An Integrated Genomic-Based Approach to Individualized Treatment of Patients With Advanced-Stage Ovarian Cancer

Metabolic-Related Gene Prognostic Index for Predicting Prognosis, Immunotherapy Response, and Candidate Drugs in Ovarian Cancer

Ovarian cancer (OC) is a highly heterogeneous disease, with patients at different tumor staging having different survival times. Metabolic reprogramming is one of the key hallmarks of cancer; however, the significance of metabolism-related genes in the prognosis and therapy outcomes of OC is unclear. In this study, we used weighted gene coexpression network analysis and differential expression analysis to screen for metabolism-related genes associated with tumor staging. We constructed the metabolism-related gene prognostic index (MRGPI), which demonstrated a stable prognostic value across multiple clinical trial end points and multiple validation cohorts. The MRGPI population had its distinct molecular features, mutational characteristics, and immune phenotypes. In addition, we investigated the response to immunotherapy in MRGPI subgroups and found that patients with low MRGPI were prone to benefit from anti-PD-1 checkpoint blockade therapy and exhibited a delayed treatment effect. Meanwhile, we identified four candidate therapeutic drugs (ABT-737, crizotinib, panobinostat, and regorafenib) for patients with high MRGPI, and we evaluated the pharmacokinetics and safety of the candidate drugs. In summary, the MRGPI was a robust clinical feature that could predict patient prognosis, immunotherapy response, and candidate drugs, facilitating clinical decision making and therapeutic strategy of OC.

SpiderLearner: An ensemble approach to Gaussian graphical model estimation

Gaussian graphical models (GGMs) are a popular form of network model in which nodes represent features in multivariate normal data and edges reflect conditional dependencies between these features. GGM estimation is an active area of research. Currently available tools for GGM estimation require investigators to make several choices regarding algorithms, scoring criteria, and tuning parameters. An estimated GGM may be highly sensitive to these choices, and the accuracy of each method can vary based on structural characteristics of the network such as topology, degree distribution, and density. Because these characteristics are a priori unknown, it is not straightforward to establish universal guidelines for choosing a GGM estimation method. We address this problem by introducing SpiderLearner, an ensemble method that constructs a consensus network from multiple estimated GGMs. Given a set of candidate methods, SpiderLearner estimates the optimal convex combination of results from each method using a likelihood-based loss function.K $$ K $$ -fold cross-validation is applied in this process, reducing the risk of overfitting. In simulations, SpiderLearner performs better than or comparably to the best candidate methods according to a variety of metrics, including relative Frobenius norm and out-of-sample likelihood. We apply SpiderLearner to publicly available ovarian cancer gene expression data including 2013 participants from 13 diverse studies, demonstrating our tool's potential to identify biomarkers of complex disease. SpiderLearner is implemented as flexible, extensible, open-source code in the R package ensembleGGM at https://github.com/katehoffshutta/ensembleGGM.

A random forest-based metabolic risk model to assess the prognosis and metabolism-related drug targets in ovarian cancer

As one of the most common gynecologic malignant tumors, ovarian cancer is usually diagnosed at an advanced and incurable stage because of its early asymptomatic onset. Increasing research into tumor biology has demonstrated that abnormal cellular metabolism precedes tumorigenesis, therefore it has become an area of active research in academia. Cellular metabolism is of great significance in cancer diagnostic and prognostic studies. In this study, we integrated The Cancer Genome Atlas dataset with multiple Gene Expression Omnibus ovarian cancer datasets, identified 17 metabolic pathways with prognostic values using the random forest algorithm, constructed a metabolic risk scoring model based on metabolic pathway enrichment scores, and classified patients with ovarian cancer into two subtypes. Then, we systematically investigated the differences between different subtypes in terms of prognosis, differential gene expression, immune signature enrichment, Hallmark signature enrichment, and somatic mutations. As well, we successfully predicted differences in sensitivity to immunotherapy and chemotherapy drugs in patients with different metabolic risk subtypes. Moreover, we identified 5 drug targets associated with high metabolic risk and low metabolic risk ovarian cancer phenotypes through the weighted correlation network analysis and investigated their roles in the genesis of ovarian cancer. Finally, we developed an XGBoost classifier for predicting metabolic risk types in patients with ovarian cancer, producing a good predictive effect. In light of the above study, the research findings will provide valuable information for prognostic prediction and personalized medical treatment of patients with ovarian cancer.

Integrating cell cycle score for precise risk stratification in ovarian cancer

Background: Ovarian cancer (OC) is a highly heterogeneous disease, of which the mesenchymal subtype has the worst prognosis, is the most aggressive, and has the highest drug resistance. The cell cycle pathway plays a vital role in ovarian cancer development and progression. We aimed to screen the key cell cycle genes that regulated the mesenchymal subtype and construct a robust signature for ovarian cancer risk stratification.

Methods: Network inference was conducted by integrating the differentially expressed cell cycle signature genes and target genes between the mesenchymal and non-mesenchymal subtypes of ovarian cancer and identifying the dominant cell cycle signature genes.

Results: Network analysis revealed that two cell cycle signature genes (POLA2 and KIF20B) predominantly regulated the mesenchymal modalities of OC and used to construct a prognostic model, termed the Cell Cycle Prognostic Signature of Ovarian Cancer (CCPOC). The CCPOC-high patients showed an unfavorable prognosis in the GSE26712 cohort, consistent with the results in the seven public validation cohorts and one independent internal cohort (BL-OC cohort, qRT-PCR, n = 51). Functional analysis, drug-sensitive analysis, and survival analysis showed that CCPOC-low patients were related to strengthened tumor immunogenicity and sensitive to the anti-PD-1/PD-L1 response rate in pan-cancer (r = −0.47, OC excluded), which indicated that CCPOC-low patients may be more sensitive to anti-PD-1/PD-L1.

Conclusion: We constructed and validated a subtype-specific, cell cycle-based prognostic signature for ovarian cancer, which has great potential for predicting the response of anti-PD-1/PD-L1.

Filtering of the Gene Signature as the Predictors of Cisplatin-Resistance in Ovarian Cancer

BACKGROUND

Gene expression profiling and prediction of drug responses based on the molecular signature indicate new molecular biomarkers which help to find the most effective drugs according to the tumor characteristics.

OBJECTIVES

In this study two independent datasets, GSE28646 and GSE15372 were subjected to meta-analysis based on Affymetrix microarrays.

MATERIAL AND METHODS

In-silico methods were used to determine differentially expressed genes (DEGs) in the previously reported sensitive and resistant A2780 cell lines to Cisplatin. Gene Fuzzy Scoring (GFS) and Principle Component Analysis (PCA) were then used to eliminate batch effects and reduce data dimension, respectively. Moreover, SVM method was performed to classify sensitive and resistant data samples. Furthermore, Wilcoxon Rank sum test was performed to determine DEGs. Following the selection of drug resistance markers, several networks including transcription factor-target regulatory network and miRNA-target network were constructed and Differential correlation analysis was performed on these networks.

RESULTS

The trained SVM successfully classified sensitive and resistant data samples. Moreover, Performing DiffCorr analysis on the sensitive and resistant samples resulted in detection of 27 and 25 significant (with correlation ≥|0.9|) pairs of genes that respectively correspond to newly constructed correlations and loss of correlations in the resistant samples.

CONCLUSIONS

Our results indicated the functional genes and networks in Cisplatin resistance of ovarian cancer cells and support the importance of differential expression studies in ovarian cancer chemotherapeutic agent responsiveness.

