Genome DASL Assay Research Articles

Potential molecular signatures in epithelial ovarian cancer by genome wide expression profiling

Ovarian cancer is the deadliest of all gynecologic cancers because of its late diagnosis and poor treatment outcomes. This study aimed to identify potential molecular signatures associated with biological processes that are implicated in epithelial ovarian cancer (EOC). Expression profiling was carried out on 16 fresh frozen EOC and normal ovarian tissue samples using the Illumina Whole Genome DASL assay (cDNA-mediated annealing, selection, extension and ligation). The differentially expressed genes were analyzed using the GeneSpring GX11.5 and Pathway Studio 8.0 software. The microarray results were validated using the immunohistochemistry analyses. Unpaired t-test identified 652 (270 up- and 382 downregulated) significant differentially expressed genes (P < 0.001 and fold change ≥2.0). Hierarchical clustering analysis displayed a distinct separation of cancer and normal samples. Gene set enrichment analysis identified alterations in the expression of genes associated with cancer development and progression. Positive immunostaining of claudin-7, ephrin receptor A1 and Forkhead Box M1 in EOC was consistent with the upregulation of these genes in the microarray result. However, the positive immunostaining of fibroblast growth factor-7 in cancer tissues was not in accordance with the downregulation of this gene in the microarray result. These results identify significant genes and their related biological processes which may contribute to the better understanding of development and progression of epithelial ovarian cancer.

Abstract 1951: miRNA expression profile of Blastic plasmacytoid dendritic cell neoplasm.

Abstract Background: Blastic plasmacytoid dendritic cell neoplasm (BPDCN) is a rare neoplasm currently included among acute myeloid leukemias in the WHO classification. BPDCN is an orphan tumor characterized by an unknown biology with clinical aggressive course and dismal prognosis. Only few studies explored its genetics, documenting a complex karyotype and sporadic genetic alterations. Unfortunately, data were based on a limited number of patients and lacked functional elucidation. Conversely, studies on the transcriptome only compared BPDCN with other leukemias, failing to identify pathogenetic events. In this study we aimed to assess whether miRNA deregulation contribute to BPDCN pathogenesis and clinical behaviour. Methods: We studied by miRNA profiling (NanoString Technologies) 30 BPDCN samples, 9 normal plasmacytoid dendritic cell (pDCs) samples and the CAL-1 tumor cell line. In addition, 25 independent BPDCN cases were studied by qRT-PCR and IHC as validation. To investigate the functional consequences of miRNA deregulation, we also performed a global gene expression profile (GEP) analysis on the same cases by using the Illumina Whole Genome DASL Assay. We applied ANOVA and Student t-test to find genes and miRNAs differentially expressed between BPDCNs and normal pDCs. The selected miRNAs were further filtered based on their predicted impact on the GEP using Linear Discriminant Analysis, Spearman correlation and Gene Set Enrichment Analysis. Finally, the functional effects of selected miRNAs were assessed ex vivo on CAL-1 cell line. Results: We found that, according to the unsupervised analysis of miRNAs, BPDCN patients have a molecular signature well distinct from their normal counterpart. The following supervised analysis identified a set of 114 miRNAs significantly deregulated whose aberrant expression might have a relevant impact on cancer-related cellular pathways. Therefore, after validating their expression in an independent set of cases, we integrated the miRNA and mRNA expression profiles on the same patients, in order to assess the functional impact of deregulated miRNAs on gene expression profile. Pathway analysis of the predicted miRNA target genes, revealed that most of them were involved in the same gene ontology category: miRNAs synergic action appeared to strongly affect the biological process of programmed cell death. We thus focused on miRNA-mRNA anti-correlations affecting the apoptotic pathway: we transfected specific mimics and inhibitors into CAL-1 cell line; we evaluated their functional impact by gene expression profile and, finally, we confirmed their relation with specific targets by Luciferase assay. In conclusion, for the first time we investigated the miRNA profile of BPDCN, defining previously uncovered molecular mechanisms underlying the pathogenesis of the disease. Specifically, we found a subset of miRNAs significantly deregulated and highly affecting the apoptotic pathway. Citation Format: Maria Rosaria Sapienza, Fabio Fuligni, Maria Antonella Laginestra, Maura Rossi, Luciano Cascione, Alessandro Laganà, Claudio Agostinelli, Maryam Etebari, Anna Gazzola, Claudia Mannu, Elena Sabattini, Francesco Bacci, Carlo Maria Croce, Stefano Aldo Pileri, Pier Paolo Piccaluga. miRNA expression profile of Blastic plasmacytoid dendritic cell neoplasm. [abstract]. In: Proceedings of the 104th Annual Meeting of the American Association for Cancer Research; 2013 Apr 6-10; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2013;73(8 Suppl):Abstract nr 1951. doi:10.1158/1538-7445.AM2013-1951

