| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Advances in Brief |
Department of Chemistry, The Scripps Research Institute, La Jolla, California 92037 [A. I. S., P. G. S.]; Genomics Institute of the Novartis Research Foundation, San Diego, California 92121 [J. B. W., L. M. S., S. G. K., P. D., H. L., P. G. S., G. M. H.]; and Departments of Medicine [S. M. P.] and Pathology [C. A. M., H. F. F.], University of Virginia Health System, Charlottesville, Virginia 22908
 |
ABSTRACT |
|---|
 Top ABSTRACT IntroductionMaterials and MethodsResults and DiscussionREFERENCES |
|---|
70% of all cancer-related deaths in the United States. The classification scheme was based on identifying gene subsets whose expression typifies each cancer class, and we quantified the extent to which these genes are characteristic of a specific tumor type by accurately and confidently predicting the anatomical site of tumor origin for 90% of 175 carcinomas, including 9 of 12 metastatic lesions. The predictor gene subsets include those whose expression is typical of specific types of normal epithelial differentiation, as well as other genes whose expression is elevated in cancer. This study demonstrates the feasibility of predicting the tissue origin of a carcinoma in the context of multiple cancer classes. |
Introduction |
|---|
|
TopABSTRACTIntroductionMaterials and MethodsResults and DiscussionREFERENCES |
|---|
4% of all patients diagnosed with cancer present with metastatic tumors for which the origin of the primary tumor has not been determined (1)
. On occasion, the primary site for a metastatic tumor is not clearly apparent even after pathological analysis. Thus, predicting the primary tumor site of origin for some of these cancers represents an important clinical objective. We have constructed a first-generation molecular classification scheme for carcinomas of the prostate, breast, colorectum, lung (adenocarcinoma and squamous cell carcinoma), liver, gastroesophagus, pancreas, ovary, kidney, and bladder/ureter, which collectively account for 70% (400,000 cases) of all cancer-related deaths in the United States (2)
. The gene expression signatures discovered by our classification approach include novel tumor-related genes whose encoded proteins may lead to new clinical reagents for successful tumor diagnosis. |
Materials and Methods |
|---|
|
TopABSTRACTIntroductionMaterials and MethodsResults and DiscussionREFERENCES |
|---|
Microarray Hybridization.
RNA extraction and hybridization on oligonucleotide microarrays (U95a GeneChip; Affymetrix Incorporated, Santa Clara, CA) was performed as described (4)
, with the exception that the arrays were hybridized at 50°C for 16–20 h. GeneChip hybridization data were processed and scaled as described (5
, 6)
. We included only those probe sets (9198) whose maximum hybridization intensity (AD) in at least one sample was >200; the other probe sets were excluded (the quantification of gene transcripts with AD values uniformly <200 are typically unreliable). All AD values <20, including negative AD values, were raised to a value of 20, and the data were log transformed. The primary hybridization data are available from our website.4
Cancer Classification and Cancer Class Prediction.
For each of the 9198 genes that passed the minimal expression threshold, a Wilcoxon rank score (7)
was calculated for the group with the highest mean expression versus samples from all other groups (implemented in Matlab version 6.0). The 100 genes with the lowest Ps in each class (total, 1100 genes) were ranked based on their predictive accuracy for discriminating one tumor class versus all others using a SVM classifier (8)
. Specifically, genes were ranked based on their LOOCV accuracy (9)
. In LOOCV for a given gene, we blinded ourselves to one sample, trained an SVM using the remaining samples, and used the SVM to predict the class identity of the blinded sample (either cancer class X, or not cancer class X). This process was repeated for all samples in the training set, and an overall prediction accuracy was calculated for each gene. The SVM procedure used here5
was implemented in the software package R v1.2.2.4. The voting scheme used the 10 genes with the highest SVM/LOOCV accuracy from each class (110 total genes across 11 tumor classes). For each class, a minimum SVM/LOOCV accuracy threshold was set such that at least 10 genes passed; because in each class multiple genes have equivalent accuracy, 216 genes were selected from the 11 classes and were iteratively bootstrapped to obtain an equal number (i.e., 10) of voting genes per class (10)
. For classifying an unknown sample, prediction scores were calculated using one set of 110 genes (calculated as described below), and final predictions were based on averaged scores over 50 iterations. Hybridization values for our 110-gene predictor set were compared to each sample in our training set. An L1 distance (sum of absolute differences) from the unknown sample to each training sample was calculated. The "class distance" was defined as the mean distance from the unknown sample to the members of that class in the training set. The class to which an unknown sample has the lowest class distance is the predicted identity. A Dixon test for outliers was used to assign a confidence score to each prediction. The Dixon metric is calculated by sorting the vector of mean distances, where Xi < Xi+1, and computing the value D = (X2 - X1)/(Xn - X1) (11)
. A Dixon threshold of D = 0.1 was empirically set as a conservative boundary for high confidence predictions.
