Evaluating Heat Shock Proteins as Biomarkers for Vaginal Fungal Infections

Objective: The purpose of this study was to determine the biological association between host-derived HSP47 and fungal-derived HSP90 in the context of vulvovaginal candidiasis (VVC) and to examine their relationships with clinical, inflammatory, and metabolic phenotypes in infected and healthy women. Methods: This study followed a six-month case-control design (February-July 2025) and was conducted at the University of Tabuk Hospital in Tabuk, Saudi Arabia. A total of 84 women aged 18-45 years were recruited, of which 42 were VVC-infected, and 42 were healthy controls. ELISA kits were used to test vaginal swabs for HSP47 and HSP90. Clinical, hematological, cytokine, and metabolic markers were also evaluated. Mann-Whitney U, Spearman correlation, and multiple linear regression tests were performed to analyze the data. Results: The levels of HSP47 and HSP90 were significantly higher among infected patients (2.29 ng/mL and 3341 ng/mL, respectively) when compared with controls (0.58 ng/mL and 1025.7 ng/mL; p < 0.001). Women who were infected were older (p = 0.02), but there were no significant differences in terms of BMI (p = 0.29). The levels of vitamin D and adiponectin were significantly decreased (p < 0.001), while pro-inflammatory cytokines (IL-6, TNF-α, IFN-γ, TGF-β, and IL-8) and WBC counts were higher compared to the control group. The hematology results were characterized by inflammation-related anemia and disturbed protein metabolism. The ROC analysis demonstrated good diagnostic performance, with an AUC of 1.0 in the case of HSP47 and 0.905 in the case of HSP90. In the case of the infected patients, the regression models were found to be weak (HSP90 R² = 0.154; HSP47 R² = 0.273), although HSP47 retained significant connections with IL-8 (p = 0.005) and IFN-γ (p = 0.028). Conclusions: High levels of HSP47 and HSP90 are observed in VVC, reflecting an epithelial stress response and fungal persistence. These HSPs have high diagnostic accuracy, which justifies their potential as biomarkers for the timely detection of VVC; they also have further implications as early biomarkers for prognostic and treatment monitoring support, despite the poor predictive models. This study has some limitations that must be addressed; in particular, the regression analyses failed to provide statistically significant predictive models, likely due to the limited sample size. In addition, the specificity of HSP90 and HSP47 for VVC in comparison with other vaginal infections was not evaluated.