Towards the Interpretability of Machine Learning Predictions for Medical Applications Targeting Personalised Therapies: A Cancer Case Survey

Artificial Intelligence is providing astonishing results, with medicine being one of its favourite playgrounds. Machine Learning and, in particular, Deep Neural Networks are behind this revolution. Among the most challenging targets of interest in medicine are cancer diagnosis and therapies but, to start this revolution, software tools need to be adapted to cover the new requirements. In this sense, learning tools are becoming a commodity but, to be able to assist doctors on a daily basis, it is essential to fully understand how models can be interpreted. In this survey, we analyse current machine learning models and other in-silico tools as applied to medicine-specifically, to cancer research-and we discuss their interpretability, performance and the input data they are fed with. Artificial neural networks (ANN), logistic regression (LR) and support vector machines (SVM) have been observed to be the preferred models. In addition, convolutional neural networks (CNNs), supported by the rapid development of graphic processing units (GPUs) and high-performance computing (HPC) infrastructures, are gaining importance when image processing is feasible. However, the interpretability of machine learning predictions so that doctors can understand them, trust them and gain useful insights for the clinical practice is still rarely considered, which is a factor that needs to be improved to enhance doctors' predictive capacity and achieve individualised therapies in the near future.

Mass-spectrometry-based proteomic correlates of grade and stage reveal pathways and kinases associated with aggressive human cancers

Proteomic signatures associated with clinical measures of more aggressive cancers could yield molecular clues as to disease drivers. Here, utilizing the Clinical Proteomic Tumor Analysis Consortium (CPTAC) mass-spectrometry-based proteomics datasets, we defined differentially expressed proteins and mRNAs associated with higher grade or higher stage, for each of seven cancer types (breast, colon, lung adenocarcinoma, clear cell renal, ovarian, uterine, and pediatric glioma), representing 794 patients. Widespread differential patterns of total proteins and phosphoproteins involved some common patterns shared between different cancer types. More proteins were associated with higher grade than higher stage. Most proteomic signatures predicted patient survival in independent transcriptomic datasets. The proteomic grade signatures, in particular, involved DNA copy number alterations. Pathways of interest were enriched within the grade-associated proteins across multiple cancer types, including pathways of altered metabolism, Warburg-like effects, and translation factors. Proteomic grade correlations identified protein kinases having functional impact in vitro in uterine endometrial cancer cells, including MAP3K2, MASTL, and TTK. The protein-level grade and stage associations for all proteins profiled-along with corresponding information on phosphorylation, pathways, mRNA expression, and copy alterations-represent a resource for identifying new potential targets. Proteomic analyses are often concordant with corresponding transcriptomic analyses, but with notable exceptions.

PlateDesigner: a web-based application for the design of microplate experiments

SUMMARY

In biological assays, systematic variability, known as a batch effect, can often confound the effects of true biological conditions and has been well documented for a variety of high-throughput technologies. In microplate-based multiplex experiments, such as Luminex or OLINK assays, researchers need to consider both position and plate effects. Those effects can be easily accounted for if the experiments are properly designed, which includes randomization of the samples across multiple experimental runs. However, doing the ad hoc randomization becomes challenging when handling multiple samples. PlateDesigner is the first web-based application that provides randomization for microplate experiments, ensuring that the main principles of the experimental design, such as grouping samples from the same biological units and balancing the distribution of experimental conditions, are applied. Creating randomizations with PlateDesigner is simple and the results can be exported in a variety of formats, and easily integrated with microplate readers and statistical analysis software.

AVAILABILITY AND IMPLEMENTATION

PlateDesigner is written in R/Shiny and is hosted online by the Center of Biostatistics at the Icahn School of Medicine at Mount Sinai. This application is freely available at platedesigner.net.

Identification of an energy metabolism‑related gene signature in ovarian cancer prognosis

Changes in energy metabolism may be potential biomarkers and therapeutic targets for cancer as they frequently occur within cancer cells. However, basic cancer research has failed to reach a consistent conclusion on the function(s) of mitochondria in energy metabolism. The significance of energy metabolism in the prognosis of ovarian cancer remains unclear; thus, there remains an urgent need to systematically analyze the characteristics and clinical value of energy metabolism in ovarian cancer. Based on gene expression patterns, the present study aimed to analyze energy metabolism-associated characteristics to evaluate the prognosis of patients with ovarian cancer. A total of 39 energy metabolism-related genes significantly associated with prognosis were obtained, and three molecular subtypes were identified by nonnegative matrix factorization clustering, among which the C1 subtype was associated with poor clinical outcomes of ovarian cancer. The immune response was enhanced in the tumor microenvironment. A total of 888 differentially expressed genes were identified in C1 compared with the other subtypes, and the results of the pathway enrichment analysis demonstrated that they were enriched in the ‘PI3K-Akt signaling pathway’, ‘cAMP signaling pathway’, ‘ECM-receptor interaction’ and other pathways associated with the development and progression of tumors. Finally, eight characteristic genes (tolloid-like 1 gene, type XVI collagen, prostaglandin F2α, cartilage intermediate layer protein 2, kinesin family member 26b, interferon inducible protein 27, growth arrest-specific gene 1 and chemokine receptor 7) were obtained through LASSO feature selection; and a number of them have been demonstrated to be associated with ovarian cancer progression. In addition, Cox regression analysis was performed to establish an 8-gene signature, which was determined to be an independent prognostic factor for patients with ovarian cancer and could stratify sample risk in the training, test and external validation datasets (P<0.01; AUC >0.8). Gene Set Enrichment Analysis results revealed that the 8-gene signature was involved in important biological processes and pathways of ovarian cancer. In conclusion, the present study established an 8-gene signature associated with metabolic genes, which may provide new insights into the effects of energy metabolism on ovarian cancer. The 8-gene signature may serve as an independent prognostic factor for ovarian cancer patients.

Anticancer SAR establishment and novel accruing signal transduction model of drug action using biscoumarin scaffold

In this paper, we have established methylenebis (4-hydroxy-2H-chromen-2-one) as a promising anticancer scaffold with kinesin spindle protein (KSP) inhibitory activity under malignant condition. A series of biscoumarin derivatives (MN1 to MN30) with different substituent were synthesized, and their anticancer activity was explored. Six biscoumarin derivatives that were found active were further selected to formulate organic nanoparticles (ONPs). Anticancer activity of both the forms (viz conventional and ONPs) was compared. MN30 was found most potent whereby MN10 showed good anticancer activity in both, i.e., conventional and ONP form; the structural activity relationship (SAR) study has been established. Computational investigation revealed biscoumarin scaffold as a suitable pharmacophore to bind against KSP protein. Molecular dynamics simulation studies revealed protein-ligand stability and dynamic behavior of biscoumarin-KSP complex. Finally, accruing signal transduction model was formulated to explain the observed MTT trend of conventional and ONP form. The model seems useful towards solving population specific varied results of chemotherapeutic agents. According to the model, MN10 and MN30 derivatives have good pharmacodynamics inertia and therefore, both the molecules were able to provide dose-dependent cytotoxic results.

Kinase Inhibitors and Ovarian Cancer

Ovarian cancer is fifth in the rankings of cancer deaths among women, and accounts for more deaths than any other gynecological malignancy. Despite some improvement in overall-(OS) and progression-free survival (PFS) following surgery and first-line chemotherapy, there is a need for development of novel and more effective therapeutic strategies. In this mini review, we provide a summary of the current landscape of the clinical use of tyrosine kinase inhibitors (TKIs) and mechanistic target of rapamycin (mTOR) inhibitors in ovarian cancer. Emerging data from phase I and II trials reveals that a combinatorial treatment that includes TKIs and chemotherapy agents seems promising in terms of PFS despite some adverse effects recorded; whereas the use of mTOR inhibitors seems less effective. There is a need for further research into the inhibition of multiple signaling pathways in ovarian cancer and progression to phase III trials for drugs that seem most promising.

