Background
Allergic rhinitis is a growing public health problem, affecting over 400 million people worldwide [
1]. Currently, allergen-specific immunotherapy (AIT) represents the only disease-modifying treatment, diminishing symptoms, improving quality of life, preventing new sensitisations and reducing the risk of asthma development [
2,
3]. The golden standard treatment is subcutaneous immunotherapy (SCIT), which shows long-term benefit for the treatment of allergic rhinitis, conjunctivitis and asthma [
3,
4]. Despite this, only 5 % of patients undergo this therapy, due to frequent injections, risk of adverse effects and the long duration of treatment [
2]. A more recent, non-invasive route of administration is sublingual immunotherapy (SLIT). Though efficacious, SLIT is associated with reduced compliance due to the long period of self-medication [
5].
Intralymphatic immunotherapy (ILIT) is an emerging form of AIT that involves three injections of allergen over a period of 12 weeks. This form of AIT directs lower doses of allergen to the highly immunocompetent lymph node, in an effort to maximise chances for tolerance induction, while minimising the risk for adverse effects. In the pioneering study in 2008, ILIT was shown to induce long-lasting allergen tolerance, equivalent to that of SCIT, but with fewer adverse events [
6]. Since then, ILIT has demonstrated clinical efficacy against allergy to cat dander and, recently, against grass-pollen induced rhinoconjunctivitis in adolescents and young adults [
7,
8].
In a recent small double-blind placebo controlled study, we too demonstrated that ILIT against grass pollen resulted in a significant improvement of patient-recorded symptoms during the pollen season, while injection-associated discomfort levels were comparable to that of SCIT [
9]. However, the clinical efficacy of ILIT is currently disputed [
10]. Consequently, the aim of the present study was to expand our first trial to determine if the safety and efficacy of ILIT persisted in a larger cohort of patients.
Methods
Study population and eligibility criteria
Study subjects were recruited amongst patients at the allergy department of Skåne University Hospital, Malmö, Sweden. Eligible patients were aged between 18 and 65 and had moderate to severe allergic birch/grass pollen-induced rhinoconjunctivitis, with symptoms including itchy nose and eyes, sneezing, nasal congestion and secretion. Allergy was verified by positive skin prick tests (SPTs), presence of serum-specific IgE antibodies towards birch and/or grass (minimum 0.35 kU/L) and positive nasal provocation tests (NPTs). Sample size was based on our previous study [
9]. All eligible participants recruited during the recruitment period were enrolled in the study.
General contraindications were pregnancy or nursing, planning for pregnancy, autoimmune and collagen disease, cardiovascular disease, current persistent asthma (not intermittent asthma), upper airway disease (non-allergic sinusitis, nasal polyps), chronic obstructive and restrictive lung disease, hepatic and renal disease, cancer, previous immune- or chemotherapy, major metabolic disease, alcohol or drug abuse, mental incapability (to cope with the study) or medication with a possible side-effect of interfering with the immune response.
Study design
This study was a parallel double-blind placebo-controlled trial, performed at the allergy department of Skåne University Hospital, Malmö, Sweden.
In total 36 patients were enrolled. Fifteen patients were recruited in the first cohort (September 2010 to September 2011) and have been previously defined [
9]. Twenty-one patients were recruited in the second cohort (September 2011 to September 2012).
At the first visit (visit 1, out of pollen season, 2010 or 2011), patient eligibility was determined, SPTs and NPTs were performed and blood was sampled (further details in Additional file
1). After approximately one week, included patients were randomly allocated to receive either placebo (
n = 15) or active (
n = 21) intralymphatic treatment. At visits 2–4 (September 2010 – January 2011 or September 2011 – January 2012), the study subjects received three 0.1 ml injections with either placebo (Alutard, ALK Abéllo, Horsholm, Denmark) or 1000 SQ-U of standardised, aluminium hydroxide adsorbed, depot birch- or grass-pollen vaccine (Alutard, ALK Abéllo) at 3- to 4-week intervals. Based on the outcome of the allergy tests, patients were challenged and vaccinated with either birch or grass (mono-sensitised). Patients returned approximately 4 weeks after the last injection (visit 5, February 2011 or February 2012) and after the consecutive pollen season (visit 6, September 2011 or September 2012), and were evaluated as per visit 1. At visit 6, patients were additionally asked to complete a questionnaire regarding their seasonal allergic symptoms as compared to the previous pollen season. One patient did not complete the treatment, due to a non-severe adverse event (local urticaria). Emergency envelopes remained unbroken.
Randomisation, allocation and blinding
Participants were randomly assigned to one of two treatment groups following a simple randomisation procedure with opaque, sealed envelopes. The randomisation procedure was generated by independent biomedical assistants. The vaccines used were pre-packed, blinded and allocated according to the randomisation sequence by independent staff with no connection to the study, and thus both patients and physicians were blind. Recruitment was performed by UPW. All those involved in the study, including participants, care-givers and those assessing outcomes, were blinded after assignment to interventions.
