Systematic investigation of genetic vulnerabilities across cancer cell lines reveals lineage-specific dependencies in ovarian cancer.
Journal: 2011/October - Proceedings of the National Academy of Sciences of the United States of America
ISSN: 1091-6490
Abstract:
A comprehensive understanding of the molecular vulnerabilities of every type of cancer will provide a powerful roadmap to guide therapeutic approaches. Efforts such as The Cancer Genome Atlas Project will identify genes with aberrant copy number, sequence, or expression in various cancer types, providing a survey of the genes that may have a causal role in cancer. A complementary approach is to perform systematic loss-of-function studies to identify essential genes in particular cancer cell types. We have begun a systematic effort, termed Project Achilles, aimed at identifying genetic vulnerabilities across large numbers of cancer cell lines. Here, we report the assessment of the essentiality of 11,194 genes in 102 human cancer cell lines. We show that the integration of these functional data with information derived from surveying cancer genomes pinpoints known and previously undescribed lineage-specific dependencies across a wide spectrum of cancers. In particular, we found 54 genes that are specifically essential for the proliferation and viability of ovarian cancer cells and also amplified in primary tumors or differentially overexpressed in ovarian cancer cell lines. One such gene, PAX8, is focally amplified in 16% of high-grade serous ovarian cancers and expressed at higher levels in ovarian tumors. Suppression of PAX8 selectively induces apoptotic cell death of ovarian cancer cells. These results identify PAX8 as an ovarian lineage-specific dependency. More generally, these observations demonstrate that the integration of genome-scale functional and structural studies provides an efficient path to identify dependencies of specific cancer types on particular genes and pathways.
Relations:
Content
Citations
(179)
References
(28)
Diseases
(1)
Conditions
(1)
Chemicals
(5)
Organisms
(1)
Processes
(4)
Anatomy
(1)
Affiliates
(1)
Similar articles
Articles by the same authors
Discussion board
Proc Natl Acad Sci U S A 108(30): 12372-12377

Systematic investigation of genetic vulnerabilities across cancer cell lines reveals lineage-specific dependencies in ovarian cancer

+21 authors

Correlation of Genetic Dependency with Properties of Cell Lines.

We then sought to understand how the vulnerabilities of cancer cell lines relate to various properties, such as mutations in a specific gene, disruption of a specific pathway, or inclusion in a specific lineage. Because human cancer cell lines are genetically and epigenetically diverse, the analysis of a large number of cell lines ensures that the relationships between cell properties and the dependence of those cells on specific genes are not particular to one context. We tested whether this large dataset permits reliable inferences about the genetic vulnerability of cancers possessing specific properties. For each cell line classification, we used a class-discrimination feature selection method called the “weight of evidence” (WoE) statistic (4, 5) to rank shRNAs by their ability to distinguish the specified classes.

Dependencies of Cell Lines with Oncogenic Mutations.

We first examined vulnerabilities of cell lines with KRAS or BRAF mutations. We defined “essential” genes by three complementary methods, including (i) WoE rank of the top shRNA targeting each gene, (ii) WoE rank of the top two shRNAs targeting each gene, or (iii) a composite score of WoE ranks for all shRNAs for each gene using the Kolmogorov–Smirnov (KS) statistic (3). The KRAS and BRAF genes themselves were ranked highly by all three methods (Fig. 2 A and D). The top scoring KRAS- and BRAF-specific shRNAs significantly discriminated between the mutant and wild-type classes (P = 1.89 × 10 and 1.89 × 10, respectively, WoE; Fig. 2 B and E).

