Genome Sequencing Is Mapping a Path to Personalized Treatment for Ovarian Cancer

Oncology Live®, Vol. 17/No. 16, Volume 17, Issue 16

Genome sequencing studies have uncovered an array of distinct genomic drivers underlying various ovarian cancer subtypes. If researchers can capitalize on these discoveries, it may offer a path to more individualized and effective treatment options.

Ovarian cancer, still known as the “silent killer,” continues to present a significant clinical challenge despite recent therapeutic advances that include several biologically targeted drugs. The universally applied treatment paradigm is clearly unsuited to the histological, clinical, and molecular heterogeneity of this disease.

The Silent Killer

Genome sequencing studies have uncovered an array of distinct genomic drivers underlying various ovarian cancer subtypes. If researchers can capitalize on these discoveries, it may offer a path to more individualized and effective treatment options.With more than 14,000 deaths expected in the United States in 2016, ovarian cancer remains the most lethal gynecologic malignancy and weighs in at No. 6 among the leading causes of cancer-related mortality. The ambiguity of early symptoms and lack of effective screening techniques result in more than 70% of patients being diagnosed in the advanced stages of disease.

The current standard of care has remained unchanged for many years and involves surgery and platinum- and taxane-based chemotherapy. Although the vast majority of patients will initially respond, disease recurrence is almost universal. Several targeted therapies have been developed for the treatment of recurrent disease. The realization that angiogenesis, the formation of new blood vessels from preexisting ones, is a hallmark of ovarian cancer led researchers to investigate the potential of bevacizumab (Avastin) and other antiangiogenic therapies.

The addition of bevacizumab to standard chemotherapy in platinum-resistant patients significantly improved progression-free survival in the phase III AURELIA trial and sealed its approval as second-line treatment for ovarian cancer in 2014.

Meanwhile, as in breast cancer, germline mutations in the breast cancer susceptibility genes BRCA1 and BRCA2 have also been discovered in ovarian cancer. These mutations impair the homologous recombination (HR) pathway used to fix DNA damage. Inhibitors of the poly(ADP)-ribose polymerase (PARP) enzyme, which plays a role in an alternative DNA repair pathway, have been developed to capitalize on the inability of ovarian cancer cells with defective HR pathways to repair damaged DNA.

In December 2014, olaparib (Lynparza) became the first PARP inhibitor to be granted approval by the FDA for the treatment of patients with BRCA mutation-positive ovarian cancer following the demonstration of response rates of more than 30% in a single-arm, phase II trial.

Histological Divisions

The addition of olaparib and bevacizumab to the ovarian cancer armamentarium marked the first new therapies approved for the treatment of ovarian cancer in almost a decade. Yet, only short-term survival gains have been achieved and long-term survival has remained stubbornly low, with 5-year rates of just 45%.Broadly, ovarian tumors are classed according to the anatomic structures from which they are presumed to originate. Between 90% and 95% of cases are epithelial ovarian carcinomas, while nonepithelial tumors such as sex cord-stromal tumors and malignant ovarian germ cell tumors are considerably rarer.

Epithelial ovarian cancers can be further divided into several different histological subtypes. Most common are serous tumors, responsible for at least 65% of epithelial ovarian cancer cases, followed by endometrioid, mucinous, and clear cell. Historically, it was thought that epithelial ovarian carcinomas were all derived from the ovarian surface epithelium but more recent studies have defined a new model of carcinogenesis that divides epithelial ovarian cancers into 2 categories, designated as type I and type II.

Type I tumors consist of low-grade serous, mucinous, endometrioid, and clear cell tumors, for which precursor lesions within the ovary have been clearly described. Type II tumors include predominantly high-grade serous cancers (HGSCs) as well as carcinosarcomas and high-grade endometrioid and poorly differentiated ovarian cancers.

Precursor ovarian lesions have not been well described for this group of tumors, and it is now thought that they can arise from the epithelium of both the ovarian surface and the fallopian tube. In addition to their histological differences, ovarian cancers also display distinct clinical behaviors.

Type II tumors are typically much more aggressive and more often found at an advanced stage, but they respond well to chemotherapy, with an average response rate of 70%. Type I tumors are more indolent, and are frequently identified earlier on in the course of disease, but are significantly more chemoresistant with a response rate of less than 30%.

Genomic Chaos in High-Grade Serous Subtype

Despite these histological and clinical differences, ovarian cancers are still tarred with the same brush when it comes to treatment. However, genome sequencing studies are beginning to reveal unique molecular underpinnings that might offer a path to more individualized treatments.Since HGSC represent the vast majority of ovarian cancer cases, these have largely been the focus of molecular studies. In 2011, The Cancer Genome Atlas (TCGA) Research Project published a comprehensive analysis of 489 HGSC samples, which included whole-exome sequencing on more than 300 tumor samples and matched normal pairs.

