Interlaboratory performance of a microarray-based gene expression test to determine tissue of origin in poorly differentiated and undifferentiated cancers.

Clinical workup of metastatic malignancies of unknown origin is often arduous and expensive and is reported to be unsuccessful in 30 to 60% of cases. Accurate classification of uncertain primary cancers may improve with microarray-based gene expression testing. We evaluated the analytical performance characteristics of the Pathwork tissue of origin test, which uses expression signals from 1668 probe sets in a gene expression microarray, to quantify the similarity of tumor specimens to 15 known tissues of origin. Sixty archived tissue specimens from poorly and undifferentiated tumors (metastatic and primary) were analyzed at four laboratories representing a wide range of preanalytical conditions (eg, personnel, reagents, instrumentation, and protocols). Cross-laboratory comparisons showed highly reproducible results between laboratories, with correlation coefficients between 0.95 to 0.97 for measurements of similarity scores, and an average 93.8% overall concordance between laboratories in terms of final tissue calls. Bland-Altman plots (mean coefficients of reproducibility of 32.48+/-3.97) and kappa statistics (kappa >0.86) also indicated a high level of agreement between laboratories. We conclude that the Pathwork tissue of origin test is a robust assay that produces consistent results in diverse laboratory conditions reflecting the preanalytical variations found in the everyday clinical practice of molecular diagnostics laboratories.

Application of a correlation correction factor in a microarray cross-platform reproducibility study.

BACKGROUND: Recent research examining cross-platform correlation of gene expression intensities has yielded mixed results. In this study, we demonstrate use of a correction factor for estimating cross-platform correlations. RESULTS: In this paper, three technical replicate microarrays were hybridized to each of three platforms. The three platforms were then analyzed to assess both intra- and cross-platform reproducibility. We present various methods for examining intra-platform reproducibility. We also examine cross-platform reproducibility using Pearson's correlation. Additionally, we previously developed a correction factor for Pearson's correlation which is applicable when X and Y are measured with error. Herein we demonstrate that correcting for measurement error by estimating the "disattenuated" correlation substantially improves cross-platform correlations. CONCLUSION: When estimating cross-platform correlation, it is essential to thoroughly evaluate intra-platform reproducibility as a first step. In addition, since measurement error is present in microarray gene expression data, methods to correct for attenuation are useful in decreasing the bias in cross-platform correlation estimates.

Data mining and clinical data repositories: Insights from a 667,000 patient data set.

Clinical repositories containing large amounts of biological, clinical, and administrative data are increasingly becoming available as health care systems integrate patient information for research and utilization objectives. To investigate the potential value of searching these databases for novel insights, we applied a new data mining approach, HealthMiner, to a large cohort of 667,000 inpatient and outpatient digital records from an academic medical system. HealthMiner approaches knowledge discovery using three unsupervised methods: CliniMiner, Predictive Analysis, and Pattern Discovery. The initial results from this study suggest that these approaches have the potential to expand research capabilities through identification of potentially novel clinical disease associations.

BCRABL transcript detection by quantitative real-time PCR : are correlated results possible from homebrew assays?

BACKGROUND: Quantitative real-time PCR has become the predominant molecular technique to monitor BCRABL levels in response to treatment in Ph(+) leukemia patients. However, without some form of standardized methodology between laboratories, the correlation of results is difficult. METHODS: Using TaqMan-based assays, parallel quantitative real-time PCR analysis was performed on 70 clinical specimens at Vanderbilt University Medical Center and Virginia Commonwealth University. While the same positive control cell line (K562) and quality control gene (BCR) were used, the RNA isolation technique, cDNA synthesis, BCR control cell line, and PCR primer and probe sequences were different. RESULTS: The detection of BCRABL-positive results spanned a dynamic range from 10(0) to 10(5)/100,000 cells. Forty-three samples were negative at both facilities. A Spearman rank correlation analysis was performed for the 22 BCRABL-positive paired results. The correlation coefficient, r(s), was 0.9435 (p < 0.00001), suggesting a strong correlation of the results. One discordant result was obtained for consecutive samples from one patient with a low BCRABL copy number as a result of a minimal RNA yield at one laboratory. CONCLUSIONS: These results suggest that quantitative real-time PCR assays for BCRABL detection can be comparable between laboratories despite significant differences in methodologies if the same positive control cell line and quality control gene are used. It is imperative that some level of assay standardization be adopted between laboratories, not only for patients who are monitored at different facilities, but also for larger investigative studies in which hematologic, cytogenetic and molecular responses are to be compared.

