorrhea and sneezing in allergic rhinitis. However, the effects of leukotriene antagonists are modest, only effective in 5 % of patients compared to placebo [17]. In contrast, intranasal corticosteroids show much greater efficacy in the treatment of seasonal allergic rhinitis when compared to antihistamines [18] and leukotriene antagonists [19]. Treatment with anti-immunoglobulin E antibody has demonstrated efficacy in seasonal allergic rhinitis. This therapy consists of a recombinant humanised monoclonal anti-IgE antibody (omalizumab) which binds to the constant region (Fc) of the IgE molecule, thereby preventing binding of IgE to the high-affinity IgE receptor on mast cells and basophils, whilst avoiding the risk of IgE cross-linking leading to anaphylaxis. It has also been shown to reduce IgE levels in serum [20].

Allergen immunotherapy (AIT) is highly effective for select individuals with IgE-mediated diseases, such as allergic rhinitis, conjunctivitis, and venom hypersensitivity [21••, 22, 23••]. AIT was initially reported by Leonard Noon in the early 20th century when he described his initial observations on the effects of grass pollen-specific injection immunotherapy resulting in suppression of ocular symptoms caused by the sensitising grass pollen allergen. AIT is traditionally administered subcutaneously (SCIT) [24] and has proven clinical benefit in adults and children who are sensitised and allergic to aeroallergens such as pollens (grasses and trees), house dust mite (HDM), animal dander, and pollen-induced allergic rhinitis with/without asthma [24]. In recent years, sublingual immunotherapy (SLIT) has emerged as an alternative to SCIT and has been shown to be effective with a superior safety profile compared to SCIT.

Sublingual immunotherapy

Clinical efficacy

Sublingual immunotherapy (SLIT) has recently emerged as an effective and safe alternative route to deliver immunotherapy. Clinical efficacy of sublingual immunotherapy has been reported in Cochrane systematic reviews and meta-analyses [25•]. Optimal high doses of grass pollen tablet treatment have been established in two separate randomised double-blind placebo-controlled studies [26, 27]. A four- to six-month pre-seasonal followed by continued seasonal treatment with SLIT tablets resulted in increased clinical efficacy when compared to two months’ pre-seasonal treatment. A recent five-year double-blind placebo-controlled study of SLIT investigated prolonged efficacy and induction of tolerance. The study consisted of three years of treatment with SQ-standardised grass SLIT tablet or placebo and two years of blinded follow-up after discontinuation of treatment. Clinical endpoints were measured, including rhinoconjunctivitis symptom and medication scores, combined scores, asthma symptom and medication scores, quality of life, days with severe symptoms, as well as immunologic endpoints and safety parameters. The mean rhinoconjunctivitis daily symptom score was significantly reduced (25–36 %) in the SLIT-treated group when compared to the placebo group over five consecutive grass pollen seasons. The rhinoconjunctivitis daily medication score was reduced by 20 % to 45 % for seasons one through four. In all three seasons during treatment and two seasons during withdrawal of treatment, the percentage of days with severe symptoms during the peak grass pollen exposure was decreased from 49 % to 63 % in the SLIT-treated group when compared to placebo-treated group. A recent double-blind placebo-controlled study of three years’ treatment with grass allergen tablets followed by two years follow-up showed that the active treatment group exhibited a sustained and persistent clinical improvement for two years after discontinuation of treatment [28, 29••].

Immunologic mechanisms of SLIT

Langerhans cells that reside in the oral mucosa constitutively express FcεRI, major histocompatibility complex (MHC) class I, II, and co-stimulatory and co-inhibitory molecules [30]. These cells are highly efficient in antigen presentation to T cells. Cross-linking of FcεRI on monocytes induces production of IL-10 [31] and indoleamine 2,3-dioxygenase [32].Indoleamine 2,3-dioxygenase is associated with reduced tryptophan levels and consequently impaired T-cell stimulatory capacity. Allam and colleagues recently demonstrated that ligation of Toll-like receptor 4 (TLR4) on isolated oral Langerhans cells resulted in increased production of IL-10 [33]. Subsequent co-culture experiments revealed a decreased T-cell proliferative response and a parallel induction of T cells with a regulatory phenotype.