Tissue Microarrays and IHC.
Tissue microarrays containing 0.6-mm cores from 265 different zinc formalin-fixed, paraffin-embedded specimens were constructed using a Tissue Microarrayer (Beecher Instruments, Silver Spring, MD). Samples consisted of 36 normal adult epithelial tissues and 229 carcinomas, which included most of the tumors whose transcripts were profiled in the study. Ovarian cancers were profiled as described previously (3)
, and 16 other independent serous papillary carcinomas of the ovary were included in the tissue microarrays. For IHC on the tissue microarrays and on a whole-tissue section of a normal ovary, the avidin-biotin immunoperoxidase method was performed. After slides had been placed in a citrate buffer and treated with microwave heat for 20 min, the polyclonal anti-WT antibody (C-19; 1:100 dilution; Santa Cruz Biotechnology, Santa Cruz, CA) was applied for 1 h at room temperature. Nuclear immunoreactivity was considered to represent true positivity.
|
Results and Discussion |
|---|
|
TopABSTRACTIntroductionMaterials and MethodsResults and DiscussionREFERENCES |
|---|
Initial analysis of the data by methods that group similarly expressed genes, as well as tumors with similar gene expression (i.e., unsupervised hierarchical clustering; Ref. 12
), showed that we could readily group cancers of some anatomical sites, such as those of the prostate and kidney, based solely on the patterns of the most variably expressed genes. In contrast, we found a high degree of similarity between cancers of the colorectum, stomach, bladder/ureter, and lung, making their histological separation difficult on the basis of unsupervised clustering (data not shown; available as supplementary Fig. 1
on our website).4
We therefore divided the process of multiclass prediction into three components: (a) filtering the large data set of gene expression (12,533 genes in 100 tumors; >1.25 x 106 data points) to exclude those genes that do not contribute to tumor distinction; (b) ranking potentially predictive genes to identify the most accurate tumor-specific classifiers; and (c) determining an optimal method by which these genes could be used to "vote" for the likely class of a blinded tumor sample in the context of multiple tumor classes.
|
. This was achieved using the Wilcoxon rank-sum test (7)
, which tests the null hypothesis that gene expression in one tumor class is not different from gene expression in any other tumor class. The genes in each class that had significant Ps represented those that disputed the null hypothesis and defined those that were most different among tumor classes. For carcinomas of all of the anatomical sites that were examined, we were able to identify many such genes (Fig. 1B)
. One hundred of the Wilcoxon-selected genes from each tumor class were then subjected to a "prediction accuracy test." Each of the genes was tested for its ability to discriminate one tumor class from all other tumor classes, using a SVM-learning algorithm (Ref. 8
; see "Materials and Methods"), which has been shown to yield good results in gene expression-based classification problems (Ref. 13
; Fig. 1
). LOOCV was used to blind ourselves sequentially to each of the 100 tumor samples; the SVM was trained on the remaining samples and then used to predict the class of the blinded specimens (9
, 10)
. This test identified >10 genes per tumor class that could predict the class of a blinded tumor in at least 91% of cases (such as lung adenocarcinomas and bladder/ureter carcinomas; Fig. 1C
). Typically, the accuracy of the classifier genes was higher, ranging up to 100% (e.g., prostate carcinoma; Fig. 1C
). A voting scheme was developed based on calculating a class distance, by which we could evaluate how molecularly related an unknown sample was to tumors of different classes (see "Materials and Methods"). We also used a confidence score to estimate the strength of each prediction and experimentally determined a confidence threshold that minimized tumor misclassification. Empirically, we determined that a small group of 110 genes, representing 10 genes per tumor class, most accurately predicted the origin of a blinded tumor sample (see "Materials and Methods"). The complete list of genes comprising the multiclass predictor is available as supplementary Table 1
from our website.4
|
; also see supplementary Table 2
on our website).4
Therefore, in the absence of the confidence threshold, we correctly predicted an anatomical origin of 97 of 100 (97%) tumors. Anecdotally, one of the tumor samples that was originally part of our training set and was labeled as a hepatocellular carcinoma was strongly predicted as a colon cancer by cross-validation (data not shown). Histological reevaluation of this poorly differentiated neoplasm revealed some minor differences between the frozen tissue section representative of the material that we had profiled and a paraffin-embedded tissue section from the patient’s tumor. DNA was isolated from both frozen and paraffin-embedded tissue sections and genotyped with a series of polymorphic microsatellite markers. The genotypes from the different sources were substantially different, suggesting that the frozen tissue sample had been mislabeled. These results underscore the use of an objective molecular classification scheme because it depends on objective molecular signatures rather than relying on morphological features of the tumor tissues.
|
, there were no obvious distinguishing features among these carcinomas versus other carcinomas in the training or test sets. Of the 11 unclassified cases, 7 were correctly classified, but with low confidence. Thus, a correct anatomical site of tumor origin was predicted for 71 of 75 (95%) cases in the test set, including 11 of the 12 (92%) metastatic lesions (Tables 1
and 2
, and supplementary Table 2
on our website).4
It is important to note that this initial test set does not fully test our class prediction model, specifically because of the lack of sufficient numbers of pancreatic, bladder, and kidney carcinomas. However, our cross-validation results strongly suggest that we would be able to correctly predict an anatomical origin for the majority of carcinomas from these tissue sites.
