| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Human Cancer Biology |
Authors' Affiliations: 1 Pathology, 2 Comprehensive Breast Cancer, 3 Molecular Oncology, 4 Biostatistics, and 5 Experimental Therapeutics Programs, H. Lee Moffitt Cancer Center and Research Institute, Department of Interdisciplinary Oncology, University of South Florida College of Medicine, Tampa, Florida
Requests for reprints: Carlos A. Muro-Cacho, Department of Orthopedics, Miller School of Medicine, University of Miami, Miami, FL 33101. Phone: 305-325-4475; Fax: 305-325-3928; E-mail: murocacho{at}moffitt.usf.edu.
 |
Abstract |
|---|
 Top Abstract Materials and MethodsResultsDiscussionReferences |
|---|
Experimental Design: Using immunohistochemistry and image analysis, we quantified the levels of phospho-Stat3 (pY-Stat3), phospho-Src (pY-Src), epidermal growth factor receptor, HER2/neu, Ki-67, estrogen receptor, Bcl-2, Bcl-xL, Survivin, and apoptosis in formalin-fixed, paraffin-embedded sections from invasive carcinomas and their paired nonneoplastic parenchyma. The levels of molecular biomarkers in nonneoplastic and tumor tissues were analyzed as continuous variables for statistically significant correlations.
Results: Levels of activated pY-Stat3 and pY-Src measured by immunohistochemistry were significantly higher in invasive carcinoma than in nonneoplastic tissue (P < 0.001). In tumors, elevated levels of pY-Stat3 correlated with those of pY-Src and Survivin. Levels of pY-Stat3 were higher in partial pathologic responders than in complete pathologic responders. In partial pathologic responders, pY-Stat3 levels correlated with Survivin expression.
Conclusions: Our findings suggest important roles for elevated activities of Stat3 and Src, as well as Survivin expression, in malignant progression of breast cancer. Furthermore, elevated Stat3 activity correlates inversely with complete pathologic response. These findings suggest that specific Stat3 or Src inhibitors could offer clinical benefits to patients with breast cancer.
In the present study, we assessed the levels of activated, tyrosine-phosphorylated Stat3 (pY-Stat3), and other proteins of potential interest to oncogenic signaling in 45 patients with stage III breast carcinoma. We also investigated changes in the expression of these molecular biomarkers in relation to pathologic response in a phase II neoadjuvant chemotherapy trial of sequential doxorubicin and docetaxel treatment. Consistent with a role for activated Stat3 in breast cancer progression, elevated Stat3 activity was detected in tumors compared with matched nonneoplastic tissues. Levels of activated Stat3 in patients who had a complete pathologic response were significantly lower than those of patients who had a partial pathologic response, indicating an inverse correlation of constitutive Stat3 activity with chemotherapy response. In patients who had partial pathologic responses, elevated levels of activated Stat3 correlated with the expression of Survivin, a member of the inhibitors of apoptosis protein family (13–16). These findings are consistent with a role of constitutively active Stat3 in mediating breast tumor cell survival and response to therapy, in part by induction of Survivin (see Gritsko et al., in this issue). Furthermore, levels of activated Src kinase (pY-Src) were significantly higher in tumors compared with matched nontumor tissues, consistent with a role of activated Src kinase in breast cancer progression. Together, our findings suggest that breast cancer patients may benefit from therapies targeting Src kinase and Stat3 signaling, either as standalone therapy or in combination with conventional chemotherapy.
|
Materials and Methods |
|---|
|
TopAbstractMaterials and MethodsResultsDiscussionReferences |
|---|
Tissue collection. Our previous studies (18) had shown that to preserve the activated phosphorylation state of signal transduction proteins, tissues have to be snap frozen in liquid nitrogen within 15 minutes from the moment of interruption of blood supply to the specimen. All tissues in this study were snap frozen or fixed in 10% neutral-buffered formalin within 15 minutes to minimize antigen loss and optimize immunohistochemical detection. The presence of normal tissue or tumor was confirmed in mirror image sections of the respective samples by examination of frozen sections immediately following collection. After chemotherapy, tumor was only available from those patients with a partial pathologic response. All data obtained were entered into a web-based database for statistical correlation of clinical, pathologic, and molecular data. Patient data were deidentified and maintained confidential by assigning each case a research specimen number.