An external file that holds a picture, illustration, etc.
Object name is pnas.1109363108fig02.jpg

Dependencies of cell lines with mutations in KRAS, BRAF, or PIK3CA. (A, D, and G) Distribution of shRNA ranks (x axis) by the WoE scores (y axis) for the class comparisons of KRAS mutant (mut) vs. KRAS wild-type (wt) cell lines (A), BRAF mutant vs. BRAF wt (D), and PIK3CA mutant vs. PIK3CA wt (G). shRNAs targeting KRAS, BRAF, and PIK3CA are highlighted in red, and their ranks are listed. Insets report the gene ranks of KRAS, BRAF, and PIK3CA for differential essentiality in the subset of cell lines with activating mutations in those respective genes. (B, E, and H) KRAS (B), BRAF (E), or PIK3CA (H) mutation status (mutant lines in green, wt lines in gray) correlates strongly with depletion of shRNAs targeting these genes. (Lower) Heatmaps report KRAS- (B), BRAF- (E), and PIK3CA-shRNA (H) fold depletion in each cell line. (C, F, and I) Subsets of the 102 cell lines were analyzed to assess convergence of the gene dependency results for the KRAS (C), BRAF (F), and PIK3CA (I) mutant vs. wt class comparison analyses as a function of the number of cell lines tested. Distributions of the scores of the top KRAS, BRAF, and PIK3CA hit shRNAs (given as the percentile of their depletion rankings, with a smaller percentage corresponding to more depleted, y axis) in the respective cell line class comparisons (using WoE) are shown for each of 100 trials, subsampling the indicated numbers of cell lines in each class (mutant and wt). The red bar indicates the median value for each group of subsamplings, boxes represent the 25th to 75th percentile of the data, and whiskers extend to the most extreme values of the group that are not considered outliers.

Interestingly, because 7/10 BRAF-mutant cell lines were derived from nonmelanoma lineages (including 5 from colon cancer), these observations suggest that cancer cell lines that harbor mutant BRAF exhibit a similar dependence on BRAF. Although initial reports suggest that clinical responses to BRAF inhibition in BRAF-mutant colon tumors are much less robust than those observed in BRAF-mutant melanomas (6), our observations indicate that BRAF is essential in colon cancer cell lines that express mutant BRAF (Fig. S4 A and B).

We next examined vulnerabilities in cell lines harboring PIK3CA mutations. PIK3CA itself strongly scored as a top differentially essential gene between PIK3CA mutant and PIK3CA wild-type cell lines in 2/3 gene-level analyses (Fig. 2 G and H). MTOR ranked highly in 2/3 analyses (23rd and 30th of 11,194; Fig. S4C); the top scoring MTOR-specific shRNA strongly discriminated the PIK3CA mutant and wild-type classes (P = 6.03 × 10, WoE; Fig. S4D), indicating that cell lines that harbor PIK3CA mutations are also dependent on mTOR. These observations confirm prior work showing that mTOR plays an important role in PI3K signaling (7).

To assess how the number of cell lines analyzed affected these analyses, we repeated our WoE scoring of a set of shRNAs using data from smaller numbers of cell lines subsampled from the entire dataset. For the top scoring KRAS, BRAF, and PIK3CA shRNAs that were able to distinguish cell lines with mutant or wild-type alleles of these same respective genes, we found that the analysis of a smaller number of cell lines (<5) decreased the likelihood of discriminating between these two classes, whereas the comparison of groups composed of >10 cell lines greatly increased our ability to distinguish the two classes (Fig. 2 C, F, and I). We concluded that the analysis of a large number of cell lines overcomes the inherent heterogeneity of cell lines and reveals robust relationships between essentiality and particular cell features that persist across different genetic and epigenetic backgrounds. With this foundation, we undertook a preliminary exploration of what can be learned about genetic vulnerabilities of cancer cells with specific properties.

Lineage-Specific Genetic Dependencies.

