Signature of a silent killer: expression profiling in epithelial ovarian cancer

Sindbis Virus Vaccine Platform: A Promising Oncolytic Virus-Mediated Approach for Ovarian Cancer Treatment

This review article provides a comprehensive overview of a novel Sindbis virus vaccine platform as potential immunotherapy for ovarian cancer patients. Ovarian cancer is the most lethal of all gynecological malignancies. The majority of high-grade serous ovarian cancer (HGSOC) patients are diagnosed with advanced disease. Current treatment options are very aggressive and limited, resulting in tumor recurrences and 50-60% patient mortality within 5 years. The unique properties of armed oncolytic Sindbis virus vectors (SV) in vivo have garnered significant interest in recent years to potently target and treat ovarian cancer. We discuss the molecular biology of Sindbis virus, its mechanisms of action against ovarian cancer cells, preclinical in vivo studies, and future perspectives. The potential of Sindbis virus-based therapies for ovarian cancer treatment holds great promise and warrants further investigation. Investigations using other oncolytic viruses in preclinical studies and clinical trials are also presented.

Exosomes as crucial emerging tools for intercellular communication with therapeutic potential in ovarian cancer

More than two-thirds of epithelial ovarian cancer (EOC) patients are diagnosed at advanced stages due to the lack of sensitive biomarkers. Currently, exosomes are intensively investigated as non-invasive cancer diagnostic markers. Exosomes are nanovesicles released in the extracellular milieu with the potential to modulate recipient cells' behavior. EOC cells release many altered exosomal cargoes that exhibit clinical relevance to tumor progression. Exosomes represent powerful therapeutic tools (drug carriers or vaccines), posing a promising option in clinical practice for curing EOC in the near future. In this review, we highlight the importance of exosomes in cell–cell communication, epithelial–mesenchymal transition (EMT), and their potential to serve as diagnostic and prognostic factors, particularly in EOC.

CircBNC2 affects epithelial ovarian cancer progression through the miR-223-3p/ LARP4 axis

Epithelial ovarian cancer (EOC) is one of the most serious cancer. Circular RNA BNC2 (circBNC2) expression was decreased in EOC tissues. However, the molecular mechanism of circBNC2 remains unknown. The expression of circBNC2, microRNA-223-3p (miR-223-3p), and La-related proteins 4 ( LARP4 ) were detected by quantitative real-time fluorescence PCR (qRT-PCR). A series of in-vitro experiments were designed to explore the function of circBNC2 in EOC cells and the regulatory mechanism between circBNC2 and miR-223-3p and LARP4 in EOC cells. Western blot examined the protein levels of Snail1, Slug, and LARP4 . The relationship between miR-223-3p and circBNC2 or LARP4 was verified by Dual-luciferase reporter assays. The xenotransplantation model was established to study the role of circBNC2 in vivo . The expression of circBNC2 and LARP4 was decreased in EOC tissues, while the expression of miR-223-3p was increased. CircBNC2 can sponge miR-223-3p, and LARP4 is the target of miR-223-3p. In-vitro complement experiments showed that overexpression of circBNC2 significantly decreased the malignant behavior of EOC, while co-transfection of miR-223-3p mimics partially upregulated this change. In addition, LARP4 knockdown increased the proliferation, migration, and invasion of EOC cells inhibited by miR-223-3p inhibitor. Mechanically, circBNC2 regulates LARP4 expression in EOC cells by spongy miR-223-3p. In addition, in-vivo studies have shown that overexpression of circBNC2 inhibits tumor growth. Overexpression of circBNC2 decreased proliferation, migration, and invasion of EOC cells by regulating the miR-223-3p/ LARP4 axis, suggesting that circBNC2/miR-223-3p/ LARP4 axis may be a potential regulatory mechanism for the treatment of EOC.

Comprehensive analysis of competitive endogenous RNAs networks reveals potential prognostic biomarkers associated with epithelial ovarian cancer

Ovarian cancer (OC) is a major health threat to females, as it has high morbidity and mortality. Evidence has increasingly demonstrated that long non-coding RNAs (lncRNAs) regulate OC progression and they may have value as early diagnostic biomarkers, prognostic biomarkers and/or therapeutic targets. In the present study, the regulatory mechanisms and prognosis associated with cancer-specific lncRNAs and their related competing endogenous (ce)RNA network in OC were investigated. The differential expression profiles and prognostic significance of lncRNAs and mRNAs were systematically explored based on data from 359 OC cases from The Cancer Genome Atlas and 180 healthy individuals from the Genotype-Tissue Expression database. Functional enrichment analyses, RNA-RNA interactome prediction, ceRNA network analysis, correlation analysis and survival analysis were utilized to identify hub lncRNAs and biomarkers associated with OC diagnosis or prognosis. A total of 1,049 differentially expressed lncRNAs and 6,516 differentially expressed mRNAs between OC and healthy tissues were detected. An lncRNA-micro (mi)RNA-mRNA regulatory network in OC was further established, containing 91 lncRNAs, 23 miRNAs and 179 mRNAs. After survival analysis based on the expression of the RNAs in the ceRNA network, 8 lncRNAs, 4 miRNAs and 11 mRNAs that were significantly associated with OC patient survival (P<0.05) were obtained. Using least absolute shrinkage and selection operator-penalized Cox regression, an eight-lncRNA risk score model was generated, which was able to readily discriminate between OC and healthy individuals and predict the survival of patients with OC. In addition, the differential expression of several key lncRNAs and mRNAs was verified by reverse transcription-quantitative PCR and western blot analysis. The current study presents a novel lncRNA-miRNA-mRNA network, which provides insight into the potential pathogenesis of OC and allows the identification of prognostic biomarkers and treatment strategies for OC.