Heterogeneity of signal across tissue sections. An important aspect of the initial experimentation process was to determine the degree of staining variability across consecutive tissue sections. Twenty consecutive sections were prepared without discarding intervening tissue and used to perform imunohistochemistry for a signaling protein not related to the project (transforming growth factor receptor type II). Quantitative image analysis revealed minimal variation in expression intensity for the first 12 consecutive sections (data not shown). Expression levels on the next set of 12 sections showed statistically significant differences when compared with the first set. Therefore, for the purposes of our study, a maximum of 11 consecutive sections were used.
Immunohistochemistry and TUNEL procedures. Consecutive 3 µm sections were prepared without discarding intervening tissue (see above). The first section was stained with H&E and the rest of the sections were used for immunohistochemistry and TUNEL assays. For all antigens, except pY-Stat3 (see procedure for pY-Stat3 detection below), the following procedure was used. Formalin-fixed, paraffin-embedded tissue sections were dried at 37°C overnight. Sections were deparaffinized by an initial warming to 60°C, followed by two xylene changes 10 minutes each, two series of 30 dips in absolute alcohol, 30 dips in 95% alcohol, and 20 dips in deionized water. Antigen retrieval or enzyme digestion procedures were done as described by the supplier of each antibody. Slides were placed for 5 minutes in TBS/Tween and processed on a DAKO Autostainer using the Dako LSAB+ peroxidase detection kit (DAKO, Carpinteria, CA). Endogenous peroxidase was blocked with 3% aqueous hydrogen peroxide followed by two 20 dips in deionized water. The anti-EGFR monoclonal antibody clone 111.6 (Signet Pathology Systems, Dedham, MA) was applied at 1:100 for 30 minutes following proteinase K digestion (25 µg/mL in TBS/Tween) for 17 minutes. The rest of the antibodies were applied for 30 minutes after microwave antigen retrieval with 0.1 mol/L citrate buffer (pH 6.0; Emerson 1,100 W microwave, high to boiling, then 20 minutes on power level 5) as follows: Bcl-2 (1:40; DAKO), Bcl-xL (1:50; Santa Cruz Biotechnology, Santa Cruz, CA), pY-Src (1:100; Cell Signaling Technology, Beverly, MA), Ki-67 (1:50; Immunotech, Norcross, GA), c-ErbB-2 (1:40; HER2/neu; Signet Pathology Systems), ER and PR (1:40; BioGenex, San Ramon, CA), and Survivin (1:100; Cell Signaling Technology). The chromogen 3,3'-diaminobenzidine was used for all proteins except for Survivin, which was detected using Nova-Red (Vector Laboratories, Burlingame, CA). Survivin expression was evaluated only in samples obtained before treatment because analysis of this antigen was added at a later time in this study based on microarray analyses (see Gritsko et al., in this issue). Counterstain was done with modified Mayer's hematoxylin. Slides were dehydrated through graded alcohol, cleared with xylene, and mounted with resinous mounting medium. Apoptosis was detected by the TUNEL assay using the Intergen Apopta G Peroxidase In situ Apoptosis detection kit as indicated by the supplier.
pY-Stat3 immunohistochemistry. After deparaffinizing, a two-stage pretreatment procedure was done as follows. First, antigen retrieval was done in a pressure cooker by placing a total of 600 mL deionized water in three containers, one containing the slides in citrate buffer (pH 6.0) and the other two containing only deionized water. The microwave oven was set on high to pressurize for 12 minutes and then at power level 4 for 10 minutes. Slides were then cooled at room temperature for 30 minutes. This was followed by limited enzymatic digestion at 37°C for 5 minutes with 0.025% trypsin in 5 mmol/L Tris-Cl (pH 7.6) with 0.05% calcium chloride. At the end of the digestion, slides were rinsed with deionized water, placed in TBS/Tween for 5 minutes, drained, and framed with an ImmunoEdge pen. Hydrogen peroxide 3% was applied for 10 minutes and 3% bovine serum albumin/PBS for 10 minutes. Sections were incubated with anti-phospho-Stat3 antibody (rabbit polyclonal P-Stat3, Cell Signaling) at 1:400 in a humid chamber at 4°C overnight and returned to the autostainer for detection and substrate development using the Dako LSAB+ detection system and 3,3'-diaminobenzidine as chromogen. Counterstaining was done for 30 seconds with modified Mayer's hematoxylin. Sections were allowed to sit in tap water for 10 to 15 minutes and dehydrated before mounting with resinous mounting medium.
