Extracellular vesicle-associated miRNAs in ovarian cancer - design of an integrated NGS-based workflow for the identification of blood-based biomarkers for platinum-resistance.

Background Extracellular vesicle (EV)-associated microRNAs (miRNAs) have been suggested as promising biomarkers for blood-based cancer diagnosis. However, one of the major limitations for the use of EVs with diagnostic purpose is the lack of standardized EV-profiling techniques. In this regard, the objective of our study was to design an integrated next-generation sequencing (NGS)-based workflow for analyzing the signature of EV-associated miRNA in the plasma of platinum-resistant ovarian cancer patients. Methods For EV-extraction, different enrichment methods were compared (ExoQuick vs. exoRNeasy). NGS was performed with the Illumina platform. Results We established an integrated NGS-based workflow, including EV-enrichment with the ExoQuick system, which resulted in an optimal RNA-yield and consistent small RNA libraries. We applied this workflow in a pilot cohort of clinically documented platinum-sensitive (n=15) vs. platinum-resistant (n=15) ovarian cancer patients, resulting in a panel of mature EV-associated miRNAs (including ovarian cancer associated miR-181a, miR-1908, miR-21, miR-486 and miR-223), which were differentially abundant in the plasma of platinum-resistant patients. Conclusions This is the first study, analyzing the profile of EV-associated miRNAs in platinum-resistant ovarian cancer patients. We provide rationale to further validate these miRNA candidates in an independent set of patients, in order to characterize their biomarker potential as predictors for platinum-resistance.

Explorative investigation of vascular endothelial growth factor receptor expression in primary ovarian cancer and its clinical relevance.

OBJECTIVES: The identification of novel molecular biomarkers, predicting outcome of ovarian cancer, is highly desirable. Considering that angiogenesis is a critical factor for ascites development and peritoneal dissemination in ovarian cancer and given that the vascular endothelial growth factor (VEGF) receptor signaling axis is a major driver of angiogenesis, we sought to analyze expression and compartmental distribution of VEGF-receptor family in ovarian cancer and to assess its clinical relevance with regard to established clinicopathological parameters, tumor cell dissemination to the bone marrow (BM) and the patient's survival. METHODS: A total of 73 patients with primary ovarian cancer were enrolled into this study. Primary tumor tissue was analyzed for the expression of VEGF-R1, VEGF-R2 and VEGF-R3 by immunohistochemistry. The presence of disseminated tumor cells (DTC) in the BM was analyzed by immunocytochemistry using the pancytokeratin antibody A45B/B3 and subsequent automatic detection based on staining and cytomorphology. RESULTS: In primary ovarian cancer tissue, VEGF-receptor expression, detected with an overall frequency of 44%, was mostly located in the vascular wall and across the stroma; positivity rates for VEGF-R1, VEGF-R2 and VEGF-R3 were 34%, 18% and 26%, respectively. Total VEGF-receptor expression correlated with residual tumor after primary debulking surgery and the presence of DTC at primary diagnosis (p=0.035, p=0.023, respectively). Interestingly, VEGF-R1 positivity significantly correlated with decreased progression-free survival (p=0.026). CONCLUSIONS: This is the first report, suggesting total VEGF-receptor status as a molecular biomarker for monitoring tumor cell spread to the BM and, particularly, revealing prognostic significance of VEGF-R1.

RASSF1A promoter methylation in high-grade serous ovarian cancer: A direct comparison study in primary tumors, adjacent morphologically tumor cell-free tissues and paired circulating tumor DNA.

The RASSF1A promoter is frequently methylated in high-grade serous ovarian cancer (HGSC). We examined RASSF1A promoter methylation in primary tumors, adjacent morphologically tumor cell-free tissues and corresponding circulating tumor DNA (ctDNA) samples of patients with HGSC, using a real-time methylation specific PCR (real-time MSP) and a methylation-sensitive high-resolution melting analysis (MS-HRMA) assay for the detection and semi-quantitative estimation of methylation, respectively. Two groups of primary HGSC tumor FFPE samples were recruited (Group A n=67 and Group B n=61), along with matched adjacent morphologically tumor cell-free tissues (n=58) and corresponding plasma samples (n=59) for group B. Using both assays, RASSF1A promoter was found highly methylated in primary tumors of both groups, and at lower percentages in the adjacent morphologically tumor cell-free tissues. Interestingly, RASSF1A promoter methylation was also observed in ctDNA by real-time MSP. Overall survival (OS) was significantly associated with RASSF1A promoter methylation in primary tumor samples using MS-HRMA (P=0.023). Our results clearly indicate that RASSF1A promoter is methylated in adjacent tissue surrounding the tumor in HGSC patients. We report for the first time that RASSF1A promoter methylation provides significant prognostic information in HGSC patients.

EMT-like circulating tumor cells in ovarian cancer patients are enriched by platinum-based chemotherapy.

