Association between fungi and postoperative recurrence in eosinophilic chronic rhinosinusitis

Eosinophilic chronic rhinosinusitis (ECRS) is a refractory subtype of chronic rhinosinusitis (CRS) characterized by type 2 (T2) inflammation and frequent postoperative recurrence.¹ Unlike allergic fungal rhinosinusitis,² ECRS is not categorized as a fungal disease, and fungi are rarely detected by culture-based methods.³ However, both conditions share an eosinophilic inflammatory background. Conventional culture has limited sensitivity, leaving unresolved whether fungi are consistently present in sinonasal tissues. To address this question, next-generation sequencing of the fungal internal transcribed spacer (ITS) region was used to assess fungal detection in CRS, its association with postoperative recurrence in ECRS, and its relationship with epithelial cytokine responses. Some of the results of these studies have been previously reported in the form of an abstract.⁴

This study was approved by the ethics committees of the participating institutions (20150070, H23-30, and 1775, respectively), and written informed consent was obtained from all patients. CRS was diagnosed according to the European Position Paper on Rhinosinusitis and Nasal Polyps (EPOS) 2020 criteria,³ and ECRS was classified using the Japanese Epidemiological Survey of Refractory Eosinophilic Chronic Rhinosinusitis (JESREC) study criteria.⁵ Control subjects were patients with nasal septum deviation, hypertrophic rhinitis, or sinonasal cysts. Middle meatus swabs were collected from 139 patients with CRS (84 ECRS and 55 non-ECRS) and 35 control subjects. After quality control, 134 patients with CRS (81 ECRS and 53 non-ECRS) and 34 control subjects were included in the analysis (Table 1). No subjects showed computed tomographic findings suggestive of fungal balls or calcifications, and no histopathological evidence of fungi was observed. DNA was extracted, ITS sequences were amplified with multiple primers and sequenced according to the Illumina-supplied protocol (1000000064940 v01), and taxonomy was assigned using the UNITE database (v01.12.2017). Fungi were consistently detected across all groups (Figure 1A). The core mycobiome, defined as taxa present at ≥1% relative abundance in ≥10% of patients with CRS, included 7 genera: Malassezia, Candida, Cladosporium, Aspergillus, Alternaria, Trametes, and Schizophyllum. A comparison of the mycobiome between ECRS and non-ECRS patients revealed no significant differences in α-diversity (P = .832) or β-diversity (P = .871). Regarding clinical phenotype, detection of Alternaria was significantly associated with the presence of nasal polyps (NPs) in the entire CRS cohort (P = .027). Subgroup analysis indicated that this association was driven by the non-ECRS group (P = .031). No significant association was observed within the ECRS subgroup (P = 1.0), which likely reflects the high prevalence of NPs among patients with ECRS (96.3%).

Table 1.

Comparison of characteristics between patients with chronic rhinosinusitis and control subjects.

Characteristic	Control	CRS			Non-ECRS vs ECRS
	(n = 34)	Total (n = 134)	Non-ECRS (n = 53)	ECRS (n = 81)	P value
Age, y, mean ± SD	39.0 ± 17.0	51.2 ± 14.7	49.3 ± 16.3	52.4 ± 13.5	.237
Sex, male	25 (73.5)	98 (73.1)	38 (71.7)	60 (74.1)	.843
Eosinophils in peripheral blood, %, median (IQR)	3.7 (1.2-6.1)	4.9 (2.8-7.1)	2.8 (1.6-4.9)	5.6 (4.2-8.7)	<.001***
Complication
Asthma	1 (2.9)	33 (24.6)	1 (1.9)	32 (39.5)	<.001***
AERD	0 (0)	7 (5.2)	0 (0)	7 (8.6)	.041*
Nasal polyps	–	109 (79.6)	31 (55.4)	78 (96.3)	<.001***
Lund–Mackay score, median (IQR)	–	11 (6-17)	6 (4-9)	14 (11-19)	<.001***
JESREC score, median (IQR)	–	11 (8-15)	7 (4-8)	15 (11-15)	<.001***
Preoperative medicine
Oral steroids	6 (17.7)	27 (20.1)	4 (7.6)	23 (28.4)	.003**
Inhaled steroids	0 (0)	29 (21.6)	2 (3.8)	27 (33.3)	<.001***
Nasal steroids	16 (47.1)	43 (32.1)	10 (18.9)	33 (40.7)	.008**
Antibiotics	2 (5.9)	70 (52.2)	36 (67.9)	34 (42.0)	.004**
Total IgE, IU/mL, median (IQR)	160.5 (46.0-521.5)	131.0 (61.1-392.0)	103.0 (39.3-295.0)	179.0 (69.1-482.5)	.083
Fungus-specific IgE, IU/mL, median (IQR)
Candida	0 (0-0.2)	0 (0-0.2)	0 (0-0.2)	0 (0-0.2)	.692
Aspergillus	0 (0-0)	0 (0-0)	0 (0-0)	0 (0-0)	.381
Alternaria	0 (0-0)	0 (0-0)	0 (0-0)	0 (0-0)	.205
Malassezia	0 (0-0)	0 (0-0)	0 (0-0.1)	0 (0-0.1)	.436
Cladosporium	0 (0-0)	0 (0-0.1)	0 (0-0)	0 (0-0)	.393

