Background
As suggested by the hygiene hypothesis, infections are of importance to the maturation of the immune system [
1]. Th1-mediated immunity may be defective in a modern clean environment resulting in facilitation of Th2 responses associated with allergic disorders [
2,
3]. Conversely, up-regulated Th1 responses, e.g. as a consequence of infections, can be associated with reduced Th2 activity and reduced responsiveness to allergen [
4]. Controlled infection-like stimulation of the immune system may in this context be beneficial, and may be achieved by the use of Toll-like receptor (TLR) agonists.
TLRs are receptors of the innate immune system that recognise conserved microbial components known as pathogen-associated molecular patterns (PAMPs) [
5]. PAMPs include the bacterial product LPS, viral single-stranded RNA, and bacterial/viral CpG DNA, acting as TLR4, TLR7, and TLR9 ligands, respectively [
6]. Activation of TLRs stimulates the innate immune system, potentially leading to down regulation of Th2 adaptive responses to allergen [
6]. The possibility of skewing the immune system away from a Th2 response, as has been attempted previously by other measures [
7‐
10], is the basis for the development of TLR agonists as treatments for allergic rhinitis and asthma.
In a murine model of allergic asthma, a TLR7 ligand (S28463) administered systemically exerted anti-allergic effects resulting in attenuated airway eosinophilia, normalized airway responsiveness, and prevention of airway remodelling [
11,
12]. In a similar model, Sel
et al. [
13] demonstrated that systemic intervention with poly(I:C) and R-848, viral ligands recognized by TLR3 and TLR7 respectively, prevented production of allergen specific IgE and IgG1 during sensitisation and subsequently alleviated experimental asthma. Moreover, administration of poly(I:C) and R-848 in established allergy markedly reduced the responsiveness to allergen [
13]. Similarly, the TLR9 ligand 1018 ISS was shown to inhibit Th2-mediated airway inflammation and hyperresponsiveness in animals [
14‐
18].
AZD8848 is a selective TLR7 agonist optimised for topical airway treatment through rapid metabolism by plasma esterases, thereby reducing systemic exposure [
19]. Observations involving peripheral blood mononuclear cells (PBMCs) and ovalbumin-sensitized splenocytes indicate that stimulation of TLR7 by AZD8848 inhibits Th2-adaptive responses to allergen via an immune response involving the induction of mediators including interferon alpha (IFN-α) [
19‐
22]. Furthermore, Ikeda
et al. [
23] reported that AZD8848 was effective against allergen-induced airway obstruction and inflammation in guinea-pig models of rhinitis and asthma with weekly as well as acute dosing.
Here, we report the results of two studies. In the first study, increasing single doses of AZD8848 were administered intranasally to healthy subjects and patients with allergic rhinitis. Indices of efficacy and tolerance were monitored. In the second study, AZD8848 was administered intranasally once weekly for five weeks to patients with allergic rhinitis: these individuals were then subjected to repeat allergen exposure and disease activity was monitored focusing on symptoms to establish proof of principle.
Discussion
In these studies we demonstrate that repeated intranasal TLR7 stimulation is associated with reduced responsiveness to allergen in patients with allergic rhinitis. It is the first observation in man where TLR7 has been successfully evaluated as a therapeutic target for allergic airway inflammation.
The TLR7 agonist AZD8848 undergoes very rapid enzymatic degradation by butyrylcholinesterase:
t1/2 in human plasma is estimated to be 20 seconds [
19]. In this study, this was reflected by undetectable plasma levels of the substance following nasal administration of doses up to 600 μg, while its acid metabolite, which is 1500 times less potent at the receptor [
19], was detectable at doses of 30 μg and above. Nevertheless, flu-like symptoms were reported, which were probably secondary to local TLR7 activation and a subsequent systemic increase in type-1 IFN. This possibility was indicated by the five-fold increase in plasma IL-1Ra following administration of AZD8848, known to reflect type-1 IFN production down-stream to TLR7 (i.e. proof of mechanism) [
21,
26]. In the context of the hygiene hypothesis, repeated administration of the TLR7 agonist AZD8848 may mimic repeated virus-like stimulation of the immune system.
In the repeat challenge/treatment study, acute symptoms in response to allergen were consistently reduced from the third day of the allergen challenge series in patients receiving the TLR7 agonist when compared to placebo, indicating a sustained effect of at least one week after the final dose of AZD8848. The same pattern was observed for the low-grade morning and evening symptoms, but these changes failed to reach statistical significance. In agreement with a symptom reducing effect, lavage fluid levels of tryptase and α
2-macroglobulin, reflecting mast cell activity and inflammatory plasma exudation [
27,
28], were also reduced. Our observations extend recent
in vitro and animal reports on anti-allergic effects of AZD8848, indicating that repeated TLR7 stimulation reduces the responsiveness to allergen [
19‐
23], and suggest that AZD8848 may be clinically effective in allergic rhinitis.