Most of the genes included in the classifier are expressed in a tissue-specific manner in the epithelium from which the tumors arose and are expressed at similar or elevated levels in the resultant carcinomas (supplementary Table 1
).4
On the basis of gene annotation alone, we recognized many well-described genes whose expression is elevated in tumors. These included MUC-2 and A33 in colon cancers, the latter of which has been used as an immunotherapeutic target in advanced colorectal carcinomas (14)
; mammaglobin-1 (MGB-1), which has been found to be a highly sensitive diagnostic marker for micrometastatic breast carcinoma (15)
; and thyroid transcription factor 1 (TTF-1), which has been proposed as a highly accurate marker for the differential diagnosis of lung adenocarcinomas (16)
. We also identified genes such as uroplakin II (UPII), whose expression in bladder carcinoma cells is likely maintained at levels similar to that of normal urothelium. Detection of UPII transcripts in circulating bladder cancer cells, however, has been proposed as a sensitive marker of micrometastasis (17)
.
We also identified genes whose annotations suggested their expression in the stromal cells that surround epithelial tumors or in inflammatory cells. In some cases we subsequently found evidence that suggests their overexpression in malignant epithelia [e.g., the fibroblast activation protein (FAP-
) in breast cancers (18)
]. In adenocarcinomas of the lung, we identified genes whose annotations indicated the presence of B cells, T cells, macrophages, and neutrophils. We suspect that many of these genes may have been selected because of the relative paucity of "lung-specific" classifiers, and not because these samples necessarily contained higher proportions of infiltrating inflammatory cells relative to the other tumor samples. Conservatively, we suggest that the most reliable classifiers of lung adenocarcinomas probably include those genes with predicted accuracies >95%, i.e., TTF-1. In pancreas cancers we identified genes whose expression is indicative of acinar cell differentiation. Although we specifically attempted to avoid normal epithelium in all of the tumor samples that we profiled, the highly diffuse nature of pancreatic cancer growth precluded an absolutely complete separation of normal and neoplastic cells. Highly expressed genes within small amounts of normal epithelia may conceivably give rise to some of the signals detected on the arrays. However, it remains a possibility that expression of some of these "acinar" genes is maintained in pancreatic tumor cells.
Because of the inherent difficulty in using gene annotation alone to judge tissue-specific versus tumor-elevated gene expression, we next sought to objectively "dissect" some of the predictor gene subsets into tissue-specific genes and tissue-specific/tumor-elevated genes. As an example, we chose 28 of the genes that were 
92% predictive of serous papillary carcinomas of the ovary and compared the expression levels of these genes in an expanded set of 24 ovarian tumor samples against 5 samples of normal ovary, 2 of which were highly enriched for surface ovarian epithelial cells (3)
. Differential expression was determined for genes whose expression was significantly different in normal and tumor tissues (P < 0.01, unpaired t test) and where the mean level of expression in tumor tissues was >3 times that in normal tissues. By these criteria, 18 of 28 genes were significantly overexpressed in the tumors (Fig. 2A)
. Among this group of genes were protease M/neurosin/kallikrein 6 (hK6), which has been proposed as a candidate serum marker for ovarian cancer (19)
, and mesothelin (CAK1), which is overexpressed in ovarian cancers and is used as a specific target for a novel therapeutic immunotoxin (20)
. The 10 tissue-specific genes, which included the WT gene (WT-1), smad6, and Hox5.1, most likely represent features of normal ovarian physiology.
|
. As expected from the analysis of classifier gene transcription in ovarian cancers, the normal serous lining epithelium of the ovary was also positive for WT protein (Fig. 2C)
. It should be noted that expression of WT has been reported in other tissues, but in the context of the tissues examined in our classification scheme, WT expression was specific for the ovary. Transcript profiles of human tumors have previously been used to predict the membership of an unknown sample into one of two, three, or at most four distinct tumor classes (21, 22, 23) . However, the use of tumor-specific genes to extend these or other discriminant methods to prediction of tumor origin in the context of multiple (>10) cancer classes has not been demonstrated and is particularly challenging. We assessed many methods for multiclass prediction during this study, based on either weighted correlation methods (21) or on other supervised learning methods (e.g., Fisher’s linear discriminant analysis). Although all of the methods that we used have performed reasonably well, we found that methods such as SVM, which do not make assumptions about the distribution of the data (8) , performed significantly better and selected for greater uniformity and specificity among the class-specific predictors. These findings have recently been corroborated (24) , although the specifics of the SVM methodology are different.
