EP041/#256 Endogenous retrovirus RNA expression differences between racial, stage and HPV cohorts offer improved prognostication among women with cervical cancer

Objectives

Endogenous human retroviruses (ERVs) are remnants of exogenous retroviruses that have been integrated into the human genome. Some ERVs may become activated allowing epigenetic alterations through DNA methylation or histone modification, which can further translate into altered gene regulation or transcription. This is a novel area of exploration in cervical cancer.

Methods

We applied ERV mapping tools to RNA-seq data from 63 cervical cancers to investigate expression of ~550,000 ERV elements from the Human Endogenous Retrovirus database (HERVd) to investigate ERV expression among various cohorts. We also investigated a prognostic model, supplementing a baseline prediction model using FIGO stage, age and HPV-positivity with ERVs.

Results

98 ERVs were differentially expressed (padj < 0.1), with Black American patients having 40 upregulated and 58 downregulated (including MER21C, HERVH-int) ERVs when compared to white American patients. Of the 138 ERVs differentially expressed between early-stage and locally advanced-stage groups, 38 were upregulated, including ERV3, and 100 were downregulated. 26,916 ERVs were differentially expressed between HPV positive and negative cohorts. There were significant differences in ERV3 protein expression (p = 0.000905). While clinical parameters are predictive of progression free survival at p = 0.06027, our supplemented model combining a 67-ERV panel and the clinical data, discriminated the two risk groups at p = 9.433 x 10¹⁵.

Conclusions

ERV RNA expression differences in cervical cancers is significantly different among racial cohorts, HPV-subgroups and disease stages. The correlation of ERV expression alongside clinical factors significantly improves prognostication when compared to clinical factors alone and may serve as future therapeutic targets.