**4. Discussion**

Being sensitized to fungi is a potential risk factor for worsening asthma outcomes, but the mechanisms underlying the fungal sensitization associated with poorer asthmatic outcomes remain unclear. The present study shows a link between cytokines and fungal sensitization associated with poorer asthmatic outcomes. Firstly, we observed that IL-6 and IL-17A, but no IgE, had a significant association with sensitization to most species of fungi in out testing panel. Only ED visit times were positively associated with IL-17A, and we

noticed that the level of IL-17A was higher in patients with *Candida albicans* sensitization who had ED visits compared with those who did not, though the difference is not significant. The mediation analysis revealed that IL-17A had positive mediation effect between *Candida albicans* sensitization and ED visit times. Overall, our finding indicated that sensitization to *Candida albicans* had a positive correlation with IL-17A levels, which were then associated with ED visits in asthmatics. Based on present results, IL-17A could be a biomarker for asthmatics with frequent emergency department visits for *Candida albicans* sensitization.

Sensitization to fungi is associated with increased asthma severity, poorer clinical outcomes, and mortality. In previous studies, the prevalence of fungal sensitization in asthma has varied. According to a UK study, 66% of patients with severe asthma were sensitized to one or more fungi, as determined by SPT or specific serum IgE testing or both [18]. Moreover, asthma patients referred to subspecialty clinics showed that 17.3% were allergic to fungi [19], but another study showed 76.3% to be allergic to at least one fungus [20]. Further, 32% of the 576 patients enrolled in the study with severe eosinophilic asthma demonstrated sensitivity to fungal allergens [21]. In our study, we found that 92.8% of asthmatic patients had fungal sensitization. These results indicate that the exact prevalence of fungal sensitization in patients with asthma remain unclear. It is likely that the large differences seen in the prevalence of fungal sensitization among asthmatic patients are related to differences in patient populations, testing methods, and geographic locations.

Fungal sensitization is associated with worse asthmatic outcomes being reported. For example, the number asthmatic patients with fungus sensitization requiring intensive care unit admission and mechanical ventilation was higher than those without fungus sensitization or nonfungal sensitization [19]. Furthermore, the lung function of patients sensitized to thermotolerant filamentous fungi was lower than that of patients not sensitized to any fungi [20]. Despite this, we were unable to find any association between fungal sensitization with measures of control, severity, or steroid use in our population. However, we found that patients with fungal sensitization had a higher prevalence of ED visits than patients without fungal sensitization (7.78% versus 0%, data not shown). These finding suggest that exposure to inhaled allergens to which patients are sensitized can increase asthma symptoms or precipitate exacerbation. According to GINA guidelines, patients with persistent symptoms and/or exacerbations should undergo allergen testing. It is possible that patients with fungal sensitization were not identified because of lack of testing [19]. Thus, greater awareness of asthmatics at risk for more difficult disease outcomes should support earlier identification of fungal sensitization.

Allergic asthma is defined by the presence of allergic sensitization and a correlation between allergen exposure and asthma symptoms. One biomarker of allergic asthma is total serum IgE level, which is more commonly elevated in allergic compared with nonallergic asthma [22], is inversely associated with lung function in asthmatics [23], and is associated with the prevalence of asthma [24]. A previous study reported that patients with fungal sensitization (*Penicillium chrysogenum*, *Cladosporium herbarum*, *Aspergillus fumigatus*, *Mucor racemosus*, *Stemphylium herbarum*, and *Alternaria alternata*) had a higher total serum IgE concentration than patients with no sensitization or nonfungal sensitization [19]. In present study, our data demonstrate that sensitization to most species of fungi had a positive relationship with IL-6 and IL-17 rather than IgE. In previous study, eosinophils and eosinophilic production of IL-23 and IL-17 were shown to be beneficial in invasive aspergillosis but detrimental in allergic disease in a mice model [12]. In addition, IL-6 is essential for mucus hypersecretion by airway epithelial cells triggered in response to inhaled *Aspergillus fumigatus* extract, which was found in a mouse model of allergic airway inflammation induced by direct airway exposure to extracts of *Aspergillus fumigatus* [25]. These findings suggest that the immune response regarding fungal sensitization is not only associated with IgE, but also other mechanisms involved that are related to IL-6/IL-17A axis. In allergic asthma, B cells may regulate the T cell response by modulating the phenotypic response. For example, in a JH−/<sup>−</sup> (B cell-deficient) murine model of fungal allergic asthma, levels of the inflammatory cytokines IL-6 and IL-17A were significantly elevated, and there