Abstract 5275: miRNA profiling of peripheral T-cell lymphomas.

Abstract Background: Peripheral T cell lymphomas not otherwise specified (PTCLs/NOS) are rare and aggressive tumors, whose molecular pathogenesis is poorly known and whose diagnosis is quite difficult. In fact, both morphology and phenotype largely overlap with those of other nodal PTCLs, including angioimmunoblastic T cell lymphoma (AITL) and anaplastic large cell lymphomas (ALCLs). Here, we performed an extensive miRNA profiling of PTCLs in order to identify differentially expressed miRNA, either involved in their pathogenesis or potentially useful for their differential diagnosis. Methods: We studied by miRNA profiling (TaqMan Array MicroRNA Cards A) 60 samples including PTCLs/NOS (N=25), AITLs (N=10), ALCLs (N=12) and normal T cells (N=13); in addition, 40 independent PTCL cases were studied by qRT-PCR as validation. Further, to assess the functional impact of deregulated miRNAs, we performed a global gene expression profile (GEP) analysis in the same cases by using the Illumina Whole Genome DASL Assay. ANOVA and Student t-test were performed to find genes and miRNAs differentially expressed between PTCLs and normal T cells or among PTCL subtypes, respectively. Differentially expressed miRNAs were further filtered based on their predicted impact on the GEP using Linear Discriminant Analysis, Spearman correlation and Gene Set Enrichment Analysis. A discriminant function based on linear combination of GEP signatures was then used to better distinguish different PTCLs subtypes. Finally, the functional effects of selected miRNA were assessed by transfection of specific mimic and inhibitors into PTCL cells and then evaluating the GEPs (functional GEP). Findings: First, unsupervised analyses clearly discriminated PTCLs and normal T cells based on the miRNA expression patterns. Second, supervised analysis identified 256 miRNA differentiating the two groups (p&amp;lt;0.05, FC ≥2, Bonferroni correction). To identify those miRNA more likely to impact the GEP of the tumors, we then performed a multi-step analysis integrating miRNA and GEP data. We identified 4 highly discriminant miRNA inversely correlated with different potential target genes. All 4 miRNA were confirmed by qPCR in a validation set, while their target genes were assessed by functional GEP confirming their impact. Thereafter, we sought miRNAs differentially expressed in the PTCL subtypes, to be used as diagnostic markers. Based on ANOVA and subsequent linear discriminant analysis we identified 3 sets of miRNA able to efficiently distinguish PTCL/NOS vs. ALK-/ALCL (N=5 miRNA), ALK+/ALCL vs. ALK-/ALCL (N=4), and PTCL/NOS vs. AITL (N=8). Noteworthy, this novel molecular diagnostic tool was validated by qPCR in independent cases, demonstrating a remarkable diagnostic accuracy ranging between 85% and 95%. In conclusion, miRNA profiling allowed to identify miRNA possibly involved in PTCL pathogenesis and to develop a novel diagnostic tool for the differential diagnosis of the commonest subtypes. Citation Format: Maria Antonella Laginestra, Fabio Fuligni, Maura Rossi, Davide Gibellini, Maria Rosaria Sapienza, Anna Gazzola, Claudia Mannu, Elena Sabattini, Francesco Bacci, Claudio Agostinelli, Stefano Aldo Pileri, Pier Paolo Piccaluga. miRNA profiling of peripheral T-cell lymphomas. [abstract]. In: Proceedings of the 104th Annual Meeting of the American Association for Cancer Research; 2013 Apr 6-10; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2013;73(8 Suppl):Abstract nr 5275. doi:10.1158/1538-7445.AM2013-5275