Extreme learning machine Cox model for high-dimensional survival analysis

Some interesting recent studies have shown that neural network models are useful alternatives in modeling survival data when the assumptions of a classical parametric or semiparametric survival model such as the Cox (1972) model are seriously violated. However, to the best of our knowledge, the plausibility of adapting the emerging extreme learning machine (ELM) algorithm for single-hidden-layer feedforward neural networks to survival analysis has not been explored. In this paper, we present a kernel ELM Cox model regularized by an L0 -based broken adaptive ridge (BAR) penalization method. Then, we demonstrate that the resulting method, referred to as ELMCoxBAR, can outperform some other state-of-art survival prediction methods such as L1 - or L2 -regularized Cox regression, random survival forest with various splitting rules, and boosted Cox model, in terms of its predictive performance using both simulated and real world datasets. In addition to its good predictive performance, we illustrate that the proposed method has a key computational advantage over the above competing methods in terms of computation time efficiency using an a real-world ultra-high-dimensional survival data.

Consensus on Molecular Subtypes of High-Grade Serous Ovarian Carcinoma

Purpose: The majority of ovarian carcinomas are of high-grade serous histology, which is associated with poor prognosis. Surgery and chemotherapy are the mainstay of treatment, and molecular characterization is necessary to lead the way to targeted therapeutic options. To this end, various computational methods for gene expression-based subtyping of high-grade serous ovarian carcinoma (HGSOC) have been proposed, but their overlap and robustness remain unknown.Experimental Design: We assess three major subtype classifiers by meta-analysis of publicly available expression data, and assess statistical criteria of subtype robustness and classifier concordance. We develop a consensus classifier that represents the subtype classifications of tumors based on the consensus of multiple methods, and outputs a confidence score. Using our compendium of expression data, we examine the possibility that a subset of tumors is unclassifiable based on currently proposed subtypes.Results: HGSOC subtyping classifiers exhibit moderate pairwise concordance across our data compendium (58.9%-70.9%; P < 10^-5) and are associated with overall survival in a meta-analysis across datasets (P < 10^-5). Current subtypes do not meet statistical criteria for robustness to reclustering across multiple datasets (prediction strength < 0.6). A new subtype classifier is trained on concordantly classified samples to yield a consensus classification of patient tumors that correlates with patient age, survival, tumor purity, and lymphocyte infiltration.Conclusions: A new consensus ovarian subtype classifier represents the consensus of methods and demonstrates the importance of classification approaches for cancer that do not require all tumors to be assigned to a distinct subtype. Clin Cancer Res; 24(20); 5037-47. ©2018 AACR.

Consensus on Molecular Subtypes of High-Grade Serous Ovarian Carcinoma

Purpose: The majority of ovarian carcinomas are of high-grade serous histology, which is associated with poor prognosis. Surgery and chemotherapy are the mainstay of treatment, and molecular characterization is necessary to lead the way to targeted therapeutic options. To this end, various computational methods for gene expression-based subtyping of high-grade serous ovarian carcinoma (HGSOC) have been proposed, but their overlap and robustness remain unknown.Experimental Design: We assess three major subtype classifiers by meta-analysis of publicly available expression data, and assess statistical criteria of subtype robustness and classifier concordance. We develop a consensus classifier that represents the subtype classifications of tumors based on the consensus of multiple methods, and outputs a confidence score. Using our compendium of expression data, we examine the possibility that a subset of tumors is unclassifiable based on currently proposed subtypes.Results: HGSOC subtyping classifiers exhibit moderate pairwise concordance across our data compendium (58.9%-70.9%; P < 10-5) and are associated with overall survival in a meta-analysis across datasets (P < 10-5). Current subtypes do not meet statistical criteria for robustness to reclustering across multiple datasets (prediction strength < 0.6). A new subtype classifier is trained on concordantly classified samples to yield a consensus classification of patient tumors that correlates with patient age, survival, tumor purity, and lymphocyte infiltration.Conclusions: A new consensus ovarian subtype classifier represents the consensus of methods and demonstrates the importance of classification approaches for cancer that do not require all tumors to be assigned to a distinct subtype. Clin Cancer Res; 24(20); 5037-47. ©2018 AACR.

Ovarian Cancers: Genetic Abnormalities, Tumor Heterogeneity and Progression, Clonal Evolution and Cancer Stem Cells

Four main histological subtypes of ovarian cancer exist: serous (the most frequent), endometrioid, mucinous and clear cell; in each subtype, low and high grade. The large majority of ovarian cancers are diagnosed as high-grade serous ovarian cancers (HGS-OvCas). TP53 is the most frequently mutated gene in HGS-OvCas; about 50% of these tumors displayed defective homologous recombination due to germline and somatic BRCA mutations, epigenetic inactivation of BRCA and abnormalities of DNA repair genes; somatic copy number alterations are frequent in these tumors and some of them are associated with prognosis; defective NOTCH, RAS/MEK, PI3K and FOXM1 pathway signaling is frequent. Other histological subtypes were characterized by a different mutational spectrum: LGS-OvCas have increased frequency of BRAF and RAS mutations; mucinous cancers have mutation in ARID1A, PIK3CA, PTEN, CTNNB1 and RAS. Intensive research was focused to characterize ovarian cancer stem cells, based on positivity for some markers, including CD133, CD44, CD117, CD24, EpCAM, LY6A, ALDH1. Ovarian cancer cells have an intrinsic plasticity, thus explaining that in a single tumor more than one cell subpopulation, may exhibit tumor-initiating capacity. The improvements in our understanding of the molecular and cellular basis of ovarian cancers should lead to more efficacious treatments.

MiR-770-5p inhibits cisplatin chemoresistance in human ovarian cancer by targeting ERCC2

In this study, we examined the role of the miRNA miR-770-5p in cisplatin chemotherapy resistance in ovarian cancer (OVC) patients. miR-770-5p expression was reduced in platinum-resistant patients. Using a 6.128-fold in expression as the cutoff value, miR-770-5p expression served as a prognostic biomarker and predicted the response to cisplatin treatment and survival among OVC patients. Overexpression of miR-770-5p in vitro reduced survival in chemoresistant cell lines after cisplatin treatment. ERCC2, a target gene of miR-770-5p that participates in the NER system, was negatively regulated by miR-770-5p. siRNA-mediated silencing of ERCC2 reversed the inhibition of apoptosis resulting from miR-770-5p downreglation in A2780S cells. A comet assay confirmed that this restoration of cisplatin chemosensitivity was due to the inhibition of DNA repair. These findings suggest that endogenous miR-770-5p may function as an anti-oncogene and promote chemosensitivity in OVC, at least in part by downregulating ERCC2. miR-770-5p may therefore be a useful biomarker for predicting chemosensitivity to cisplatin in OVC patients and improve the selection of effective, more personalized, treatment strategies.

