Expert Sees Major Genomic Knowledge Gaps in Cervical Cancer

Murty Vundavalli, PhD

OncLive: What are the central genetic events underlying the development of cervical cancer?

More than 20 years ago, Murty Vundavalli, PhD, was among the first researchers to characterize genetic abnormalities in cervical cancer.¹ Today, in his laboratory at Columbia University, Vundavalli continues working to understand the genetic and epigenetic aspects of cervical cancer in an attempt to elucidate prognostic markers of response to treatment. Through genomewide studies, Vundavalli and colleagues have identified a variety of genomic alterations specific to cervical cancer, including amplification of the HER2 gene in squamous cell carcinoma of the cervix.¹ Currently, they are evaluating a predictive biomarker of apoptotic response to combination drug therapy.²Vundavalli: The development of invasive cervical cancer proceeds by a distinct series of morphological changes in squamous epithelium called cervical intraepithelial neoplasia (CIN). CINs have varying potential to progress to invasive cancer. Genetic changes that occur in CIN can be considered as primary or central events that cause cervical cancer. Identifying a list of such genetic hits in CINs and invasive cervical cancer provides tools for early detection, predicting risk of progression of CINs, and in testing as targets for treatment.

Recent high-throughput whole-genome single nucleotide polymorphism (SNP) arrays and sequencing methods have provided some clues about the recurrent genomic aberrations in cervical cancer. Copy number alterations of chromosomal regions involving gains of 3q, 5p, and 20q have been shown to be some of the earliest genomic changes. These are currently being used as biomarkers of early detection and risk prediction of CINs.

How has our understanding of the genomics of cervical cancer evolved in the past decade?

The mutation spectrum of genes identified by recent next-generation sequencing approaches seems to represent relatively late events. These recurrent somatic mutations would serve as central targets against specific gene(s)/pathway(s) for developing therapies.The genomic alterations that contribute to the development of cervical cancer are relatively less understood compared with other epithelial cancers such as breast and ovarian carcinomas. This shortage of knowledge about causal genetic mutations in cervical cancer has been overcome in the last decade by extraordinary progress due to innovations in array and next-generation sequencing technologies. We now have a broad understanding of focal copy number alterations (such as gains of chromosomal regions 1q, 3q, 5p, 20q, and losses on 3p, 2q, 11q), gene expression signatures, gene fusions, HPV integration sites, mutation spectrum in genes and, to a lesser extent, of epigenetic modifications.

How has this knowledge affected development of novel targeted therapies?

What are the most significant unanswered questions relating to the genomics of cervical cancer?

Further understanding of relationships between this mutational spectrum and the role it plays in specific genetic pathways will facilitate identification of targets for therapies.Among the genomic changes identified so far, the newly identified oncogenic driver mutations in genes such as PIK3CA, MAPK1, EP300, FBXW7, ERRB2, HLA-B, PTEN, TP53, STK11, KRAS, NFE2L2, ELF3, and CBFB, may eventually benefit from targeted therapies. However, it is currently unclear as to which of these will serve as biomarkers of therapy response to cervical cancer pending functional and clinical investigations. For example, mutations in PI3K/AKT pathway and ERBB2 may be targetable by specific tyrosine kinase inhibitor agents as in other tumor types. So far, no such clearly identified targeted therapies exist for cervical cancer.Although the recent unfolding of the molecular characterization of cervical cancer forms a backbone in our understanding of the genetics of cervical cancer, we are still limited in a number of areas of research. The incidence of cervical cancer differs geographically, likely accounting for etiological factors such as tobacco and oral contraceptive use, differences in incidence of high-risk HPV types, promiscuity, and parity.

However, differences in the genomic landscape of cervical cancer among various populations with specific etiologic exposures are still not known. Whether there are differences in genomic alterations in high- versus low-prevalence countries remains to be clarified.

Although several drugs are approved by the FDA to treat cervical cancer, these were not based on genomic changes. To successfully achieve personalized treatments in cervical cancer, a better understanding based on genetic mutations is needed. For example, functional characterization of target mutated genes/pathways and the development of patient-derived xenografts (PDX) to serve as preclinical models are required.

Unfortunately, no clinical data are available on treatment outcomes and survival based on genomic aberrations. To successfully target specific patients, treatment response in relation to one or more combination of genomic aberrations is essential. For example, PIK3CA mutations are the most common in cervical cancers. Since these mutations are outside of the kinase domain, whether the response to PIK3CA/AKT/ mTOR inhibitors will be similar to kinase domain mutations is not known.

Another major obstacle in our understanding of cervical cancer genomics is the lack of data on intratumor heterogeneity. Current genomic studies in cervical cancer represent the entire tumor and lack accounting for the intratumor heterogeneity. Such heterogeneity has been proven to play a key role in response to conventional as well as targeted therapies.

What influence does HPV have on the genetic landscape of cervical cancers?

One other challenging task ahead in cervical cancer is understanding the role epigenetic alterations play. Ultimately, our success in gaining insights into targeted therapies in cervical cancer relies on multidimensional approaches involving genomics, epigenetics, and functional models combined with integrative systems biology tools.Although the etiologic role of HPV is well established in cervical carcinogenesis, its influence on genome modifications seems relatively complex and poorly understood. High-risk HPV genomes can replicate episomally and integrate into the host genome at late stages of cancer, resulting in expression of E6 and E7 genes. Decades ago, it was shown that this overexpression of E6 and E7 serves as oncoproteins by inactivating TP53 and RB cellular pathways, respectively. Recent genomic investigations have pointed out that high-risk HPV types integrate into specific sites in the host genome, mainly targeting tumor suppressor genes (eg, RAD51B), or leading to increased copies of nearby oncogenic sequences (amplification). Correlation of HPV integration sites with interferon-gamma or HLA-A/B genes indicates its role in immune response.

Overall, whole-genome sequencing approaches by Ojesina et al3 and others have shown that the HPV integration causes host genomic instability resulting in disrupting or overexpressing nearby host genes that may play role in early stages of cervical carcinogenesis. A better and comprehensive understanding of the role of HPV on the host genome would likely help in designing better targeted therapies in the management of precancerous lesions to avoid progression to higher grades as well as invasive cervical cancer.

References

1. Mitra AB, Murty VV, Pratap M, et al. ERBB2 (HER2/neu) oncogene is frequently amplified in squamous cell carcinoma of the uterine cervix. Cancer Res. 1994;54(3):637-639.
2. Narayan G, Xie D, Ishdorj G, et al. Epigenetic inactivation of TRAIL decoy receptors at 8p12-21.3 commonly deleted region confers sensitivity to Apo2L/trail-cisplatin combination therapy in cervical cancer [published online November 6, 2015]. Genes Chromosomes Cancer. doi:10.1002/gcc.22325.
3. Ojesina AI, Lichtenstein L, Freeman SS, et al. Landscape of genomic alterations in cervical carcinomas. Nature. 2014;506(7488):371-375.