We hypothesized that a subset of genes showing enhanced dependencies in specific lineages would also be aberrantly activated in tumors due to amplification or overexpression. Recent studies have identified oncogenic transcription factors that are amplified, overexpressed, and essential in subsets of tumors from cancers of specific lineages, including NKX2-1 in lung adenocarcinoma (8), MITF in melanoma (9), and SOX2 in squamous cell carcinomas (10). To identify lineage-specific dependencies, we ranked shRNAs by their ability to discriminate cell lines of one lineage from cell lines from all other lineages (Fig. 3A). We selected the top 150 genes (1.3% of those screened) based on ranking of the top-ranked shRNAs, the top 300 genes (2.7%) by the second-best ranking shRNAs, and the top 300 genes (2.7%) as assessed by the KS statistic (ref. 3; Table S3). Three categories of essential genes were considered for further analysis: (i) genes scoring by all three methods from individual lineage analyses; (ii) genes scoring by any method that also were amplified in primary tumors (essential and amplified; Table S3); and (iii) genes scoring by any method that also were differentially up-regulated in cell lines derived from that lineage (essential and overexpressed; Tables S3 and S4). An overview of these results from analyses performed across six cancer lineages is displayed in Table S5.

An external file that holds a picture, illustration, etc.
Object name is pnas.1109363108fig03.jpg

Lineage-specific dependencies. (A) Heatmap of differentially antiproliferative shRNAs in cell lines from individual cancer lineages in comparison with all others. The top 20 shRNAs that distinguish each lineage from the others are displayed. (B) Ovarian-specific dependencies. Three complementary methods of gene scoring [ranking by (i) best or (ii) second best scoring shRNA or (iii) composite of all shRNAs for the gene using a KS statistic identified 582 (5.2%) genes that were selectively required for ovarian cancer cell proliferation. Fifty of these were among the 1,825 recurrently amplified genes in primary high-grade serous ovarian tumors (1). Among the 200 genes that were differentially overexpressed in ovarian cancer cell lines, 114 genes were included in the shRNA pool, and 5 genes showed enhanced essentiality in ovarian cancer lines. Twenty-two of these genes that were scored by all three gene-scoring methods were considered as high-confidence essential genes. (C) Distributions of the scores of PAX8 shRNA (given as the percentile of their rankings, y axis) after 100 trials of the ovarian vs. nonovarian WoE comparison, for equal class sizes of 1–25 (x axis; colors indicate PAX8 shRNAs 1–5). The red bar indicates the median value for each group of subsamplings, boxes represent the 25th to 75th percentile of the data, and whiskers extend to the most extreme values of the group that are not considered outliers.

In colon cancer cell lines, we found KRAS, CTNNB1, and BRAF among 23 essential genes that scored by all three methods and found KRAS and IGF1R (11) among 35 essential and amplified genes. In pancreatic cancer cell lines, we identified KRAS among 23 essential genes that scored in all three analyses (Table S5). In NSCLC, we found NKX2-1 as the only essential gene that is both amplified and overexpressed and found CDK6 among 7 essential and amplified genes (8). These observations provide evidence that such integrative lineage analyses identify both known oncogenes and other relevant lineage-restricted dependencies.

In addition, we identified a number of particularly interesting, previously undescribed candidate lineage-specific dependencies. For example, we found MAP2K4, an activator of JNK and p38 (12), among 10 genes that showed selective essentiality and expression in NSCLC. We found MYB (13) and AXIN2 (14) among 9 genes that were essential and differentially expressed in colon cancer, and we identified SOX9 (15) among 18 genes in pancreatic cancer that emerged as lineage-specific dependencies nominated by all three gene-scoring methods.

To extend these observations, we selected the ovarian lineage for deeper analysis. Of the 582 genes (5.2%) nominated as candidates for enhanced dependency in ovarian cancer cells, we identified 22 essential genes that scored in all three analytical methods (Fig. 3B and Table S5) and found 5 essential and overexpressed genes (Tables S4 and S5). TCGA identified 1,825 genes residing on recurrently amplified regions in ovarian tumors, and we identified 50 amplified genes as also essential (Fig. 3B and Table S5). The set of amplified and essential genes included the known oncogene CCNE1 (16) and candidates including the FRS2 adaptor protein (17), the PRKCE protein kinase (18), RPTOR (19), and the PAX8 paired box transcription factor. Similarly to NKX2-1 in NSCLC, MITF in melanoma, and MYB in colon cancer, PAX8 was not only essential and amplified but also overexpressed in a lineage-specific manner.