The study involved hundreds of researchers from more than 80 different institutions and established that the most common molecular alteration in HGSC was a mutation in the guardian of the genome, the TP53 gene, which was present in 96% of all samples. The pivotal role of TP53 in maintaining genome stability, through its effects on cell cycle progression, apoptosis, DNA repair, and other important cellular functions, means that a mutation in this gene can generate genomic chaos.

Beyond TP53, few genes are recurrently mutated in HGSC. The TCGA identified just 8 other recurrently mutated genes and the frequency of mutations for individual genes were low. Germline mutations in BRCA1/2 were observed in 9% of patients, with somatic mutations in a further 3%, while RB1 (2%), NF1 (4%), FAT3 (6%), CSMD3 (6%), GABRA6 (2%), and CDK12 (3%) were also recurrently mutated.

Notably, large structural changes to the genome, in which genes are accidentally deleted or duplicated as a result, appear to be a major driver of ovarian cancer development, making it a so-called C-class tumor (a tumor driven by copy number alterations).

DNA Repair Defects

The most commonly amplified genes, each found in more than 20% of cases, were CCNE1, MYC, and MECOM. CCNE1 encodes the cyclin E1 gene, with a major role in cell-cycle regulation. MECOM and MYC are transcription factors that regulate the activity of numerous cellular signaling pathways involved in a variety of cancer hallmark processes. The most frequent losses were in the NF1, RB1, and PTEN genes, all known tumor suppressors.It has been suggested that cancers that share the molecular features of BRCA-mutant tumors may respond similarly to PARP inhibitors and other drugs designed to take advantage of defective HR pathways—a property dubbed BRCAness.

Another significant finding of the TCGA Research Project’s study of ovarian cancer was that up to half of all cases of HGSC display HR defects. Although these are predominantly caused by germline and somatic mutations as well as epigenetic alterations in the BRCA1/2 genes, other types of BRCAness also existed.

These included mutations in other components of the HR pathway such as in Fanconi anemia genes, principally PALB2, FANCA, FANCI, FANCL, FANCC; the RAD genes, including RAD50 and RAD51; and DNA damage response genes ATM, ATR, CHEK1, and CHEK2.

In addition to HR pathway alterations, genes in other DNA repair pathways such as the nucleotide excision repair and mismatch repair pathways were mutated or deleted in a further 8% of HGSC samples. Interestingly, a recent study that used genome sequencing to analyze the possible mechanisms of resistance and recurrence in ovarian cancer found that restoration of the HR pathway was a common means of achieving chemoresistance— for example, through the reversion to wild-type BRCA1/2 genes or the removal of BRCA gene promoter methylation.

Other Ways of Classifying HGSC

The other half of HGSC samples that do not display HR deficiency remain poorly characterized, although up to one-fifth of these tumors have a focal amplification of CCNE1. According to the study on resistance mechanisms, CCNE1 amplification is the most frequent cause of primary resistance to therapy, which could in part explain why patients with BRCA1/2 mutations have a more favorable response to chemotherapy.The TCGA also performed gene expression profiling on HGSC samples to examine how sets of genes are expressed in ovarian cancer and what implications this could have for therapy or what it might mean for patient survival.

Researchers identified 4 transcriptional subtypes within ovarian cancers. An “immunoreactive” group was characterized by expression of the T-cell chemokine ligands CXCL11 and CXCL10 and the receptor CXCR3. So-called “differentiated” tumors were associated with an increased expression of MUC16 and MUC1 and expression of the secretory fallopian tube marker SLP1.

A “proliferative” subgroup displayed high expression of genes involved in proliferation, including MCM2 and PCNA, while those classed as “mesenchymal” had markers that suggested a high level of stromal components, such as fibroblasts, indicated by the expression of the FAP gene.

Molecular Distinctions

The different gene expression groups did not translate into variations in patient survival. However, the researchers did identify a 193-gene expression signature that was predictive of overall survival.Although other types of ovarian cancer have not been as well characterized genomically as HGSC, what has emerged from genome sequencing studies is that the subtypes of ovarian cancer are as distinct in their genome as in their histology and clinical behavior. Since type I ovarian cancers are significantly less responsive to chemotherapy than type II tumors, therapies designed to target the underlying genomic drivers may be particularly important.

A key driver of clear cell carcinoma is the ARID1A gene, which encodes a member of SWI/ SNF family of proteins that acts as a chromatin remodeling complex. ARID1A is mutated or deleted in up to half of clear cell ovarian cancers. Clear cell cancers are also characterized by mutations in the PIK3CA gene in 20% to 33% of cases, driving activation of the phosphatidylinositol- 3-kinase (PI3K)/AKT pathway. Other features of this subtype are overexpression of the proinflammatory cytokine interleukin-6 (IL-6) and of hypoxia inducible factor 1 alpha (HIF1α).