Ethics, consent and permissions
The study was approved by the local ethics committee (ref. ID 2009/714) and all participants gave their written informed consent.
Intralymphatic injections
A superficial inguinal lymph node in either the left or right groin was aseptically injected using a 25-gauge needle and ultrasound guidance. The superficial lymph nodes in the groin were identified as hypoechoic nodules with a diameter of 0.5 to 1.5 cm. The same side and, as far as possible the same node, was targeted during all three injections. Aspirations were made before the injections to avoid inadvertent intravascular administration. The peak expiratory flow (PEF) was measured before and after each injection, and all patients were monitored at the ward for no less than 60 min after each injection. The trial staff recorded all signs of local and/or systemic reactions in conjunction with the injections. The patients were subsequently asked to record and report all indications of late reactions for the following 24 h.
Trial outcomes
Primary outcome measures
The primary outcome measure for the study was the change in pollen season-associated allergic symptoms. At the end of the first allergy season, after treatment had been given (visit 6), patients were asked to compare their most recent seasonal allergic symptoms with the symptoms they experienced during the pollen season prior to treatment. To this end, a visual analogue scale (VAS), ranging from 0 (unchanged symptoms, no improvement) to ten (total symptom relief, complete recovery), was used, as previously described [
11].
Secondary outcome measures
The secondary outcome measures of the study were 1) the safety of intralymphatic injections; 2) the change in nasal symptom score (NSS) following NPT (comparisons of NSS after NPT before treatment, after treatment and after the consecutive pollen season), as previously described [
9]; 3) the change in circulating immunoglobulin (IgE and IgG
4) levels (comparisons of levels before treatment, after treatment and after the consecutive pollen season); and 4) change in circulating inflammatory cells (comparisons of levels before treatment, after treatment and after the consecutive pollen season).
Additional secondary outcomes
Blood samples acquired at visit 6 (after the pollen season following treatment) were re-analysed following unblinding of the study, to determine alterations in IgG
4 affinity. This was due to the appearance of two distinct sub-populations in the active ILIT group (“improved” and “non-improved”; see
Results), which demonstrated no difference in IgG
4 levels. Further analysis was consequently performed, with the aim to immunologically discriminate between these groups.
Tertiary outcomes
As a tertiary outcome, patients were asked to report their usage of allergy medication during the pollen season immediately after ILIT, with their use in the allergy season prior to the start of treatment. Patients were asked to report a reduction in use, an increase in use or a no change in use, in terms of their use of anti-histamines, corticosteroid nasal spray and eye drops.
Methods used for assessment of secondary and additional trial outcomes
Methods used for the assessment of trial outcomes, including nasal provocation test, flow cytometry, and assessment of IgG
4 affinity can be found in the supplementary material (Additional file
1).
Discussion
In a recent study, we demonstrated that ILIT is a safe and effective administration route for the induction of allergen tolerance and treatment of nasal allergic inflammation [
9]. In the present study, we have added a new cohort of study subjects and together they reinforce the hypothesis that ILIT is effective in reducing allergic symptoms and that it is not associated with severe adverse events. In addition, we highlight a role for increased allergen-specific IgG
4 affinity in successful ILIT.
During the past few years, ILIT has come forward as a less time-consuming and more cost-effective alternative to conventional subcutaneous allergen-specific immunotherapy (SCIT) for treatment of pollen-induced rhinoconjunctivitis. It has provided promising clinical results in combination with an excellent safety profile [
6,
9], likely related to the drastic reduction in amount of administered allergen. This present study demonstrates that, as compared to placebo, active ILIT treatment results in improvement of seasonal allergic symptoms, without any moderate or severe adverse reactions. This study further supports ILIT as a safe and effective treatment strategy for patients with pollen-induced rhinoconjunctivitis. Indeed, the 60 % success rate is comparable to the approximate 70 % success rate for SCIT [
12]. Although it is evident that five out of eight placebo patients responded to treatment, it is worth noting that the number needed to treat (NNT) value for ILIT, which takes into account responders and non-responders in both active and placebo groups, is 7.5. This NNT value is much lower than those calculated for more common treatments of allergic rhinitis, such as anti-histamines [
13].
The somewhat lower success rate of ILIT as compared to SCIT may be related to the technical aspects of allergen injections. Unlike SCIT, ILIT uses complex medical equipment and involves higher technical precision, as targeted nodes are relatively small (0.5–1.5 cm). Despite the use of a fine needle and ultrasound guidance, it can on occasion be difficult to penetrate the capsule surrounding the lymph node, which may lead to inadequate administration of allergen.
Despite repeated promising results, the overall efficacy of ILIT against grass-pollen-induced allergic rhinitis has recently been questioned [
10]. In a double-blind placebo-controlled trial, 15 patients receiving active ILIT against grass pollen showed immunological indications of tolerance, but no clinical improvement. As recently argued [
14,
15], these results were likely due to the interval between injections. Witten et al [
10] injected allergen every 2 weeks, which, unlike a 4-week interval, does not allow sufficient time for the development of non-interfering waves of antigen-specific immune responses [
16]. The present study, where injections were given at 4-week intervals, again highlights the importance of correct time interval between injections.
