Discussion and Conclusions
The results of this study demonstrate that targeting the IL-5 pathway with mAbs is a potent and effective strategy to prevent in a concentration dependent manner the AHR in passively sensitized airways, a condition leading to a massive production of IL-5 in bronchial tissue.
Benralizumab and mepolizumab reached the primary endpoint since both the agents flattened and shifted rightward the CRC to histamine. Lower concentrations of benralizumab than mepolizumab significantly reduced the AHR to histamine, with benralizumab being also most effective than mepolizumab by reducing the myogenic response at lower levels than that detected in non-hyperresponsive bronchi. The advantage of blocking IL-5Rα to IL-5 was further confirmed by the analysis of the potency, since benralizumab was ≃0.7 logarithm more potent than mepolizumab in reducing the contractile plateau generated by histamine in hyperresponsive airways.
Concerning the secondary endpoints, although the efficacy in preventing the AHR to parasympathetic stimulation was similar between the investigated mAbs, benralizumab was more potent than mepolizumab when the tissues were stimulated by higher EFS frequencies. Furthermore, lower concentrations of benralizumab than mepolizumab were sufficient to significantly counteract the FRC to EFS in hyperresponsive bronchi. Benralizumab and mepolizumab also prevented the hyperresponsive myogenic tone in response to QS, although only benralizumab reduced the ASM contractility at a level lower than that detectable in non-sensitized airways. Again, at lower concentrations (10 μg/ml) benralizumab but not mepolizumab was effective in inhibiting the AHR to QS.
As previously reported (Cazzola et al., 2016b), also in this study the passive sensitization altered the concentration of cAMP in hyperresponsive airways, a condition that was reverted in a concentration dependent manner by both benralizumab and mepolizumab. Benralizumab, but not mepolizumab, improved the cAMP concentrations at a level greater that that detected in non-sensitized tissue.
Certainly several complex indirect mechanisms (Molfino et al., 2012; Mukherjee et al., 2014) are involved in the prevention of AHR elicited by the inhibition of the IL-5/IL-5Rα axis in hyperresponsive airways. Whatever the pathways implicated, we have found that benralizumab and mepolizumab can restore at physiological levels the cAMP concentrations in passively sensitized airways. Increased intracellular concentrations of cAMP in ASM promote bronchodilation via activation of cAMP-dependent protein kinase (A kinase) (Rogliani et al., 2016). However, conflicting findings are currently available regarding the real impact of cAMP on airway inflammation, with some evidences suggesting for anti-inflammatory effects and other reporting increased cytokine synthesis and T cell adhesion to ASM cells induced by enhanced concentrations of cAMP (Black et al., 2012). Isolated airways include several residential cells such as ASM cells, fibroblasts, parasympathetic cells, epithelial cells, and inflammatory cells such as T-helper type 2 (Th2) lymphocytes, eosinophils, and mast cells, most of them stimulated or activated in response to passive sensitization (Schmidt et al., 2000). Therefore, in the isolated airways used in our experiments it was not possible to discern the exact origin of cAMP enhancement induced by the investigated mAbs. However, the strong correlation between the cAMP levels and the inhibition of AHR contractility induced by benralizumab and mepolizumab suggests that the improvement in cAMP concentrations represents a key mechanism on which the inhibition of IL-5/IL-5Rα axis may converge and, thus, protect ASM from hyperresponsiveness.
Unexpectedly, in passively sensitized airways benralizumab but not mepolizumab improved AHR and cAMP at levels greater than those detected in untreated non-sensitized tissue. Isolated airways are characterized by a certain degree of intrinsic tone mediated by the spontaneous generation of histamine and leukotrienes. Therefore, since eosinophil secretory granules contain both histaminase and leukotrienes (Bagnasco et al., 2017; Bandeira-Melo et al., 2002), it was expected that preventing the activation and degranulation of eosinophils by administering either benralizumab or mepolizumab would have led to the same effect. Perhaps, the difference in the efficacy between benralizumab than mepolizumab can be explained by considering that although both these mAbs counteract the IL-5 pathway, only benralizumab induces ADCC leading to the apoptosis of eosinophils and mast cells caused by natural killer (NK) cells (Kolbeck et al., 2010). Thus, the combination of IL5Rα blockade and ADCC activity of benralizumab may also support the greater potency of this mAb compared to mepolizumab against the AHR to histamine. Furthermore, while at low concentrations benralizumab can antagonize IL-5Rα and elicit ADCC (Kolbeck et al., 2010), low concentrations of mepolizumab could not be sufficient to neutralize the massive release of IL-5 in the sensitized tissue.
Such a condition can be translated to asthmatic patients before the next dose administration of a mAb. Accordingly with pharmacokinetic (PK) studies in healthy subjects (Martin et al., 2019; Shabbir et al., 2019) that received approved doses of mAbs (European Medicines Agency, 2015; European Medicines Agency, 2018; US Food and Drug Administration, 2015; US Food and Drug Administration, 2017), while the maximum plasma concentrations of benralizumab and mepolizumab were ≃3 μg/ml and ≃12 μg/ml respectively, the trough concentrations were ≃1 μg/ml and ≃5 μg/ml respectively. Indeed, our results indicate that at these concentrations both the agents are effective in submaximally inhibit the AHR to histamine in isolated airways. However, while the efficacy of the circulating levels of mepolizumab at trough could be neutralized by the cytokine storm during an asthma exacerbation (Borish, 2016), the protective effect of low concentrations of benralizumab can be preserved as it is directed on IL-5Rα and ADCC regardless of the the amount of circulating IL-5.