Characterization of PAX8 Dependency in Ovarian Cancer.

PAX8 was the only gene that was (i) identified as an essential gene in all three scoring methods, (ii) amplified in primary high-grade serous ovarian tumors, and (iii) differentially expressed in ovarian cancer cell lines (Table S5). Cell line subsampling analysis revealed that the large number of ovarian cell lines screened enabled the identification of this previously undescribed dependency (Fig. 3C).

PAX8 is a lineage-restricted transcription factor that plays an essential role in organogenesis of the Müllerian system (20), the thyroid, and the kidney (21). PAX8 was previously found to be overexpressed in ovarian cancers (22) and implicated in follicular thyroid cancer development (23). We observed that PAX8 was the most differentially expressed gene when we compared ovarian cell lines to all of the other cancer cell lines (Fig. 4A). Furthermore, we found that PAX8 was amplified in 16% of primary ovarian tumors [log2(copy number ratio) > 0.3; n = 345] in a peak (2q13) that also contains PSD4 and LOC654433 (Fig. 4B).

An external file that holds a picture, illustration, etc.
Object name is pnas.1109363108fig04.jpg

PAX8 is essential for ovarian cancer cell proliferation and survival. (A) PAX8 is the top-ranked differentially expressed gene between ovarian and nonovarian cancer cell lines. Arrow indicates PAX8. (B) SNP array colorgrams depict genomic amplification of PAX8 in primary high-grade serous ovarian cancers (1). Regions of genomic amplification and deletion are denoted in red and blue, respectively. SNP array profiles derived from primary ovarian tumors were sorted based on the degree of amplification of each gene. Black vertical lines denote the boundaries of the PAX8 gene. (C) Boxplot showing significant difference in the degree of depletion of a PAX8-specific shRNA in 63 cell lines with low levels of PAX8 compared with 20 lines with high levels of PAX8 (P = 2.14 × 10, t test). Cell lines were divided into high- and low-expressing groups. The red line indicates the median value for each group, boxes represent the 25th to 75th percentile of the data, and whiskers extend to the most extreme values of the group that are not considered outliers. Ovarian cancer cell lines are plotted with red circles; cell lines from all other lineages are plotted with green circles. (D Upper) Effects of PAX8 suppression on proliferation in eight ovarian cancer cell lines; dotted line indicates 50% relative proliferation. (Lower) Immunoblot of PAX8 in a panel of eight ovarian cancer cell lines and in immortalized IOSE-T80 cells. Cell lines with amplification of 2q13 (log2 copy number ratio > 0.3) are marked in red. * denotes nonspecific band. (E Upper) Effects of PAX8 suppression on proliferation of cell lines from indicated cancer types. (Lower) Immunoblot of PAX8. Error bars indicate SD of six replicate measurements. (F) Immunoblot of poly(ADP-ribose) polymerase after PAX8 suppression in two 2q13-amplified cell lines. * denotes nonspecific band.

We further examined the relationships among PAX8 amplification, PAX8 expression, and dependence on PAX8 in ovarian cancer cells. The PAX8-specific shRNA that scored 7th out of 54,020 shRNAs and the 2nd-ranking PAX8 shRNA both suppressed PAX8 (Fig. S4E). The sensitivity of cell lines to inhibition by the highest ranked PAX8-specific shRNA correlated with the level of PAX8 expression, based on comparison of cell lines with high vs. low PAX8 levels (P = 2.14 × 10, t test; Fig. 4C). Cell lines expressing high levels of PAX8 included the vast majority (21/25) of ovarian cancer cell lines as well as a renal and an endometrial cancer line. These observations suggested that the expression of PAX8 is selectively required for the proliferation/survival of cell lines expressing PAX8.

