Proteogenomic studies in epithelial ovarian cancer: established knowledge and future needs

Periostin promotes ovarian cancer metastasis by enhancing M2 macrophages and cancer-associated fibroblasts via integrin-mediated NF-κB and TGF-β2 signaling

BACKGROUND

Ovarian cancer has the highest mortality among gynecological cancers due to late diagnosis and lack of effective targeted therapy. Although the study of interplay between cancer cells with their microenvironment is emerging, how ovarian cancer triggers signaling that coordinates with immune cells to promote metastasis is still elusive.

METHODS

Microarray and bioinformatics analysis of low and highly invasive ovarian cancer cell lines were used to reveal periostin (POSTN), a matrix protein with multifunctions in cancer, with elevated expression in the highly invasive cells. Anchorage independent assay, Western blot, RNA interference, confocal analysis and neutralizing antibody treatment were performed to analyze the effects of POSTN on tumor promotion and to explore the underlying mechanism. Chemotaxis, flow cytometry and cytokine array analyses were undertaken to analyze the involvement of POSTN in cancer-associated fibroblast (CAF) and macrophage modulation. Correlations between POSTN expression levels and clinical characteristics were analyzed using the Oncomine, commercial ovarian cancer cDNA and China Medical University Hospital patient cohort. In vivo effect of POSTN on metastasis was studied using a mouse xenograft model.

RESULTS

Expression of POSTN was found to be elevated in highly invasive ovarian cancer cells. We observed that POSTN was co-localized with integrin β3 and integrin β5, which was important for POSTN-mediated activation of ERK and NF-κB. Ectopic expression of POSTN enhanced whereas knockdown of POSTN decreased cancer cell migration and invasion in vitro, as well as tumor growth and metastasis in vivo. POSTN enhanced integrin/ERK/NF-κB signaling through an autocrine effect on cancer cells to produce macrophage attracting and mobilizing cytokines including MIP-1β, MCP-1, TNFα and RANTES resulting in increased chemotaxis of THP-1 monocytes and their polarization to M2 macrophages in vitro. In agreement, tumors derived from POSTN-overexpressing SKOV3 harbored more tumor-associated macrophages than the control tumors. POSTN induced TGF-β2 expression from ovarian cancer cells to promote activation of adipose-derived stromal cells to become CAF-like cells expressing alpha smooth muscle actin and fibroblast activation protein alpha. Consistently, increased CAFs were observed in POSTN overexpressing SKOV3 cells-derived metastatic tumors. In clinical relevance, we found that expression of POSTN was positively correlated with advanced-stage diseases and poor overall survival of patients.

CONCLUSIONS

Our study revealed a POSTN-integrin-NF-κB-mediated signaling and its involvement in enhancing M2 macrophages and CAFs, which could potentially participate in promoting tumor growth. Our results suggest that POSTN could be a useful prognosis marker and potential therapeutic target.

Bayesian Hierarchical Varying-sparsity Regression Models with Application to Cancer Proteogenomics

Identifying patient-specific prognostic biomarkers is of critical importance in developing personalized treatment for clinically and molecularly heterogeneous diseases such as cancer. In this article, we propose a novel regression framework, Bayesian hierarchical varying-sparsity regression (BEHAVIOR) models to select clinically relevant disease markers by integrating proteogenomic (proteomic+genomic) and clinical data. Our methods allow flexible modeling of protein-gene relationships as well as induces sparsity in both protein-gene and protein-survival relationships, to select ge-nomically driven prognostic protein markers at the patient-level. Simulation studies demonstrate the superior performance of BEHAVIOR against competing method in terms of both protein marker selection and survival prediction. We apply BEHAV-IOR to The Cancer Genome Atlas (TCGA) proteogenomic pan-cancer data and find several interesting prognostic proteins and pathways that are shared across multiple cancers and some that exclusively pertain to specific cancers.

Blood Plasma-Derived Anti-Glycan Antibodies to Sialylated and Sulfated Glycans Identify Ovarian Cancer Patients