The observed reduction in responsiveness to allergen might reflect that the immune system was functionally skewed away from a Th2 response. If so, it did probably not represent repolarisation of T-lymphocytes (i.e. a change from Th2 to Th1 phenotype), as this study involved atopic individuals with established populations of memory T-cells with a life span of at least two to three years [
29]. Arguably, the outcome was more likely a consequence of a functionally reduced responsiveness of memory Th2 lymphocytes. In this context, for future studies, it would be of interest to examine if prolonged treatment, i.e. a time period sufficient to induce true repolarisation of T-lymphocytes, could produce a more marked anti-allergic effect.
While the repeat challenge/treatment study demonstrated that 60 μg of AZD8848 administered intranasally once weekly for five weeks induced a desired hyporesponsiveness to allergen, likely through activation of TLR7, further studies are warranted to optimize the effect. Preclinical data have indicated that more frequent administration of AZD8848 can produce more marked anti allergic effects (AstraZeneca: data on file). Furthermore, as demonstrated in a Brown Norway rat model of allergic rhinitis/asthma, both nasal and bronchial administration of AZD8848 can reduce the ability of a bronchial allergen challenge to produce bronchial airway eosinophilia and generate IL-13 [
19], suggesting the possibility that nasal administration of AZD8848 may be effective in the treatment of asthma.
In this study, safety and tolerability of intranasal AZD8848 was evaluated parallel to the exploration of its anti-allergic effects. None of the patients treated with AZD8848 discontinued the study prematurely due to drug-related AE. Furthermore, standard laboratory indices (haematology, clinical chemistry, and urine analysis) were unaffected by the treatment, except for the anticipated transient reductions in blood lymphocyte counts. Moreover, vital signs (blood pressure, pulse, and body temperature) and continuous ECG were unaffected. However, dose-dependent local side effects were common, albeit of mild intensity. These were dominated by blood-admixed nasal secretions and in these cases nasal inspection revealed superficial mucosal irritations/ulcerations. This effect, and the temporary flu-like symptoms that were experienced by a third of the patients, needs to be further evaluated in order to assess overall tolerability of intranasal AZD8848 as a potential treatment.
The body of knowledge on TLRs is increasing as their distribution and functions are outlined, along with potential associations with specific allergic and airway conditions [
30,
31] and their treatment, notably specific immunotherapy [
32,
33]. In the context of established allergic airway conditions, animal observations suggest that stimulation of TLRs (i.e. TLR3, TLR4, TLR7, TLR8, and TLR9) has a general potential to reduce allergen responsiveness. However, focusing on human conditions available observations are scarce. In patients with allergic asthma, a synthetic oligonucleotide containing immunostimulatory CpG motifs (acting on TLR9) was reported to increase the expression of IFN-γ and IFN inducible genes without affecting allergen challenge induced changes [
34], possibly reflecting that the dose employed was too low to produce an anti-allergic effect. In a report by Casale
et al. [
35], which focused on tolerability to topical CRX-675 (acting on TLR4), data on efficacy was not given in detail, but a decrease in allergen-induced nasal symptoms was reported in one of four treatment groups compared with placebo. Moreover, it was recently reported that nasal administration of a TLR8 agonist (i.e. VTZ-1463) improved symptoms of allergic rhinitis [
36]. Taken together, available information suggests that TLR agonists are valid treatment targets for allergic airway disease.
The natural ligand for TLR7 is single stranded RNA. Accordingly, various respiratory viruses, e.g. influenza, corona, and potentially rhinovirus [
37], may activate the receptor. In this context, it is of interest to consider the evidence indicating that respiratory viral infections often heighten the responsiveness to allergen and produce asthma exacerbations [
38,
39]. The possibility that acute pro-inflammatory effects of TLR7 stimulation may heighten the responsiveness to allergen while later effects may reduce allergen responsiveness suggests that the timing of interventions with TLR7 agonists in relation to allergen exposure is important. Further studies are warranted to explore this and to outline the benefits and risks of treatment with TLR7 agonists in allergic airway disease.
Competing interests
The work described in this manuscript was supported and funded through a collaboration with AstraZeneca and Dainippon Sumitomo Pharma. In the past five years, LG and MA have received project related financial support from AstraZeneca, Schering-Plough, Orexo, HealthCap (Biolipox/Orexo/LTB4 Sweden/CC10 Sweden), Bioglan, and Nares. LG and MA are shareholders in Nares (a company active in the field of allergic rhinitis). In the past five years, AC has received financial support from AstraZeneca and Mediplast. In the past five years, EH has received financial support from AstraZeneca, Grünenthal, and NovoNordisk. GA and AK are employed by AstraZeneca. LE and PN were employed by AstraZeneca when the study was conducted. CAE, JD, ILS, and HW declare no competing interests.
Authors’ contributions
Conception, design and management: LG, GA, AC, LE, EH, PN. Patient examination: LG, CAE, MA, AC, JD, EH, PN, ILS, HW. Data input, drafting, revision and/or approval of manuscript: All. Statistics: LG, AC, AK. All authors read and approved the manuscript.