MicroRNA-520g promotes epithelial ovarian cancer progression and chemoresistance via DAPK2 repression

The lack of efficient tumor progression and chemoresistance indicators leads to high mortality in epithelial ovarian cancer (EOC) patients. Dysregulated miR-520g expression is involved in these processes in hepatic and colorectal cancers. In this study, we found that miR-520g expression gradually increased across normal, benign, borderline and EOC tissues. High miR-520g expression promoted tumor progression and chemoresistance to platinum-based chemotherapy, and reduced survival in EOC patients. miR-520g upregulation increased EOC cell proliferation, induced cell cycle transition and promoted cell invasion, while miR-520g downregulation inhibited tumor-related functions. In vivo, overexpression or downregulation of miR-520g respectively generated larger or smaller subcutaneous xenografts in nude mice. Death-associated protein kinase 2 (DAPK2) was a direct target of miR-520g. In 116 EOC tissue samples, miR-520g expression was significantly lower following DAPK2 overexpression. DAPK2 overexpression or miR-520g knockdown reduced EOC cell proliferation, invasion, wound healing and chemoresistance. This study suggests that miR-520g contributes to tumor progression and drug resistance by post-transcriptionally downregulating DAPK2, and that miR-520g may be a valuable therapeutic target in patients with EOC.

CONCORD biomarker prediction for novel drug introduction to different cancer types

Many cancer therapeutic agents have shown to be effective for treating multiple cancer types. Yet major challenges exist toward introducing a novel drug used in one cancer type to different cancer types, especially when a relatively small number of patients with the other cancer type often benefit from anti-cancer therapy with the drug. Recently, many novel agents were introduced to different cancer types together with companion biomarkers which were obtained or biologically assumed from the original cancer type. However, there is no guarantee that biomarkers from one cancer can directly predict a therapeutic response in another. To tackle this challenging question, we have developed a concordant expression biomarker-based technique ("CONCORD") that overcomes these limitations. CONCORD predicts drug responses from one cancer type to another by identifying concordantly co-expressed biomarkers across different cancer systems. Application of CONCORD to three standard chemotherapeutic agents and two targeted agents demonstrated its ability to accurately predict the effectiveness of a drug against new cancer types and predict therapeutic response in patients.

Decentralized Learning Framework of Meta-Survival Analysis for Developing Robust Prognostic Signatures

PURPOSE

A significant hurdle in developing reliable gene expression-based prognostic models has been the limited sample size, which can cause overfitting and false discovery. Combining data from multiple studies can enhance statistical power and reduce spurious findings, but how to address the biologic heterogeneity across different datasets remains a major challenge. Better meta-survival analysis approaches are needed.

MATERIAL AND METHODS

We presented a decentralized learning framework for meta-survival analysis without the need for data aggregation. Our method consisted of a series of proposals that together alleviated the influence of data heterogeneity and improved the performance of survival prediction. First, we transformed the gene expression profile of every sample into normalized percentile ranks to obtain platform-agnostic features. Second, we used Stouffer's meta-z approach in combination with Harrell's concordance index to prioritize and select genes to be included in the model. Third, we used survival discordance as a scale-independent model loss function. Instead of generating a merged dataset and training the model therein, we avoided comparing patients across datasets and individually evaluated the loss function on each dataset. Finally, we optimized the model by minimizing the joint loss function.

RESULTS

Through comprehensive evaluation on 31 public microarray datasets containing 6,724 samples of several cancer types, we demonstrated that the proposed method has outperformed (1) single prognostic genes identified using conventional meta-analysis, (2) multigene signatures trained on single datasets, (3) multigene signatures trained on merged datasets as well as by other existing meta-analysis methods, and (4) clinically applicable, established multigene signatures.

CONCLUSION

The decentralized learning approach can be used to effectively perform meta-analysis of gene expression data and to develop robust multigene prognostic signatures.

Molecular analysis of high-grade serous ovarian carcinoma with and without associated serous tubal intra-epithelial carcinoma

Many high-grade serous carcinomas (HGSCs) of the pelvis are thought to originate in the distal portion of the fallopian tube. Serous tubal intra-epithelial carcinoma (STIC) lesions are the putative precursor to HGSC and identifiable in ~ 50% of advanced stage cases. To better understand the molecular etiology of HGSCs, we report a multi-center integrated genomic analysis of advanced stage tumors with and without STIC lesions and normal tissues. The most significant focal DNA SCNAs were shared between cases with and without STIC lesions. The RNA sequence and the miRNA data did not identify any clear separation between cases with and without STIC lesions. HGSCs had molecular profiles more similar to normal fallopian tube epithelium than ovarian surface epithelium or peritoneum. The data suggest that the molecular features of HGSCs with and without associated STIC lesions are mostly shared, indicating a common biologic origin, likely to be the distal fallopian tube among all cases.High-grade serous carcinomas (HGSCs) are associated with precursor lesions (STICs) in the fallopian epithelium in only half of the cases. Here the authors report the molecular analysis of HGSCs with and without associated STICs and show similar profiles supporting a common origin for all HGSCs.

Identification of common oncogenic and early developmental pathways in the ovarian carcinomas controlling by distinct prognostically significant microRNA subsets

Background

High-grade serous ovarian carcinoma (HG-SOC) is the dominant tumor histologic type in epithelial ovarian cancers, exhibiting highly aberrant microRNA expression profiles and diverse pathways that collectively determine the disease aggressiveness and clinical outcomes. However, the functional relationships between microRNAs, the common pathways controlled by the microRNAs and their prognostic and therapeutic significance remain poorly understood.

Methods

We investigated the gene expression patterns of microRNAs in the tumors of 582 HG-SOC patients to identify prognosis signatures and pathways controlled by tumor miRNAs. We developed a variable selection and prognostic method, which performs a robust selection of small-sized subsets of the predictive features (e.g., expressed microRNAs) that collectively serves as the biomarkers of cancer risk and progression stratification system, interconnecting these features with common cancer-related pathways.

Results

Across different cohorts, our meta-analysis revealed two robust and unbiased miRNA-based prognostic classifiers. Each classifier reproducibly discriminates HG-SOC patients into high-confidence low-, intermediate- or high-prognostic risk subgroups with essentially different 5-year overall survival rates of 51.6-85%, 20-38.1%, and 0-10%, respectively. Significant correlations of the risk subgroup’s stratification with chemotherapy treatment response were observed. We predicted specific target genes involved in nine cancer-related and two oocyte maturation pathways (neurotrophin and progesterone-mediated oocyte maturation), where each gene can be controlled by more than one miRNA species of the distinct miRNA HG-SOC prognostic classifiers.

Conclusions

We identified robust and reproducible miRNA-based prognostic subsets of the of HG-SOC classifiers. The miRNAs of these classifiers could control nine oncogenic and two developmental pathways, highlighting common underlying pathologic mechanisms and perspective targets for the further development of a personalized prognosis assay(s) and the development of miRNA-interconnected pathway-centric and multi-agent therapeutic intervention.

Electronic supplementary material

The online version of this article (doi:10.1186/s12864-017-4027-5) contains supplementary material, which is available to authorized users.

The Prognostic 97 Chemoresponse Gene Signature in Ovarian Cancer

Patient diagnosis and care would be significantly improved by understanding the mechanisms underlying platinum and taxane resistance in ovarian cancer. Here, we aim to establish a gene signature that can identify molecular pathways/transcription factors involved in ovarian cancer progression, poor clinical outcome, and chemotherapy resistance. To validate the robustness of the gene signature, a meta-analysis approach was applied to 1,020 patients from 7 datasets. A 97-gene signature was identified as an independent predictor of patient survival in association with other clinicopathological factors in univariate [hazard ratio (HR): 3.0, 95% Confidence Interval (CI) 1.66–5.44, p = 2.7E-4] and multivariate [HR: 2.88, 95% CI 1.57–5.2, p = 0.001] analyses. Subset analyses demonstrated that the signature could predict patients who would attain complete or partial remission or no-response to first-line chemotherapy. Pathway analyses revealed that the signature was regulated by HIF1α and TP53 and included nine HIF1α-regulated genes, which were highly expressed in non-responders and partial remission patients than in complete remission patients. We present the 97-gene signature as an accurate prognostic predictor of overall survival and chemoresponse. Our signature also provides information on potential candidate target genes for future treatment efforts in ovarian cancer.