In this study, two distinct sub-populations could be observed among the actively treated patients. Patients in the first sub-population reported a clear improvement of their allergic symptoms, as characterised by reduced nasal allergen reactivity, decreased usage of allergy medication (anti-histamines, local steroids, eye drops) and self-recorded improvement (VAS-scale). In contrast, patients in the second sub-population experienced no resolution of symptoms. The reason for the emergence of two sub-populations is unknown and may, at least in part, be related to technical difficulties during intralymphatic injection, resulting in extra-nodal deposits of allergen and an ineffective tolerance induction. Nevertheless, the sub-populations firstly demonstrate the usefulness of the nasal provocation test (NPT) as a read-out of clinical efficacy, as nine out of 12 patients with self-reported improvement in season allergic symptoms (VAS-scale) also reported a reduction in symptoms following NPT. Secondly, these populations give insights into the underlying immunological mechanisms associated with efficacious ILIT.
The present study, as well as the aforementioned study by Witten et al [
10], both suggest that immunological alterations associated with tolerance induction, including elevated T-regulatory cells and cytokines, and IgG
4 levels in blood, are not prerequisites for clinical efficacy of ILIT. Witten et al [
10] noted increases in T-regulatory cells, IL-10 production and IgG
4 following active ILIT, but demonstrated no clinical improvement when compared to placebo. In contrast, we demonstrated clinical improvement in the absence of increases in circulating T-regulatory cells and total serum IgG
4. The reason for this is unclear, but may be related to the short time interval between vaccination and blood sampling. However, we demonstrated that clinical improvement was associated and significantly correlated with increased allergen-specific IgG
4 affinity. Indeed, as previously suggested [
14] affinity of B- and T-cell receptors is likely a very important parameter for the induction of tolerance to allergen and subsequent efficacious therapeutic responses.
There is an intricate interaction between B- and T-cells in the specialised lymph node compartment. B-cells bind the injected allergen via the cell-surface B-cell receptor (BCR), which leads to receptor mediated endocytosis, processing and peptide loading. Subsequently, allergen-derived peptides are presented via the MHC class II pocket to allergen-specific CD4
+ T-helper cells. This B-cell induced activation of T-cells, reciprocally, allows for T-helper cell-licensed class switching and affinity maturation of IgG [
17]. In the case of the present study, the injected composition resulted in increased affinity of IgG
4, avoiding IgE production. The reason for IgG
4 preferences in the present study is likely the use of adjuvant, as well as the compartment for the delivery of allergens, lymph nodes being more prone to IgG responses, as compared to airway responses to pollen allergens.
Furthermore, competition for antigen, which occurs more effectively when low concentrations of antigen is present in lymph follicles, is vital for the positive selection of high affinity B-cells, and this phenomenon has similarly been suggested for selection of high affinity T-cells [
18]. This suggests that a 2-week-interval between injections, where concentrations of allergen in lymph nodes is consistently high, may induce clear B- and T-cell responses, as seen by increased serum levels of total IgG
4 levels and allergen-induced intracellular T-cell cytokine production [
10], but with a low affinity maturation, resulting in poor clinical efficacy. Indeed, previous studies have noted that functionality of IgG
4, rather than levels, more closely relate to clinical efficacy in AIT [
19,
20]
Although this study clearly indicated that ILIT does improve clinical symptoms of pollen-induced rhinoconjunctivitis, particularly in the improved sub-population, there are a number of limitations. Firstly, the use of a VAS scale for patient recorded outcome has been debated, in particular in long-term studies, due to recall bias. However, several publications have demonstrated that using VAS to measure self-recorded outcome following immunotherapy parallels the more objective, and more regularly monitored, symptom and medication score (SMS), up to five years following treatment [
21‐
23]. In addition, with the small sample size of 35 patients used in this study, it is impossible determine whether ILIT is an effective treatment for the general allergic population. However, previous studies with smaller [
9] and larger [
6] sample sizes show similar effects, which suggests that ILIT has a generalised efficacy. Further, as this study only spanned over one year, long-term efficacy of ILIT is unknown. Nevertheless, previous studies have demonstrated that efficacy of ILIT against grass-pollen-induced rhinoconjunctivitis persists for at least 3 years following treatment [
6].
Competing interests
The authors declare that they have no competing interests.
Authors’ contributions
TH was involved in study design and responsible for acquisition, analysis and interpretation of data, as well as drafting of the manuscript. OL was involved in analysis and interpretation of data, as well as drafting of the manuscript. UPW was involved in study design, coordination of study and acquisition of data. ME was involved in acquisition of data. SKG was involved in coordination of the study, interpretation of data and drafting of the manuscript. LOC designed and coordinated the study, was involved in interpretation of data and drafting of the manuscript. All authors read and approved the final manuscript.