Development of a 3D functional assay and identification of biomarkers, predictive for response of high-grade serous ovarian cancer (HGSOC) patients to poly-ADP ribose polymerase inhibitors (PARPis): targeted therapy

Background

Poly(ADP-ribose) polymerase inhibitors (PARPis) specifically target homologous recombination deficiency (HRD) cells and display good therapeutic effect in women with advanced-stage BRCA1/2-mutated breast and epithelial ovarian cancer (EOC). However, about 50% of high grade serous ovarian cancers (HGSOC) present with HRD due to epigenetic BRCA1 inactivation, as well as genetic/epigenetic inactivation(s) of other HR genes, a feature known as “BRCAness”. Therefore, there is a potential for extending the use of PARPis to these patients if HR status can be identified.

Methods

We have developed a 3D (spheroid) functional assay to assess the sensitivity of two PARPis (niraparib and olaparib) in ascites-derived primary cell cultures (AsPCs) from HGSOC patients. A method for AsPCs preparation was established based on a matrix (agarose), allowing for easy isolation and successive propagation of monolayer and 3D AsPCs. Based on this method, we performed cytotoxicity assays on 42 AsPCs grown both as monolayers and spheroids.

Results

The response to PARPis treatment in monolayer AsPCs, was significantly higher, compared to 3D AsPCs, as 88% and 52% of the monolayer AsPCs displayed sensitivity to niraparib and olaparib respectively, while 66% of the 3D AsPCs were sensitive to niraparib and 38% to olaparib, the latter being more consistent with previous estimates of HRD (40%–60%) in EOC. Moreover, niraparib displayed a significantly stronger cytotoxic effect in both in 3D and monolayer AsPCs, which was confirmed by consecutive analyses of the HR pathway activity (γH2AX foci formation) in PARPis-sensitive and resistant AsPCs. Global gene expression comparison of 6 PARPi-resistant and 6 PARPi-sensitive 3D AsPCs was indicative for the predominant downregulation of numerous genes and networks with previously demonstrated roles in EOC chemoresistance, suggesting that the PARPis-sensitive AsPCs could display enhanced sensitivity to other chemotherapeutic drugs, commonly applied in cancer management. Microarray data validation identified 24 potential gene biomarkers associated with PARPis sensitivity. The differential expression of 7 selected biomarkers was consecutively confirmed by immunohistochemistry in matched EOC tumor samples.

Conclusion

The application of this assay and the potential biomarkers with possible predictive significance to PARPis therapy of EOC patients now need testing in the setting of a clinical trial.

Utility Of Plasma circBNC2 As A Diagnostic Biomarker In Epithelial Ovarian Cancer

BACKGROUND

Epithelial ovarian cancer (EOC) is the fifth most common cause of cancer death in women. Due to a lacking of early detection method, its five-year survival rate is only 30%. Nevertheless, novel biomarkers for diagnosis remain to be discovered. The potential of microRNA signatures in the diagnosis of EOC has been especially described in recent years. In our previous experiments, we identified that circBNC2 was downregulated in EOC specimens, and was associated with advanced tumor stage and lymph node metastasis (LNM) by performing circRNA-sequencing analysis. The aim of this study was to explore the diagnostic value of circBNC2 in patients with epithelial ovarian cancer (EOC).

METHODS

Plasma from 249 age and menopause-matched women (83 with EOC; 83 with benign ovarian cyst; 83 were healthy volunteers) was collected prior to surgery. CircBNC2 was analyzed using reverse transcription-quantitative polymerase chain reaction (RT-qPCR). Cancer antigen 125 (CA125) and human epididymis protein 4 (HE4) were analyzed using enzyme-linked immunosorbent assay (ELISA). Receiver operating curve (ROC), the area under the curve (AUC), sensitivity and specificity were estimated.

RESULTS

CircBNC2 was downregulated in EOC and had higher ROC AUC in comparing EOC to benign (ROC AUC 0.879, sensitivity 96.4%, specificity 80.7%) or healthy (ROC AUC 0.923, sensitivity 95.2%, specificity 85.5%) cohorts than HE4 (ROC AUC: 0.742, benign cohort; 0.779, healthy cohort) and CA125 (ROC AUC: 0.373, benign cohort; 0.713, healthy cohort). Early stage EOC vs benign (ROC AUC 0.864, sensitivity 92.0%, specificity 80.7%) and healthy (ROC AUC 0.908, sensitivity 92.0%, specificity 85.5%) cohorts could be significantly separated by circBNC2. CircBNC2 performed alike in pre- and postmenopausal women, within EOC compared to the benign or healthy cohort.