We found that classification of tumors arising in certain anatomical sites was relatively straightforward because of the large number of unequivocal predictor genes (e.g., 19 genes with 100% predictive accuracy for prostate cancer). In contrast, prediction of other tumors, such as those of the lung, bladder/ureter, or gastroesophagus, was more difficult because of the relative paucity of highly predictive classifier genes. The difficulty in selecting genes whose expression is specific to these cancers reflects a high degree of molecular relatedness, which we had observed in initial analyses of tumor gene expression.4 For example, blinded gastroesophageal cancers that could not be predicted by our method were assigned as lung tumors (albeit with confidence scores close to zero). Analysis of the entire human transcriptome may uncover tumor-specific genes for those neoplasms that we have shown to have a high similarity in expression profiles.
A striking conclusion from the data presented here is that we could identify subsets of genes with highly restricted, tumor-specific expression for as many as 11 distinct tumor classes, despite well-described tumor heterogeneity and obvious molecular similarities among many divergent tumor classes. The fact that we could successfully use these gene subsets to predict the origin of a given tumor in a majority of cases underscores how strongly characteristic these genes must be for specific histopathological subtypes of cancer. In that regard, it is worth noting that, using as few as 11 genes (i.e., 1 gene per tumor class), we could predict the anatomical origin of up to 91 and 83% of the training and blinded tumor samples, respectively (in the absence of a strict confidence threshold). These results suggest that we can construct custom DNA microarrays for a molecular classification of solid tumors, a resource that will augment traditional site-specific and histopathological classification schemes. The extension of these and other discriminant methods to identify molecular correlates with tumor grade, stage, response to therapy, and outcome will further contribute to the optimal management of patients with cancer.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
1 Andrew I. Su acknowledges the ARCS Foundation of San Diego and the La Jolla Interfaces in Science Program for predoctoral support. 
2 To whom requests for reprints should addressed, at Genomics Institute of the Novartis Research Foundation, 3115 Merryfield Row, San Diego, CA 92121. Phone: (858) 812-1522; Fax: (858) 812-1746; E-mail: garret_hampton{at}yahoo.com
3 The abbreviations used are: IHC, immunohistochemistry; AD, average difference; SVM, support vector machine; LOOCV, leave-out-one cross-validation; WT, Wilms’ tumor.
5 Developed by E. Dimitriadou, K. Hornik, F. Leisch, D. Meyer, and A. Weingessel.
Received 7/23/01. Accepted 8/31/01.
|
REFERENCES |
|---|
|
TopABSTRACTIntroductionMaterials and MethodsResults and DiscussionREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  F. A. Monzon, M. Lyons-Weiler, L. J. Buturovic, C. T. Rigl, W. D. Henner, C. Sciulli, C. I. Dumur, F. Medeiros, and G. G. Anderson Multicenter Validation of a 1,550-Gene Expression Profile for Identification of Tumor Tissue of Origin J. Clin. Oncol., May 20, 2009; 27(15): 2503 - 2508. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 N. L. Guo, Y.-W. Wan, K. Tosun, H. Lin, Z. Msiska, D. C. Flynn, S. C. Remick, V. Vallyathan, A. Dowlati, X. Shi, et al. Confirmation of Gene Expression-Based Prediction of Survival in Non-Small Cell Lung Cancer Clin. Cancer Res., December 15, 2008; 14(24): 8213 - 8220. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. A. Oien and T.R. J. Evans Raising the Profile of Cancer of Unknown Primary J. Clin. Oncol., September 20, 2008; 26(27): 4373 - 4375. [Full Text] [PDF]
|
|
|
|
|
|
H. M. Horlings, R. K. van Laar, J.-M. Kerst, H. H. Helgason, J. Wesseling, J. J.M. van der Hoeven, M. O. Warmoes, A. Floore, A. Witteveen, J. Lahti-Domenici, et al. Gene Expression Profiling to Identify the Histogenetic Origin of Metastatic Adenocarcinomas of Unknown Primary J. Clin. Oncol., September 20, 2008; 26(27): 4435 - 4441. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