NUAK1 (ARK5) Is Associated with Poor Prognosis in Ovarian Cancer

Background and objective

Nua kinase 1 (NUAK1) was identified in multigene signatures of survival and suboptimal debulking in high-grade serous ovarian cancer (HGSOC). This study investigates the individual clinical and biologic contributions of NUAK1 in HGSOC patients and cell lines.

Methods

Public transcript expression, clinical, and outcome data were used to interrogate the relationship between NUAK1 and clinicopathologic factors and patient outcomes including progression-free survival (PFS) and molecular subtypes using logistic and Cox modeling. Analysis of NUAK1 transcript expression was performed in primary tumors from 34 HGSOC patients with < or ≥2 years PFS. The impact of silencing NUAK1 by RNA interference (RNAi) on the migratory potential and chemosensitivity of SOC cells was assessed in vitro.

Results

Elevated NUAK1 transcript expression was associated with worse PFS (hazard ratio = 1.134), advanced stage (odds ratio, OR = 1.7), any residual disease (OR = 1.58), and mesenchymal disease subtype (OR = 7.79 ± 5.89). Elevated NUAK1 transcript expression was observed in HGSOC patients with < vs. ≥2 years PFS (p < 0.045). RNAi-mediated silencing of NUAK1 expression attenuated migration of OV90 and E3 HGSOC cells in vitro, but did not modulate sensitivity to cisplatin or paclitaxel.

Conclusion

Elevated NUAK1 was associated with poor survival as well as advanced stage, residual disease after cytoreductive surgery and mesenchymal molecular subtype. NUAK1 impacted migration, but not chemosensitivity, in vitro. Additional studies are needed to further develop the concept of NUAK1 as a clinically deployable biomarker and therapeutic target in HGSOC.

Minimax Rate-optimal Estimation of High-dimensional Covariance Matrices with Incomplete Data

Missing data occur frequently in a wide range of applications. In this paper, we consider estimation of high-dimensional covariance matrices in the presence of missing observations under a general missing completely at random model in the sense that the missingness is not dependent on the values of the data. Based on incomplete data, estimators for bandable and sparse covariance matrices are proposed and their theoretical and numerical properties are investigated. Minimax rates of convergence are established under the spectral norm loss and the proposed estimators are shown to be rate-optimal under mild regularity conditions. Simulation studies demonstrate that the estimators perform well numerically. The methods are also illustrated through an application to data from four ovarian cancer studies. The key technical tools developed in this paper are of independent interest and potentially useful for a range of related problems in high-dimensional statistical inference with missing data.

Structured Matrix Completion with Applications to Genomic Data Integration

Matrix completion has attracted significant recent attention in many fields including statistics, applied mathematics and electrical engineering. Current literature on matrix completion focuses primarily on independent sampling models under which the individual observed entries are sampled independently. Motivated by applications in genomic data integration, we propose a new framework of structured matrix completion (SMC) to treat structured missingness by design. Specifically, our proposed method aims at efficient matrix recovery when a subset of the rows and columns of an approximately low-rank matrix are observed. We provide theoretical justification for the proposed SMC method and derive lower bound for the estimation errors, which together establish the optimal rate of recovery over certain classes of approximately low-rank matrices. Simulation studies show that the method performs well in finite sample under a variety of configurations. The method is applied to integrate several ovarian cancer genomic studies with different extent of genomic measurements, which enables us to construct more accurate prediction rules for ovarian cancer survival.

ADAMTS16 mutations sensitize ovarian cancer cells to platinum-based chemotherapy

Ovarian cancer is one of the most lethal malignant tumors in women. The prognosis of ovarian cancer patients depends, in part, on their response to platinum-based chemotherapy. Our recent analysis of genomics and clinical data from the Cancer Genome Atlas demonstrated that somatic mutations of ADAMTS 1, 6, 8, 9, 15, 16, 18 and L1 genes were associated with higher sensitivity to platinum and longer progression-free survival, overall survival, and platinum-free survival duration in 512 patients with high-grade serous ovarian carcinoma. Among the ADAMTS mutations, ADAMTS16 is the most commonly affected gene in ovarian cancer. However, the functional role of these mutations in ovarian cancer cells is largely unknown. We performed in vitro studies to compare the functional effects of the six identified ADAMTS missense mutations on the platinum sensitivity of ovarian cancer cells. We also used a well-characterized in vivo mouse model to evaluate the response of ovarian cancer cells with ADAMTS16 mutations to platinum-based therapy. Our results showed that exogenously expressed ADAMTS16 missense mutations inhibited cell growth or sensitized tumor cells to cisplatin and inhibited tumor growth in vivo. Orthotopic xenograft experiments showed that mice injected with ovarian cancer cells that exogenously expressed ADAMTS16 mutations had a better response to cisplatin treatment. Thus, these functional studies provide evidence that mutations of ADAMTS16 actively contribute to therapeutic response in ovarian cancer.

KCNN4 and S100A14 act as predictors of recurrence in optimally debulked patients with serous ovarian cancer

Approximately 50-75% of patients with serous ovarian carcinoma (SOC) experience recurrence within 18 months after first-line treatment. Current clinical indicators are inadequate for predicting the risk of recurrence. In this study, we used 7 publicly available microarray datasets to identify gene signatures related to recurrence in optimally debulked SOC patients, and validated their expressions in an independent clinic cohort of 127 patients using immunohistochemistry (IHC). We identified a two-gene signature including KCNN4 and S100A14 which was related to recurrence in optimally debulked SOC patients. Their mRNA expression levels were positively correlated and regulated by DNA copy number alterations (CNA) (KCNN4: p=1.918e-05) and DNA promotermethylation (KCNN4: p=0.0179; S100A14: p=2.787e-13). Recurrence prediction models built in the TCGA dataset based on KCNN4 and S100A14 individually and in combination showed good prediction performance in the other 6 datasets (AUC:0.5442-0.9524). The independent cohort supported the expression difference between SOC recurrences. Also, a KCNN4 and S100A14-centered protein interaction subnetwork was built from the STRING database, and the shortest regulation path between them, called the KCNN4-UBA52-KLF4-S100A14 axis, was identified. This discovery might facilitate individualized treatment of SOC.

Expression signature distinguishing two tumour transcriptome classes associated with progression-free survival among rare histological types of epithelial ovarian cancer

BACKGROUND

The mechanisms of recurrence have been under-studied in rare histologies of invasive epithelial ovarian cancer (EOC) (endometrioid, clear cell, mucinous, and low-grade serous). We hypothesised the existence of an expression signature predictive of outcome in the rarer histologies.

METHODS

In split discovery and validation analysis of 131 Mayo Clinic EOC cases, we used clustering to determine clinically relevant transcriptome classes using microarray gene expression measurements. The signature was validated in 967 EOC tumours (91 rare histological subtypes) with recurrence information.

RESULTS

We found two validated transcriptome classes associated with progression-free survival (PFS) in the Mayo Clinic EOC cases (P=8.24 × 10(-3)). This signature was further validated in the public expression data sets involving the rare EOC histologies, where these two classes were also predictive of PFS (P=1.43 × 10(-3)). In contrast, the signatures were not predictive of PFS in the high-grade serous EOC cases. Moreover, genes upregulated in Class-1 (with better outcome) were showed enrichment in steroid hormone biosynthesis (false discovery rate, FDR=0.005%) and WNT signalling pathway (FDR=1.46%); genes upregulated in Class-2 were enriched in cell cycle (FDR=0.86%) and toll-like receptor pathways (FDR=2.37%).

