Samuel Mok

Dicer, Drosha, and outcomes in patients with ovarian cancer.

BACKGROUND We studied Dicer and Drosha, components of the RNA-interference machinery, in ovarian cancer. METHODS We measured messenger RNA (mRNA) levels of Dicer and Drosha in specimens of invasive epithelial ovarian cancer from 111 patients, using a quantitative reverse-transcriptase-polymerase-chain-reaction assay, and compared the results with clinical outcomes. Validation was performed with the use of published microarray data from cohorts of patients with ovarian, breast, and lung cancer. Mutational analyses of genomic DNA from the Dicer and Drosha genes were performed in a subgroup of ovarian-cancer specimens. Dicer-dependent functional assays were performed by means of in vitro transfection with small interfering RNA (siRNA) and short hairpin RNA (shRNA). RESULTS Levels of Dicer and Drosha mRNA correlated with the levels of expression of the corresponding protein and were decreased in 60% and 51% of ovarian-cancer specimens, respectively. Low Dicer expression was significantly associated with advanced tumor stage (P=0.007), and low Drosha expression with suboptimal surgical cytoreduction (P=0.02). Cancer specimens with both high Dicer expression and high Drosha expression were associated with increased median survival (>11 years, vs. 2.66 years for other subgroups; P<0.001). We found three independent predictors of reduced disease-specific survival in multivariate analyses: low Dicer expression (hazard ratio, 2.10; P=0.02), high-grade histologic features (hazard ratio, 2.46; P=0.03), and poor response to chemotherapy (hazard ratio, 3.95; P<0.001). Poor clinical outcomes among patients with low Dicer expression were validated in additional cohorts of patients. Rare missense mutations were found in the Dicer and Drosha genes, but their presence or absence did not correlate with the level of expression. Functional assays indicated that gene silencing with shRNA, but not siRNA, may be impaired in cells with low Dicer expression. CONCLUSIONS Our findings indicate that levels of Dicer and Drosha mRNA in ovarian-cancer cells have associations with outcomes in patients with ovarian cancer.

DOC-2, a candidate tumor suppressor gene in human epithelial ovarian cancer

Using RNA fingerprinting (RAP) strategy and Northern blot analysis, we identified a differentially expressed sequence DOC-2 which is detectable in all normal human ovarian surface epithelial (HOSE) cell cultures but not in ovarian cancer cell lines and tissues. Subsequent cloning of DOC-2 from a cDNA library generated from the HOSE cells was carried out using the 3′ and 5′ RACE approach. A 3268 base pair full length cDNA of DOC-2 was isolated and sequenced. The predicted protein has a length of 770 amino acids. Homology search of all NCBI sequences indicated that the amino acid sequence of DOC-2 shares 93% homology with the mouse p96/mDab2 phosphoprotein and has a phosphotyrosine interacting domain (PID) and multiple SH3 binding motifs. Chromosomal localization by FISH showed that the DOC-2 gene is located on 5p13. Western blot analysis showed that the 105 kDa DOC-2 protein was down-regulated in all the carcinoma cell lines. In-situ immunohistochemistry performed on normal ovaries, and benign, borderline and invasive ovarian tumor tissues showed down regulation of DOC-2 protein particularly in serous ovarian tumor tissues. When DOC-2 was transfected into the ovarian carcinoma cell line SKOV3, the stable transfectants showed significantly reduced growth rate and ability to form tumors in nude mice. These data suggest that down-regulation of DOC-2 may play an important role in ovarian carcinogenesis.