Altered levels of naturally occurring anti-glycan antibodies (AGA) circulating in human blood plasma are found in different pathologies including cancer. Here the levels of AGA directed against 22 negatively charged (sialylated and sulfated) glycans were assessed in high-grade serous ovarian cancer (HGSOC, n = 22) patients and benign controls (n = 31) using our previously developed suspension glycan array (SGA). Specifically, the ability of AGA to differentiate between controls and HGSOC, the most common and aggressive type of ovarian cancer with a poor outcome was determined. Results were compared to CA125, the commonly used ovarian cancer biomarker. AGA to seven glycans that significantly (P<0.05) differentiated between HGSOC and control were identified: AGA to top candidates SiaTn and 6-OSulfo-TF (both IgM) differentiated comparably to CA125. The area under the curve (AUC) of a panel of AGA to 5 glycans (SiaTn, 6-OSulfo-TF, 6-OSulfo-LN, SiaLea, and GM2) (0.878) was comparable to CA125 (0.864), but it markedly increased (0.985) when combined with CA125. AGA to SiaTn and 6-OSulfo-TF were also valuable predictors for HGSOC when CA125 values appeared inconclusive, i.e. were below a certain threshold. AGA-glycan binding was in some cases isotype-dependent and sensitive to glycosidic linkage switch (α2-6 vs. α2-3), to sialylation, and to sulfation of the glycans. In conclusion, plasma-derived AGA to sialylated and sulfated glycans including SiaTn and 6-OSulfo-TF detected by SGA present a valuable alternative to CA125 for differentiating controls from HGSOC patients and for predicting the likelihood of HGSOC, and may be potential HGSOC tumor markers.

Integration of Proteomics and Transcriptomics Data Sets for the Analysis of a Lymphoma B-Cell Line in the Context of the Chromosome-Centric Human Proteome Project

A comprehensive study of the molecular active landscape of human cells can be undertaken to integrate two different but complementary perspectives: transcriptomics, and proteomics. After the genome era, proteomics has emerged as a powerful tool to simultaneously identify and characterize the compendium of thousands of different proteins active in a cell. Thus, the Chromosome-centric Human Proteome Project (C-HPP) is promoting a full characterization of the human proteome combining high-throughput proteomics with the data derived from genome-wide expression profiling of protein-coding genes. Here we present a full proteomic profiling of a human lymphoma B-cell line (Ramos) performed using a nanoUPLC-LTQ-Orbitrap Velos proteomic platform, combined to an in-depth transcriptomic profiling of the same cell type. Data are available via ProteomeXchange with identifier PXD001933. Integration of the proteomic and transcriptomic data sets revealed a 94% overlap in the proteins identified by both -omics approaches. Moreover, functional enrichment analysis of the proteomic profiles showed an enrichment of several functions directly related to the biological and morphological characteristics of B-cells. In turn, about 30% of all protein-coding genes present in the whole human genome were identified as being expressed by the Ramos cells (stable average of 30% genes along all the chromosomes), revealing the size of the protein expression-set present in one specific human cell type. Additionally, the identification of missing proteins in our data sets has been reported, highlighting the power of the approach. Also, a comparison between neXtProt and UniProt database searches has been performed. In summary, our transcriptomic and proteomic experimental profiling provided a high coverage report of the expressed proteome from a human lymphoma B-cell type with a clear insight into the biological processes that characterized these cells. In this way, we demonstrated the usefulness of combining -omics for a comprehensive characterization of specific biological systems.

Meta-analysis of incomplete microarray studies

Meta-analysis of microarray studies to produce an overall gene list is relatively straightforward when complete data are available. When some studies lack information-providing only a ranked list of genes, for example-it is common to reduce all studies to ranked lists prior to combining them. Since this entails a loss of information, we consider a hierarchical Bayes approach to meta-analysis using different types of information from different studies: the full data matrix, summary statistics, or ranks. The model uses an informative prior for the parameter of interest to aid the detection of differentially expressed genes. Simulations show that the new approach can give substantial power gains compared with classical meta-analysis and list aggregation methods. A meta-analysis of 11 published studies with different data types identifies genes known to be involved in ovarian cancer and shows significant enrichment.

Interactions between osteopontin and vascular endothelial growth factor: Implications for cancer

For this comprehensive review, 257 publications with the keywords "osteopontin" or "OPN" and "vascular endothelial growth factor" or "VEGF" in PubMed were screened (time frame from year 1996 to year 2014). 37 articles were excluded because they were not focused on the interactions between these molecules, and papers relevant for transformation-related phenomena were selected. Osteopontin (OPN) and vascular endothelial growth factor (VEGF) are characterized by a convergence in function for regulating cell motility and angiogenesis, the response to hypoxia, and apoptosis. Often, they are co-expressed or one molecule induces the other, however, in some settings OPN-associated pathways and VEGF-associated pathways are distinct. Their relationships affect the pathogenesis in cancer, where they contribute to progression and angiogenesis and serve as markers for poor prognosis. The inhibition of OPN may reduce VEGF levels and suppress tumor progression. In vascular pathologies, these two cytokines mediate remodeling, but may also perpetuate inflammation and narrowing of the arteries. OPN and VEGF are elevated and contribute to vascularization in inflammatory diseases.