Hepatic artery thrombosis after liver transplantation and genetic factors: prothrombin G20210A polymorphism.

Hepatic artery thrombosis (HAT) after liver transplantation is associated with a high incidence of graft failure. The incidence ranges between 2% and 25%, with an overall incidence of approximately 7%. Different risk factors have been associated, but the participation of genetic factors in the cause of HAT is less well studied. A single-base change (G to A) at position 20210 in the 3' untranslated region of the prothrombin gene is associated with increased plasma levels of prothrombin and might therefore increase the risk for thrombosis. We reviewed our HAT experience in 11 years at Medical College of Virginia hospitals of 491 patients undergoing 533 liver transplantations. There were 14 liver grafts with documented HAT (2.62%) in 13 patients. Prothrombin G20210A polymorphism was found in the DNA obtained from 2 of 14 liver allograft tissues (14.2%) but not in the DNA from leukocytes obtained from the peripheral blood of recipients with HAT.

Real-time quantitative analysis of telomerase activity in breast tumor specimens using a highly specific and sensitive fluorescent-based assay.

Telomerase activity has been associated with almost 90% of malignant human cancers from a variety of tissue sources, making it one of the most prominent molecular cancer markers known to date. As such, telomerase has become a very attractive diagnostic and therapeutic target. The advent of the telomeric repeat amplification protocol (TRAP) has allowed for the semiquantitative detection of telomerase from limiting sample amounts. Both the standard TRAP assay and a real-time assay using Amplifluor technology with primers designed specifically for telomerase activity amplification were used to quantitatively assess telomerase activity in primary tumors and tumor-derived cell lines. We have adapted the recently developed TRAPeze XL telomerase detection kit (Intergen, Gaithersburg, MD) for use with real-time polymerase chain reaction for more accurate quantification and high-throughput capabilities. In doing so, the reliability, assay time, and accuracy of quantitation have all been dramatically improved. A comparison of the quantitative analysis for the standard TRAP assay versus the real-time assay using 19 breast tumors revealed telomerase quantitation and standardization using the real-time assay was superior to the standard assay. Our data suggest that this assay will be useful for clinical and research studies involving detection of telomerase activity as it relates to cancer diagnosis.

Clinical verification of the performance of the pathwork tissue of origin test: utility and limitations.

Gene expression-based assays have been introduced into the clinical arena to assist in the diagnosis of poorly differentiated or undifferentiated tumors. The US Food and Drug Administration has cleared the microarray-based Pathwork Tissue of Origin (TOO) Test (Pathwork Diagnostics, Sunnyvale, CA) for the molecular characterization of such challenging specimens. We aimed at verifying the analytic and clinical performance of this test on 43 poorly differentiated and undifferentiated tumor samples, including 6 off-panel cases and 7 cancers of unknown primary (CUP). Our results showed 97% (95% confidence interval, 80.4%-99.8%) agreement between the Pathwork TOO Test result and the complete diagnosis, which included clinical correlations and immunohistochemical staining, after the original diagnosis. We concluded that for off-panel and CUP samples, the tissue type and the cell type may be confounded by the Pathwork TOO Test and that careful clinicopathologic assessment is needed when interpreting results from this helpful ancillary tool for pathologists.

Preservation of fine-needle aspiration specimens for future use in RNA-based molecular testing.