Clinical studies of SLIT are heterogeneous, and a wide range of laboratory techniques have been used to measure putative immunological mechanisms, which may at least partly explain the variability of results. Tracer studies of radioiodine-labeled allergen have shown that allergen is retained within the mouth for at least 2 hours and up to 18–20 hours following sublingual administration, affording opportunities for both local as well as systemic effects on the immune system [34]. Few studies have considered assessing the immune cell infiltration of the oral mucosa tissue before and after SLIT in humans. A study reported equal numbers of local CD1a+ dendritic cells (DCs) and CD3+ T cells before and after SLIT [35]. The cellular distribution of various immune cells during SLIT has been studied recently in more detail. Epithelial FoxP3+ regulatory T cells were elevated in SLIT patients when compared with the placebo group and were associated with reduced subepithelial CD1c+ myeloid DCs (mDCs) and mast cells [36]. The effects of SLIT on allergic effector cells have been demonstrated in patients treated with parietaria-specific SLIT. A significant reduction in the number of eosinophils, neutrophils, and ICAM-1 expression in the nasal mucosa was reported [37]. A reduction of eosinophil cationic protein [38, 39] and number of eosinophils has been observed in several but not all studies [40].

T cell responses during SLIT

Studies of peripheral T cell responses to inhalant allergens before and after SLIT have been highly variable. Decreased T cell proliferative responses in birch [41] and grass pollen-treated patients [42] have been observed in some but not all studies [43, 44]. Similarly, results for T cell cytokine production at both messenger RNA (mRNA) and protein levels have been highly variable, with some studies showing an increase in interferon gamma and/or decrease in Th2 cytokines [4547] and others showing no changes [43, 48, 49]. A more consistent finding (as found in SCIT) has been increases in IL-10 production in peripheral T cells, which have been observed at protein [41, 48, 49] and mRNA levels [47] in several but not all studies [43].

A well-designed mechanistic study by Bohle and colleagues on small numbers of birch-treated patients revealed a reduction in T cell proliferative responses to Bet v1 that was accompanied by increases in IL-10 [41]. This suppression was reversed by using neutralising anti-IL-10 antibody or depletion of CD25+ cells from the cultures, which implied involvement of regulatory T cells (Treg). In another study of HDM-SLIT, allergen-induced CD4+ T cell proliferation and IL-5 production were significantly decreased after active treatment in contrast to no change in the placebo–treated group [50]. The allergen-induced T cell proliferative responses were sTGF-βRII-dependent at six months. In this study, Foxp3+ Tregs were increased in the HDM-SLIT-treated but not placebo-treated group. Another study of grass pollen-SLIT also showed increased in Foxp3+ T regulatory cells in the nasal mucosa [36]. A recent study investigated the epigenetic modification of memory Treg cells during dual HDM and grass pollen SLIT. Methylated CpG sites within the Foxp3 locus of enriched memory CD45RO+Treg cells were enumerated before and 12 months following immunotherapy. DNA methylation was decreased in CD45RO+Treg cells at the FOXP3 locus in subjects after 12 months treatment with dual AIT, whereas no changes in FOXP3 locus DNA methylation were observed in CD45RO+ Treg cells in either placebo-treated allergic participants or healthy normal controls [51]. These observations support the theory that functional FoxP3+ Tregs are induced during SLIT.

Dendritic cell responses during SLIT

In vivo murine and in vitro human studies have shown that proinflammatory epithelial-derived mediators and cytokines, such as TSLP, IL-25, IL-33, prime DCs to polarise naïve T cell responses towards a pro-allergic Th2 phenotype [52, 53]. AIT may dampen these inflammatory epithelial responses, resulting in induction of tolerogenic DCs, which are able to polarise T cells towards an IL-10-producing Treg phenotype. The inducible IL-10+ Tregs, in turn, may suppress pro-inflammatory DCs and modulate Th2 responses [54, 55].