Image analysis and quantification. The Optimas 6.5 (Media Cybernetics, Silver Springs, MD) software was used to quantify protein expression. Regions of interest were identified on the H&E-stained slide and the same areas were marked on the consecutive sections used for each of the biomarkers and TUNEL assay. Digital images of these areas were obtained using identical magnifications (x400) and camera settings with a Leica DM microscope (Leica Microscopes, Bannockburn, IL) with neutral density 6 and 12 filters, coupled to a SPOT Digital Camera System (Diagnostic Instruments, Sterling Heights, MI) and SPOT software set as AutoGain, RGB filter color, no binning, full chip area, and adjustment factor set to 1. Before photography, Koehler epi-illumination was done. Optimal light conditions for each objective were stored as software settings to be replicated in each measurement session. Image acquisition was done after 1 hour of microscope lamp warm up. White balance was done on an area of the slide with no tissue, and background values were subtracted using a negative control slide. Images of the selected areas were stored as TIFF images. A macro was specifically set in the software to automate the process and transfer mathematical calculations to a Microsoft Excel spreadsheet. These were converted to SAS data sets for statistical analysis. Quantitative image analysis was done on all pretreatment samples and in those posttreatment samples for which tumor tissue was still identifiable (partial pathologic responders).
HER2/neu clinical testing. HER2/neu status was also assessed in all tumors at an independent laboratory using a Food and Drug Administration–approved immunohistochemistry procedure. Results of this test are reported negative when intensity scores are 0 or 1+ and positive when the intensity score is 3+. Cases with intermediate intensity score of 2+ are further evaluated by fluorescence in situ hybridization. Using this method, 11 cases (24.4%) were positive, 31 (68.9%) were negative, and 3 (6.7%) remained undetermined. These percentages are consistent with previous reports (27). In the group of tumors that were positive by the Food and Drug Administration–approved test, the average HER2/neu index calculated by computerized image analysis in this study was 50 and the SD 14.5. In the group of tumors that were negative by the Food and Drug Administration–approved test, the average HER2/neu index by computerized image analysis was 35.5 and the SD was 30. The difference was not statistically significant but suggests a similar trend of HER2/neu detection with both methods.
Electrophoretic mobility shift assay. Details of this procedure are described in by Gritsko et al., in this issue. Briefly, a 32P-radiolabeled oligonucleotide probe containing a high-affinity cis-inducible element (hSIE) derived from the c-fos promoter that binds Stat3 proteins was incubated with nuclear extract. Protein-DNA complexes were resolved by nondenaturing PAGE and detected by autoradiography. Quantification of Stat3 DNA-binding activity was determined relative to internal control standards, which were serial dilutions of nuclear extract prepared from the MDA-MB-468 breast cancer cell line [frozen aliquots of the exact same extract were used in all electrophoretic mobility shift assays (EMSA) for consistent comparisons across tissue specimens and experiments] and measured by PhosphorImager (Molecular Dynamics, Sunnyvale, CA). EMSA was done in triplicate, independent analyses for all matched tumor and nontumor specimens in this trial, and Stat3 DNA-binding activities were compared with the levels of phospho-Stat3 as measured by immunohistochemistry of mirror image tissue samples.
Statistical considerations. Before statistical evaluation, pathologic response was classified as either complete pathologic response or partial pathologic response based on the size of residual tumor after treatment (complete pathologic response if 0 cm, partial pathologic response if >0 cm). Biomarker values were analyzed by descriptive statistics, including mean, SD, and range (minimal-maximal), values in all patients as a group and in the complete pathologic response and partial pathologic response groups separately both in nonneoplastic tissue and in tumor tissue. Comparison of pretreatment samples between the complete pathologic response and partial pathologic response groups was examined using the Student's t test if the data followed a normal distribution and the Wilcoxon Mann-Whitney test if the normality assumption was not met. In addition, pretreatment to posttreatment changes were evaluated using the paired t test in the partial pathologic response group. Correlations between pretreatment values were assessed by the Spearman's rank correlation coefficient. No adjustments were made for multiple testing owing to the exploratory nature of this study. All tests were two-sided and declared significant at the 5% level.