CONCLUSIONS

These findings provide important biological insights into the rarer EOC histologies that may aid in the development of targeted treatment options for the rarer histologies.

Molecular diagnosis and molecular profiling to detect treatment-resistant ovarian cancer

INTRODUCTION

Epithelial ovarian cancer remains the gynecologic tumor with the highest rate of recurrence after initial optimal cytoreductive surgery followed by adjuvant chemotherapy. Unfortunately, with the development of recurrent ovarian cancer often comes the discovery of chemo-resistant disease. The absence of improvement in long term survival, notwithstanding the use of newer agents as is seen in other cancers, emphasizes the need for improved understanding of the processes that lead to chemo-resistant disease.

AREAS COVERED

This review will cover the following topics: 1. Molecular and cellular mechanisms in platinum and paclitaxel resistance 2. Other molecular mediators of chemo-resistance 3. Expression of stem cell markers in ovarian cancer and relationship to chemo-resistance 4. MicroRNA and long non-coding RNA expression in chemo-resistant ovarian cancer 5. Determination of chromosomal aberrations as markers of chemo-resistance 6. Molecular profiling in chemo-resistant disease. A standard MEDLINE search was performed using the key words; ovarian cancer, chemo-resistant disease, molecular profiling, cancer stem cells and chemotherapy. Expert Commentary: Over the next few years the challenge remains to precisely determine the mechanisms responsible for the onset and maintenance of chemo-resistance and to effectively target these mechanisms.

Prediction of Optimal Cytoreductive Surgery of Serous Ovarian Cancer With Gene Expression Data

OBJECTIVES

Cytoreductive surgery is the cornerstone of ovarian cancer (OVCA) treatment. Detractors of initial maximal surgical effort argue that aggressive tumor biology will dictate survival, not the surgical effort. We investigated the role of biology in achieving optimal cytoreduction in serous OVCA using microarray gene expression analysis.

METHODS

For the initial model, we used a gene expression signature from a microarray expression analysis of 124 women with serous OVCA, defining optimal cytoreduction as removal of all disease greater than 1 cm (with 64 women having optimal and 60 suboptimal cytoreduction). We then applied this model to 2 independent data sets: the Australian Ovarian Cancer Study (AOCS; 190 samples) and The Cancer Genome Atlas (TCGA; 468 samples). We performed a second analysis, defining optimal cytoreduction as removal of all disease to microscopic residual, using data from AOCS to create the gene signature and validating results in TCGA data set.

RESULTS

Of the 12,718 genes included in the initial analysis, 58 predicted accuracy of cytoreductive surgery 69% of the time (P = 0.005). The performance of this classifier, measured by the area under the receiver operating characteristic curve, was 73%. When applied to TCGA and AOCS, accuracy was 56% (P = 0.16) and 62% (P = 0.01), respectively, with performance at 57% and 65%, respectively. In the second analysis, 220 genes predicted accuracy of cytoreductive surgery in the AOCS set 74% of the time, with performance of 73%. When these results were validated in TCGA set, accuracy was 57% (P = 0.31) and performance was at 62%.

CONCLUSION

Gene expression data, used as a proxy of tumor biology, do not predict accurately nor consistently the ability to perform optimal cytoreductive surgery. Other factors, including surgical effort, may also explain part of the model. Additional studies integrating more biological and clinical data may improve the prediction model.

The limitations of simple gene set enrichment analysis assuming gene independence

Since its first publication in 2003, the Gene Set Enrichment Analysis method, based on the Kolmogorov-Smirnov statistic, has been heavily used, modified, and also questioned. Recently a simplified approach using a one-sample t-test score to assess enrichment and ignoring gene-gene correlations was proposed by Irizarry et al. 2009 as a serious contender. The argument criticizes Gene Set Enrichment Analysis's nonparametric nature and its use of an empirical null distribution as unnecessary and hard to compute. We refute these claims by careful consideration of the assumptions of the simplified method and its results, including a comparison with Gene Set Enrichment Analysis's on a large benchmark set of 50 datasets. Our results provide strong empirical evidence that gene-gene correlations cannot be ignored due to the significant variance inflation they produced on the enrichment scores and should be taken into account when estimating gene set enrichment significance. In addition, we discuss the challenges that the complex correlation structure and multi-modality of gene sets pose more generally for gene set enrichment methods.

Studying platinum sensitivity and resistance in high-grade serous ovarian cancer: Different models for different questions

High-grade serous ovarian cancer (HGSOC) has the highest mortality rate among all gynecological cancers. Patients are generally diagnosed in an advanced stage with the majority of cases displaying platinum resistant relapses. Recent genomic interrogation of large numbers of HGSOC patient samples indicated high complexity in terms of genetic aberrations, intra- and intertumor heterogeneity and underscored their lack of targetable oncogenic mutations. Sub-classifications of HGSOC based on expression profiles, termed 'differentiated', 'immunoreactive', 'mesenchymal' and 'proliferative', were shown to have prognostic value. In addition, in almost half of all HGSOC patients, a deficiency in homologous recombination (HR) was found that potentially can be targeted using PARP inhibitors. Developing precision medicine requires advanced experimental models. In the current review, we discuss experimental HGSOC models in which resistance to platinum therapy and the use of novel therapeutics can be carefully studied. Panels of better-defined primary cell lines need to be established to capture the full spectrum of HGSOC subtypes. Further refinement of cell lines is obtained with a 3-dimensional culture model mimicking the tumor microenvironment. Alternatively, ex vivo ovarian tumor tissue slices are used. For in vivo studies, larger panels of ovarian cancer patient-derived xenografts (PDXs) are being established, encompassing all expression subtypes. Ovarian cancer PDXs grossly retain tumor heterogeneity and clinical response to platinum therapy is preserved. PDXs are currently used in drug screens and as avatars for patient response. The role of the immune system in tumor responses can be assessed using humanized PDXs and immunocompetent genetically engineered mouse models. Dynamic tracking of genetic alterations in PDXs as well as patients during treatment and after relapse is feasible by sequencing circulating cell-free tumor DNA and analyzing circulating tumor cells. We discuss how various models and methods can be combined to delineate the molecular mechanisms underlying platinum resistance and to select HGSOC patients other than BRCA1/2-mutation carriers that could potentially benefit from the synthetic lethality of PARP inhibitors. This integrated approach is a first step to improve therapy outcomes in specific subgroups of HGSOC patients.

Pareto Optimization Identifies Diverse Set of Phosphorylation Signatures Predicting Response to Treatment with Dasatinib

Multivariate biomarkers that can predict the effectiveness of targeted therapy in individual patients are highly desired. Previous biomarker discovery studies have largely focused on the identification of single biomarker signatures, aimed at maximizing prediction accuracy. Here, we present a different approach that identifies multiple biomarkers by simultaneously optimizing their predictive power, number of features, and proximity to the drug target in a protein-protein interaction network. To this end, we incorporated NSGA-II, a fast and elitist multi-objective optimization algorithm that is based on the principle of Pareto optimality, into the biomarker discovery workflow. The method was applied to quantitative phosphoproteome data of 19 non-small cell lung cancer (NSCLC) cell lines from a previous biomarker study. The algorithm successfully identified a total of 77 candidate biomarker signatures predicting response to treatment with dasatinib. Through filtering and similarity clustering, this set was trimmed to four final biomarker signatures, which then were validated on an independent set of breast cancer cell lines. All four candidates reached the same good prediction accuracy (83%) as the originally published biomarker. Although the newly discovered signatures were diverse in their composition and in their size, the central protein of the originally published signature — integrin β4 (ITGB4) — was also present in all four Pareto signatures, confirming its pivotal role in predicting dasatinib response in NSCLC cell lines. In summary, the method presented here allows for a robust and simultaneous identification of multiple multivariate biomarkers that are optimized for prediction performance, size, and relevance.