Abstract 2997: Peripheral T-cell lymphomas not otherwise specified correspond to distinct mature T-cell populations basing on the global gene expression profile

Abstract Peripheral T-cell lymphomas not otherwise specified (PTCLs/NOS) are the commonest T-cell neoplasms. Recent studies suggested that different PTCL/NOS subgroups might be identified according to a diverse cellular derivation or correspondence to various cellular counterparts. However, so far, it was possible only to discriminate T-helper (TH) vs. T-cytotoxic (TC) tumors. Noteworthy, due to their typical aberrant phenotype, such distinction could be effectively performed only by gene expression profiling (GEP). In addition, further distinctions (i.e. TH subtypes) were only occasionally described in scattered cases. In this study, we aimed to define the counterparts of PTCLs/NOS based on their global GEP. We generated and analyzed GEP from 108 PTCLs/NOS by using either Affymetrix HG-U133 2.0 plus microarray (N=28 cases) or Illumina whole genome DASL assay (N=80). The two groups corresponded to fresh/frozen samples and formalin fixed paraffin embedded (FFPE) ones, respectively. Tissue microarrays (TMAs) were built including the same cases, and immunohistochemistry (IHC) was used for candidate markers validation. The molecular signatures of T-cell subsets (Th1, Th2, Th17, T-follicular helper/Tfh, Treg, T-central memory/Tcm, T-effector memory/Tem, and T-cytotoxic/TC) were studied in the tumor samples using previously validated cell classifiers as well as gene set enrichment analysis (GSEA). First, we studied the 28 samples for which fresh/frozen biopsies were available. We found that 23/28 corresponded to TH and 5/28 to TC cells. Moreover, the TH were more related to Th1 (18/23) rather then to Th2 (1/23) cells, in 4 instances not being a significant enrichment in neither one signature. When further signatures were tested, we found that PTCLs/NOS were more frequently related to Tcm cells (12/23), than to Tem (5/23), Tfh (2/23), Treg (0/23), or Th17 (0/23). Four again presented no significant similarity with the studied populations. We then tested whether reliable profiles could be obtained from FFPE samples. Notably, we could assign 66/80 (83%) cases to a specific counterpart. In particular, 42/80 were TH, while 24/80 were TC. At further analysis, the TH cases were first classified as Th1 (N=21) or Th2 (N=13) and then as Tcm (N=20), Tem (N=7), Tfh (N=4) or Treg (N=2), 1 not being further classified. IHC confirmed the genomic data as far as Tfh and Treg markers were concerned, not being available, at present, consistent IHC markers for Th1, Th2, Tcm and Tem. Conversely, it was inadequate to distinguish TH and TC, due to the frequent lack of CD4, CD8 and other cytotoxic molecules. In conclusion, we showed that PTCL/NOS heterogeneity is largely due to the correspondence to different counterparts and confirmed our previous observation that GEP is probably more effective than IHC for the sub classification of this tumor type. Citation Format: {Authors}. {Abstract title} [abstract]. In: Proceedings of the 103rd Annual Meeting of the American Association for Cancer Research; 2012 Mar 31-Apr 4; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2012;72(8 Suppl):Abstract nr 2997. doi:1538-7445.AM2012-2997