To confirm these findings, we introduced two distinct shRNAs targeting PAX8 into 17 cell lines. We found that suppression of PAX8 resulted in a >50% reduction in the viability in six of eight ovarian cancer cell lines (Fig. 4D), but failed to affect the proliferation of immortalized human ovarian surface epithelial cells (IOSE-T80; ref. 24) and eight other cell lines that did not express PAX8 (Fig. 4E). The six sensitive ovarian cell lines included three cell lines with amplification of 2q13 in which the PAX8 locus resides (Fig. S4F) and three additional cell lines that express higher levels of PAX8 protein compared with IOSE-T80 cells (Fig. 4D). Suppression of PAX8 induced apoptosis in these ovarian cancer cell lines (Fig. 4F). In contrast, the two cell lines (COV504 and OV-90) least sensitive to PAX8 suppression did not harbor the 2q13 amplification and expressed relatively low levels of PAX8 (Fig. 4D). These observations suggest that PAX8 represents a lineage-specific essential gene in a significant subset of ovarian cancer.

Pooled shRNA Screening.

Lentiviral pLKO.1- shRNA constructs were obtained from the RNAi Consortium, and the human 54K pool of 54,020 shRNA plasmids was assembled by combining 16 normalized subpools of ∼3400 shRNA plasmids. The list of 54,020 shRNAs can be found at http://www.broadinstitute.org/igp. Genome-scale pooled shRNA screens to identify genes essential for proliferation in 102 cancer cell lines were performed (3) using a lentivirally delivered pool of 54,020 shRNAs targeting 11,194 genes. The culture conditions for all cancer cell lines are listed in Table S1. Each cell line was infected in quadruplicate and propagated for at least 16 population doublings. The abundance of shRNA constructs relative to the initial DNA plasmid pool was measured by microarray hybridization (3) and analyzed by using a uniform pipeline. Detailed descriptions of each procedure can be found in SI Methods.

Data Processing, Class Comparison, and Gene Ranking.

Raw .CEL files from custom Affymetrix barcode arrays were processed with a modified version of dCHIP software (3). The GenePattern modules shRNAscores and NormalizeCellLines were used to calculate the log fold change in shRNA abundances for each cell line at the conclusion of the screening relative to the initial plasmid DNA reference pool and to normalize these depletion values by using peak median absolute deviation normalization, a variation of Z score with median absolute deviation (3). Class definition files (.cls) were made by using the GenePattern module SubsetGctandCls; definitions included cell line lineage (e.g., ovarian cancer, NSCLC, etc.) or genetic alterations (28, 29). To compute the statistical evidence that a given shRNA contributes to the observed essentiality phenotype between two classes of interest, we used a WoE approach (4, 5). The GENE-E program (http://www.broadinstitute.org/cancer/software/GENE-E; ref. 3) was used to collapse shRNAscores to gene rankings by three complementary methods. These methods included (i) ranking genes by their highest shRNA depletion score, (ii) ranking genes based on the P value rank (correcting for different set sizes of shRNA targeting different genes) of their second best ranked shRNA, and (iii) ranking genes using a KS statistic in an approach similar to gene set enrichment analysis (RNAi gene enrichment ranking; ref. 3). Detailed descriptions of each procedure can be found in SI Methods. All data files, accessory files, and GenePattern modules can be found through the Integrative Genomics Portal (http://www.broadinstitute.org/igp).

Competition Assay.

OVCAR-8 (5 × 10) cells were seeded into each well of a 96-well plate and spin-infected with 2 or 4 μL of lentiviruses (in duplicate) at 930 × g for 2 h at 30 °C in the presence of 4 μg/mL polybrene to transduce ∼50% of the cells. Cells were then trypsinized and replated into 24-well plates. The percent of GFP+ cells at 3 and 7 d after infection was measured using BD LSR II flow cytometry system equipped with a high-throughput sampler (BD Biosciences). The fraction of GFP+ cells 7 d after infection relative to 3 d after infection was calculated. Data represent mean ± SD of duplicate infections.

