Differential expression of VEGFR2 protein in HER2 positive primary human breast cancer: potential relevance to anti-angiogenic therapies.

BACKGROUND: Clinically relevant predictive biomarkers to tailor anti-angiogenic therapies to breast cancer (BRC) patient subpopulations are an unmet need. METHODS: We analyzed tumor vascular density and VEGFR2 protein expression in various subsets of primary human BRCs (186 females; Mean age: 59 years; range 33-88 years), using a tissue microarray. Discrete VEGFR2+ and CD34+ tumor vessels were manually scored in invasive ductal, lobular, mixed ductal-lobular and colloid (N = 139, 22, 18, 7) BRC cores. RESULTS: The observed CD34+ and VEGFR2+ tumor vascular counts in individual cases were heterogeneous. Mean CD34+ and VEGFR2+ tumor vessel counts were 11 and 3.4 per tumor TMA core respectively. Eighty-nine of 186 (48%) cases had >10 CD34+ tumor vessels, while 97/186 (52%) had fewer CD34+ vessels in each TMA core. Of 169 analyzable cores in the VEGFR2 stained TMA, 90 (53%) showed 1-5 VEGFR2+ tumor vessels/TMA core, while 42/169 (25%) cores had no detectable VEGFR2+ tumor vessels. Thirteen of 169 (8%) cases also showed tumor cell (cytoplasmic/membrane) expression of VEGFR2. Triple-negative breast cancers (TNBCs) appeared to be less vascular (Mean VD = 9.8, range 0-34) than other breast cancer subtypes. Overall, VEGFR2+ tumor vessel counts were significantly higher in HER2+ as compared to HR+ (p = 0.04) and TNBC (p = 0.02) tissues. Compared to HER2- cases, HER2+ breast cancers had higher VEGFR2+ tumor vessel counts (p = 0.007). CONCLUSION: Characterization of pathologic angiogenesis in HER2+ breast cancer provides scientific rationale for future investigation of clinical activity of agents targeting the VEGF/VEGFR2 axis in this clinically aggressive breast cancer subtype.

Expression profiling of human renal carcinomas with functional taxonomic analysis.

BACKGROUND: Molecular characterization has contributed to the understanding of the inception, progression, treatment and prognosis of cancer. Nucleic acid array-based technologies extend molecular characterization of tumors to thousands of gene products. To effectively discriminate between tumor sub-types, reliable laboratory techniques and analytic methods are required. RESULTS: We derived mRNA expression profiles from 21 human tissue samples (eight normal kidneys and 13 kidney tumors) and two pooled samples using the Affymetrix GeneChip platform. A panel of ten clustering algorithms combined with four data pre-processing methods identified a consensus cluster dendrogram in 18 of 40 analyses and of these 16 used a logarithmic transformation. Within the consensus dendrogram the expression profiles of the samples grouped according to tissue type; clear cell and chromophobe carcinomas displayed distinctly different gene expression patterns. By using a rigorous statistical selection based method we identified 355 genes that showed significant (p < 0.001) gene expression changes in clear cell renal carcinomas compared to normal kidney. These genes were classified with a tool to conceptualize expression patterns called "Functional Taxonomy". Each tumor type had a distinct "signature," with a high number of genes in the categories of Metabolism, Signal Transduction, and Cellular and Matrix Organization and Adhesion. CONCLUSIONS: Affymetrix GeneChip profiling differentiated clear cell and chromophobe carcinomas from one another and from normal kidney cortex. Clustering methods that used logarithmic transformation of data sets produced dendrograms consistent with the sample biology. Functional taxonomy provided a practical approach to the interpretation of gene expression data.

The influence of tumor size and environment on gene expression in commonly used human tumor lines.

BACKGROUND: The expression profiles of solid tumor models in rodents have been only minimally studied despite their extensive use to develop anticancer agents. We have applied RNA expression profiling using Affymetrix U95A GeneChips to address fundamental biological questions about human tumor lines. METHODS: To determine whether gene expression changed significantly as a tumor increased in size, we analyzed samples from two human colon carcinoma lines (Colo205 and HCT-116) at three different sizes (200 mg, 500 mg and 1000 mg). To investigate whether gene expression was influenced by the strain of mouse, tumor samples isolated from C.B-17 SCID and Nu/Nu mice were also compared. Finally, the gene expression differences between tissue culture and in vivo samples were investigated by comparing profiles from lines grown in both environments. RESULTS: Multidimensional scaling and analysis of variance demonstrated that the tumor lines were dramatically different from each other and that gene expression remained constant as the tumors increased in size. Statistical analysis revealed that 63 genes were differentially expressed due to the strain of mouse the tumor was grown in but the function of the encoded proteins did not link to any distinct biological pathways. Hierarchical clustering of tissue culture and xenograft samples demonstrated that for each individual tumor line, the in vivo and in vitro profiles were more similar to each other than any other profile. We identified 36 genes with a pattern of high expression in xenograft samples that encoded proteins involved in extracellular matrix, cell surface receptors and transcription factors. An additional 17 genes were identified with a pattern of high expression in tissue culture samples and encoded proteins involved in cell division, cell cycle and RNA production. CONCLUSIONS: The environment a tumor line is grown in can have a significant effect on gene expression but tumor size has little or no effect for subcutaneously grown solid tumors. Furthermore, an individual tumor line has an RNA expression pattern that clearly defines it from other lines even when grown in different environments. This could be used as a quality control tool for preclinical oncology studies.