Taken together the data from PK studies with those of our study, it is also evident that the concentrations of benralizumab and mepolizumab detectable in plasma can only partially prevent the AHR induced by parasympathetic activation. This translational evidence is of certain interest to optimize the pharmacological treatment of severe asthmatics in which a high intrinsic parasympathetic tone has been documented (Liccardi et al., 2016). Probably the current Global Initiative for asthma (GINA) recommendations (GINA, 2019) should be improved by considering this specific asthma phenotype in Step 5, in which combining an anti-IL-5/IL-5Rα mAb with a long-acting muscarinic antagonist (LAMA) as preferred controller could lead to clinical and functional benefits.
Airway sensitization contributes to the adhesion of eosinophils to parasympathetic nerves, leading to their priming, activation and degranulation with consequent release of major basic protein (MBP). MBP increases acetylcholine release due to loss of function of the neuronal M2 muscarinic autoreceptor expressed on postganglionic parasympathetic neurons (Drake et al., 2018). Since such an intimate interaction between eosinophils and airway cholinergic nerves contributes to the AHR in the course of tissue sensitization, favourable synergistic interaction could result by combining a LAMA, that inhibits the cholinergic transmission, with an anti-IL-5/IL-5Rα mAb that protects vagal fibres from the deleterious influence of activated eosinophils.
By a strict pharmacological viewpoint, the findings of our study suggest that the profile of loss of Emax to histamine and EFS in passively sensitized airways treated with either benralizumab and mepolizumab is due to an indirect inhibition of endogenous bronchoconstricting intermediaries release that, in turn, lead to the AHR. This evidence is supported by the fact that the AHR induced by histamine in passively sensitized bronchi is mediated not only by the direct activation of histaminergic receptors expressed on ASM, but also by the indirect facilitator effect of the acetylcholine release from the parasympathetic nerve (Cazzola et al., 2016a). Moreover, the AHR elicited by EFS is prevalently indirect and mediated by the sensitization of vagal fibres leading to increased release of endogenous acetylcholine from postganglionic parasympathetic nerves that in turn activate muscarinic M3 receptor expressed on ASM (Ichinose et al., 1996; Mitchell et al., 1993).
This research provides also ancillary results, mainly concerning the validity of passive sensitization of human isolated airways as a suitable model of AHR in asthma. Although already demonstrated at human ASM cellular level (Grunstein et al., 2002; Hakonarson et al., 1999a; Hakonarson et al., 1999b), here we provide for the first time the evidence that passive sensitization of the whole human bronchial tissue induces an extensive synthesis of IL-5, and that IL-5 represents a key factor leading to the AHR in human airways. Certainly targeting the IL-5 pathway counteracts the AHR in human subsegmental bronchi, nevertheless Manson et al. (Manson et al., 2019) have recently demonstrated that IL-5 does not induces hyperresponsiveness in human small airways. Perhaps the lack of IL-5 mediated AHR in small airways could be due to the preponderance of ASM in the bronchioles wall and the significantly lower, and almost absent, number of eosinophils in bronchioles compared to medium bronchi we have used in our experiments (Faul et al., 1997; Hyde et al., 2009). Furthermore, a possible bias leading to the absence of AHR in the bronchioles used by Manson et al. (Manson et al., 2019) is that their tissues were not passively sensitized, a procedure that induces AHR via IgE in the presence of further serum factors (Ichinose et al., 1996; Mitchell et al., 1997; Schmidt et al., 2000; Schmidt et al., 1999). In our study we have also found that the passive sensitization of subsegmental bronchi (4-6 mm inner diameter) elicits AHR in response to EFS. Since no augmentation to EFS was detected in previous studies in smaller airways (2-3 mm inner diameter) regardless of the frequency delivered (Mitchell et al., 1997), it can be assumed that the distribution of functional parasympathetic innervation at the level of airways ≤3 mm is sparse and/or hyporesponsive to passive sensitization (Calzetta et al., 2018a), making these smaller airways not appropriate to reproduce ex vivo the AHR mediated by vagal activation.
The main limitation of this study is intrinsic to the isolated bronchial model itself, as it permitted to characterize the effect of benralizumab and mepolizumab against AHR in a sub-acute ex vivo experimental setting but not after a long-term exposure. In this regard, since AHR has been proposed to be a main treatable trait toward precision medicine in eosinophilic asthmatic patients (Agusti et al., 2016; Bel et al., 2017), well designed head-to-head clinical trials are needed to compare the efficacy of chronic treatment with benralizumab and mepolizumab specifically against AHR in severe asthma.
Taken together, the findings of this study demonstrate that passive sensitization induces a massive release of IL-5 from human airways, and that targeting the IL-5/IL-5Rα axis with mAbs prevent in a concentration dependent manner the AHR in response to histamine, parasympathetic activation, and mechanical stress. Benralizumab, by blocking the IL-5Rα, resulted more potent than the anti-IL-5 mepolizumab, and the beneficial effects of both these agents were correlated with improved levels of cAMP in hyperresponsive airways. These observations also indicate that IL-5 is a key factor in determining AHR in passively sensitized airways. Further head-to-head studies evaluating the impact of long term treatment with benralizumab and mepolizimab against AHR are needed in severe asthmatic subjects, with specific focus on patients characterized by a high intrinsic parasympathetic tone.