Identification of Differentially Expressed miRNAs in Peripheral t-Cell Lymphomas

Abstract 773Peripheral T-cell lymphomas (PTCLs) are rare and heterogeneous tumors whose biology is largely unknown. Interestingly, the commonest subtypes (i.e. PTCL not otherwise specified, NOS; angioimmunoblastic T-cell lymphoma, AITL; and anaplastic large cell lymphoma, ALCL) present on one hand few disease-specific molecular features and, on the other hand, several apparently common abnormalities. So far, no data are available regarding miRNA expression in these tumors.In order to identify miRNA deregulated in PTCLs, we performed an extensive miRNA profiling (by studying 379 targets on the TaqMan Array MicroRNA Cards) of 44 PTCLs (including 23 PTCLs/NOS, 12 ALCLs, and 9 AITLs) and 13 sample representative of normal T-cell sub-populations (CD4+ and CD8+, both resting and activated). In addition, for all these cases, gene expression profiles (GEPs) were generated by the Ilumina whole genome DASL-assay. TaqMan Quantitative-PCR (qPCR) was then used for validation.First, we found that PTCLs and normal T-cells could be easily distinguished based on their miRNA profile, by both unsupervised and supervised analysis. Specifically, the latter identified 91 miRNA differentially expressed in PTCLs vs. T-cells with a fold change ≥2 and a pvalue<0.01. Interestingly, the predicted target genes of such miRNA were involved in relevant cellular functions (such as cell proliferation, apoptosis, signal transduction etc.) and were found to be differentially expressed in tumor vs. controls, indicating a functional impact of miRNA deregulation in the molecular phenotype of neoplastic T-cells. Intriguingly, several drugs were estimated in silico to be able to interfere with pathways controlled by the de-regulated miRNA, including flavopiridol, everolimus/ temsirolimus, and forodesine.Secondly, by supervised analysis (ANOVA), we identified miRNA differentially expressed among PTCL/NOS, AITL and ALCL. Indeed, basing on such miRNA it was possible to roughly separate the different entities by hierarchical clustering (HC) (chi-square, p=0.0005). Specifically, we found 7 miRNA differentially expressed in PTCL/NOS vs. AITL, 26 in PTCL/NOS vs. ALK- ALCL, and 16 in ALK- vs. ALK+ ALCL. Selected miRNA were then studied by qPCR in an independent set of PTCLs (including 10 PTCL/NOS, 5 AITL, 5 ALK- ALCL, and 5 ALK+ ALCL) as for validation.Subsequently, in order to assess the impact of miRNA on the GEP, we picked up the miRNA most efficient in differentiating the diverse tumors (miR-431, miR-887, miR-889, and miR-155), and we studied the expression of their predicted target genes. Indeed, we found that miR-431 target genes were differentially expressed in PTCL/NOS vs. AITL (GSEA, q-value=0.45), and basing on their expression, HC could efficiently divide the two diseases (chi-square, p=0.004). Similarly, miR-889 targets were differentially expressed in PTCL/NOS vs. ALK- ALCL (GSEA, q-value=0; chi-square for HC, p=0.01), while miR-155 targets were differentially expressed in ALK+ vs. ALK- ALCL (GSEA, q-value=0; chi-square for HC, p<0.001).In conclusion, we performed for the first time an extensive miRNA profiling of PTCLs and combined it with global GEP. Importantly we identified miRNAs deregulated in PTCLs with possible pathogenetic implications. In addition, we found miRNAs, whose detection might be relevant for the differential diagnosis of PTCL subtypes. Finally, we provided in silico evidences that the cellular pathway controlled by de-regulated miRNA in PTCLs might serve as potential therapeutic targets, thus warranting further functional investigation. Disclosures:No relevant conflicts of interest to declare.

Gene expression profiling of human whole blood samples with the Illumina WG-DASL assay

BackgroundMicroarray-based gene expression analysis of peripheral whole blood is a common strategy in the development of clinically relevant biomarker panels for a variety of human diseases. However, the results of such an analysis are often plagued by decreased sensitivity and reliability due to the effects of relatively high levels of globin mRNA in whole blood. Globin reduction assays have been shown to overcome such effects, but they require large amounts of total RNA and may induce distinct gene expression profiles. The Illumina whole genome DASL assay can detect gene expression levels using partially degraded RNA samples and has the potential to detect rare transcripts present in highly heterogeneous whole blood samples without the need for globin reduction. We assessed the utility of the whole genome DASL assay in an analysis of peripheral whole blood gene expression profiles.ResultsWe find that gene expression detection is significantly increased with the use of whole genome DASL compared to the standard IVT-based direct hybridization. Additionally, globin-probe negative whole genome DASL did not exhibit significant improvements over globin-probe positive whole genome DASL. Globin reduction further increases the detection sensitivity and reliability of both whole genome DASL and IVT-based direct hybridization with little effect on raw intensity correlations. Raw intensity correlations between total RNA and globin reduced RNA were 0.955 for IVT-based direct hybridization and 0.979 for whole genome DASL.ConclusionsOverall, the detection sensitivity of the whole genome DASL assay is higher than the IVT-based direct hybridization assay, with or without globin reduction, and should be considered in conjunction with globin reduction methods for future blood-based gene expression studies.