CONCLUSION

CircBNC2 is downregulated in EOC (both in tissue and plasma samples) and might present promising novel biomarker for EOC. Further studies are needed to verify our results.

IMPACT

CircBNC2 is downregulated in EOC and warrants investigation in a screening study in females at risk for EOC.

Prognostic value of the tumor-specific ceRNA network in epithelial ovarian cancer

Epithelial ovarian cancer is one of the leading causes of cancer-related death worldwide. Growing evidence indicates that multiple complex altered pathways play important regulatory roles in the development and progression of a variety of cancers, including epithelial ovarian cancer. However, the underlying mechanisms remain unclear. First, we identified differentially expressed messenger RNAs (mRNAs), long noncoding RNAs (lncRNAs), and microRNAs (miRNAs) in epithelial ovarian cancer by comparing the expression profiles between epithelial ovarian cancer samples and normal tissue samples in different GEO datasets. Then, GO- and KEGG-pathway-enrichment analyses were applied to investigate the primary functions of the overlapped differentially expressed mRNAs. Moreover, the primary enriched genes were used to construct the signal-network with Cytoscape software. In addition, we integrated the relationship among lncRNAs-miRNAs-mRNAs to create a competing endogenous RNA network. Finally, mRNAs that were associated with patient prognosis in epithelial ovarian cancer were selected using univariate Cox regression analysis. A total of 2,225 mRNAs, 336 lncRNAs, and 14 miRNAs were shown to be differentially expressed in epithelial ovarian cancer compared with normal tissues. The dysregulated mRNAs were primarily enriched in cell division and signal transduction, according to Gene Ontology, whereas, according to KEGG, they were primarily enriched in metabolic pathways and pathways in cancer. A total of 10 mRNAs were associated with patient prognosis in ovarian cancer. This study identifies a novel lncRNA-miRNA-mRNA network, which may suggest potential molecular mechanisms underlying the development of epithelial ovarian cancer, providing new insights for survival prediction and interventional strategies for epithelial ovarian cancer.

Novel high-grade serous epithelial ovarian cancer cell lines that reflect the molecular diversity of both the sporadic and hereditary disease

Few cell line models of epithelial ovarian cancer (EOC) have been developed for the high-grade serous (HGS) subtype, which is the most common and lethal form of gynaecological cancer. Here we describe the establishment of six new EOC cell lines spontaneously derived from HGS tumors (TOV2978G, TOV3041G and TOV3291G) or ascites (OV866(2), OV4453 and OV4485). Exome sequencing revealed somatic TP53 mutations in five of the cell lines. One cell line has a novel BRCA1 splice-site mutation, and another, a recurrent BRCA2 nonsense mutation, both of germline origin. The novel BRCA1 mutation induced abnormal splicing, mRNA instability, resulting in the absence of BRCA1 protein. None of the cell lines harbor mutations in KRAS or BRAF, which are characteristic of other EOC subtypes. SNP arrays showed that all of the cell lines exhibited structural chromosomal abnormalities, copy number alterations and regions of loss of heterozygosity, consistent with those described for HGS. Four cell lines were able to produce 3D-spheroids, two exhibited anchorage-independent growth, and three (including the BRCA1 and BRCA2 mutated cell lines) formed tumors in SCID mice. These novel HGS EOC cell lines and their detailed characterization provide new research tools for investigating the most common and lethal form of EOC.

Engineered microenvironments provide new insights into ovarian and prostate cancer progression and drug responses

Tissue engineering technologies, which have originally been designed to reconstitute damaged tissue structure and function, can mimic not only tissue regeneration processes but also cancer development and progression. Bioengineered approaches allow cell biologists to develop sophisticated experimentally and physiologically relevant cancer models to recapitulate the complexity of the disease seen in patients. Tissue engineering tools enable three-dimensionality based on the design of biomaterials and scaffolds that re-create the geometry, chemistry, function and signalling milieu of the native tumour microenvironment. Three-dimensional (3D) microenvironments, including cell-derived matrices, biomaterial-based cell culture models and integrated co-cultures with engineered stromal components, are powerful tools to study dynamic processes like proteolytic functions associated with cancer progression, metastasis and resistance to therapeutics. In this review, we discuss how biomimetic strategies can reproduce a humanised niche for human cancer cells, such as peritoneal or bone-like microenvironments, addressing specific aspects of ovarian and prostate cancer progression and therapy response.