Dysregulation of microRNA-204 mediates migration and invasion of endometrial cancer by regulating FOXC1

MicroRNAs (miRNAs) regulate mRNA stability and protein expression, and certain miRNAs have been demonstrated to act either as oncogenes or tumor suppressors. Differential miRNA expression signatures have been documented in many human cancers but the role of miRNAs in endometrioid endometrial cancer (EEC) remains poorly understood. This study identifies significantly dysregulated miRNAs of EEC cells, and characterizes their impact on the malignant phenotype. We studied the expression of 365 human miRNAs using Taqman low density arrays in EECs and normal endometriums. Candidate differentially expressed miRNAs were validated by quantitative real‐time PCR. Expression of highly dysregulated miRNAs was examined in vitro through the effect of anti‐/pre‐miRNA transfection on the malignant phenotype. We identified 16 significantly dysregulated miRNAs in EEC and 7 of these are novel findings with respect to EEC. Antagonizing the function of miR‐7, miR‐194 and miR‐449b, or overexpressing miR‐204, repressed migration, invasion and extracellular matrix‐adhesion in HEC1A endometrial cancer cells. FOXC1 was determined as a target gene of miR‐204, and two binding sites in the 3′‐untranslated region were validated by dual luciferase reporter assay. FOXC1 expression was inversely related to miR‐204 expression in EEC. Functional analysis revealed the involvement of FOXC1 in migration and invasion of HEC1A cells. Our results present dysfunctional miRNAs in endometrial cancer and identify a crucial role for miR‐204‐FOXC1 interaction in endometrial cancer progression. This miRNA signature offers a potential biomarker for predicting EEC outcomes, and targeting of these cancer progression‐ and metastasis‐related miRNAs offers a novel potential therapeutic strategy for the disease.

Expression Profiling of Mucinous Tumors of the Ovary Identifies Genes of Clinicopathologic Importance

Purpose: To elucidate the molecular mechanisms contributing to the unique clinicopathologic characteristics of mucinous ovarian carcinoma, global gene expression profiling of mucinous ovarian tumors was carried out. Experimental Design: Gene expression profiling was completed for 25 microdissected mucinous tumors [6 cystadenomas, 10 low malignant potential (LMP) tumors, and 9 adenocarcinomas] using Affymetrix U133 Plus 2.0 oligonucleotide microarrays. Hierarchical clustering and binary tree prediction analysis were used to determine the relationships among mucinous specimens and a series of previously profiled microdissected serous tumors and normal ovarian surface epithelium. PathwayAssist software was used to identify putative signaling pathways involved in the development of mucinous LMP tumors and adenocarcinomas. Results: Comparison of the gene profiles between mucinous tumors and normal ovarian epithelial cells identified 1,599, 2,916, and 1,765 differentially expressed in genes in the cystadenomas, LMP tumors, and adenocarcinomas, respectively. Hierarchical clustering showed that mucinous and serous LMP tumors are distinct. In addition, there was a close association of mucinous LMP tumors and adenocarcinomas with serous adenocarcinomas. Binary tree prediction revealed increased heterogeneity among mucinous tumors compared with their serous counterparts. Furthermore, the cystadenomas coexpressed a subset of genes that were differentially regulated in LMP and adenocarcinoma specimens compared with normal ovarian surface epithelium. PathwayAssist highlighted pathways with expression of genes involved in drug resistance in both LMP and adenocarcinoma samples. In addition, genes involved in cytoskeletal regulation were specifically up-regulated in the mucinous adenocarcinomas. Conclusions: These data provide a useful basis for understanding the molecular events leading to the development and progression of mucinous ovarian cancer.

Effect of a c-Met-specific, ATP-competitive small-molecule inhibitor SU11274 on human ovarian carcinoma cell growth, motility, and invasion