AIT has been shown to augment peripheral DC TLR9-mediated innate immune function [56]. A three- to fivefold increase in IFN-α production by plasmacytoid dendritic cells (pDCs) in response to in vitro CpG stimulation was demonstrated in subjects who received HDM-specific AIT. A recent study revealed that different subsets of human DCs enriched from peripheral blood could preferentially induce IL-10+ regulatory T cells and subsequently suppress in vitro allergen-driven Th2 responses [57••].

A novel inhibitory cytokine, Interleukin-27 (IL-27), which is produced by dendritic cells following in vitro TLR4 stimulation by LPS, has been shown to suppress T helper 2 responses in patients with seasonal allergic rhinitis [58••]. IL-27 belongs to the IL-12 family and consists of Epstein Bar inducible gene 3 (EBI-3) [59] and IL-12p28 [60]. In vitro IL-27 was found to suppress grass pollen allergen-induced PBMC proliferation in a dose-dependent manner. IL-27 upregulated mRNA T-bet and c-Maf expression and downregulated GATA-3. Additionally, IL-27 downregulated IL-4 and IL-5 and upregulated IL-10 and IFN-g mRNA expression. Moreover, T effector cell proliferation was suppressed when grass pollen-stimulated IL-27-primed DCs were cultured with T effector cells. The immunomodulatory effects of IL-27 during AIT remain to be further determined and validated in other SLIT studies.

Immunoglobulin responses during SLIT

During pollen SLIT, increases in allergen-specific IgE occur within weeks and do not appear to be associated with adverse reactions. These early increases are followed by blunting of seasonal rises in IgE. There follows an increase in allergen-specific IgG and IgG4. These elevations are both time- and allergen dose-dependent [27] and are progressive for at least two years [61] although of lower magnitude than observed during SCIT [62]. These protective IgG4 antibodies persist after discontinuation of treatment [29••].

Some studies have shown increases in specific IgG4 in the absence of demonstrable efficacy [63], whereas others have shown no difference in IgG levels, likely related to the lower allergen doses employed [43], particularly in relation to HDM SLIT [40, 44, 64]. These findings raise the issue of causality versus bystander effects. In functional assays, sera obtained after grass pollen SLIT was able to inhibit IgE-binding in-vitro [36, 61]. The heterogeneity of immunological responses in particular in relation to allergen-specific IgE and IgG4 antibodies may be largely explained on the basis of the allergen dose, duration of treatment, and types of allergens used.

Potential surrogate/predictive biomarker of SLIT

In order to evaluate potential biomarkers of effective SLIT to predict responders from non-responders or to inform whether to discontinue or recommence treatment, various attempts have been undertaken to assess the relationship between T and B cell responses in the periphery and/or target organs. Hitherto, only serum-based biomarkers have been employed in large clinical trials, as current assays assessing cellular T and B cell responses are complex and challenging to standardise between laboratories for multicentre studies or for monitoring individual subjects.

Immunoreactive immunoglobulin G

Elevation in allergen-specific IgG1 and IgG4 levels has been observed in both serum and local target organs following SLIT [28, 29••, 65••] and SCIT [23••, 66, 67]. Despite an increase in IgG4 antibody levels, however, these studies failed to establish a relationship between IgG4 antibodies and clinical efficacy. Thus, quantitative allergen-specific IgG4 may serve only as a marker of allergen exposure and not clinical outcome. Moreover, it has also been reported that IgG4 antibodies induced following SLIT may inhibit basophil activation and prevent CD23-mediated IgE-facilitated allergen presentation by competing with allergen-specific IgE in B cells to T cell clones [68, 69].