A multiscale road map of cancer spheroids – incorporating experimental and mathematical modelling to understand cancer progression

Computational models represent a highly suitable framework, not only for testing biological hypotheses and generating new ones but also for optimising experimental strategies. As one surveys the literature devoted to cancer modelling, it is obvious that immense progress has been made in applying simulation techniques to the study of cancer biology, although the full impact has yet to be realised. For example, there are excellent models to describe cancer incidence rates or factors for early disease detection, but these predictions are unable to explain the functional and molecular changes that are associated with tumour progression. In addition, it is crucial that interactions between mechanical effects, and intracellular and intercellular signalling are incorporated in order to understand cancer growth, its interaction with the extracellular microenvironment and invasion of secondary sites. There is a compelling need to tailor new, physiologically relevant in silico models that are specialised for particular types of cancer, such as ovarian cancer owing to its unique route of metastasis, which are capable of investigating anti-cancer therapies, and generating both qualitative and quantitative predictions. This Commentary will focus on how computational simulation approaches can advance our understanding of ovarian cancer progression and treatment, in particular, with the help of multicellular cancer spheroids, and thus, can inform biological hypothesis and experimental design.

Allelic transcripts dosage effect in morphologically normal ovarian cells from heterozygous carriers of a BRCA1/2 French Canadian founder mutation

We hypothesized that the transcriptome of primary cultures of morphologically normal ovarian surface epithelial cells could be altered by the presence of a heterozygous BRCA1 or BRCA2 mutation. We aimed to discover early events associated with ovarian carcinogenesis, which could represent putative targets for preventive strategies of this silent killer tumor. We identified the first molecular signature associated with French Canadian BRCA1 or BRCA2 founder mutations in morphologically normal ovarian epithelial cells. We discovered that wild-type and mutated BRCA2 allelic transcripts were expressed not only in morphologically normal but also in tumor cells from BRCA2-8765delAG carriers. Further analysis of morphologically normal ovarian and tumor cells from BRCA1-4446C>T carriers lead to the same observation. Our data support the idea that one single hit in BRCA1 or BRCA2 is sufficient to alter the transcriptome of phenotypically normal ovarian epithelial cells. The highest level of BRCA2-mutated allele transcript expression was measured in cells originating from the most aggressive ovarian tumor. The penetrance of the mutation and the aggressiveness of the related tumor could depend on a dosage effect of the mutated allele transcript.

The chemiluminescence based Ziplex automated workstation focus array reproduces ovarian cancer Affymetrix GeneChip expression profiles

Background

As gene expression signatures may serve as biomarkers, there is a need to develop technologies based on mRNA expression patterns that are adaptable for translational research. Xceed Molecular has recently developed a Ziplex^® technology, that can assay for gene expression of a discrete number of genes as a focused array. The present study has evaluated the reproducibility of the Ziplex system as applied to ovarian cancer research of genes shown to exhibit distinct expression profiles initially assessed by Affymetrix GeneChip^® analyses.

Methods

The new chemiluminescence-based Ziplex^® gene expression array technology was evaluated for the expression of 93 genes selected based on their Affymetrix GeneChip^® profiles as applied to ovarian cancer research. Probe design was based on the Affymetrix target sequence that favors the 3' UTR of transcripts in order to maximize reproducibility across platforms. Gene expression analysis was performed using the Ziplex Automated Workstation. Statistical analyses were performed to evaluate reproducibility of both the magnitude of expression and differences between normal and tumor samples by correlation analyses, fold change differences and statistical significance testing.

Results

Expressions of 82 of 93 (88.2%) genes were highly correlated (p < 0.01) in a comparison of the two platforms. Overall, 75 of 93 (80.6%) genes exhibited consistent results in normal versus tumor tissue comparisons for both platforms (p < 0.001). The fold change differences were concordant for 87 of 93 (94%) genes, where there was agreement between the platforms regarding statistical significance for 71 (76%) of 87 genes. There was a strong agreement between the two platforms as shown by comparisons of log₂ fold differences of gene expression between tumor versus normal samples (R = 0.93) and by Bland-Altman analysis, where greater than 90% of expression values fell within the 95% limits of agreement.

Conclusion

Overall concordance of gene expression patterns based on correlations, statistical significance between tumor and normal ovary data, and fold changes was consistent between the Ziplex and Affymetrix platforms. The reproducibility and ease-of-use of the technology suggests that the Ziplex array is a suitable platform for translational research.

Reprogramming of the transcriptome in a novel chromosome 3 transfer tumor suppressor ovarian cancer cell line model affected molecular networks that are characteristic of ovarian cancer