Supplementary Material

Supporting Information:
Department of Medical Oncology,
Center for Cancer Genome Discovery and
Center for Molecular Oncologic Pathology, Dana-Farber Cancer Institute, Boston, MA 02115;
Broad Institute of Harvard and MIT, Cambridge, MA 02142;
Harvard-MIT Division of Health Sciences and Technology and
Electrical Engineering and Computer Science, David H. Koch Institute for Integrative Cancer Research, Massachusetts Institute of Technology, Cambridge, MA 02139;
Departments of Pathology and
Medicine, Brigham and Women's Hospital, Harvard Medical School, Boston, MA 02115; and
Howard Hughes Medical Institute, Chevy Chase, MD 20815
To whom correspondence may be addressed. E-mail: gro.etutitsnidaorb@rednal, gro.etutitsnidaorb@toord, or ude.dravrah.icfd@nhah_mailliw.
Contributed by Eric S. Lander, June 13, 2011 (sent for review April 13, 2011)

Author contributions: H.W.C., G.S.C., B.A.W., J.S.B., E.S.L., D.E.R., and W.C.H. designed research; H.W.C., G.S.C., B.A.W., J.S.B., S.R., J.A.S., A.E., L.D.A., P.H.L., T.C.W., G.J., and J.H. performed research; C.H.M., G.G., J.B., S.G., P.T., A.T., N.S., B.L., Y.R., R.D., S.N.B., and L.A.G. contributed new reagents/analytic tools; B.A.W., S.G., P.T., J.G., A.T., J.P.M., M.M., E.S.L., D.E.R., and W.C.H. analyzed data; and H.W.C., G.S.C., B.A.W., J.S.B., E.S.L., D.E.R., and W.C.H. wrote the paper.

H.W.C., G.S.C., B.A.W., and J.S.B. contributed equally to this work.
Contributed by Eric S. Lander, June 13, 2011 (sent for review April 13, 2011)
Freely available online through the PNAS open access option.

Abstract

A comprehensive understanding of the molecular vulnerabilities of every type of cancer will provide a powerful roadmap to guide therapeutic approaches. Efforts such as The Cancer Genome Atlas Project will identify genes with aberrant copy number, sequence, or expression in various cancer types, providing a survey of the genes that may have a causal role in cancer. A complementary approach is to perform systematic loss-of-function studies to identify essential genes in particular cancer cell types. We have begun a systematic effort, termed Project Achilles, aimed at identifying genetic vulnerabilities across large numbers of cancer cell lines. Here, we report the assessment of the essentiality of 11,194 genes in 102 human cancer cell lines. We show that the integration of these functional data with information derived from surveying cancer genomes pinpoints known and previously undescribed lineage-specific dependencies across a wide spectrum of cancers. In particular, we found 54 genes that are specifically essential for the proliferation and viability of ovarian cancer cells and also amplified in primary tumors or differentially overexpressed in ovarian cancer cell lines. One such gene, PAX8, is focally amplified in 16% of high-grade serous ovarian cancers and expressed at higher levels in ovarian tumors. Suppression of PAX8 selectively induces apoptotic cell death of ovarian cancer cells. These results identify PAX8 as an ovarian lineage-specific dependency. More generally, these observations demonstrate that the integration of genome-scale functional and structural studies provides an efficient path to identify dependencies of specific cancer types on particular genes and pathways.

Keywords: high throughput, RNAi, oncogene, lineage
Abstract

The application of whole-genome approaches to identify genetic alterations in cancer genomes is providing new insights into the spectrum of molecular events that occur in human tumors. Although in some cases this knowledge immediately illuminates a path toward clinical implementation, the long lists of genes with aberrant sequence, copy number, or expression in tumors already found make it clear that complementary information from systematic functional studies will be essential to obtain a comprehensive molecular understanding of cancer and to convert this knowledge into therapeutic strategies.