Data are presented as No. (%) unless otherwise indicated.

Abbreviations: AERD, aspirin-exacerbated respiratory disease; CRS, chronic rhinosinusitis; ECRS, eosinophilic chronic rhinosinusitis; IQR, interquartile range; JESREC, Japanese Epidemiological Survey of Refractory Eosinophilic Chronic Rhinosinusitis; SD, standard deviation.

^* P <.05.

^** P <.01.

^*** P <.001.

Figure 1.

Association between sinonasal mycobiome and postoperative recurrence risk in eosinophilic chronic rhinosinusitis (ECRS). (A) Genus-level differences in the relative abundances of internal transcribed spacer sequences. The bar charts represent the relative abundance of core fungal genera detected at ≥1% relative abundance in ≥10% of patients with chronic rhinosinusitis. Each stacked bar represents the fungal composition at the individual level. (B) Kaplan–Meier curves showing postoperative recurrence-free survival among patients with Alternaria-detected vs -undetected ECRS, and among patients with Candida-detected vs -undetected ECRS. (C) Comparison of α-diversity (Shannon index) and β-diversity (principal coordinates analysis) between patients with Alternaria-detected vs -undetected ECRS. (D) Gene expression levels of IL-25, IL-33, and thymic stromal lymphopoietin (TSLP) in nasal polyps between patients with Alternaria-detected vs -undetected CRS. *P <.05.

Next, the relationship between fungi and postoperative recurrence was evaluated in patients with ECRS. Postoperative recurrence was defined as a nasal polyp score ≥1 on endoscopic examination during follow-up.⁶ Kaplan–Meier analysis showed that patients with Alternaria-detected ECRS had significantly shorter recurrence-free survival than did patients with Alternaria-undetected ECRS (P <.001, Figure 1B). Candida detection was also associated with recurrence (P = .015). After false discovery rate correction, the association between Alternaria detection and recurrence remained statistically significant (adjusted P = .002), whereas the association with Candida was not statistically significant (adjusted P = .053). In a multivariate Cox proportional hazards model including Alternaria detection, Candida detection, age, and sex, Alternaria remained significantly associated with recurrence in patients with ECRS (hazard ratio [HR], 2.81 [95% CI, 1.28-6.16]), whereas Candida showed no significant association. In contrast, in the non-ECRS group, Candida detection, but not Alternaria detection, was associated with recurrence (HR, 5.88 [95% CI, 1.10-31.34]). These findings indicate that the presence of Alternaria may independently be associated with an increased risk of recurrence in patients with ECRS.