|
Results |
|---|
|
TopAbstractMaterials and MethodsResultsDiscussionReferences |
|---|
|
|
|
0.001) and when the analysis was done in the complete pathologic response and partial pathologic response groups separately. In contrast, pretreatment levels of ER in the complete pathologic response group, and Bcl-2 in both complete pathologic response and partial pathologic response groups, were significantly higher in nonneoplastic tissues than in tumors (compare Tables 1 and 2). Similar relationships were found in tissues after chemotherapy treatment, with higher levels of pY-Stat3, pY-Src, and Ki-67 in tumors and higher levels of Bcl-2 in nonneoplastic tissues.
|
|
Posttreatment. In the posttreatment partial pathologic response group, a direct correlation between HER2/neu and Ki-67 and an inverse correlation between EGFR and pY-Src were observed.
Pretreatment versus posttreatment (partial pathologic responders only). Levels of Ki-67 were higher in pretreatment samples than in posttreatment samples (P < 0.02), and levels of Bcl-2 were higher in posttreatment samples than in pretreatment samples (P < 0.02).
Stat3 activation determined by EMSA
Nuclear extract preparations from snap frozen, mirror image samples of the same tissues used for immunohistochemistry were analyzed by EMSA for Stat3 DNA-binding activity. EMSA results show constitutive activation of Stat3 homodimers in the majority of the breast tumors tested (Fig. 3). These results are consistent with our previous findings (18). However, we did not observe a direct correlation between the values for activated Stat3 DNA-binding levels obtained by EMSA and those for immunohistochemistry of pY-Stat3 levels in the same matched patient pairs (data not shown). This may be due, in large part, to the presence of nontumor stromal cells, which contain reduced levels of activated Stat3 (e.g., Fig. 1; Table 1), in the nuclear extract preparations of whole tumor specimens that would affect overall DNA-binding activity. Thus, we consider immunohistochemistry to more accurately reflect levels of activated Stat3 in tumor cells specifically. Furthermore, activated Stat1:Stat3 heterodimers and Stat1 homodimers were also detected by EMSA in some cases (top). It is not clear from this analysis whether the activated Stat1 arises predominantly from tumor or nontumor cells (stromal and/or immune cells) present in the tissue specimens. Importantly, both immunohistochemistry and EMSA showed a similar overall trend of higher levels of activated Stat3 in tumors compared with nonneoplastic tissues (bottom).
|
|
Discussion |
|---|
|
TopAbstractMaterials and MethodsResultsDiscussionReferences |
|---|
An earlier report suggests that nuclear expression of pY-Stat3 is associated with a better prognosis for overall survival in node-negative breast cancer (34). Although this apparent discrepancy with our findings is not well understood at the present time, we note the critical influence of processing time following tissue resection to preserve pY-Stat3 levels. Our experience indicates that prolonged delay over 15 minutes in fixing or freezing tissue specimens results in reduced phosphorylation levels of STATs (18), probably due to tyrosine phosphatase activity. In this context, the use of retrospective archived specimens in the earlier study (34) may in part explain some of the discrepancy with our study because the archived specimens might not have been processed within 15 minutes. For the present study, we used only prospectively collected specimens for which a tissue processing time of <15 minutes was documented. Furthermore, the earlier study (34) correlated pY-Stat3 levels with prognosis, whereas our study focused on response to chemotherapy. The present study is consistent with a previous report that elevated Stat1 activity in breast cancer is a prognostic indicator of longer and relapse-free survival (35), reflecting the known function of Stat1 in inducing growth arrest and apoptosis in response to physiologic signals, such as IFN-
(36). In particular, evidence indicates that Stat1 functions antagonistically to Stat3, which has intrinsic oncogenic potential, whereas Stat1 has opposing tumor suppressor–like properties (37, 38). Our findings reported here show that elevated levels of nuclear pY-Stat3 are associated with invasive breast cancer.