G. R. Varadhachary, D. Talantov, M. N. Raber, C. Meng, K. R. Hess, T. Jatkoe, R. Lenzi, D. R. Spigel, Y. Wang, F. A. Greco, et al. Molecular Profiling of Carcinoma of Unknown Primary and Correlation With Clinical Evaluation J. Clin. Oncol., September 20, 2008; 26(27): 4442 - 4448. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. Li, W. Liu, Z. Liu, J. Wang, Q. Liu, Y. Zhu, and F. He PRINCESS, a Protein Interaction Confidence Evaluation System with Multiple Data Sources Mol. Cell. Proteomics, June 1, 2008; 7(6): 1043 - 1052. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. I. Dumur, M. Lyons-Weiler, C. Sciulli, C. T. Garrett, I. Schrijver, T. K. Holley, J. Rodriguez-Paris, J. R. Pollack, J. L. Zehnder, M. Price, et al. Interlaboratory Performance of a Microarray-Based Gene Expression Test to Determine Tissue of Origin in Poorly Differentiated and Undifferentiated Cancers J. Mol. Diagn., January 1, 2008; 10(1): 67 - 77. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. Zhang, J. Huang, N. Yang, J. Greshock, S. Liang, K. Hasegawa, A. Giannakakis, N. Poulos, A. O'Brien-Jenkins, D. Katsaros, et al. Integrative Genomic Analysis of Phosphatidylinositol 3'-Kinase Family Identifies PIK3R3 as a Potential Therapeutic Target in Epithelial Ovarian Cancer Clin. Cancer Res., September 15, 2007; 13(18): 5314 - 5321. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. K. Lee, D. M. Havaleshko, H. Cho, J. N. Weinstein, E. P. Kaldjian, J. Karpovich, A. Grimshaw, and D. Theodorescu A strategy for predicting the chemosensitivity of human cancers and its application to drug discovery PNAS, August 7, 2007; 104(32): 13086 - 13091. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 X. Wu, Z. Cai, X.-F. Wan, T. Hoang, R. Goebel, and G. Lin Nucleotide composition string selection in HIV-1 subtyping using whole genomes Bioinformatics, July 15, 2007; 23(14): 1744 - 1752. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 X. Shi, Y.-Q. Ma, Y. Tu, K. Chen, S. Wu, K. Fukuda, J. Qin, E. F. Plow, and C. Wu The MIG-2/Integrin Interaction Strengthens Cell-Matrix Adhesion and Modulates Cell Motility J. Biol. Chem., July 13, 2007; 282(28): 20455 - 20466. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. C. Smith, G. Oxford, A. S. Baras, C. Owens, D. Havaleshko, D. L. Brautigan, M. K. Safo, and D. Theodorescu Expression of Ral GTPases, Their Effectors, and Activators in Human Bladder Cancer Clin. Cancer Res., July 1, 2007; 13(13): 3803 - 3813. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
X. Zhou and D. P. Tuck MSVM-RFE: extensions of SVM-RFE for multiclass gene selection on DNA microarray data Bioinformatics, May 1, 2007; 23(9): 1106 - 1114. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 G. Pentheroudakis, E. Briasoulis, and N. Pavlidis Cancer of Unknown Primary Site: Missing Primary or Missing Biology? Oncologist, April 1, 2007; 12(4): 418 - 425. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 G. Ismael, E. de Azambuja, and A. Awada Molecular profiling of a tumor of unknown origin. N. Engl. J. Med., September 7, 2006; 355(10): 1071 - 1072. [Full Text] [PDF]
|
|
|
|
 |
|
 G. J. Gordon, L. A. Deters, M. D. Nitz, B. C. Lieberman, B. Y. Yeap, and R. Bueno Differential diagnosis of solitary lung nodules with gene expression ratios J. Thorac. Cardiovasc. Surg., September 1, 2006; 132(3): 621 - 627. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. Talantov, J. Baden, T. Jatkoe, K. Hahn, J. Yu, Y. Rajpurohit, Y. Jiang, C. Choi, J. S. Ross, D. Atkins, et al. A Quantitative Reverse Transcriptase-Polymerase Chain Reaction Assay to Identify Metastatic Carcinoma Tissue of Origin J. Mol. Diagn., July 1, 2006; 8(3): 320 - 329. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
F. Hube, Y. Myal, E. Leygue, J. Rollin, Y. Gruel, Y. Yatabe, and on behalf of the authors Expression of Two Breast-Specific Molecules in the Lung J. Mol. Diagn., July 1, 2006; 8(3): 390 - 393. [Full Text] [PDF]
|
|
|
|
|
|