Given the association between Alternaria and recurrence in ECRS, the relationship between Alternaria presence and distinct microbial profiles or inflammatory responses was further examined. The α-diversity was significantly higher in patients with Alternaria-detected ECRS than in patients with Alternaria-undetected ECRS (P = .037, Figure 1C). The β-diversity showed distinct clustering between the 2 groups (P = .001, Figure 1C). Aspergillus, Candida, and Cladosporium were significantly increased in patients with Alternaria-detected ECRS (Aspergillus: adjusted P <.001, linear discriminant analysis (LDA) score = 4.25; Candida: adjusted P = .004, LDA score = 4.32; Cladosporium: adjusted P = .040, LDA score = 3.99). This pattern suggests that Alternaria detection is associated with differences in fungal community composition. Finally, consistent with the broader association between Alternaria and NPs observed in the overall cohort, the expression of epithelial cytokine genes was quantified by real-time PCR using all available NP tissues. While no differences were found in IL-25 and thymic stromal lymphopoietin (TSLP) expression, IL-33 expression was significantly higher in patients with Alternaria-detected CRS compared to patients with Alternaria-undetected CRS (IL-25, P >.05; IL-33, P = .037; TSLP, P >.05; Figure 1D), although this significance was attenuated after multiple testing correction (adjusted P = .070). Stratification by ECRS status demonstrated a borderline trend in patients with ECRS (P = .066 vs P = .510 in non-ECRS), likely reflecting limited statistical power. These exploratory findings suggest a potential link between Alternaria and epithelial activation in the eosinophilic environment, warranting validation in larger cohorts.

Alternaria is well-documented as an inducer of T2 inflammation in airway mucosa and stimulates epithelial cytokines including IL-25, IL-33, and TSLP.⁷ Moreover, group 2 innate lymphoid cells (ILC2s), a known source of T2 cytokines, have been implicated in the pathogenesis of CRS with NPs⁸ and respond to Alternaria-induced IL-33 expression from airway epithelium, amplifying T2 inflammation.⁹ A recent study using an allergen detection method in NPs reported that Alternaria allergen was detected in NPs and that higher Alternaria allergen levels in NPs correlated with increased postoperative recurrence, elevated T2 cytokine expression, and eosinophil activation.¹⁰ In the present cohort, approximately 20% of patients with ECRS tested positive for Alternaria (Figure 1A), and recurrence rates were notably higher in this group. In addition, patients with Alternaria-detected CRS exhibited increased expression of IL-33 irrespective of eosinophilic inflammation status. Collectively, these findings suggest that Alternaria detection is associated with increased epithelial IL-33 expression and a T2-related inflammatory environment. Preoperative antibiotic use did not significantly affect fungal α-diversity (P = .815) or β-diversity (P = .211), nor Alternaria detection (P = 1.0). These results suggest that preoperative antibiotic use was unlikely to be a major confounder. In addition, the detection of fungal genera, including Alternaria, in control subjects without CRS indicates that sinonasal fungal colonization is not necessarily indicative of pathogenic infection. However, the direction of this relationship cannot be determined.

In conclusion, ECRS, conventionally classified as a nonfungal CRS subtype, may contain distinct fungal signatures associated with postoperative recurrence. Recognizing Alternaria as a potential risk factor for recurrence highlights the need to revise the diagnostic framework for ECRS. Establishing criteria for Alternaria-associated ECRS and exploring its clinical management could improve outcomes for this distinct clinical phenotype.

Author contributions

Designed the study: M.K., E.N. Involved in data acquisition: N.A., K.K., S.A., K.O. Involved in methodology: N.A., M.K., Y.I., E.N. Conducted analysis: N.A., M.K., R.I., Y.I., K.K., E.N. Interpreted the data: All authors. Drafted the manuscript: N.A., M.K., A.K., E.N.

Supplementary material

Supplementary material is available at American Journal of Respiratory and Critical Care Medicine online.

Conflicts of interest

S.F. is an adviser for GlaxoSmithKline, Kyowa Kirin, and Sanofi; received honoraria from Kyorin Pharmaceutical, Taiho Pharma, and Mitsubishi Tanabe Pharma; and received research funding from Maruho, Tsumura, Mitsubishi Tanabe Pharma, and Sanofi. The other authors have no conflicts of interest to declare.

Funding

This study was funded by a grant from the Japan Society for the Promotion of Science (KAKENHI grant numbers 20K22998, 21H03086, 21K16827, and 21KK0287), a Health Labour Sciences Research Grant (H30-Nanchitou (nan)-Ippan-016), the Japan Agency for Medical Research and Development (AMED, numbers JP22ek0410077 and JP23jf0126005), and the MSD Life Science Foundation, Public Interest Foundation Grant Number RA-017. It was also partially supported by the Society for Promotion of International Oto-Rhino-Laryngology (SPIO), Japan, through the SODA TOYOJI SPIO Scholarship 2022 (Number SS22005), and by research fellowships from the Uehara Memorial Foundation and the Mochida Memorial Foundation for Medical and Pharmaceutical Research.