Increased expression of the receptor tyrosine kinase c-Met has been shown to correlate with enhanced cell proliferation, motility, and invasion. The objectives of this study were to characterize total and activated c-Met expression in both normal and malignant human ovarian epithelial cells and to determine the effects of inhibiting the activation of c-Met on ovarian epithelial cell growth, motility, and invasion. Total c-Met was overexpressed in 82 (68%) of 119 ovarian carcinomas, as shown by immunohistochemistry. Quantitative reverse transcription–polymerase chain reaction and Western blot analyses revealed that ovarian carcinoma cell lines had higher levels of c-Met messenger RNA, total protein, and activated protein expression compared to normal ovarian epithelial cell cultures. Using a specific adenosine triphosphate-competitive small-molecule inhibitor, SU11274, activated c-Met was decreased in normal and ovarian carcinoma cell lines. 3-(4,5-Dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) assays showed that cell growth inhibition directly correlated to the level of activated c-Met detected in each cell line (r=−0.87, P= 0.012). Using modified Boyden chamber assays, ovarian carcinoma cells treated with SU11274 demonstrated significantly decreased cell motility and invasion compared to untreated cells (P= 0.003 and P< 0.001, respectively). These data indicate that c-Met is overexpressed in the majority of malignant ovarian epithelial cells both In vivo and in vitro and that decreasing activated c-Met in vitro can significantly decrease ovarian carcinoma cell growth, motility, and invasion. Developing therapies that specifically inhibit the activation of c-Met may represent a novel therapeutic modality for patients with ovarian carcinomas expressing high levels of c-Met.

A one centimorgan deletion unit on chromosome Xq12 is commonly lost in borderline and invasive epithelial ovarian tumors

We have used polymerase chain reaction (PCR) amplification of tandem repeats to study the pattern of allelic loss on chromosome X11.2-q12 in borderline and invasive epithelial ovarian tumors. Using eight microsatellite markers spanning Xq11.2-q12, 41 borderline and 65 invasive ovarian tumors, together with their corresponding normal tissues, were analysed. The highest percentage of loss of heterozygosity (LOH) was observed at the DXS1194 locus in borderline tumors (four of 16 informative cases, 25%) and at the androgen receptor (AR) locus in invasive epithelial ovarian tumors (18 of 47 informative cases, 38%). X chromosome activation studies performed in cases with LOH at the AR locus showed that the allelic loss at the AR locus is not confined to the inactive allele. A one centimorgan region including the AR locus and flanked by the primers DXS1161 and PGK1P1 was identified as the smallest common loss region in both borderline and invasive epithelial ovarian tumors.

Comparison of cyclin expression between endometrioid-type endometrial cancer (EEC) and endometrioid ovarian cancer (EOC)

5013 Background: The biological behavior of EEC and EOC differs despite their histologic similarity. While EEC is usually diagnosed at an early stage and has a good prognosis, EOC is generally advanced at diagnosis and shows a poor prognosis. Cyclins are reported to regulate the cell cycle and influence the prognosis of several cancers. We compared the pattern of cyclin expression between EEC and EOC, and evaluated the relationship among cyclin expression, estrogen receptor (ER) status, and the Ki67 labeling index (Ki67-LI). METHODS There were 36 EECs and 37 EOCs. Immunohistochemistry (IHC) was used to detect cyclin A, B1, D1, E and F (CNA, CNB1, CND1, CNE, CNF), Ki67, and ER. The percentage of positive nuclear staining for cyclins was calculated. Quantitative PCR was performed for CNA, CND1 and CNE. RESULTS (1) When EEC and EOC were compared, CNA expression was higher in EEC. In contrast, CND1 and CNE expression were higher in EOC. The Ki67-LI was also higher in EOC than EEC. Conversely, ER positivity was higher in EEC than EOC (80.6% vs 35.1%, p=0.0007). (2) EEC: Only CNA expression was significantly correlated with the Ki67-LI (r=0.55, p=0.0004). ER status was not related to cyclin expression. Protein and mRNA levels were significantly correlated for CNA and CNE. (3) EOC: Expression of CNA and CND1 was weakly correlated with the Ki67-LI (r=0.54, p=0.001; r=0.38, p=0.04). CNA expression was higher in ER-negative tumors than ER-positive tumors (23.7 vs 9.6, p=0.027). CND1 expression was also higher in ER-negative than ER-positive tumors (12.1 vs 0, p=0.065). CND1, and CNE protein and mRNA levels were significantly correlated. CONCLUSIONS The pattern of cyclin expression differed between EEC and EOC. [Figure: see text] No significant financial relationships to disclose.