was significantly more robust airway eosinophilia and neutrophilia [26]. The evidence suggests that the IL-6/IL-17A axis is associated with fungal allergic asthma in conditions with B cell deficiency. Thus, the IL-6/IL-17A axis would be a possible mechanism associated with fungal sensitization other than IgE.

Accumulating evidence now suggests that Th17 cells and their related cytokines are also involved in the pathophysiology of allergic asthma. IL-17 expression is increased in the lung, sputum, bronchoalveolar lavage fluid (BALF), and sera in patients with asthma, and the severity of AHR is positively correlated with IL-17 expression levels [27,28]. In our study, IL-17, but not IL-6, had a positive correlation with ED visits and had a significant mediation effect on the association between *Candida albicans* sensitization and ED visit. A previous study found that IL-17 increased in *Aspergillus fumigatus*-sensitized mice [29]. In addition, IL-17A response is associated strongly with acute ABPA, and its specific decline in response to ABPA treatment suggests that this atypical Th17 response plays an active role in manifesting and/or exacerbating disease [30]. The evidence supports the hypothesis that IL-17A has a mediated effect on the relationship between *Candida albicans* sensitization and poorer clinical outcomes, particularly ED visits. A previous study identified the role of IL-17-mediated immunity in Candidiasis, and the implications for clinical therapies for both autoimmune conditions and fungal infections [31]. In asthma, to our best knowledge, our findings firstly provide evidence that IL-17A is a potential mediator to link *Candida albicans* sensitization and poorer clinical outcomes. Despite the underlying cell-mediated mechanism is unclear, a study report that protective lung Th2 and Th17 cell responses against the common mucosa-associated fungus *Candida albicans* are coordinated through lung megakaryocytes and platelets [32]. The findings pointed suggested that Th17, an IL-17-producing cell, may have a protective effect against allergy to *Candida albicans* in the lungs.

Th17 immune responses differ between the sexes due to suppressive and enhancing effects of sex hormones [33]. A previous study indicated testosterone is associated with expansion of Th17 populations in a murine model of experimental autoimmune encephalomyelitis (EAE) [34]. However, a recent study indicated that the increased production of IL-17A by 17β-estradiol and progesterone from TH17 cells may provide a potential mechanism for the increased prevalence of severe asthma in women compared with men [35]. In contrast, estrus levels of estradiol downregulated the Th17 response to *Candida albicans* in in vivo vaginal infection models [36]. In our finding, we found that the IL-17 levels in asthmatic patients were 3.28 and 4.17 in males and females, respectively, but there is no significance between sex differences (*p* = 0.369) (Supplemental Table S1). The correlation between IL-17 and *Candida albicans* was 0.233 and 0.458 in males and females, respectively, but there are no significant differences between males and females in the correlation between IL-17 and *Candida albicans* (*p*interaction = 0.166) (Supplemental Table S1). The finding demonstrates that sex difference has no influence on the correlation between IL-17 and *Candida albicans* in our population. Taken together, we suggest that the influence of sex difference on IL-17 is inconsistent that may depend on disease-specific characteristics or severity; thus, the mediation effect of IL-17 between poor outcomes and *Candida albicans* across sex difference is needs further study.

There are some limitations of our study. Firstly, further evaluation in a multiple center setting is needed to expand the generalizability. Secondly, we lacked information regarding immune cell type; therefore, the cell-mediated immune response related to IL-17A in fungal sensitization, especially in *Candida albicans*, needs to be further investigation. Thirdly, since some of our asthma patients were without ED visits, the causal effect is difficult to clarify due to limitation of statistical method.