Flexible and accessible workflows for improved proteogenomic analysis using the Galaxy framework

Proteogenomics combines large-scale genomic and transcriptomic data with mass-spectrometry-based proteomic data to discover novel protein sequence variants and improve genome annotation. In contrast with conventional proteomic applications, proteogenomic analysis requires a number of additional data processing steps. Ideally, these required steps would be integrated and automated via a single software platform offering accessibility for wet-bench researchers as well as flexibility for user-specific customization and integration of new software tools as they emerge. Toward this end, we have extended the Galaxy bioinformatics framework to facilitate proteogenomic analysis. Using analysis of whole human saliva as an example, we demonstrate Galaxy’s flexibility through the creation of a modular workflow incorporating both established and customized software tools that improve depth and quality of proteogenomic results. Our customized Galaxy-based software includes automated, batch-mode BLASTP searching and a Peptide Sequence Match Evaluator tool, both useful for evaluating the veracity of putative novel peptide identifications. Our complex workflow (approximately 140 steps) can be easily shared using built-in Galaxy functions, enabling their use and customization by others. Our results provide a blueprint for the establishment of the Galaxy framework as an ideal solution for the emerging field of proteogenomics.

Reliable in vitro studies require appropriate ovarian cancer cell lines

Ovarian cancer is the fifth most common cause of cancer death in women and the leading cause of death from gynaecological malignancies. Of the 75% women diagnosed with locally advanced or disseminated disease, only 30% will survive five years following treatment. This poor prognosis is due to the following reasons: limited understanding of the tumor origin, unclear initiating events and early developmental stages of ovarian cancer, lack of reliable ovarian cancer-specific biomarkers, and drug resistance in advanced cases. In the past, in vitro studies using cell line models have been an invaluable tool for basic, discovery-driven cancer research. However, numerous issues including misidentification and cross-contamination of cell lines have hindered research efforts. In this study we examined all ovarian cancer cell lines available from cell banks. Hereby, we identified inconsistencies in the reporting, difficulties in the identification of cell origin or clinical data of the donor patients, restricted ethnic and histological type representation, and a lack of tubal and peritoneal cancer cell lines. We recommend that all cell lines should be distributed via official cell banks only with strict guidelines regarding the minimal available information required to improve the quality of ovarian cancer research in future.

Increased WD-repeat containing protein 1 in interstitial fluid from ovarian carcinomas shown by comparative proteomic analysis of malignant and healthy gynecological tissue

We aimed to identify differentially expressed proteins in interstitial fluid from ovarian cancer employing multiple fractioning and high resolution mass spectrometry-based proteomic analysis, and asked whether specific proteins that may serve as biomarker candidates or therapeutic targets could be identified. High throughput proteomics was conducted on immunodepleted and fractioned interstitial fluid from pooled samples of ovarian carcinomas, using endometrial carcinomas and healthy ovarian tissue as controls. Differential analysis revealed the up-regulation of extracellular proteasomes in tumor interstitial fluid compared to the healthy control. Moreover, a number of differentially expressed proteins in interstitial fluid from ovarian carcinomas compared with control tissues were identified. Detection of proteasome 20S related proteins in TIF compared to IF from healthy tissue indicates that the 20S proteasome can have a role in the tumor microenvironment. Six selected proteins, CEACAM5, FREM2, MUC5AC, TFF3, PYCARD and WDR1, were independently validated in individual tumor lysates from ovarian carcinomas by multiple reaction monitoring initiated detection and sequence analysis, Western blot and/or selected reaction monitoring. Quantification of specific proteins revealed substantial heterogeneity between individual samples. Nevertheless, WD repeat-containing protein 1 was confirmed as being significantly overexpressed in interstitial fluid from ovarian carcinomas compared to healthy ovarian tissue by Orbitrap analysis of individual native interstitial fluid from ovarian and endometrial carcinomas and healthy ovarian tissue. We suggest that this protein should be explored as a therapeutic target in ovarian carcinomas. This article is part of a Special Issue entitled: An Updated Secretome.

Serum antiglycan antibody detection of nonmucinous ovarian cancers by using a printed glycan array

Epithelial ovarian cancer has the highest mortality rate among gynecological cancers. Altered glycosylation is associated with oncogenic transformation producing tumor-associated carbohydrate antigens. We investigated the potential of natural occurring antiglycan antibodies in the diagnosis of ovarian cancer by using printed glycan array. Antiglycan antibodies bound to 203 chemically synthesized printed glycans were detected via biotin-streptavidin fluorescence system in serum of women with normal operative findings (healthy controls; n = 24) and nonmucinous borderline or ovarian cancer of various FIGO stages (n = 33). Data were validated measuring blood group associated di-, tri and tetrasaccharide antigens on known ABO blood groups. Antiglycan antibodies demonstrated high reproducibility (r(c) > 0.9). Cluster analysis identified repetitive patterns of specific core carbohydrate structures: 11 N-linked glycans, 3 O-linked glycans and 2 glycosphingolipids. Biomarker detection revealed 24 glycans including P(1) (Galα1-4Galβ1-4GlcNAcβ; p < 0.001) significantly discriminating between (low-) malignant tumors and healthy controls. Comparable sensitivity and specificity with tumor marker CA125 was achieved by a panel of multivariate selected and linear combined antiglycan antibody signals (79.2 and 84.8%, respectively). Our findings demonstrate the potential of glycan arrays in the development of a new generation of biomarkers for ovarian cancer.