Investigating fixative-induced changes in RNA quality and utility by microarray analysis.

Described herein is a detailed analysis of the impact of three fixatives (10% neutral buffered formalin, modified methacarn and 70% ethanol) on RNA quality and utility using microarray analysis compared to OCT-embedded and flash frozen tissue. From rat livers fixed and stored in paraffin blocks for 1 month or 1 year, RNA was isolated and applied to rat whole genome microarrays. At both time points, RNA isolated from OCT-embedded tissue lost up to 5% of the information contained in snap frozen control liver. Of the fixatives used, modified methacarn was associated with the smallest loss of RNA information content (approximately 10%), while liver fixed in 70% ethanol and 10% neutral buffered formalin lost roughly 25% and 80%, respectively. We conclude that when optimum morphology is required for techniques such as laser microdissection, modified methacarn is the fixative least harmful to nucleic acids of the three tested in this study. In contrast, using traditional isolation techniques, RNA derived from tissue fixed in 10% NBF will not give reliable results on microarray studies, and should be reserved for techniques less affected by the fragmentation and modification of the template RNA, such as quantitative RT-PCR.

Limited influence of UGT1A1*28 and no effect of UGT2B7*2 polymorphisms on UGT1A1 or UGT2B7 activities and protein expression in human liver microsomes.

AIMS: UGT1A1 and UGT2B7 are enzymes that commonly contribute to drug glucuronidation. Since genetic factors have been suggested to contribute to variability in activities and expression levels of these enzymes, a quantitative assessment of the influence of the major genotypes (UGT1A1*28 or UGT2B7*2) on enzyme activities was conducted. METHODS: Using a bank of microsomal samples from 59 human livers, the effect of UGT1A1*28 or UGT2B7*2 polymorphisms were investigated on rates of estradiol 3-glucuronidation (a marker of UGT1A1 enzyme activity) or zidovudine glucuronidation (a marker of UGT2B7 enzyme activity) and levels of immunoreactive protein for each enzyme. Glucuronidation rates for both enzymes were measured at K(m)/S(50) and 10 times K(m)/S(50) concentrations. RESULTS: UGT1A1 and UGT2B7 enzyme activities varied up to 16-fold and sixfold, respectively. Rates at K(m)/S(50) concentration closely correlated with rates at 10 times K(m)/S(50) concentration for both enzymes (but not at 1/10th K(m) for UGT2B7). Enzyme activities correlated with relative levels of immunoreactive protein for UGT1A1 and UGT2B7. Furthermore, rates of zidovudine glucuronidation correlated well with rates of glucuronidation of the UGT2B7 substrate gemcabene, but did not correlate with UGT1A1 enzyme activities. For the UGT1A1*28 polymorphism, consistent with levels of UGT1A1 immunoreactive protein, mean UGT1A1 activity was 2.5- and 3.2-fold lower for TA(6)/TA(7) (P < 0.05) and TA(7)/TA(7) (P < 0.001) genotypes in comparison with the TA(6)/TA(6) genotype. CONCLUSIONS: Relative to the observed 16-fold variability in UGT1A1 activity, these data indicate only a partial (approximately 40%) contribution of the UGT1A1*28 polymorphism to variability of interindividual differences in UGT1A1 enzyme activity. For the UGT2B7*2 polymorphism, genotype had no influence on immunoreactive UGT2B7 protein or the rate of 3'-azido-3'-deoxythymidine glucuronidation.

Evaluating methods for classifying expression data.

An attractive application of expression technologies is to predict drug efficacy or safety using expression data of biomarkers. To evaluate the performance of various classification methods for building predictive models, we applied these methods on six expression datasets. These datasets were from studies using microarray technologies and had either two or more classes. From each of the original datasets, two subsets were generated to simulate two scenarios in biomarker applications. First, a 50-gene subset was used to simulate a candidate gene approach when it might not be practical to measure a large number of genes/biomarkers. Next, a 2000-gene subset was used to simulate a whole genome approach. We evaluated the relative performance of several classification methods by using leave-one-out cross-validation and bootstrap cross-validation. Although all methods perform well in both subsets for a relative easy dataset with two classes, differences in performance do exist among methods for other datasets. Overall, partial least squares discriminant analysis (PLS-DA) and support vector machines (SVM) outperform all other methods. We suggest a practical approach to take advantage of multiple methods in biomarker applications.