Skip to main page content
Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2017 Feb 21;8(8):13375-13386.
doi: 10.18632/oncotarget.14533.

Identification and Characterization of HPV-independent Cervical Cancers

Free PMC article

Identification and Characterization of HPV-independent Cervical Cancers

Carolyn E Banister et al. Oncotarget. .
Free PMC article


Background: Human papillomavirus (HPV) initiates cervical cancer, and continuous expression of HPV oncogenes E6 and E7 is thought to be necessary to maintain malignant growth. Current therapies target proliferating cells, rather than specific pathways, and most experimental therapies specifically target E6/E7. We investigated the presence and expression of HPV in cervical cancer, to correlate HPV oncogene expression with clinical and molecular features of these tumors that may be relevant to new targeted therapies.

Results: While virtually all cervical cancers contained HPV DNA, and most expressed E6/E7 (HPV-active), a subset (8%) of HPV DNA-positive cervical cancers did not express HPV transcripts (HPV-inactive). HPV-inactive tumors occurred in older women (median 54 vs. 45 years, p = 0.02) and were associated with poorer survival (median 715 vs 3046 days, p = 0.0003). Gene expression profiles of HPV-active and -inactive tumors were distinct. HPV-active tumors expressed E2F target genes and increased AKT/MTOR signaling. HPV-inactive tumors had increased WNT/β-catenin and Sonic Hedgehog signaling. Substantial genome-wide differences in DNA methylation were observed. HPV-inactive tumors had a global decrease in DNA methylation; however, many promoter-associated CpGs were hypermethylated. Many inflammatory response genes showed promoter methylation and decreased expression. The somatic mutation landscapes were significantly different. HPV-active tumors carried few somatic mutations in driver genes, whereas HPV-inactive tumors were enriched for non-synonymous somatic mutations (p-value < 0.0000001) specifically targeting TP53, ARID, WNT, and PI3K pathways.

Materials and methods: The Cancer Genome Atlas (TCGA) cervical cancer data were analyzed.

Conclusions: Many of the gene expression changes and somatic mutations found in HPV-inactive tumors alter pathways for which targeted therapeutics are available. Treatment strategies focused on WNT, PI3K, or TP53 mutations may be effective against HPV-inactive tumors and could improve survival for these cervical cancer patients.

Keywords: APOBEC; CTNNB1; HPV; TP53; cervical cancer.

Conflict of interest statement




Figure 1
Figure 1. Unsupervised classification of cervical cancers by HPV gene expression
HPV-active (circles) and HPV-inactive (squares) differ by total gene and oncogene expression levels.
Figure 2
Figure 2. Diagnosis (A) and Survival (B) are compared between HPV-active (dashed line) and HPV-inactive (solid line) patients
Patients with HPV-inactive tumors were on average 9 years older at diagnosis and died on average 6.4 years earlier.
Figure 3
Figure 3. Gene expression (A) and DNA methylation (B) in HPV-active (horizontal axis) and HPV-inactive (vertical axis) cervical tumors
The majority of significantly different genes had increased expression in HPV-inactive tumors. The majority of significantly different CpG loci were decreased in DNA methylation in HPV-inactive tumors.
Figure 4
Figure 4. Interaction between DNA methylation and gene expression landscapes
(A) Distribution of the DNA methylation test statistics (HPV-inactive vs. HPV-active) of the significant differently methylated CpGs. The majority of gene-associated and unclassified CpGs were hypomethylated in HPV-inactive tumors (negative test statistic), whereas promoter-associated CpGs were equally divided between hypermethylated (positive test statistic) and hypomethylated (negative test statistic). (B) Distribution of the gene expression test statistics (HPV-inactive vs. HPV-active) of the genes associated with significant differently methylated CpGs. Test statistics (and expression levels) of genes near gene-associated or unclassified CpGs were not correlated with CpG methylation levels, whereas test statistics (and expression levels) of genes near promoter-associated CpGs were inversely correlated with CpG methylation levels.
Figure 5
Figure 5. Cancers clustered by somatic mutation profiles
Samples were clustered using the –log transformation of the p-value obtained from MutSigCV. Unsupervised clustering was performed using Pearson dissimilarity with complete linkage. A total of 7882 genes were significantly mutated (MutSigCV v0.9 p-value < 0.05) in at least one cohort.
Figure 6
Figure 6. Somatic mutation differences between HPV-active and HPV-inactive cervical cancers
Pink bars highlight significant genes with odds ratios larger than the 95% confidence interval.

Similar articles

See all similar articles

Cited by 21 articles

See all "Cited by" articles


    1. Walboomers JM, Jacobs MV, Manos MM, Bosch FX, Kummer JA, Shah KV, Snijders PJ, Peto J, Meijer CJ, Munoz N. Human papillomavirus is a necessary cause of invasive cervical cancer worldwide. J Pathol. 1999;189:12–19. - PubMed
    1. Ho GY, Bierman R, Beardsley L, Chang CJ, Burk RD. Natural history of cervicovaginal papillomavirus infection in young women. N Engl J Med. 1998;338:423–428. - PubMed
    1. Akagi K, Li J, Broutian TR, Padilla-Nash H, Xiao W, Jiang B, Rocco JW, Teknos TN, Kumar B, Wangsa D, He D, Ried T, Symer DE, et al. Genome-wide analysis of HPV integration in human cancers reveals recurrent, focal genomic instability. Genome Res. 2014;24:185–199. - PMC - PubMed
    1. Ojesina AI, Lichtenstein L, Freeman SS, Pedamallu CS, Imaz-Rosshandler I, Pugh TJ, Cherniack AD, Ambrogio L, Cibulskis K, Bertelsen B, Romero-Cordoba S, Trevino V, Vazquez-Santillan K, et al. Landscape of genomic alterations in cervical carcinomas. Nature. 2014;506:371–375. - PMC - PubMed
    1. Tang KW, Alaei-Mahabadi B, Samuelsson T, Lindh M, Larsson E. The landscape of viral expression and host gene fusion and adaptation in human cancer. Nat Commun. 2013;4:2513. - PMC - PubMed