Inhibition of IgE-facilitated allergen presentation

In the mid-1930s, Cooke and colleagues were the first to report the induction of serum inhibitory antibody activity following AIT [70].Subsequent studies further characterised these antibodies, which were found to be retained within the IgA and IgG fraction in serum [71, 72]. Later, using serum obtained from subjects that had undergone birch pollen AIT, Van Neerven and colleagues demonstrated that the inhibitory activity for IgE-facilitated antigen presentation by B cells to T cell clones could translate into a reduction in allergen-driven T cell proliferation and cytokine production [68]. These findings were subsequently reproduced in subjects who had undergone grass pollen AIT [69] using a simplified flow cytometry-based assay in which the cooperative binding of allergen-IgE complexes binding to low-affinity IgE receptor FcεRII (CD23) on the surface of B cells was detected (IgE-FAB). The binding of allergen-IgE complexes to B cells [69] was shown to correlate with the vigour of T cell clone proliferation. The IgE-FAB is an in vitro biofunctional cellular assay that is highly reproducible, with within-assay and between-assay reproducibility of 4 % and 12 %, respectively, that can be utilised to detect IgG-associated serum inhibitory activity during AIT. Furthermore, grass pollen-specific IgG4 antibodies with inhibitory activity for IgE-FAB are found to be elevated in subjects treated with SCIT. Interestingly, this inhibitory activity for IgE-facilitated allergen binding was demonstrated in post-immunotherapy serum and co-purified with IgG4-containing fractions [23••, 73••, 74, 75••]. In a randomised double-blind placebo-controlled study of SLIT that included 238 moderate to severe grass pollen–allergics with or without asthma, clinical efficacy following SLIT grass treatment was associated with sustained increases in serum grass pollen-specific IgG4 antibodies. This finding was associated with parallel increases in serum IgE-blocking factor and serum inhibitory activity for the binding of allergen-IgE complexes to B cells. Notably, the immunologic changes persisted two years after the completion of treatment, consistent with long-term clinical benefit [29••].

At this time, it is necessary to further validate whether IgE-FAB inhibitory activity could be used as a surrogate and/or predictive marker of clinical efficacy within individual patients. The primary deficiency in previous studies has been the lack of data from a baseline pollen season in order to be able to define individuals as ‘responders’ or ‘non-responders’. Nonetheless, a modest inverse correlation has been demonstrated between combined symptom and rescue medication scores and inhibition of IgE-facilitated allergen binding following SCIT [23••]. This association needs to be tested following SLIT.

IgE-blocking factor

The extent to which IgE is hindered from binding to allergens and thus not eliciting clinical symptoms [29••, 76, 77] is termed IgE-blocking factor. A functional solid-phase assay has been developed and can be utilised to examine this aspect. In North American subjects with seasonal allergic rhinitis, increases in Phl p5-specific IgG4 and IgE-blocking factor have been demonstrated in the actively-treated group when compared to those who received placebo tablets following grass pollen SLIT treatment [78••]. The inhibition of facilitated allergen presentation and binding, as well as basophil histamine release, has now also been linked to IgE-blocking factor by Würtzen and colleagues, wherein they observed a significant reduction of IgE-FAB, IgE-blocking factor, and basophil histamine release after one year of treatment and noted that this effect was sustained during the second year of treatment [76]. This established a clear correlation between CD23-mediated IgE-facilitated allergen binding as a measure of T cell activation and serum IgE-blocking factor and basophil histamine release. IgE-FAB, therefore, represents a more physiologic readout of functional IgG1 and IgG4 antibodies correlating with combined symptom and rescue medication scores [73••]. This is in contrast to IgE-blocking factor, which accounts for all serum blocking antibodies that compete with IgE for the allergen on a solid-phase matrix in vitro.

Serum allergen-specific IgA

Similar to IgG4, immunoglobulin A is also a noninflammatory isotype that is unable to activate, complement, or partake in immune complex formation [79]. An elevation in allergen-specific IgA2 antibodies and polymeric IgA2 levels has been reported following grass pollen-specific SCIT. Moreover, it has also been reported that passive sensitisation of monocytes using purified polymeric IgA2 from IgA-containing serum obtained post immunotherapy, followed by in vitro cross-linking of IgA on monocytes by antigen or anti-IgA, can result in IL-10 production [80]. Hence, the secretion of IL-10 from accessory cells may sequentially favour immunoglobulin isotype class-switching in favour of IgG4 antibody production. Allergen-specific IgA2 has been shown to increase following grass pollen SLIT, and these findings further highlight the role for IgA antibodies in the induction of tolerance. Their use as a predictive biomarker, however, warrants further investigations.