Most ovarian cancer patients present at an advanced stage with widely disseminated disease in the peritoneal cavity. Despite advances in surgery and chemotherapy, the majority of ovarian cancer patients re-present with relapsed and progressively chemotherapy-resistant and lethal disease. The Cancer Genome Atlas (TCGA) Project has characterized nearly 500 primary high-grade serous ovarian cancer genomes to identify recurrent genetic alterations in ovarian cancer (1).

A major feature of ovarian cancers is recurrent regions of copy-number alteration (1). A small number of these recurrent genomic events harbor known oncogenes and tumor suppressor genes, such as MYC, CCNE1, and RB (2). However, as in other cancers, the specific genes out of the ∼1,800 genes targeted by recurrent amplification events remain undefined. Here, we report a genome-scale functional study to identify genes that are essential for the survival of 102 human cancer cell lines. The interrogation of a large number of cancer cell lines provides increased power to identify relationships between gene expression and dependence. To demonstrate the utility of integrating these data with results from cancer genome characterization, we focused on ovarian cancer and identified ovarian cancer lineage-specific dependencies.

Click here to view.

Acknowledgments

We thank M. Reich, A. Derr, and T. Liefield for assistance with analysis and data portal creation. This work was supported in part by National Institutes of Health Grants R33 CA128625, U54 CA112962, and RC2 CA148268, Novartis Pharmaceuticals, Richard Hunt, and the H. L. Snyder Foundation.

Acknowledgments

Footnotes

Conflict of interest statement: R.D., L.A.G., M.M., and W.C.H. are consultants for Novartis Pharmaceuticals.

This article contains supporting information online at www.pnas.org/lookup/suppl/doi:10.1073/pnas.1109363108/-/DCSupplemental.