Tumor suppression as a consequence of the transfer of chromosome 3p fragments was previously observed in a novel epithelial ovarian cancer (EOC) OV-90 cell line model harboring loss of 3p. Microarray analysis revealed that tumor suppression was associated with a modified transcriptome. To investigate the relevance of the altered transcriptome, the differentially expressed genes identified by Affymetrix analysis in the 3p transfer studies, were integrated with a comparative microarray analysis of normal ovarian surface epithelial (NOSE) cells and malignant ovarian (TOV) cancers. Data from 219 significantly differentially expressed genes exhibited patterns in the direction predicted by the analysis of 3p transfer study. The 30 genes with the highest statistically significant differences (P < 1 x 10(-8)) in expression were found consistently differentially expressed between NOSE and TOV samples. The investigation of these genes in benign serous ovarian tumors and EOC cell lines also exhibited predictable expression patterns. Within the group of differentially expressed genes were SPARC, DAB2, CP, EVI1, ELF3, and EHD2, known to play a role in ovarian cancer, genes implicated in other cancers, such as GREM1 and GLIPR1, as well as genes not previously reported in a cancer context such as AKAP2 and ATAD4. A number of the differentially expressed genes are implicated in the TGF-beta signaling pathway. These findings suggest that the reprogramming of the transcriptome that occurred as a consequence of the chromosome 3 transfer and tumor suppression affected molecular networks that are characteristic of ovarian carcinogenesis thus validating our novel ovarian cancer cell line model.

BMP-2 signaling in ovarian cancer and its association with poor prognosis

BACKGROUND

We previously observed the over-expression of BMP-2 in primary cultures of epithelial ovarian cancer (EOC) cells as compared to normal epithelial cells based on Affymetrix microarray profiling 1. Here we investigate the effect of BMP-2 on several parameters of ovarian cancer tumorigenesis using the TOV-2223, TOV-1946 and TOV-112D EOC cell lines.

METHODS

We treated each EOC cell line with recombinant BMP-2 and assayed various parameters associated with tumorigenesis. More specifically, cell signaling events induced by BMP-2 treatment were investigated by western-blot using anti-phosphospecific antibodies. Induction of Id1, Snail and Smad6 mRNA expression was investigated by real time RT-PCR. The ability of cells to migrate was tested using the scratch assay. Cell-cell adhesion was analyzed by the ability of cells to form spheroids. We also investigated BMP-2 expression in tissue samples from a series of EOC patients.

RESULTS

Treatment of these cell lines with recombinant BMP-2 induced a rapid phosphorylation of Smad1/5/8 and Erk MAPKs. Increased expression of Id1, Smad6 and Snail mRNAs was also observed. Only in the TOV-2223 cell line were these signaling events accompanied by an alteration in cell proliferation. We also observed that BMP-2 efficiently increased the motility of all three cell lines. In contrast, BMP-2 treatment decreased the ability of TOV-1946 and TOV-112D cell lines to form spheroids indicating an inhibition of cell-cell adhesion. The expression of BMP-2 in tumor tissues from patients was inversely correlated with survival.

CONCLUSION

These results suggest that EOC cell secretion of BMP-2 in the tumor environment contributes to a modification of tumor cell behavior through a change in motility and adherence. We also show that BMP-2 expression in tumor tissues is associated with a poorer prognosis for ovarian cancer patients.

Influence of monolayer, spheroid, and tumor growth conditions on chromosome 3 gene expression in tumorigenic epithelial ovarian cancer cell lines

Background

Expression microarray analyses of epithelial ovarian cancer (EOC) cell lines may be exploited to elucidate genetic and epigenetic events important in this disease. A possible variable is the influence of growth conditions on discerning candidates. The present study examined the influence of growth conditions on the expression of chromosome 3 genes in the tumorigenic EOC cell lines, OV-90, TOV-21G and TOV-112D using Affymetrix GeneChip^® HG-U133A expression microarray analysis.

Methods

Chromosome 3 gene expression profiles (n = 1147 probe sets, representing 735 genes) were extracted from U133A expression microarray analyses of the EOC cell lines OV-90, TOV-21G and TOV-112D that were grown as monolayers, spheroids or nude mouse xenografts and monolayers derived from these tumors. Hierarchical cluster analysis was performed to compare chromosome 3 transcriptome patterns of each growth condition. Differentially expressed genes were identified and characterized by two-way comparative analyses of fold-differences in gene expression between monolayer cultures and each of the other growth conditions, and between the maximum and minimum values of expression of all growth conditions for each EOC cell line.

Results

An overall high degree of similarity (> 90%) in gene expression was observed when expression values of alternative growth conditions were compared within each EOC cell line group. Two-way comparative analysis of each EOC cell line grown in an alternative condition relative to the monolayer culture showed that overall less than 15% of probe sets exhibited at least a 3-fold difference in expression profile. Less than 23% of probe sets exhibited greater than 3-fold differences in gene expression in comparisons of the maximum and minimum value of expression of all growth conditions within each EOC cell line group. The majority of these differences were less than 5-fold. There were 17 genes in common which were differentially expressed in all EOC cell lines. However, the patterns of expression of these genes were not necessarily the same for each growth condition when one cell line was compared with another.

Conclusion

The various alternative in vivo and in vitro growth conditions of tumorigenic EOC cell lines appeared to modestly influence the global chromosome 3 transcriptome supporting the notion that the in vitro cell line models are a viable option for testing gene candidates.