Analysis of biomarker expression in pretreatment samples revealed numerous significant correlations (Tables 2 and 3) of potential relevance to breast oncogenesis. As might be expected, the cell proliferation marker Ki-67 was significantly higher in tumor compared with normal tissues and correlated well with Src activation. Paradoxically, however, levels of Bcl-2 were higher in normal tissues compared with tumors, consistent with other studies (39). In addition, levels of pY-Stat3, HER2/neu, and pY-Src are correlated with each other in the group of complete pathologic responders. These observations are consistent with the known link between Src and Stat3 activation in fibroblasts and breast carcinoma cells (18, 40). Furthermore, Src activation has been implicated in HER2/neu-mediated migratory potential of breast cancer cell lines (17) and in the maintenance of the transformed phenotype of cells overexpressing HER2/neu (41). Other studies indicate activation of Stat3 signaling by the HER2/neu pathway (42, 43). Thus, our present study and those of previous reports suggest cooperation between HER2/neu, Src, and Stat3 signaling pathways in breast oncogenesis although the relevance of this interaction to chemotherapy response is unclear. Our studies did not reveal a good correlation between EGFR and constitutively active pY-Stat3, despite the fact that EGF is a potent physiologic inducer of Stat3 signaling (44). This finding may in part reflect induction of constitutive Stat3 activation in breast carcinoma cells by other tyrosine kinases independent of EGFR, including Src (18, 45) and Janus kinase family tyrosine kinases (18, 46). Further study is required to better understand the potential relevance of all the correlations noted among the various molecular biomarkers, some of which are consistent with expectations, whereas others are surprising in the context of our current knowledge.
A clinically important correlation was observed between pY-Stat3 and Survivin, an antiapoptotic protein of the inhibitors of apoptosis protein family (13–16). The findings revealed a strong correlation between pY-Stat3 and Survivin levels in those patients who were partial pathologic responders to chemotherapy.6 Further support for a link between Stat3 and Survivin comes from studies showing that Survivin is a direct downstream target of Stat3 (see Gritsko et al., in this issue). Together, these findings suggest that elevated levels of Stat3 activity and consequently Survivin expression in breast tumor cells are key factors in promoting the survival of tumor cells and determining their response to chemotherapy. Our conclusion is consistent with reports that inhibition of persistently active Stat3 signaling enhance sensitivity to chemotherapy in B-cell non-Hodgkin's lymphoma and myeloma cells (47, 48).
The observation that elevated Stat3 activity is associated with breast oncogenesis and correlates inversely with complete pathologic response to chemotherapy suggests that the Stat3 signaling pathway is a potential therapeutic target in breast cancer. Furthermore, inhibitors of Src tyrosine kinase activity block constitutively active Stat3 signaling and induce apoptosis in breast cancer cells (18, 49). Our present study provides support for the development of small molecule inhibitors of Stat3 as well as those of Src kinase for cancer therapeutics (3, 12, 50). This conclusion is further strengthened by our finding that inhibition of Stat3 and downstream Survivin expression induces apoptosis in human breast cancer cell lines harboring constitutively active Stat3 (see Gritsko et al., in this issue). Small molecule inhibitors of Stat3 and Src could be used in combination with conventional chemotherapy or with other signal transduction blockers for increased therapeutic benefits in breast cancer patients.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
Note: N. Diaz and S. Minton contributed equally to this work.
6 S. Minton et al., submitted for publication. 
Received 8/27/04; revised 6/30/05; accepted 9/20/05.
|
References |
|---|
|
TopAbstractMaterials and MethodsResultsDiscussionReferences |
|---|
-dependent tumor surveillance system in immunocompetent mice. Proc Natl Acad Sci U S A 1998;95:7556–61.