E. Dahl, G. Kristiansen, K. Gottlob, I. Klaman, E. Ebner, B. Hinzmann, K. Hermann, C. Pilarsky, M. Durst, M. Klinkhammer-Schalke, et al. Molecular Profiling of Laser-Microdissected Matched Tumor and Normal Breast Tissue Identifies Karyopherin {alpha}2 as a Potential Novel Prognostic Marker in Breast Cancer. Clin. Cancer Res., July 1, 2006; 12(13): 3950 - 3960. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 O. Mendez, B. Martin, R. Sanz, R. Aragues, V. Moreno, B. Oliva, V. Stresing, and A. Sierra Underexpression of transcriptional regulators is common in metastatic breast cancer cells overexpressing Bcl-xL Carcinogenesis, June 1, 2006; 27(6): 1169 - 1179. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. Guo, Y. Ma, R. Ward, V. Castranova, X. Shi, and Y. Qian Constructing Molecular Classifiers for the Accurate Prognosis of Lung Adenocarcinoma. Clin. Cancer Res., June 1, 2006; 12(11): 3344 - 3354. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. Zhang, J. Huang, N. Yang, S. Liang, A. Barchetti, A. Giannakakis, M. G. Cadungog, A. O'Brien-Jenkins, M. Massobrio, K. F. Roby, et al. Integrative Genomic Analysis of Protein Kinase C (PKC) Family Identifies PKC{iota} as a Biomarker and Potential Oncogene in Ovarian Carcinoma. Cancer Res., May 1, 2006; 66(9): 4627 - 4635. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S.-H. Yeh, D.-C. Wu, C.-Y. Tsai, T.-J. Kuo, W.-C. Yu, Y.-S. Chang, C.-L. Chen, C.-F. Chang, D.-S. Chen, and P.-J. Chen Genetic Characterization of Fas-Associated Phosphatase-1 as a Putative Tumor Suppressor Gene on Chromosome 4q21.3 in Hepatocellular Carcinoma Clin. Cancer Res., February 15, 2006; 12(4): 1097 - 1108. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. M. Barnes, S. Huang, A. Kaipainen, D. Sanoudou, E. J. Chen, G. S. Eichler, Y. Guo, Y. Yu, D. E. Ingber, J. B. Mulliken, et al. Evidence by molecular profiling for a placental origin of infantile hemangioma PNAS, December 27, 2005; 102(52): 19097 - 19102. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. C. Tan, D. Q. Naiman, L. Xu, R. L. Winslow, and D. Geman Simple decision rules for classifying human cancers from gene expression profiles Bioinformatics, October 15, 2005; 21(20): 3896 - 3904. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. Espana, B. Martin, R. Aragues, C. Chiva, B. Oliva, D. Andreu, and A. Sierra Bcl-xL-Mediated Changes in Metabolic Pathways of Breast Cancer Cells: From Survival in the Blood Stream to Organ-Specific Metastasis Am. J. Pathol., October 1, 2005; 167(4): 1125 - 1137. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. M. Eder, X. Sui, D. G. Rosen, L. K Nolden, K. W. Cheng, J. P. Lahad, M. Kango-Singh, K. H. Lu, C. L. Warneke, E. N. Atkinson, et al. Atypical PKC{iota} contributes to poor prognosis through loss of apical-basal polarity and Cyclin E overexpression in ovarian cancer PNAS, August 30, 2005; 102(35): 12519 - 12524. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C.-T. R. Yu, J.-M. Hsu, Y.-C. G. Lee, A.-P. Tsou, C.-K. Chou, and C.-Y. F. Huang Phosphorylation and Stabilization of HURP by Aurora-A: Implication of HURP as a Transforming Target of Aurora-A Mol. Cell. Biol., July 15, 2005; 25(14): 5789 - 5800. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. Mijalski, A. Harder, T. Halder, M. Kersten, M. Horsch, T. M. Strom, H. V. Liebscher, F. Lottspeich, M. H. de Angelis, and J. Beckers Identification of coexpressed gene clusters in a comparative analysis of transcriptome and proteome in mouse tissues PNAS, June 14, 2005; 102(24): 8621 - 8626. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. W. Tothill, A. Kowalczyk, D. Rischin, A. Bousioutas, I. Haviv, R. K. van Laar, P. M. Waring, J. Zalcberg, R. Ward, A. V. Biankin, et al. An Expression-Based Site of Origin Diagnostic Method Designed for Clinical Application to Cancer of Unknown Origin Cancer Res., May 15, 2005; 65(10): 4031 - 4040. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. A. Shedden, M. P. Kshirsagar, D. R. Schwartz, R. Wu, H. Yu, D. E. Misek, S. Hanash, H. Katabuchi, L. H. Ellenson, E. R. Fearon, et al. Histologic Type, Organ of Origin, and Wnt Pathway Status: Effect on Gene Expression in Ovarian and Uterine Carcinomas Clin. Cancer Res., March 15, 2005; 11(6): 2123 - 2131. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Statnikov, C. F. Aliferis, I. Tsamardinos, D. Hardin, and S. Levy A comprehensive evaluation of multicategory classification methods for microarray gene expression cancer diagnosis Bioinformatics, March 1, 2005; 21(5): 631 - 643. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. Zhang, N. Yang, J. Huang, R. J. Buckanovich, S. Liang, A. Barchetti, C. Vezzani, A. O'Brien-Jenkins, J. Wang, M. R. Ward, et al. Transcriptional Coactivator Drosophila Eyes Absent Homologue 2 Is Up-Regulated in Epithelial Ovarian Cancer and Promotes Tumor Growth Cancer Res., February 1, 2005; 65(3): 925 - 932. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. E. Burczynski, N. C. Twine, G. Dukart, B. Marshall, M. Hidalgo, W. M. Stadler, T. Logan, J. Dutcher, G. Hudes, W. L. Trepicchio, et al. Transcriptional Profiles in Peripheral Blood Mononuclear Cells Prognostic of Clinical Outcomes in Patients with Advanced Renal Cell Carcinoma Clin. Cancer Res., February 1, 2005; 11(3): 1181 - 1189. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J.-M. Kim, H.-Y. Sohn, S. Y. Yoon, J.-H. Oh, J. O. Yang, J. H. Kim, K. S. Song, S.-M. Rho, H. S. Yoo, Y. S. Kim, et al. Identification of Gastric Cancer-Related Genes Using a cDNA Microarray Containing Novel Expressed Sequence Tags Expressed in Gastric Cancer Cells Clin. Cancer Res., January 15, 2005; 11(2): 473 - 482. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. S. Lam, A. S. Belldegrun, and R. A. Figlin Tissue Array-Based Predictions of Pathobiology, Prognosis, and Response to Treatment for Renal Cell Carcinoma Therapy Clin. Cancer Res., September 15, 2004; 10(18): 6304S - 6309S. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. M. Mazzanti, A. Tandle, D. Lorang, N. Costouros, D. Roberts, G. Bevilacqua, and S. K. Libutti Early Genetic Mechanisms Underlying the Inhibitory Effects of Endostatin and Fumagillin on Human Endothelial Cells Genome Res., August 1, 2004; 14(8): 1585 - 1593. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. Hibbs, K. M. Skubitz, S. E. Pambuccian, R. C. Casey, K. M. Burleson, T. R. Oegema Jr, J. J. Thiele, S. M. Grindle, R. L. Bliss, and A. P.N. Skubitz Differential Gene Expression in Ovarian Carcinoma: Identification of Potential Biomarkers Am. J. Pathol., August 1, 2004; 165(2): 397 - 414. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. M. Mintzer, M. Warhol, A.-M. Martin, and G. Greene Cancer of Unknown Primary: Changing Approaches. A Multidisciplinary Case Presentation from the Joan Karnell Cancer Center of Pennsylvania Hospital Oncologist, June 1, 2004; 9(3): 330 - 338. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. I. Su, T. Wiltshire, S. Batalov, H. Lapp, K. A. Ching, D. Block, J. Zhang, R. Soden, M. Hayakawa, G. Kreiman, et al. A gene atlas of the mouse and human protein-encoding transcriptomes PNAS, April 20, 2004; 101(16): 6062 - 6067. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. I. Dorrell, E. Aguilar, C. Weber, and M. Friedlander Global Gene Expression Analysis of the Developing Postnatal Mouse Retina Invest. Ophthalmol. Vis. Sci., March 1, 2004; 45(3): 1009 - 1019. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
G. Bloom, I. V. Yang, D. Boulware, K. Y. Kwong, D. Coppola, S. Eschrich, J. Quackenbush, and T. J. Yeatman Multi-Platform, Multi-Site, Microarray-Based Human Tumor Classification Am. J. Pathol., January 1, 2004; 164(1): 9 - 16. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. A. Shedden, J. M. G. Taylor, T. J. Giordano, R. Kuick, D. E. Misek, G. Rennert, D. R. Schwartz, S. B. Gruber, C. Logsdon, D. Simeone, et al. Accurate Molecular Classification of Human Cancers Based on Gene Expression Using a Simple Classifier with a Pathological Tree-Based Framework Am. J. Pathol., November 1, 2003; 163(5): 1985 - 1995. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. L. Tanyi, Y. Hasegawa, R. Lapushin, A. J. Morris, J. K. Wolf, A. Berchuck, K. Lu, D. I. Smith, K. Kalli, L. C. Hartmann, et al. Role of Decreased Levels of Lipid Phosphate Phosphatase-1 in Accumulation of Lysophosphatidic Acid in Ovarian Cancer Clin. Cancer Res., September 1, 2003; 9(10): 3534 - 3545. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. Pusztai, M. Ayers, J. Stec, E. Clark, K. Hess, D. Stivers, A. Damokosh, N. Sneige, T. A. Buchholz, F. J. Esteva, et al. Gene Expression Profiles Obtained from Fine-Needle Aspirations of Breast Cancer Reliably Identify Routine Prognostic Markers and Reveal Large-Scale Molecular Differences between Estrogen-negative and Estrogen-positive Tumors Clin. Cancer Res., July 1, 2003; 9(7): 2406 - 2415. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Boussioutas, H. Li, J. Liu, P. Waring, S. Lade, A. J. Holloway, D. Taupin, K. Gorringe, I. Haviv, P. V. Desmond, et al. Distinctive Patterns of Gene Expression in Premalignant Gastric Mucosa and Gastric Cancer Cancer Res., May 15, 2003; 63(10): 2569 - 2577. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. L. Neuhouser, R. E. Patterson, M. D. Thornquist, G. S. Omenn, I. B. King, and G. E. Goodman Fruits and Vegetables Are Associated with Lower Lung Cancer Risk Only in the Placebo Arm of the {beta}-Carotene and Retinol Efficacy Trial (CARET) Cancer Epidemiol. Biomarkers Prev., April 1, 2003; 12(4): 350 - 358. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. A. Giuliano High-Content Profiling of Drug-Drug Interactions: Cellular Targets Involved in the Modulation of Microtubule Drug Action by the Antifungal Ketoconazole J Biomol Screen, April 1, 2003; 8(2): 125 - 135. [Abstract] [PDF]
|
|
|
|
|
|
J. B. Welsh, L. M. Sapinoso, S. G. Kern, D. A. Brown, T. Liu, A. R. Bauskin, R. L. Ward, N. J. Hawkins, D. I. Quinn, P. J. Russell, et al. Large-scale delineation of secreted protein biomarkers overexpressed in cancer tissue and serum PNAS, March 18, 2003; 100(6): 3410 - 3415. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. G. Covell, A. Wallqvist, A. A. Rabow, and N. Thanki Molecular Classification of Cancer: Unsupervised Self-Organizing Map Analysis of Gene Expression Microarray Data Mol. Cancer Ther., March 1, 2003; 2(3): 317 - 332. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. J. Giordano, D. G. Thomas, R. Kuick, M. Lizyness, D. E. Misek, A. L. Smith, D. Sanders, R. T. Aljundi, P. G. Gauger, N. W. Thompson, et al. Distinct Transcriptional Profiles of Adrenocortical Tumors Uncovered by DNA Microarray Analysis Am. J. Pathol., February 1, 2003; 162(2): 521 - 531. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. J. Yeatman The Future of Cancer Management: Translating the Genome, Transcriptome, and Proteome Ann. Surg. Oncol., January 1, 2003; 10(1): 7 - 14. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. L. Dennis, J. K. Vass, E. C. Wit, W. N. Keith, and K. A. Oien Identification from Public Data of Molecular Markers of Adenocarcinoma Characteristic of the Site of Origin Cancer Res., November 1, 2002; 62(21): 5999 - 6005. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. Ginestier, E. Charafe-Jauffret, F. Bertucci, F. Eisinger, J. Geneix, D. Bechlian, N. Conte, J. Adelaide, Y. Toiron, C. Nguyen, et al. Distinct and Complementary Information Provided by Use of Tissue and DNA Microarrays in the Study of Breast Tumor Markers Am. J. Pathol., October 1, 2002; 161(4): 1223 - 1233. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H. F. Frierson Jr, A. K. El-Naggar, J. B. Welsh, L. M. Sapinoso, A. I. Su, J. Cheng, T. Saku, C. A. Moskaluk, and G. M. Hampton Large Scale Molecular Analysis Identifies Genes with Altered Expression in Salivary Adenoid Cystic Carcinoma Am. J. Pathol., October 1, 2002; 161(4): 1315 - 1323. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P. F. Macgregor and J. A. Squire Application of Microarrays to the Analysis of Gene Expression in Cancer Clin. Chem., August 1, 2002; 48(8): 1170 - 1177. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. Karan, D. L. Kelly, A. Rizzino, M.-F. Lin, and S. K. Batra Expression profile of differentially-regulated genes during progression of androgen-independent growth in human prostate cancer cells Carcinogenesis, June 1, 2002; 23(6): 967 - 976. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Ramaswamy and T. R. Golub DNA Microarrays in Clinical Oncology J. Clin. Oncol., April 1, 2002; 20(7): 1932 - 1941. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Ramaswamy, P. Tamayo, R. Rifkin, S. Mukherjee, C.-H. Yeang, M. Angelo, C. Ladd, M. Reich, E. Latulippe, J. P. Mesirov, et al. Multiclass cancer diagnosis using tumor gene expression signatures PNAS, December 6, 2001; (2001) 211566398. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Ramaswamy, P. Tamayo, R. Rifkin, S. Mukherjee, C.-H. Yeang, M. Angelo, C. Ladd, M. Reich, E. Latulippe, J. P. Mesirov, et al. Multiclass cancer diagnosis using tumor gene expression signatures PNAS, December 18, 2001; 98(26): 15149 - 15154. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Cancer Research | Clinical Cancer Research |
| Cancer Epidemiology Biomarkers & Prevention | Molecular Cancer Therapeutics |
| Molecular Cancer Research | Cancer Prevention Research |
| Cancer Prevention Journals Portal | Cancer Reviews Online |
| Annual Meeting Education Book | Meeting Abstracts Online |