Abstract 4830: Identification of differentially expressed genes and protein in bone invasive non-invasive meningiomas

Abstract Introduction: Meningiomas are cranial tumors arising from the meningeal coverings of the brain and while the majority grow intradurally, some can invade the underlying bone causing hyperostosis and invasion into neuronal structures. The surgical resection of such bone-invading tumors is challenging and repeat surgery is often required, resulting in significant patient morbidity. To date there has been very limited studies focused on the molecular pathophysiology of bone invading meningiomas. Our aim was to use an integrative analysis approach, performing RNA microarray and tissue microarray analysis to identify differentially expressed protein and genes involved in bone tropism of meningioma cells. Methods: We chose radiological features to define two distinct bone invading meningioma population and their control counterpart. 1) spheno-orbital meningioma and control counterpart non-bone invading sphenoid wing meningioma and 2) transbasal meningioma and control counterpart anterior skull base meningioma with no bone invasion. We identified 57 patients satisfying these bone invading classifications operated on in the past ten years at our institution. RNA was extracted for microarray analysis from paraffin-embedded tissue sections of invasive and non-invasive meningiomas and processed on Illumina Whole Genome DASL assay. Data were analyzed using Multi Expression Viewer Software (MEV). Quantitative real-time PCR (RT-qPCR) was used to verify micro-array data. We further verified our data using Tissue microarray (TMA) and examined commercially available antibodies involved in bone invasion (osteopontin, MMP2 and integrin-β1). TMA scoring was carried out with two independent observers using percentage and intensity staining.Three different meningioma cell lines (IOMMA-Lee, CH157-MN and F5 cells) were used to verify microarray results and in vitro and in vivo functional studies. Results: RNA microarray data analysis identified 222 differentially expressed genes (92 genes over-expressed and 130 genes under expressed), amongst which we can refer to over expression of PDGFRα, MMP16, MMP19, Matrilin4 and ADAMTS4 in bone-invasive relative to non-invasive meningiomas. Upregulation of these genes were verified using quantitative real-time PCR (RT-qPCR) analysis in both tumor specimens and meningioma cell lines.TMA analysis identified increased expression of both MMP2 and integrin-β1 in tumor cells of non-invasive meningiomas, and increase of vascular MMP2 expression in non-invasive compared to invasive meningiomas. Conclusions: Our results identify novel differentially expressed proteins and genes in bone-invading meningiomas compared to non-invasive meningiomas. Our baseline in-vivo mechanistic results will help design new therapeutic strategies that can control bone-invading meningioma progression and provide the basis for translation to clinical studies. Citation Format: {Authors}. {Abstract title} [abstract]. In: Proceedings of the 102nd Annual Meeting of the American Association for Cancer Research; 2011 Apr 2-6; Orlando, FL. Philadelphia (PA): AACR; Cancer Res 2011;71(8 Suppl):Abstract nr 4830. doi:10.1158/1538-7445.AM2011-4830

Analyses and interpretation of whole-genome gene expression from formalin-fixed paraffin-embedded tissue: an illustration with breast cancer tissues