Molecular description of a 3D in vitro model for the study of epithelial ovarian cancer (EOC)

Epithelial ovarian cancer (EOC) cell lines are useful tools for the molecular and biological characterization of ovarian cancer. The use of an in vitro multidimensional (3-D) culture model recapitulates some of the growth conditions encountered by tumor cells in vivo. Here we describe a molecular comparison of spheroid based 3D EOC models versus monolayer cultures and xenografts using cell lines from malignant ovarian tumors (TOV-21G and TOV-112D) and ascites (OV-90) previously established and characterized in our laboratory. Gene expression analyses of the three models were performed using the Affymetrix HG-U133A high density DNA array. Cluster analysis identified a set of genes that stratified expression profiles from the EOC cell lines grown as spheroids and xenografts from that of monolayer cultures. The gene expression analysis results were validated by Q-PCR analyses on an independent set of RNAs. Differential expression observed for the S100A6 gene between the monolayer, spheroid cultures and xenografts was confirmed at the protein level by immunohistochemistry. The analysis was extended to various ovarian tumor tissues using an EOC tissue array. This result represents an example of a gene that, if studied in vitro, is more representative of the in vivo disease in a 3D model rather than the monolayer culture. Identification of genes in spheroid models that mimic the in vivo tumor gene expression patterns may allow a better understanding of the community effect observed in human disease that is determined by direct or indirect interactions of cells with their environment or other surrounding cells.

An integrated clinical-genomics approach identifies a candidate multi-analyte blood test for serous ovarian carcinoma

PURPOSE

Cancer of the ovary confers the worst prognosis among women with gynecologic malignancies, underscoring the need to develop new biomarkers for detection of early disease, particularly those that can be readily monitored in the blood.

EXPERIMENTAL DESIGN

We developed an algorithm to identify secreted proteins encoded among approximately 22,500 genes on commercial oligonucleotide arrays and applied it to gene expression profiles of 67 stage I to IV serous papillary carcinomas and 9 crudely enriched normal ovarian tissues, to identify putative diagnostic markers. ELISAs were used to validate increased levels of secreted proteins in patient sera encoded by genes with differentially high expression.

RESULTS

We identified 275 genes predicted to encode secreted proteins with increased/decreased expression in ovarian cancers (<0.5- or >2-fold, P < 0.001). The serum levels of four of these proteins (matrix metalloproteinase-7, osteopontin, secretory leukoprotease inhibitor, and kallikrein 10) were significantly elevated in a series of 67 independent patients with serous ovarian carcinomas compared with 67 healthy controls (P < 0.001, Wilcoxon rank sum test). Optimized support vector machine classifiers with as few as two of these markers (osteopontin or kallikrein 10/matrix metalloproteinase-7) in combination with CA-125 yielded sensitivity and specificity values ranging from 96% to 98.7% and 99.7% to 100%, respectively, with the ability to discern early-stage disease from normal, healthy controls.

CONCLUSIONS

Our data suggest that this assay combination warrants further investigation as a multi-analyte diagnostic test for serous ovarian adenocarcinoma.

Gene expression profile of ovarian serous papillary carcinomas: identification of metastasis-associated genes

OBJECTIVE

The purpose of this study was to identify genes that are highly differentially expressed in metastatic serous papillary ovarian tumors (MET) when compared with primary ovarian serous carcinomas (OSPC).

STUDY DESIGN

An oligonucleotide microarray with probe sets complementary to >14,500 human genes was used to determine whether patterns of gene expression may differentiate OSPC from MET in 31 snap-frozen serous papillary ovarian carcinomas (ie, 14 primary OSPC and 17 omental metastasis [MET]).

RESULTS

Hierarchic cluster analysis of gene expression in OSPC and MET identified 156 genes that exhibited > 2-fold differences (P < .05) and that distinguished OSPC from MET. A number of invasion and metastasis predictive genes (including plasminogen activator; matrix metalloproteinase; matrix structural constituent genes encoding products with collagen, heparin, and hyaluronic acid binding activity; genes encoding receptors for insulin-like growth factors; vascular endothelial growth factor; endothelin type A; fibroblast growth factor; thrombospondin 1 and 2; type A and B integrins, and chemokines [stromal cell-derived factor 1 (CXCL12)]) were found among the 120 genes that were highly differentially overexpressed in MET, when compared with OSPC. Down-regulated genes in MET compared with OSPC included hepsin and testisin. Overexpression of CXCL12, matrix metalloproteinase, plasminogen activator, type A and B integrins, and hepsin genes was validated by quantitative real-time polymerase chain reaction in all samples. Finally, overexpression of CXCL12 in MET, when compared with OSPC, was validated at the protein level by immunohistochemistry.

CONCLUSION

Gene expression profiling may differentiate metastatic ovarian cancer from primary OSPC. The identification of metastasis-associated genes may provide a foundation for the development of new type-specific diagnostic strategies and treatment for metastatic ovarian cancer.