in a Src and JAK2-dependent manner. J Biol Chem 2002;277:38486–93.This article has been cited by other articles:
|
|
 |
|
  R. S. Finn Targeting Src in breast cancer Ann. Onc., May 16, 2008; (2008) mdn291v1. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Vultur, R. Buettner, C. Kowolik, W. Liang, D. Smith, F. Boschelli, and R. Jove SKI-606 (bosutinib), a novel Src kinase inhibitor, suppresses migration and invasion of human breast cancer cells Mol. Cancer Ther., May 1, 2008; 7(5): 1185 - 1194. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H Kothmaier, F Quehenberger, I Halbwedl, P Morbini, F Demirag, H Zeren, C E Comin, B Murer, P T Cagle, R Attanoos, et al. EGFR and PDGFR differentially promote growth in malignant epithelioid mesothelioma of short and long term survivors Thorax, April 1, 2008; 63(4): 345 - 351. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. R. Seethala, W. E. Gooding, P. N. Handler, B. Collins, Q. Zhang, J. M. Siegfried, and J. R. Grandis Immunohistochemical Analysis of Phosphotyrosine Signal Transducer and Activator of Transcription 3 and Epidermal Growth Factor Receptor Autocrine Signaling Pathways in Head and Neck Cancers and Metastatic Lymph Nodes Clin. Cancer Res., March 1, 2008; 14(5): 1303 - 1309. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Z. Duan, J. Bradner, E. Greenberg, R. Mazitschek, R. Foster, J. Mahoney, and M. V. Seiden 8-Benzyl-4-oxo-8-azabicyclo[3.2.1]oct-2-ene-6,7-dicarboxylic Acid (SD-1008), a Novel Janus Kinase 2 Inhibitor, Increases Chemotherapy Sensitivity in Human Ovarian Cancer Cells Mol. Pharmacol., November 1, 2007; 72(5): 1137 - 1145. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H.-W. Lo, S.-C. Hsu, W. Xia, X. Cao, J.-Y. Shih, Y. Wei, J. L. Abbruzzese, G. N. Hortobagyi, and M.-C. Hung Epidermal Growth Factor Receptor Cooperates with Signal Transducer and Activator of Transcription 3 to Induce Epithelial-Mesenchymal Transition in Cancer Cells via Up-regulation of TWIST Gene Expression Cancer Res., October 1, 2007; 67(19): 9066 - 9076. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. Azare, K. Leslie, H. Al-Ahmadie, W. Gerald, P. H. Weinreb, S. M. Violette, and J. Bromberg Constitutively Activated Stat3 Induces Tumorigenesis and Enhances Cell Motility of Prostate Epithelial Cells through Integrin {beta}6 Mol. Cell. Biol., June 15, 2007; 27(12): 4444 - 4453. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Wang, L. Zhang, X. Han, J. Yang, J. Qian, S. Hong, F. Samaniego, J. Romaguera, and Q. Yi Atiprimod inhibits the growth of mantle cell lymphoma in vitro and in vivo and induces apoptosis via activating the mitochondrial pathways Blood, June 15, 2007; 109(12): 5455 - 5462. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
X. Ling, M. Konopleva, Z. Zeng, V. Ruvolo, L. C. Stephens, W. Schober, T. McQueen, M. Dietrich, T. L. Madden, and M. Andreeff The Novel Triterpenoid C-28 Methyl Ester of 2-Cyano-3, 12-Dioxoolen-1, 9-Dien-28-Oic Acid Inhibits Metastatic Murine Breast Tumor Growth through Inactivation of STAT3 Signaling Cancer Res., May 1, 2007; 67(9): 4210 - 4218. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
Z. Duan, J. E. Bradner, E. Greenberg, R. Levine, R. Foster, J. Mahoney, and M. V. Seiden SD-1029 Inhibits Signal Transducer and Activator of Transcription 3 Nuclear Translocation. Clin. Cancer Res., November 15, 2006; 12(22): 6844 - 6852. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. K. Lee, K. Man, R. T.P. Poon, C. M. Lo, A. P. Yuen, I. O. Ng, K. T. Ng, W. Leonard, and S. T. Fan Signal Transducers and Activators of Transcription 5b Activation Enhances Hepatocellular Carcinoma Aggressiveness through Induction of Epithelial-Mesenchymal Transition. Cancer Res., October 15, 2006; 66(20): 9948 - 9956. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. Liby, N. Voong, C. R. Williams, R. Risingsong, D. B. Royce, T. Honda, G. W. Gribble, M. B. Sporn, and J. J. Letterio The Synthetic Triterpenoid CDDO-Imidazolide Suppresses STAT Phosphorylation and Induces Apoptosis in Myeloma and Lung Cancer Cells. Clin. Cancer Res., July 15, 2006; 12(14): 4288 - 4293. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. E. Cox, J. M. Cox, L. B. White, N. G. Stowell, J. D. Clark, N. Allred, M. Meyers, E. Dupont, B. Furman, and S. Minton Sentinel Node Biopsy Before Neoadjuvant Chemotherapy for Determining Axillary Status and Treatment Prognosis in Locally Advanced Breast Cancer Ann. Surg. Oncol., April 1, 2006; 13(4): 483 - 490. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||