Inhibition of TMEM45A suppresses proliferation, induces cell cycle arrest and reduces cell invasion in human ovarian cancer cells

The association of TMEM45A with various cancers has been recently reported. However, the biological function of TMEM45A in ovarian cancer remains unclear. The present study aimed to elucidate the role of TMEM45A in regulating the biological behavior of ovarian cancer cells. We compared the expression of TMEM45A between ovarian cancer tissues and normal tissues based on RNA-sequencing data of the ovarian cancer cohort from The Cancer Genome Atlas (TCGA) project and our real-time PCR data from 25 pairs of ovarian cancer and their matched non-cancerous tissue samples. The expression of TMEM45A was then suppressed in two ovarian cancer cell lines, HO-8910 and A2780, by RNA interference. Cell proliferation, cell cycle distribution, adhesion and invasive ability were then detected using the Cell Counting Kit-8 assay (CCK-8), propidium iodide (PI) staining, and cell adhesion and Transwell assays, respectively. In addition, the mRNA and protein levels of transforming growth factor-β (TGF-β1 and TGF-β2), Ras homolog family member A (RhoA) and Rho-associated kinase 2 (ROCK2) were detected with real-time PCR and western blotting, respectively. TCGA data and our real-time PCR results demonstrated the overexpression of TMEM45A in ovarian cancer. Silencing of TMEM45A significantly inhibited cell proliferation and significantly increased the cell population in the G1 phase. Moreover, knockdown of TMEM45A also inhibited cell adhesion as well as cell invasion. More importantly, suppression of TMEM45A notably downregulated the expression of TGF-β1, TGF-β2, RhoA and ROCK2. In conclusion, TMEM45A may function as an oncogene for ovarian cancer, and inhibition of TMEM45A may be a therapeutic strategy for ovarian cancer.

A reproducible approach to high-throughput biological data acquisition and integration

Modern biological research requires rapid, complex, and reproducible integration of multiple experimental results generated both internally and externally (e.g., from public repositories). Although large systematic meta-analyses are among the most effective approaches both for clinical biomarker discovery and for computational inference of biomolecular mechanisms, identifying, acquiring, and integrating relevant experimental results from multiple sources for a given study can be time-consuming and error-prone. To enable efficient and reproducible integration of diverse experimental results, we developed a novel approach for standardized acquisition and analysis of high-throughput and heterogeneous biological data. This allowed, first, novel biomolecular network reconstruction in human prostate cancer, which correctly recovered and extended the NFκB signaling pathway. Next, we investigated host-microbiome interactions. In less than an hour of analysis time, the system retrieved data and integrated six germ-free murine intestinal gene expression datasets to identify the genes most influenced by the gut microbiota, which comprised a set of immune-response and carbohydrate metabolism processes. Finally, we constructed integrated functional interaction networks to compare connectivity of peptide secretion pathways in the model organisms Escherichia coli, Bacillus subtilis, and Pseudomonas aeruginosa.

Elevated AKAP12 in paclitaxel-resistant serous ovarian cancer cells is prognostic and predictive of poor survival in patients

A majority of high-grade (HG) serous ovarian cancer (SOC) patients develop resistant disease despite high initial response rates to platinum/paclitaxel-based chemotherapy. We identified shed/secreted proteins in preclinical models of paclitaxel-resistant human HGSOC models and correlated these candidate proteins with patient outcomes using public data from HGSOC patients. Proteomic analyses of a HGSOC cell line secretome was compared to those from a syngeneic paclitaxel-resistant variant and from a line established from an intrinsically chemorefractory HGSOC patient. Associations between the identified candidate proteins and patient outcome were assessed in a discovery cohort of 545 patients and two validation cohorts totaling 795 independent SOC patients. Among the 81 differentially abundant proteins identified (q < 0.05) from paclitaxel-sensitive vs -resistant HGSOC cell secretomes, AKAP12 was verified to be elevated in all models of paclitaxel-resistant HGSOC. Furthermore, elevated AKAP12 transcript expression was associated with worse progression-free and overall survival. Associations with outcome were observed in three independent cohorts and remained significant after adjusted multivariate modeling. We further provide evidence to support that differential gene methylation status is associated with elevated expression of AKAP12 in taxol-resistant ovarian cancer cells and ovarian cancer patient subsets. Elevated expression and shedding/secretion of AKAP12 is characteristic of paclitaxel-resistant HGSOC cells, and elevated AKAP12 transcript expression is a poor prognostic and predictive marker for progression-free and overall survival in SOC patients.

Microarray analysis of differentially expressed genes in ovarian and fallopian tube epithelium from risk-reducing salpingo-oophorectomies

Mutations in the BRCA1 and BRCA2 genes confer an increased lifetime risk for breast and ovarian cancer. Ovarian cancer risk can be decreased by risk-reducing salpingo-oophorectomy (RRSO). Studies on RRSO material have altered the paradigm of serous ovarian cancer pathogenesis. The purpose of this study was to identify candidate genes possibly involved in the pathogenesis of serous ovarian cancer by carrying out a microarray analysis of differentially expressed genes in BRCA1/2- mutation positive ovarian and fallopian tube epithelium derived from RRSO surgery. Freshly frozen ovarian and fallopian tube samples from nine BRCA1/2 mutation carriers scheduled for RRSO were prospectively collected together with five mutation-negative control patients undergoing salpingo-oophorectomy for benign indications. Microarray analysis of genome-wide gene expression was performed on ovarian and fallopian tube samples from the BRCA1/2 and control patients. The validation of microarray data was performed by quantitative real-time polymerase chain reaction (qRT-PCR) in selected cases of RRSO samples and also in high grade serous carcinoma samples collected from patients with a BRCA phenotype. From 22,733 genes, 454 transcripts were identified that were differentially expressed in BRCA1/2 mutation carriers when compared with controls, pooling all ovarian and fallopian tube samples together. Of these, 299 genes were statistically significantly downregulated and 155 genes upregulated. Differentially expressed genes in BRCA1/2 samples reported here might be involved in serous ovarian carcinogenesis and provide interesting targets for further studies.

Association between morphologic CT imaging traits and prognostically relevant gene signatures in women with high-grade serous ovarian cancer: a hypothesis-generating study

PURPOSE

To investigate associations among imaging traits observed on computed tomographic (CT) images, Classification of Ovarian Cancer (CLOVAR) gene signatures, and survival in women with high-grade serous ovarian cancer (HGSOC).

MATERIALS AND METHODS

The institutional review board approved this HIPAA-compliant retrospective study of CT images obtained before cytoreductive surgery in 46 women with HGSOC, whose tumors were subjected to molecular analysis performed by the Cancer Genome Atlas Research Network. Two readers independently evaluated the CT features of the primary ovarian mass and sites of metastatic spread if present, including size, outline, and texture. Fisher exact test was used to examine the relationship between imaging traits and CLOVAR subtypes (CLOVAR differentiated, immunoreactive, mesenchymal, and proliferative). Kaplan-Meier and Cox proportional hazards regression survival analyses were performed.