Differential gene expression identifies subgroups of ovarian carcinoma

Papillary serous ovarian carcinoma, the most common type of ovarian cancer, displays different biological behavior in different patients. This heterogeneity cannot be recognized by light microscopy. In this study, gene expression in 29 papillary serous ovarian carcinoma samples (21 invasive tumors and 8 borderline tumors), and 17 nonmalignant tissue types comprising 512 samples, was determined using Affymetrix U_133 oligonucleotide microarrays (Affymetrix, Inc., Santa Clara, Calif) representing approximately 40,000 known genes and expression sequence tags (ESTs). Differences in gene expression were quantified as the fold change in gene expression between the various sets of samples. A set of genes was identified that was over-expressed in the invasive ovarian carcinoma samples compared with the normal ovary samples. Principle component analysis of the set of invasive ovarian carcinomas using this set of genes revealed the existence of 2 major subgroups among the invasive ovarian carcinomas. A series of principle component analyses of the ovarian carcinomas using different gene sets composed of genes involved in different metabolic pathways also revealed the same 2 major subgroups of the invasive ovarian carcinomas. Review of the pathology by a single pathologist in a blinded manner suggested that these 2 subgroups differed in pathologic grade. Genes differentially expressed between the 2 ovarian carcinoma subsets were identified. Examination of gene expression in each ovarian carcinoma subset compared with that in 17 different normal tissue types (512 samples) revealed genes specifically over-expressed in ovarian carcinoma compared with these normal tissues. It is concluded that gene expression patterns may be useful in helping to further classify subtypes of papillary serous ovarian carcinoma that may have clinical significance. In addition, the genes identified as over-expressed in each set of serous ovarian carcinoma compared with normal tissues may represent potential biomarkers and/or targets for therapy.

From gene profiling to diagnostic markers: IL-18 and FGF-2 complement CA125 as serum-based markers in epithelial ovarian cancer

We used an oligonucleotide-based DNA microarray to identify potential markers in 39 primary cultures of ovarian cancer specimens compared with 11 primary cultures of normal ovarian epithelia. Differential gene expression of IL-18 and FGF-2 was validated on a subset of samples by quantitative PCR and by IHC, using an independent tissue array of 90 cores of 20 normal ovarian surface epithelia and 70 EOCs representing different grades and pathologies of ovarian disease. We further compared, by ELISA, these two markers with CA125 in sera from 25 cancer-free and 47 ovarian cancer patients. IL-18 and FGF-2 proteins were significantly elevated in tumor tissues (p<0.04) and sera (p<0.05) from patients with ovarian cancer. In combination, the three markers (IL-18, FGF-2, and CA125) showed similar sensitivity in scoring for ovarian cancer (35/45 patients) compared to that of CA125 alone (37/45) and significantly improved the specificity of detection (20/25 patients) compared to each marker individually (15/25 for CA125; 18/25 FGF-2; 16/25 for IL-18). In conclusion we show that a combination of the three serum markers (IL-18, FGF-2 and CA125) is associated with EOC, with higher specificity than CA125 alone. Prospective studies with a large cohort of susceptible ovarian cancer patients will be required to expand these findings.

Gene expression profiling of primary cultures of ovarian epithelial cells identifies novel molecular classifiers of ovarian cancer

In order to elucidate the biological variance between normal ovarian surface epithelial (NOSE) and epithelial ovarian cancer (EOC) cells, and to build a molecular classifier to discover new markers distinguishing these cells, we analysed gene expression patterns of 65 primary cultures of these tissues by oligonucleotide microarray. Unsupervised clustering highlights three subgroups of tumours: low malignant potential tumours, invasive solid tumours and tumour cells derived from ascites. We selected 18 genes with expression profiles that enable the distinction of NOSE from these three groups of EOC with 92% accuracy. Validation using an independent published data set derived from tissues or primary cultures confirmed a high accuracy (87–96%). The distinctive expression pattern of a subset of genes was validated by quantitative reverse transcription–PCR. An ovarian-specific tissue array representing tissues from NOSE and EOC samples of various subtypes and grades was used to further assess the protein expression patterns of two differentially expressed genes (Msln and BMP-2) by immunohistochemistry. This study highlights the relevance of using primary cultures of epithelial ovarian cells as a model system for gene profiling studies and demonstrates that the statistical analysis of gene expression profiling is a useful approach for selecting novel molecular tumour markers.