Footnotes

References

  • 1. TCGA-NetworkIntegrated genomic analyses of ovarian carcinoma. Nature. 2011;474:609–615.[Google Scholar]
  • 2. Bast RC, Jr, Hennessy B, Mills GBThe biology of ovarian cancer: New opportunities for translation. Nat Rev Cancer. 2009;9:415–428.[Google Scholar]
  • 3. Luo B, et al Highly parallel identification of essential genes in cancer cells. Proc Natl Acad Sci USA. 2008;105:20380–20385.[Google Scholar]
  • 4. Good IJ Good Thinking: The Foundations of Probability and Its Applications. Minneapolis: University of Minnesota Press; 1983. [PubMed][Google Scholar]
  • 5. Tamayo P, et al Predicting relapse in patients with medulloblastoma by integrating evidence from clinical and genomic features. J Clin Oncol. 2011;29:1415–1423.[Google Scholar]
  • 6. Flaherty KT, et al Inhibition of mutated, activated BRAF in metastatic melanoma. N Engl J Med. 2010;363:809–819.[Google Scholar]
  • 7. Weigelt B, Warne PH, Downward JPIK3CA mutation, but not PTEN loss of function, determines the sensitivity of breast cancer cells to mTOR inhibitory drugs. Oncogene. 2011 in press. [[PubMed][Google Scholar]
  • 8. Weir BA, et al Characterizing the cancer genome in lung adenocarcinoma. Nature. 2007;450:893–898.[Google Scholar]
  • 9. Garraway LA, et al Integrative genomic analyses identify MITF as a lineage survival oncogene amplified in malignant melanoma. Nature. 2005;436:117–122.[PubMed][Google Scholar]
  • 10. Bass AJ, et al SOX2 is an amplified lineage-survival oncogene in lung and esophageal squamous cell carcinomas. Nat Genet. 2009;41:1238–1242.[Google Scholar]
  • 11. Lopez T, Hanahan DElevated levels of IGF-1 receptor convey invasive and metastatic capability in a mouse model of pancreatic islet tumorigenesis. Cancer Cell. 2002;1:339–353.[PubMed][Google Scholar]
  • 12. Khatlani TS, et al c-Jun N-terminal kinase is activated in non-small-cell lung cancer and promotes neoplastic transformation in human bronchial epithelial cells. Oncogene. 2007;26:2658–2666.[PubMed][Google Scholar]
  • 13. Ramsay RG, Gonda TJMYB function in normal and cancer cells. Nat Rev Cancer. 2008;8:523–534.[PubMed][Google Scholar]
  • 14. Lammi L, et al Mutations in AXIN2 cause familial tooth agenesis and predispose to colorectal cancer. Am J Hum Genet. 2004;74:1043–1050.[Google Scholar]
  • 15. Lynn FC, et al Sox9 coordinates a transcriptional network in pancreatic progenitor cells. Proc Natl Acad Sci USA. 2007;104:10500–10505.[Google Scholar]
  • 16. Etemadmoghadam D, et al. Australian Ovarian Cancer Study Group. Amplicon-dependent CCNE1 expression is critical for clonogenic survival after cisplatin treatment and is correlated with 20q11 gain in ovarian cancer. PLoS ONE. 2010;5:e15498.
  • 17. Gotoh NRegulation of growth factor signaling by FRS2 family docking/scaffold adaptor proteins. Cancer Sci. 2008;99:1319–1325.[PubMed][Google Scholar]
  • 18. Okhrimenko H, et al Protein kinase C-epsilon regulates the apoptosis and survival of glioma cells. Cancer Res. 2005;65:7301–7309.[Google Scholar]
  • 19. Kim DH, Sabatini DMRaptor and mTOR: Subunits of a nutrient-sensitive complex. Curr Top Microbiol Immunol. 2004;279:259–270.[PubMed][Google Scholar]
  • 20. Mittag J, Winterhager E, Bauer K, Grümmer RCongenital hypothyroid female pax8-deficient mice are infertile despite thyroid hormone replacement therapy. Endocrinology. 2007;148:719–725.[PubMed][Google Scholar]
  • 21. Mansouri A, Chowdhury K, Gruss PFollicular cells of the thyroid gland require Pax8 gene function. Nat Genet. 1998;19:87–90.[PubMed][Google Scholar]
  • 22. Bowen NJ, et al Emerging roles for PAX8 in ovarian cancer and endosalpingeal development. Gynecol Oncol. 2007;104:331–337.[PubMed][Google Scholar]
  • 23. Wang Q, et al Pax genes in embryogenesis and oncogenesis. J Cell Mol Med. 2008;12(6A):2281–2294.[PubMed][Google Scholar]
  • 24. Liu J, et al A genetically defined model for human ovarian cancer. Cancer Res. 2004;64:1655–1663.[PubMed][Google Scholar]
  • 25. Carlson JW, et al Serous tubal intraepithelial carcinoma: Its potential role in primary peritoneal serous carcinoma and serous cancer prevention. J Clin Oncol. 2008;26:4160–4165.[Google Scholar]
  • 26. Köbel M, et al Ovarian carcinoma subtypes are different diseases: Implications for biomarker studies. PLoS Med. 2008;5:e232.[Google Scholar]
  • 27. Kaelin WG., Jr Synthetic lethality: A framework for the development of wiser cancer therapeutics. Genome Med. 2009;1:99.
  • 28. Thomas RK, et al High-throughput oncogene mutation profiling in human cancer. Nat Genet. 2007;39:347–351.[PubMed][Google Scholar]
  • 29. Forbes SA, et al The Catalogue of Somatic Mutations in Cancer (COSMIC) Curr Protoc Hum Genet. 2008;57:10.11.1–10.11.26.[Google Scholar]
Collaboration tool especially designed for Life Science professionals.Drag-and-drop any entity to your messages.