Abstract 4249: Evidence of APOBEC3 editing in the HPV16 genome

Abstract

Abstract Human papillomavirus (HPV) is a very common sexually transmitted infection, however only a small proportion of women progress to cervical precancer or cancer. HPV16 is the most carcinogenic type, causing more than half of the cervical cancer globally. The HPV16 genome is 7,906base-pairs coding for 8 genes (E6, E7, E1, E2, E4, E5, L2, L1) and one upstream regulatory region (URR). Within HPV16 there are 4 main lineages (A, B, C, D) that are strongly associated with disease risk. Human APOBEC3A (hA3A) cytidine deaminases have been shown to have antiviral effects. The APOBEC mutational process results in a C to T base change at specific motifs (5’ [C/T]•C>T•W 3’). Previous studies established that there was evidence of APOBEC3 editing on a small number of HPV16 samples. It’s unknown how these mutations are related to infection clearance or the long-term accumulation of genomic mutations that contribute to HPV-associated cancers. We conducted detailed analyses to comprehensively evaluate APOBEC3 editing on HPV16 genomes using HPV16 whole-genome sequencing data from 3,215 HPV16-infected women in the NCI-HPV Persistence and Progression (PaP) cohort. Cases were defined as women with cervical precancer (CIN3, N = 1,093) or cancer (N = 109) and controls were women with no histologic evidence of precancer or cancer (<CIN2, N = 1,107). HPV16 DNA was extracted from banked specimens and whole-genome sequenced using a high-throughput assay. We evaluated all rare variants, defined as having a minor allele frequency of <1%, for matching an APOBEC3-associated variant. An APOBEC3-associated variant was defined as a variant having one of the eight possible motifs (5’ [C/T]•C>T•W 3’) out of 96 potential motifs of 3 base-pairs. Using logistic regression, we compared the number of APOBEC3-associated variants in cases and controls, among HPV16 variant lineages, and among genome regions of the virus. We discovered that there is evidence of APOBEC3 editing throughout the HPV16 genome. Specifically, we observed that women with precancer or cancer had less APOBEC3-associated variants compared to the controls (OR = 0.84, p-value = 0.06). We further showed that women with an HPV16 A lineage infection have more APOBEC3-associated variants compared to those with a non-A lineage infection (OR = 1.35, p-value = 0.02). After controlling for the number of APOBEC3 vulnerable loci, we observed that the L1 (OR = 0.23, p-value = 0.04) and E7 (OR = 0.29, p-value = 0.07) genes have less APOBEC3 footprints overall, and particularly in the cases compared to the controls in these regions, compared to the viral non-coding upstream regulatory region (URR). Overall, we found that APOBEC3 is not affecting the HPV16 genome in a uniform way, and instead, it appears to be targeting specific regions which could suggest antiviral activity. Importantly, we determined that APOBEC3-associated variants are less prevalent in cases which could be related to disease progression in these individuals. Further evaluation is underway. Citation Format: Yanzi Xiao, Bin Zhu, Meredith Yeager, Michael Cullen, Joseph Boland, Nicolas Wentzensen, Tina Raine-Bennett, Zigui Chen, Kai Yu, Qi Yang, Mia Steinberg, David Roberson, Sara Bass, Laurie Burdette, Thomas Lorey, Philip Castle, Robert Burk, Mark Schiffman, Lisa Mirabello. Evidence of APOBEC3 editing in the HPV16 genome [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2017; 2017 Apr 1-5; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2017;77(13 Suppl):Abstract nr 4249. doi:10.1158/1538-7445.AM2017-4249