BackgroundWe evaluated (a) the feasibility of whole genome cDNA-mediated Annealing, Selection, extension and Ligation (DASL) assay on formalin-fixed paraffin-embedded (FFPE) tissue and (b) whether similar conclusions can be drawn by examining FFPE samples as proxies for fresh frozen (FF) tissues. We used a whole genome DASL assay (addressing 18,391 genes) on a total of 72 samples from paired breast tumor and surrounding healthy tissues from both FF and FFPE samples.ResultsGene detection was very good with comparable success between the FFPE and FF samples. Reproducibility was also high (r2 = 0.98); however, concordance between the two types of samples was low. Only one-third of the differentially expressed genes in tumor tissues (compared to corresponding normal) from FF samples could be detected in FFPE samples and conversely only one-fourth of the differentially expressed genes from FFPE samples could be detected in FF samples. GO-enrichment analysis, gene set enrichment analysis (GSEA) and GO-ANOVA analyses also suggested small overlap between the lead functional groups that were differentially expressed in tumor detectable by examining FFPE and FF samples. In other words, FFPE samples may not be ideal for picking individual target gene(s), but may be used to identify some of the lead functional group(s) of genes that are differentially expressed in tumor. The differentially expressed genes in breast cancer found in our study were biologically meaningful. The "cell cycle" & "cell division" related genes were up-regulated and genes related to "regulation of epithelial cell proliferation" were down-regulated.ConclusionsGene expression experiments using the DASL assay can efficiently handle fragmentation issues in the FFPE tissues. However, formalin fixation seems to change RNA and consequently significantly alters gene expression in a number of genes which may not be uniform between tumor and normal tissues. Therefore, considerable caution needs to be taken when interpreting gene expression data from FFPE tissues, especially in relation to specific genes.

Gene expression profiling of formalin‐fixed, paraffin‐embedded familial breast tumours using the whole genome‐DASL assay

Tissue sample acquisition is a limiting step in many studies. There are many thousands of formalin-fixed, paraffin-embedded archival blocks collected around the world, but in contrast relatively few fresh frozen samples in tumour banks. Once samples are fixed in formalin, the RNA is degraded and traditional methods for gene expression profiling are not suitable. In this study, we have evaluated the ability of the whole genome DASL (cDNA-mediated Annealing, Selection, extension, and Ligation) assay from Illumina to perform transcriptomic analysis of archived breast tumour tissue in formalin-fixed, paraffin-embedded (FFPE) blocks. We profiled 76 familial breast tumours from cases carrying a BRCA1, BRCA2 or ATM mutation, or from non-BRCA1/2 families. We found that replicate samples correlated well with each other (r(2) = 0.9-0.98). In 12/15 cases, the matched formalin-fixed and frozen samples predicted the same tumour molecular subtypes with confidence. These results demonstrate that the whole genome DASL assay is a valuable tool to profile degraded RNA from archival FFPE material. This assay will enable transcriptomic analysis of a large number of archival samples that are stored in pathology archives around the globe and consequently will have the potential to improve our understanding and characterization of many diseases.

Comments on "Transcriptional profiling of oral squamous cell carcinoma using formalin-fixed paraffin-embedded samples" by Saleh et al., Oral Oncol 46 (2010) 379-386 Reply