RESULTS

The presence of mesenteric infiltration and diffuse peritoneal involvement by tumor at CT were significantly associated with CLOVAR subtype (P = .002-.004 for reader 1 and P = .005-.012 for reader 2). Mesenteric infiltration at CT was associated with CLOVAR mesenchymal subtype. Patients with mesenteric infiltration had shorter median progression-free survival than patients without mesenteric involvement (14.7 months vs 25.6 months according to both readers; P = .019 for reader 1 and .015 for reader 2) and overall survival (49.0 vs 58.2 months; P = .014 [reader 1] and 50.0 vs 59.1 months; P = .015 [reader 2]). No other imaging features were significantly associated with CLOVAR subtype or survival.

CONCLUSION

Specific CT imaging traits were associated with the CLOVAR subtypes and survival in patients with HGSOC.

Differential antigen expression profile predicts immunoreactive subset of advanced ovarian cancers

The presence and composition of lymphocytes characterizing an immune response has been connected to prognosis in advanced ovarian cancer. Our aim is to establish novel associations between prognosis and the expression of immune-related genes through a focused screen utilizing publicly available high-throughput assays. We consider transcriptome profiles from n=1137 advanced ovarian cancer patients observed in four separate studies divided into discovery/validation sets (n=503/n=634). We focus on a subset of lymphocyte markers, antigen presentation and processing genes, T cell receptor associated co-stimulatory/repressor genes and cancer testis (CT) antigens. We modeled differential expression and co-expression using these subsets and tested for association with overall survival. Fifteen of 64 immune-related genes are associated with survival of which 5 are reproduced in the validation set. The expression of these genes defines an immunoreactive (IR) subgroup of patients with a favorable prognosis. Phenotypic characterization of the immune compartment signal includes upregulation of markers of CD8+ T-cell activation in these patients. Using multivariate model building, we find that the expression of 6 CT antigens can predict IR status in the discovery and validation sets. These analyses confirm that a genomic approach can reproducibly detect lymphocyte signals in tumor tissue suggesting a novel way to study the tumor microenvironment. Our search has identified new candidate prognostic markers associated with immune components and uncovered preliminary evidence of prognostic subgroups associated with different immune mechanisms.

Ovarian cancer biomarker discovery based on genomic approaches

Ovarian cancer presents at an advanced stage in more than 75% of patients. Early detection has great promise to improve clinical outcomes. Although the advancing proteomic technologies led to the discovery of numerous ovarian cancer biomarkers, no screening method has been recommended for early detection of ovarian cancer. Complexity and heterogeneity of ovarian carcinogenesis is a major obstacle to discover biomarkers. As cancer arises due to accumulation of genetic change, understanding the close connection between genetic changes and ovarian carcinogenesis would provide the opportunity to find novel gene-level ovarian cancer biomarkers. In this review, we summarize the various gene-based biomarkers by genomic technologies, including inherited gene mutations, epigenetic changes, and differential gene expression. In addition, we suggest the strategy to discover novel gene-based biomarkers with recently introduced next generation sequencing.

Más-o-menos: a simple sign averaging method for discrimination in genomic data analysis

MOTIVATION

The successful translation of genomic signatures into clinical settings relies on good discrimination between patient subgroups. Many sophisticated algorithms have been proposed in the statistics and machine learning literature, but in practice simpler algorithms are often used. However, few simple algorithms have been formally described or systematically investigated.

RESULTS

We give a precise definition of a popular simple method we refer to as más-o-menos, which calculates prognostic scores for discrimination by summing standardized predictors, weighted by the signs of their marginal associations with the outcome. We study its behavior theoretically, in simulations and in an extensive analysis of 27 independent gene expression studies of bladder, breast and ovarian cancer, altogether totaling 3833 patients with survival outcomes. We find that despite its simplicity, más-o-menos can achieve good discrimination performance. It performs no worse, and sometimes better, than popular and much more CPU-intensive methods for discrimination, including lasso and ridge regression.

AVAILABILITY AND IMPLEMENTATION

Más-o-menos is implemented for survival analysis as an option in the survHD package, available from http://www.bitbucket.org/lwaldron/survhd and submitted to Bioconductor.

Gynecologic biopsy for molecular profiling: a review for the interventional radiologist

The interventional radiologist is often asked to obtain multiple biopsies of gynecological malignancies for genetic profiling. This article reviews the current indications for gynecological biopsy as well as how the information gained contributes to a personalized medicine plan for the individual patient. The specific focus of this review is the current knowledge and practice of molecular profiling for gynecological malignancies.

Risk prediction for late-stage ovarian cancer by meta-analysis of 1525 patient samples

BACKGROUND

Ovarian cancer causes more than 15000 deaths per year in the United States. The survival of patients is quite heterogeneous, and accurate prognostic tools would help with the clinical management of these patients.

METHODS

We developed and validated two gene expression signatures, the first for predicting survival in advanced-stage, serous ovarian cancer and the second for predicting debulking status. We integrated 13 publicly available datasets totaling 1525 subjects. We trained prediction models using a meta-analysis variation on the compound covariable method, tested models by a "leave-one-dataset-out" procedure, and validated models in additional independent datasets. Selected genes from the debulking signature were validated by immunohistochemistry and quantitative reverse-transcription polymerase chain reaction (qRT-PCR) in two further independent cohorts of 179 and 78 patients, respectively. All statistical tests were two-sided.

RESULTS

The survival signature stratified patients into high- and low-risk groups (hazard ratio = 2.19; 95% confidence interval [CI] = 1.84 to 2.61) statistically significantly better than the TCGA signature (P = .04). POSTN, CXCL14, FAP, NUAK1, PTCH1, and TGFBR2 were validated by qRT-PCR (P < .05) and POSTN, CXCL14, and phosphorylated Smad2/3 were validated by immunohistochemistry (P < .001) as independent predictors of debulking status. The sum of immunohistochemistry intensities for these three proteins provided a tool that classified 92.8% of samples correctly in high- and low-risk groups for suboptimal debulking (area under the curve = 0.89; 95% CI = 0.84 to 0.93).

CONCLUSIONS

Our survival signature provides the most accurate and validated prognostic model for early- and advanced-stage high-grade, serous ovarian cancer. The debulking signature accurately predicts the outcome of cytoreductive surgery, potentially allowing for stratification of patients for primary vs secondary cytoreduction.

Comparative meta-analysis of prognostic gene signatures for late-stage ovarian cancer

BACKGROUND

Ovarian cancer is the fifth most common cause of cancer deaths in women in the United States. Numerous gene signatures of patient prognosis have been proposed, but diverse data and methods make these difficult to compare or use in a clinically meaningful way. We sought to identify successful published prognostic gene signatures through systematic validation using public data.

METHODS

A systematic review identified 14 prognostic models for late-stage ovarian cancer. For each, we evaluated its 1) reimplementation as described by the original study, 2) performance for prognosis of overall survival in independent data, and 3) performance compared with random gene signatures. We compared and ranked models by validation in 10 published datasets comprising 1251 primarily high-grade, late-stage serous ovarian cancer patients. All tests of statistical significance were two-sided.

RESULTS

Twelve published models had 95% confidence intervals of the C-index that did not include the null value of 0.5; eight outperformed 97.5% of signatures including the same number of randomly selected genes and trained on the same data. The four top-ranked models achieved overall validation C-indices of 0.56 to 0.60 and shared anticorrelation with expression of immune response pathways. Most models demonstrated lower accuracy in new datasets than in validation sets presented in their publication.

CONCLUSIONS

This analysis provides definitive support for a handful of prognostic models but also confirms that these require improvement to be of clinical value. This work addresses outstanding controversies in the ovarian cancer literature and provides a reproducible framework for meta-analytic evaluation of gene signatures.