The activity of HIF1α supports the use of antiangiogenic therapy in ovarian cancer since one of the targets of HIF1α is vascular endothelial growth factor (VEGF). Gene expression profiling of clear cell carcinomas has further reinforced this idea, demonstrating activation of pathways involved in the response to hypoxia and oxidative stress.

Mucinous adenocarcinomas and low-grade serous carcinomas are characterized by aberrant activation of the mitogen-activated protein kinase (MAPK) pathway via activating mutations in the KRAS and BRAF genes. These mutations are found in up to 33% of low-grade serous tumors and between 50% and 75% of mucinous tumors.

The human epidermal growth factor receptor 2 (HER2) protein is also significantly overexpressed in around 20% of mucinous ovarian cancers. Similar to clear cell carcinomas, the ARID1A gene is also recurrently mutated in around 30% of endometrioid tumors. Endometrioid ovarian cancers are also distinguished by mutations in CTNNB1, particularly in low-grade tumors, a gene which encodes the beta-catenin protein that serves as the central regulator of the Wnt/beta-catenin pathway that plays a significant role in numerous cellular processes.

Personalized Therapies

Both PIK3CA and PTEN mutations are present in around a fifth of endometrioid tumors, implicating the PI3K/AKT pathway as a significant regulator of this type of ovarian cancer. BRAF mutations are also present in up to 24% of endometrioid tumors.A greater understanding of the genomic basis of ovarian cancer and the differences between the various subtypes of this disease is steadily beginning to inform the development of targeted therapies. Given the importance of DNA repair defects in HGSC, drugs targeting these pathways have remained a central focus of research efforts. Beyond the approval of olaparib and the continued development of other PARP inhibitors, the CHEK1 and CHEK2 proteins also have emerged as promising targets.

Forming part of the DNA damage signaling response (DDR) pathway, CHEK1 and CHEK2 are activated by the ATR and ATM kinases in response to breaks in a cell’s DNA and subsequently regulate a number of cellular processes, including checkpoint activation, cell cycle arrest, and initiation of DNA repair.

Various CHEK1 inhibitors have been developed, including GDC-0425, MK-8776, and AZD7762. Unfortunately, few clinical responses have been observed in patients treated with single-agent CHEK1 inhibitor therapy. Only Eli Lilly’s LY2606368 continues to be evaluated in clinical trials (NCT02203513).

The fact that TP53 is so highly mutated in HGSC also makes this an extremely promising therapeutic target. However, targeting p53 has proved to be a significant clinical challenge. Based on conventional drug development processes, the p53 protein, which is a tumor suppressor and transcription factor, is not an ideal target. Nevertheless, some progress has been made in developing drugs that reactivate the mutant protein. APR-246, an analog of PRIMA-1 (p53 reactivation and induction of massive apoptosis) is being evaluated in a phase I/II clinical trial in patients with recurrent high-grade serous ovarian cancer (NCT02098343).

The PI3K/AKT and MAPK pathways have been shown to be dysregulated through mutations in the genes encoding their component proteins, including PIK3CA, BRAF, and PTEN. Small-molecule inhibitors of some of these proteins are already available and are being evaluated in ovarian cancer, including BRAF and MEK inhibitors and PI3K and mammalian target of rapamycin (mTOR) inhibitors.

Jane de Lartigue, PhD, is a freelance medical writer and editor based in New Haven, Connecticut.

Key Research

  • Cancer Genome Atlas Research Network. Integrated genomic analyses of ovarian carcinoma. Nature. 2011;474(7353):609-615.
  • Coward JIG, Middleton K, Murphy F. New perspectives on targeted therapy in ovarian cancer. Int J Womens Health. 2015;7:189-203.
  • Fadare O, Khabele D. Molecular profiling of epithelial ovarian cancer. My Cancer Genome. https://goo.gl/4kjiJM. Updated January 26, 2016. Accessed August 5, 2016.
  • Krzystyniak J, Ceppi L, Dizon DS, Birrer MJ. Epithelial ovarian cancer: the molecular genetics of epithelial ovarian cancer. Ann Oncol. 2016;27(suppl 1):i4-i10.
  • Mittempergher L. Genomic characterization of high-grade serous ovarian cancer: dissecting its molecular heterogeneity as a road towards effective therapeutic strategies. Curr Oncol Rep. 2016;18(7):44-52.
  • Patch AM, Christie EL, Etemadmoghadam D, et al. Whole-genome characterization of chemoresistant ovarian cancer [published correction appears in Nature. 2015;521(7578):398]. Nature. 2015;521(7553):489-493.
  • Verschraegen C, Lounsbury K, Howe A, Greenblatt M. Therapeutic implications for ovarian cancer emerging from the tumor Cancer Genome Atlas. Transl Cancer Res. 2015;4(1):40-59.
  • Wei W, Dizon D, Vathipadiekal V, Birrer MJ. Ovarian cancer: genomic analysis. Ann Oncol. 2013;24(suppl 10):x7-x15.