Dear Editor, We thank Rentoft and colleagues for commenting on our paper ?Transcriptional profiling of oral squamous cell carcinoma using formalin-fixed paraffin-embedded samples?, and agreeing that formalin-fixed paraffin embedded (FFPE) tissue specimens can be used for gene expression studies using microarrays, to identify genes that are significantly involved in oral carcinogenesis. The similarities and high concordance between the study by Rentoft et al. 1 and ours 2 is comforting and clearly supports the use of FFPE tissues in such experiments. More importantly, these studies act as independent validation for one another and strongly suggests that genes that were found to be up- or down-regulated in oral squamous cell carcinoma (OSCC) do indeed play a role in these cancers and therefore warrant further investigation to determine their utility as biomarkers and therapeutic targets for OSCC. We wanted to point out however, that the similarities between these 2 studies are not completely unexpected despite previous reports highlighting that the concordance between microarray studies are hard to achieve 3. The main disparities between datasets from microarray experiments have been attributed to the use of different microarray platforms and the heterogeneity of the tissue specimens that were used 4 and 5. Indeed, many microarray studies reported for the head and neck, used tissues from several distinct areas which have been reported to be genetically heterogeneous, and associated with different aetiologies 6 and 7. Given that both our studies used the DASL assay and tissues from the oral cavity (albeit from different sites-explained further below) the consistency of the genes that were identified should not come as a complete surprise. However, it is still intriguing that the similarities between these two studies were so close despite previous reports indicating that there are distinct differences between oral cancers associated with different aetiology 8 and 9, and reports describing the distinct genetic differences between subsites of the oral cavity 10, 11 and 12. Our previous study describing the differences in gene-expression patterns between oral cancers associated with betel quid chewing and smoking demonstrated that despite the differences seen, genetic changes common to all the cancers were also observed suggesting that there are core events and pathways that are important regardless of the aetiology or site of the cancer 8. Consistently, upon close examination of the genes that are most differentially expressed between our study and that of Rentoft et al. 1 5 of the top 10 over-lapping genes most differentially expressed were matrix metalloproteinases, whilst other genes include IL-8, CXCL-9 and BCL2A1, genes that were consistently up-regulated in many other microarray studies of the head and neck independent of the subsites of cancer 8, 12, 13 and 14. An unprecedented large overlap of genes between our study 2 and that of Rentoft and colleagues 1 may also be due to the smaller number of genes on the arrays that were used in these studies (502 genes), and that these genes were pre-selected based on their involvement in cancer development, whereas the majority of previous studies used platforms consisting of larger arrays and hence could capture much more of the heterogeneity typically observed in cancers. More recently, Illumina launched its whole genome DASL assay and it would be interesting to see if the similarities between buccal and tongue cancers still prevails. In conclusion, increasing number studies including the 2 compared here, strongly supports the use of FFPE tissues for gene expression studies using microarrays. The development of new technologies and statistical methods that addresses the challenges associated with using FFPE tissues, combined with the clinical information available with these specimens, will indeed facilitate the identification and discovery of clinically relevant gene signatures and biomarkers and therapeutic targets to improve the management of cancers in general.

Analysis of gene expression in stage I serous tumors identifies critical pathways altered in ovarian cancer

Objective Despite recent advances in the conceptual understanding of the pathogenesis of ovarian cancer, it remains the foremost cause of death from gynecologic malignancies in developed countries. The main reason for such a high rate of mortality is the lack of sensitive and specific biomarkers and imaging techniques for early detection of ovarian cancer. Additional biological insights into early-stage ovarian carcinogenesis are needed to help speed the development of markers for early detection of ovarian cancer. The objective of this study was to characterize differentially expressed genes in high-grade stage I serous carcinoma of the ovary. Methods We analyzed gene expression in macrodissected formalin-fixed, paraffin-embedded samples from 5 high-grade stage I serous carcinomas and 5 stage I borderline tumors of the ovary using the Illumina Whole Genome DASL assay (c DNA-mediated annealing, selection, extension, and ligation) corresponding to 24,000 genes. Significance Analysis of Microarrays was performed to determine differentially expressed genes in stage I serous carcinoma, and class prediction analysis was performed to determine the predictive value of differentially expressed gene sets to correctly classify serous carcinoma from borderline tumors in 3 independent data sets. Altered transcription factor pathways and biological pathways unique to stage I serous carcinoma were identified through class comparison of differentially expressed genes. Results Unsupervised cluster analysis of gene expression correctly classified stage I serous carcinomas from serous borderline tumors. Supervised analysis identified several known, as well as novel, genes differentially expressed in stage I ovarian cancer. Using a differentially expressed gene set, class comparison prediction analysis correctly identified serous carcinomas from serous borderline tumors in 3 independent data sets at over 80% accuracy, sensitivity, and specificity. Pathway analysis demonstrated the significance of p53 and E2F pathways in serous carcinogenesis and significant involvements of cell cycle and immune response pathways in stage I serous epithelial ovarian cancer. Conclusion We have identified differentially expressed genes associated with the carcinogenesis of high-grade stage I serous EOC. Furthermore, integrative analysis of biological and transcription pathway data contributed to the confirmation of important biological pathways and discovery of additional unique genes and pathways that may have potential importance in ovarian pathogenesis and biomarker development.