Erlotinib or gefitinib for the treatment of relapsed platinum pretreated non-small cell lung cancer and ovarian cancer: a systematic review

BACKGROUND

Platinum-based chemotherapy is the standard of care for ovarian cancer and non-small cell lung cancer (NSCLC). However, resistance to platinum agents invariably develops. Targeted therapies, such as tyrosine kinase inhibitors (TKIs), have great potential here as they exert their anti-tumour effect via alternative mechanisms to platinum-based drugs and as such may remain unaffected by emergent resistance to platinum.

METHODS

A systematic review was conducted to investigate whether two EGFR-TKIs, erlotinib and gefitinib, have efficacy in the platinum-resistance setting. Preclinical studies of platinum-resistant cancer cell lines, which had been subsequently treated with EGFR-TKIs, were sought to establish proof-of-concept. Clinical trials reporting administration of EGFR-TKIs to ovarian cancer and NSCLC patients relapsed after therapy with platinum drugs were investigated to determine sensitivity of these cohorts to EGFR-TKI treatment. The role of EGFR mutation, copy number and protein expression on response to EGFR-TKIs after failure of platinum chemotherapy were also investigated.

RESULTS

Preclinical models of platinum-resistant cancer were found which display a spectrum of cross-resistance profiles to EGFR-TKIs. Sensitivity to EGFR-TKIs is dependent on the activation of the EGFR pathway or EGFR interacting proteins such as HER-2. EGFR-TKIs show favourable response rates in platinum-pretreated NSCLC, 11.14% and 15.25% for 150mg/day erlotinib and 250mg/day gefitinib, respectively. These response rates significantly improve in patients of Asian descent (28.3% and 29.17%, respectively) and patients with EGFR activation mutations (41.6% and 63.89%, respectively) or increased copy number (33.3% and 45.45%, respectively). Gefitinib significantly outperformed erlotinib and should therefore be the EGFR-TKI of choice in platinum-pretreated relapsed NSCLC. In contrast, response rates are very poor to both erlotinib and gefitinib in platinum pretreated ovarian cancer, 0-5.9% and they should not be used in this cohort of patients. Preclinical models demonstrate that, while cross resistance can occur between platinums and EGFR-TKIs, there is not a generalised cross-resistance phenotype. Erlotinib and gefitinib are suitable for the treatment of platinum-pretreated NSCLC, particularly in patients with EGFR mutations or increases in copy number. Unfortunately, the high rates of EGFR protein overexpression in ovarian cancer are not translating to a clinically useful therapeutic target for EGFR-TKIs; EGFR mutations are rare in ovarian cancer. Newer TKIs may improve response rates in these cohorts and future clinical trials need to collect tumour biopsies from all patients to ensure the success of personalised chemotherapy.

No benefit from combining HE4 and CA125 as ovarian tumor markers in a clinical setting

OBJECTIVE

About 70% of epithelial ovarian cancer patients (EOC) are diagnosed at advanced stage with a five-year survival rate of only 30%. Whilst CA125 detects peritoneally-spread disease, it has limited sensitivity for early cancers, many of which are potentially curable.

METHODS

We compared the new commercially available tumor marker HE4 with CA125 individually, in combination, within the risk of malignancy index (RMI) and the newly defined risk of malignancy algorithm (ROMA). Our prospectively-collected cohort of 160 patients consisted of healthy controls, benign diseases, and borderline tumors/adenocarcinomas of ovarian, tubal, peritoneal and endometrial origin. HE4 and CA125 were measured in serum using standardized ELISA.

RESULTS

Both markers showed similar diagnostic performance in the detection of EOC at clinically defined thresholds (CA125 35U/ml; HE4 70pM) but HE4 was not elevated in endometriosis. Comparison of non-malignant diagnoses (n=71) versus early stage ovarian and tubal cancers (n=19) revealed that HE4 and ROMA displayed the best diagnostic performance (AUC 0.86/0.87, specificity 85.9%/87.3% and sensitivity 78.9%/78.9%, respectively). Whilst RMICA125 detects peritoneal cancer better than all other models (AUC 0.99, specificity 97.2%, sensitivity 80.0%), there is no other detection benefit from RMI compared to HE4 alone or included in ROMA.

CONCLUSIONS

The major advantage of HE4 lies in its specificity and improved detection of borderline tumors and early stage ovarian and tubal cancers. HE4 is superior to CA125 with or without RMI and ROMA indices. However, we see no benefit from combining both markers in clinical practice.