Fibroblast growth factor receptor 4 (FGFR4) expression as a prognostic indicator in high-grade serous ovarian carcinoma (HGSC): Signaling pathway, mechanism, and implication for therapeutic targeting.

5049 Background: Poor survival associated with the amplicon 5q31-35(previously reported) in HGSC may be due to subsequent overexpression of FGFR4. This work aims to correlate FGFR4 with survival in HGSC, identify its role, molecular mechanism & test feasibility of therapeutic targeting. METHODS QT-PCR validated amplification of FGFR4 on 5q31-35 in 52 cases. IHC quantified FGFR4 expression in 181 patients. Western blots(WB) quantified FGFR4 expression in HGSC cell lines. Role of FGFR4 in proliferation, survival, migration and invasion potential of HGSC in vitro was quantified in HGSC cells transfected with FGFR4 siRNA and control scramble sequence siRNA. HGSC cell lines expressing response elements linked to luciferase were treated with Fibroblast growth factor 1(FGF1) in the presence and absence of FGFR4 siRNA or FGFR4 trap protein to identify relevant pathways. Reverse phase protein array analysis and WB were also performed on HGSC cell lines subjected to similar treatments. Mutational analysis on FGFR4was performed in 43 HGSC samples. RESULTS FGFR4 staining intensity correlated with poor OS in HGSC patient (median survival 28 vs 55 months p<0.001) and increase risk of death (HR 1.22 p<0.001). FGFR4 was over-expressed in HGSC cell lines compared to ovarian surface epithelium. FGFR4 and FGF1 DNA copy number were significantly correlated (r=0.4, p=0.04). Significant decreases in cell proliferation (p<0.001), survival (p<0.01) and migration (p<0.05) were noted in cells treated with FGFR4 siRNA compared to controls. RPPA, the luciferase reporter assay and western blot analysis demonstrated activation of multiple pathways (MAPK,WNT and NFkB) in response to FGF1 but the effect was inhibited by silencing FGFR4 or by FGFR4 trap treatment. Sequencing of FGFR4 did not reveal mutations. CONCLUSIONS Over-expression of FGFR4 is associated with poor survival in late stage HGSC. In vitro data provides evidence that binding of ligand to FGFR4 activates multiple signaling pathways leading to a more aggressive phenotype. Targeting FGFR4 over-expression in HGSC may be a therapeutic strategy with potential for improving survival in HGSC.

ABCF2 is a new biomarker of ovarian clear cell adenocarcinoma

5050 Background: Epithelial ovarian cancer (EOC) can be subdivided into five major histologic types (Serous(SC), Mucinous(MC), Endometrioid(EC), Clear cell(CC), Undifferentiated (UC)). Among them, ...

Global RNA expression analysis of primary and recurrent ovarian tumors

Many antineoplastic agents fail because of intrinsic or acquired resistance developed by the cancer cells; drug resistance is therefore a major obstacle to successful chemotherapy of human cancers, including ovarian cancer. Understanding the mechanisms by which drug resistance arises and identification of the molecular factors affecting new drug target development is of paramount importance. In a study to compare the global RNA expression profiles of primary and recurrent ovarian tumors from the same patient by means of oligonucleotide microarray analysis, we have identified XIST (inactive X [Xi] chromosome-specific transcripts) as the most differentially expressed gene that was downregulated in the recurrent tumor. The XIST gene resides within the X inactivation center and is unique in being expressed exclusively from the inactive X chromosome. XIST is both necessary and sufficient for X inactivation. Preliminary studies of the cell line derived from the recurrent tumor showed that the line is resistant to the antineoplastic agent taxol, one of the anticancer drugs included in the treatment of the patients under study. The recurrent cell line has at least one copy of XIST in the genome, although the gene is not expressed. Further characterization of XIST expression in a panel of ten ovarian cell lines and six breast cancer cell lines showed that the expression levels of XIST correlate significantly with taxol sensitivity. These data indicate a possible correlation between X inactivation and taxol resistance in ovarian cancer. We have launched a detailed mechanistic study to characterize the underlying mechanism.