Ratio of specific IgE/Total IgE

A recent clinical study evaluated the ratio of serum-specific IgE/total IgE in timothy grass pollen immunotherapy as a potential predictive marker for clinical efficacy [81], which involved in vitro analysis of patients who had received four years of conventional or sublingual immunotherapy. Serum total IgE (t-IgE) and specific IgE (s-IgE) levels, blood eosinophil counts, and serum s-IgE/t-IgE ratios were examined and assessed for correlation with clinical response as measured by visual analogue scores (VAS). Receiver operating characteristic (ROC) analysis for each in vitro biomarker showed a greater diagnostic and predictive value in favour of baseline serum s-IgE/t-IgE ratios. A diagnostic/predictive cutoff value of 16.2 % to predict successful SCIT outcome was established in both treatment regimes, and revealed clinical sensitivity and clinical specificity of 97.2 % and 88.1 %, respectively [81]. However, these results could not be replicated in a randomised controlled open-label three-parallel-group study of house dust mite in monosensitised asthmatic/rhinitis children to HDM who received SLIT, SCIT, or pharmacotherapy. The results from this study displayed no significant correlation between combined symptoms and medication scores or VAS and the ratio of serum s-IgE/t-IgE, which could be attributed to the small sample size in each treated group. Therefore, further assessment of serum s-IgE/t-IgE ratios in large randomised double-blind placebo-controlled studies are necessary to further evaluate the utility of s-IgE/t-IgE as a biomarker for effective immunotherapy. Use of recombinant allergens for accurate in vitro analysis of both individual major and minor allergens and the targeted use of a known recombinant or mixture of recombinant allergens for therapy will likely provide further clarification [82].

STAB1 and C1Q RNA expression

Proteomic analysis and mass spectroscopy of peripheral human DCs identified two potential candidate proteins, namely stabilin-1 (STAB1) and the complement component C1Q as potentially representing a tolerogenic signature of DCs. Ex vivo studies involving quantitative polymerase chain reaction of peripheral blood mononuclear cells purified from blood drawn before/after grass pollen SLIT revealed elevations in STAB1 and C1Q RNA expression that correlated with the clinical response to immunotherapy [57••]. These proteins may be relevant for inducing tolerogenic T regulatory responses.

Conclusions

Allergen immunotherapy is effective and induces long-lasting immunological and clinical tolerance that persists for years following cessation of treatment. Immunotherapy is associated with suppression of allergic inflammation in target organs and increased IgG4− and IgA2-associated blocking antibodies. The induction of blocking antibodies is accompanied by suppression of undesired allergen-specific Th2 cell responses (Fig. 1). This suppression occurs within weeks or months as a result of the induction of regulatory T cells that exert their effects by mechanisms involving cell–cell contact and also by release of immunomodulatory cytokines such as IL-10 and TGF-β. The more delayed appearance of antigen-specific Th1 responses and alternative mechanisms such as Th2-cell anergy and/or apoptosis may also be involved. A greater understanding of these mechanisms has provided potential surrogate clinical/immunological efficacy biomarkers and has led to novel immunotherapy strategies, although the mechanisms of long-term clinical tolerance remain to be further fully elucidated.

Figure 1
figure 1

Immunological mechanisms of sublingual immunotherapy. Low-dose allergen exposure at mucosal surfaces in atopic individuals results in type I IgE-mediated allergic responses. High-allergen dose through sublingual oral route readdresses the balance between TH2/TH1 responses in favour of TH1 responses. This is accompanied by an increase in the ratio of IFN-g (Th1 cytokines) to IL-4, IL-5 and IL-13 (Th2 cytokines). The induction of inducible Treg cells (iTreg) and natural Treg cells (nTreg) and associated cytokines such as IL-10 and TGF-b following SLIT play an important role in suppressing pro-inflammatory Th2 responses and contributes towards the induction of allergen-specific IgA1, IgA2 and in particular IgG4 antibodies with inhibitory activity. IgG4 antibodies are able to suppress FceRI-and CD23-mediated IgE-facilitated allergen presentation and basophil histamine release.