Proteome profiling of human epithelial ovarian cancer cell line TOV-112D

A proteome profiling of the epithelial ovarian cancer cell line TOV-112D was initiated as a protein expression reference in the study of ovarian cancer. Two complementary proteomic approaches were used in order to maximise protein identification: two-dimensional gel electrophoresis (2DE) protein separation coupled to matrix assisted laser desorption/ionisation time-of-flight mass spectrometry (MALDI-TOF MS) and one-dimensional gel electrophoresis (1DE) coupled to liquid-chromatography tandem mass spectrometry (LC MS/MS). One hundred and seventy-two proteins have been identified among 288 spots selected on two-dimensional gels and a total of 579 proteins were identified with the 1DE LC MS/MS approach. This proteome profiling covers a wide range of protein expression and identifies several proteins known for their oncogenic properties. Bioinformatics tools were used to mine databases in order to determine whether the identified proteins have previously been implicated in pathways associated with carcinogenesis or cell proliferation. Indeed, several of the proteins have been reported to be specific ovarian cancer markers while others are common to many tumorigenic tissues or proliferating cells. The diversity of proteins found and their association with known oncogenic pathways validate this proteomic approach. The proteome 2D map of the TOV-112D cell line will provide a valuable resource in studies on differential protein expression of human ovarian carcinomas while the 1DE LC MS/MS approach gives a picture of the actual protein profile of the TOV-112D cell line. This work represents one of the most complete ovarian protein expression analysis reports to date and the first comparative study of gene expression profiling and proteomic patterns in ovarian cancer.

Ovarian cancer diagnosis by hippocampus and neocortex-inspired learning memory structures

Early detection and accurate staging of ovarian cancer are the keys to improving survival rate. However, at present there is no single diagnosis modality that is sufficiently sensitive. DNA microarray analysis is an emerging technique that has potential for ameliorating the hardship in early detection and staging of ovarian disease. However, microarray data is ultra-huge and difficult to analyze. Hence, computational intelligence methods are often utilized to assist in the diagnosis and analysis process. Fuzzy Neural Networks (FNN) are more suitable for this task as FNN provides not only the accuracy, but also the interpretability of its reasoning process. Hippocampus-inspired Complementary Learning FNN (CLFNN) is able to rapidly derive fuzzy sets and formulate fuzzy rules. CLFNN uses positive and negative learning, and hence it reduces the effect of the curse of dimensionality and is capable of modeling the dynamics of the problem space with relatively good classification performance. One of its successors, a hybrid of complementary hippocampal learning and associative neocortical learning called Pseudo Associative Complementary Learning (PACL), is a structure that seeks to functionally model the memory consolidation process. Both PACL and CLFNN have human-like reasoning that allows physicians to examine their computation using familiar terms. They can construct intuitive fuzzy rules autonomously to justify their reasoning, which is important to generate trust among the users. Hence, CLFNN and PACL are applied as a diagnostic decision support system in ovarian cancer diagnosis. The experimental results are encouraging.

High-resolution mapping of genomic imbalance and identification of gene expression profiles associated with differential chemotherapy response in serous epithelial ovarian cancer

Array comparative genomic hybridization (aCGH) and microarray expression profiling were used to subclassify DNA and RNA alterations associated with differential response to chemotherapy in ovarian cancer. Two to 4 Mb interval arrays were used to map genomic imbalances in 26 sporadic serous ovarian tumors. Cytobands 1p36, 1q42-44, 6p22.1-p21.2, 7q32.1-q34 9q33.3-q34.3, 11p15.2, 13q12.2-q13.1, 13q21.31, 17q11.2, 17q24.2-q25.3, 18q12.2, and 21q21.2-q21.3 were found to be statistically associated with chemotherapy response, and novel regions of loss at 15q11.2-q15.1 and 17q21.32-q21.33 were identified. Gene expression profiles were obtained from a subset of these tumors and identified a group of genes whose differential expression was significantly associated with drug resistance. Within this group, five genes (GAPD, HMGB2, HSC70, GRP58, and HMGB1), previously shown to form a nuclear complex associated with resistance to DNA conformation-altering chemotherapeutic drugs in in vitro systems, may represent a novel class of genes associated with in vivo drug response in ovarian cancer patients. Although RNA expression change indicated only weak DNA copy number dependence, these data illustrate the value of molecular profiling at both the RNA and DNA levels to identify small genomic regions and gene subsets that could be associated with differential chemotherapy response in ovarian cancer.

Discrimination between serous low malignant potential and invasive epithelial ovarian tumors using molecular profiling

Tumors of low malignant potential (LMP) represent 20% of epithelial ovarian cancers (EOCs) and are associated with a better prognosis than the invasive tumors (TOV). Defining the relationship between LMPs and TOVs remains an important goal towards understanding the molecular pathways that contribute to prognosis, as well as providing molecular markers, for these EOCs. To this end, DNA microarray analyses were performed either in a primary culture or a tumor tissue model system and selected candidate genes showing a distinctive expression profile between LMPs and TOVs were identified using a class prediction approach based on three statistical methods of analysis. Both model systems appear relevant as candidate genes identified by either model allowed the proper reclassification of samples as either LMPs or TOVs. Selected candidate genes (CAS, CCNE1, LGALS8, ITGbeta3, ATP1B1, FLIP, KRT7 and KRT19) were validated by real-time quantitative PCR analysis and show differential expression between LMPs and TOVs. Immunohistochemistry analyses showed that the two tumor classes were distinguishable by their expression of CAS, TNFR1A, FLIP, CKS1 and CCNE1. These results define signature patterns for gene expression of LMPs and TOVs and identify gene candidates that warrant further study to deepen our understanding of the biology of EOC.