Risk stratification of cervical disease using detection of human papillomavirus (HPV) E4 protein and cellular MCM protein in clinical liquid based cytology samples

Stevenson, Andrew and Kavanagh, Kim and Pan, Jiafeng and Stevenson, Lynne and Griffin, Heather and Doorbar, John and Scott, Evelyn and Deeny, Miriam and Cuschieri, Kate and Graham, Sheila V. (2018) Risk stratification of cervical disease using detection of human papillomavirus (HPV) E4 protein and cellular MCM protein in clinical liquid based cytology samples. Journal of Clinical Virology, 108. pp. 19-25. ISSN 1386-6532 (https://doi.org/10.1016/j.jcv.2018.08.011)

[thumbnail of Stevenson-etal-JCV-2018-Risk-stratification-of-cervical-disease-using-detection-of-human-papillomavirus]
Preview
Text. Filename: Stevenson_etal_JCV_2018_Risk_stratification_of_cervical_disease_using_detection_of_human_papillomavirus.pdf
Final Published Version
License: Creative Commons Attribution 4.0 logo

Download (1MB)| Preview

Abstract

Background: While human papillomavirus (HPV) DNA testing offers high sensitivity for the detection of significant cervical disease, its specificity is suboptimal given the high prevalence of transient HPV infections (CIN1 or less). Biomarkers to identify those suffering from low grade disease from those with high grade disease could save healthcare costs and reduce patient anxiety. Objective: The objective of the present work was to develop and test an immunohistochemistry (IHC)-based dual viral and cellular biomarker strategy which was applicable to liquid based cytology (LBC) samples. Study design: We developed a novel IHC assay for detection of HPV E4 and cellular minichromosome maintenance (MCM) proteins in routinely taken cervical LBC samples using cytospin-prepared slides. The assay was applied to a prospective cohort of Scottish women referred to a colposcopy clinic due to preceding cytological abnormalities. The performance of the biomarkers for detection of clinically insignificant (CIN1 or less) versus significant disease was determined. Results: A total of 81 women were recruited representing 64 cases of<=CIN1 and 28 of CIN2+. Biomarker performance relative to histopathology outcomes showed high levels of MCM detection was significantly associated with CIN2+ (p=0.03) while E4 was detected more frequently in<=CIN1 (p=0.06). Conclusions: Combined detection of a host proliferation marker and a marker of viral gene expression could allow triage of cases of clinically insignificant disease prior to colposcopy. However, there was overlap between distributions of MCM levels in CIN2+ and<=CIN1 suggesting that additional biomarkers would be required for improved specificity. Combined with cytospin-prepared slides this approach could provide a means of risk stratification of disease in low resource settings.