Background: Persistent infection with high-risk human papillomavirus (HR-HPV) is the primary etiological factor for cervical cancer, yet HPV DNA testing alone lacks specificity for predicting disease progression. Host protein alterations associated with HPV-driven epithelial transformation may provide complementary biomarkers for early detection and triage. Proteomic profiling of cervical exfoliated cells offers a minimally invasive approach suitable for community-based screening, particularly in low-resource settings.
Objectives: To identify HR-HPV–associated host protein signatures in cervical exfoliated cells using discovery proteomics and bioinformatics to understand pathways involved in HPV-driven epithelial transformation.To validate and assess the triage potential of selected protein biomarkers (transferrin, fibrinogen beta chain, and cytokeratin-13) across different HPV genotype groups using ELISA for improving specificity of community-based cervical cancer screening.
Methods: Cervical exfoliated cell samples collected from a South Indian community screening program were tested for HR-HPV using the cobas® 4800 platform. For discovery proteomics, 40 samples were selected by convenience sampling and grouped into HPV16-positive, HPV18-positive, other high-risk HPV (OHR)-positive, and HPV-negative (n = 10 per group). Pooled protein extracts were analyzed using two-dimensional gel electrophoresis (2D-GE) followed by MALDI-TOF mass spectrometry. Differentially expressed proteins were subjected to functional annotation, gene ontology, and STRING network analysis. For validation, ELISA was performed on 84 individual samples (HPV16: n = 42; HPV18: n = 8; OHR: n = 14; mixed infections: n = 10; HPV-negative: n = 10) to quantify serotransferrin (TF), cytokeratin-13 (CK13), and fibrinogen beta chain (FGβ). ELISA concentrations were interpolated using Morgan–Mercer–Flodin (MMF) regression models. Group comparisons were performed using the Mann–Whitney U test.
Results: Proteomic profiling identified 14 significantly upregulated protein spots corresponding to 9 unique proteins in HPV-positive samples, including TF, FGB, KRT13, KRT6A, KRT6B, KRT4, CALCOCO1, ZSCAN26, and ZNF462. Network analysis revealed two major functional clusters: (i) cytoskeletal and epithelial differentiation pathways dominated by keratins, and (ii) metabolic, inflammatory, and stress-response pathways involving TF, FGB, and CALCOCO1. ELISA validation showed significantly elevated levels of all three biomarkers in HPV-positive compared to HPV-negative samples (p < 0.000001). Median FGβ concentrations ranged from 11.29–14.89 ng/µL in HPV-positive groups versus 1.43 ng/µL in controls. Median transferrin levels were ~23.8 ng/µL in HPV-positive samples versus ~1.3 ng/µL in controls. CK13 levels were markedly higher in HPV-positive groups (median ~336 pg/µL) compared to HPV-negative samples (median ~54 pg/µL). Elevated biomarker levels were observed across HPV16, HPV18, OHR, and mixed-infection groups.
Conclusion: This study demonstrates that cervical exfoliated cell–based proteomic profiling can identify host protein alterations strongly associated with HR-HPV infection. The consistent upregulation and independent ELISA validation of transferrin, fibrinogen beta chain, and cytokeratin-13 highlight their potential as complementary biomarkers for HPV-positive triage and early cervical cancer risk assessment. These findings support the feasibility of integrating protein biomarker assays with HPV DNA testing to improve screening specificity and enable scalable community-level screening strategies in resource-limited settings.