BACKGROUND: Assuming that tumor cell dissemination requires a shift to a mesenchymal phenotype, we analyzed the incidence of epithelial-to-mesenchymal-transition (EMT)-like circulating tumor cells (CTCs) in ovarian cancer patients and inquired, how their molecular phenotypes respond to platinum-based chemotherapy and influence outcome. RESULTS: Before surgery, overall detection rate for epithelial CTCs was 18%. EMT-like CTCs were more frequently observed (30%) and were mutually exclusive to epithelial CTCs in the majority of patients (82%). After chemotherapy, EMT-like CTCs increased up to 52%, accompanied by the "de novo" emergence of PI3Kα+/Twist+ EMT-like CTCs. Before surgery, PI3K+ EMT-like CTCs in combination with epithelial CTCs indicated decreased OS (p = 0.02) and FIGO I-III patients with residual tumor burden after surgery were more likely to be positive for EMT-like CTCs after chemotherapy (p = 0.02). In the latter group, epithelial CTCs alone significantly correlated with decreased PFS and OS (p = 0.02, p = 0.002), supported by an additional inclusion of PI3K+ CTCs (OS, p = 0.001). MATERIALS AND METHODS: Blood samples of 91 ovarian cancer patients before surgery and 31 matched samples after adjuvant chemotherapy were evaluated for CTCs with the AdnaTest ovarian cancer and EMT-1, analyzing the epithelial-associated transcripts EpCAM, Muc-1 and CA125 and the EMT-associated transcripts PI3Kα, Akt-2 and Twist. CONCLUSIONS: Platinum-based chemotherapy seems to select for EMT-like CTCs in ovarian cancer patients and provokes a shift towards PI3Kα and Twist expressing CTCs, which may reflect clonal tumor evolution towards therapy resistance. It has to be determined, whether this CTC subgroup may serve as a biomarker to identify patients at high risk.

ERCC1-expressing circulating tumor cells as a potential diagnostic tool for monitoring response to platinum-based chemotherapy and for predicting post-therapeutic outcome of ovarian cancer.

BACKGROUND: We recently showed that the presence of ERCC1+CTCs is an independent predictive biomarker for platinum-resistance and poor prognosis of ovarian cancer. The goal of our current research was to determine how the auxiliary assessment of ERCC1-transcripts influences overall CTC-detection rate. We extended this investigation from an initially predictive setting to paired pre- and post-therapeutic blood analysis in order to see, whether ERCC1+CTCs dynamics mirror response to chemotherapy. METHODS: 65 Paired blood samples (10ml) of primary ovarian cancer patients at primary diagnosis and after chemotherapy were studied for CTCs with the AdnaTest Ovarian Cancer (QIAGEN Hannover GmbH). We analyzed the tumor-associated transcripts EpCAM, MUC-1 and CA-125. ERCC1-transcripts were investigated in a separate approach by singleplex RT-PCR. RESULTS: Auxiliary assessment of ERCC1-transcripts enhanced the overall CTC-detection rate up to 17%. ERCC1+CTCs (defined as positive for one of the AdnaTest markers plus ERCC1-positivity) were detected in 15% of patients at primary diagnosis and in 12% after chemotherapy. The presence of ERCC1+CTCs after chemotherapy correlated with platinum-resistance (P=0.01), reduced PFS (P=0.0293) and OS (P=0.0008) and their persistence indicated poor post-therapeutic outcome (PFS: P=0.005; OS: P=0.0058). Interestingly, the assessment of ERCC1-transcripts alone was sufficient for the detection of prognostic relevant ERCC1-expressing CTCs. CONCLUSION: Auxiliary assessment of ERCC1-transcripts expands the phenotypic spectrum of CTC detection and defines an additional overlapping fraction of ERCC1-expressing CTCs, which are potentially selected by platinum-based chemotherapy. Specifically, we suggest that ERCC1+CTCs could additionally be useful as a surrogate for monitoring platinum-based chemotherapy and to assess the post-therapeutic outcome of ovarian cancer.

Analysis of disseminated tumor cells before and after platinum based chemotherapy in primary ovarian cancer. Do stem cell like cells predict prognosis?

BACKGROUND: We recently reported that the presence of disseminated tumor cells (DTCs) in the bone marrow (BM) of primary ovarian cancer patients (POC pts) correlated with reduced progression free survival (PFS) and overall survival (OS). Here we analyzed whether the negative prognostic influence was related to DTC persistence after platinum based chemotherapy and/or due to DTCs associated with stem cell character. RESULTS: DTCs were detected in 33/79 pts (42%) before and in 32/79 pts (41%) AT. Persistent DTCs were found in 13 pts, 20 pts were only positive BT, 19 pts AT and 27 pts had no DTCs. Whereas the presence of DTCs BT significantly correlated with reduced OS (p = 0.02), pts initially DTCneg BT but DTCpos AT had a significantly shorter PFS (p = 0.03). DTC persistence resulted in a shorter PFS and OS reaching borderline significance (p = 0.06; p = 0.07). LIN-28-and SOX-2 positive cells were detected in all eight pts AT. PATIENTS AND METHODS: 79 POC pts were studied for DTCs before therapy (BT) and after therapy (AT) using immunocytochemistry. Eight pts harboring at least five DTCs AT were further analyzed on two additional slides by four-fold immunofluorescence staining for DAPI, Cytokeratin (CK), SOX-2 or LIN-28, CD45 and CD34 (Cy5). A stem-like tumor cell was classified as Dapipos, CD45neg, CD34neg, SOX-2pos/LIN-28pos and CKpos or CKneg. CONCLUSIONS: Stem cell associated proteins are expressed in DTCs that are present AT and their presence seem to be correlated with a worse outcome. Additional therapeutic regimens may be necessary to eliminate these cells.