BACKGROUND: The application of ancillary molecular testing is becoming more important for the diagnosis and classification of disease. The use of fine-needle aspiration (FNA) biopsy as the means of sampling tumors in conjunction with molecular testing could be a powerful combination. FNA is minimally invasive, cost effective, and usually demonstrates accuracy comparable to diagnoses based on excisional biopsies. Quality control (QC) and test validation requirements for development of molecular tests impose a need for access to pre-existing clinical samples. Tissue banking of excisional biopsy specimens is frequently performed at large research institutions, but few have developed protocols for preservation of cytologic specimens. This study aimed to evaluate cryopreservation of FNA specimens as a method of maintaining cellular morphology and ribonucleic acid (RNA) integrity in banked tissues. METHODS: FNA specimens were obtained from fresh tumor resections, processed by using a cryopreservation protocol, and stored for up to 27 weeks. Upon retrieval, samples were made into slides for morphological evaluation, and RNA was extracted and assessed for integrity by using the Agilent Bioanalyzer (Agilent Technologies, Santa Clara, Calif). RESULTS: Cryopreserved specimens showed good cell morphology and, in many cases, yielded intact RNA. Cases showing moderate or severe RNA degradation could generally be associated with prolonged specimen handling or sampling of necrotic areas. CONCLUSIONS: FNA specimens can be stored in a manner that maintains cellular morphology and RNA integrity necessary for studies of gene expression. In addition to addressing quality control (QC) and test validation needs, cytology banks will be an invaluable resource for future molecular morphologic and diagnostic research studies.

Human Oncoprotein MDM2 Up-regulates Expression of NF-κB2 Precursor p100 Conferring a Survival Advantage to Lung Cells.

The current model predicts that MDM2 is primarily overexpressed in cancers with wild-type (WT) p53 and contributes to oncogenesis by degrading p53. Following a correlated expression of MDM2 and NF-κB2 transcripts in human lung tumors, we have identified a novel transactivation function of MDM2. Here, we report that in human lung tumors, overexpression of MDM2 was found in approximately 30% of cases irrespective of their p53 status, and expression of MDM2 and NF-κB2 transcripts showed a highly significant statistical correlation in tumors with WT p53. We investigated the significance of this correlated expression in terms of mechanism and biological function. Increase in MDM2 expression from its own promoter in transgenic mice remarkably enhanced expression of NF-κB2 compared with its non-transgenic littermates. Knockdown or elimination of endogenous MDM2 expression in cultured non-transformed or lung tumor cells drastically reduced expression of NF-κB2 transcripts, suggesting a normal physiological role of MDM2 in regulating NF-κB2 transcription. MDM2 could up-regulate expression of NF-κB2 transcripts when its p53-interaction domain was blocked with Nutlin-3, indicating that the MDM2-p53 interaction is dispensable for up-regulation of NF-κB2 expression. Consistently, analysis of functional domains of MDM2 indicated that although the p53-interaction domain of MDM2 contributes to the up-regulation of the NFκB2 promoter, MDM2 does not require direct interactions with p53 for this function. Accordingly, MDM2 overexpression in non-transformed or lung cancer cells devoid of p53 also generated a significant increase in the expression of NF-κB2 transcript and its targets CXCL-1 and CXCL-10, whereas elimination of MDM2 expression had the opposite effects. MDM2-mediated increase in p100/NF-κB2 expression reduced cell death mediated by paclitaxel. Furthermore, knockdown of NF-κB2 expression retarded cell proliferation. Based on these data, we propose that MDM2-mediated NF-κB2 up-regulation is a combined effect of p53-dependent and independent mechanisms and that it confers a survival advantage to lung cancer cells.

Genes involved in radiation therapy response in head and neck cancers.

OBJECTIVES: This is a pilot study designed to identify gene expression profiles able to stratify head and neck squamous cell carcinoma (HNSCC) tumors that may or may not respond to chemoradiation or radiation therapy. STUDY DESIGN: We prospectively evaluated 14 HNSCC specimens, arising from patients undergoing chemoradiotherapy or radiotherapy alone with curative intent. A complete response was assessed by clinical evaluation with no evidence of gross tumor after a 2-year follow-up period. METHODS: Residual biopsy samples from eight complete responders (CR) and six nonresponders (NR) were evaluated by genome-wide gene expression profiling using HG-U133A 2.0 arrays. Univariate parametric t-tests with proportion of false discoveries controlled by multivariate permutation tests were used to identify genes with significantly different gene expression levels between CR and NR cases. Six different prediction algorithms were used to build gene predictor lists. Three representative genes showing 100% crossvalidation support after leave-one-out crossvalidation (LOOCV) were further validated using real-time QRT-PCR. RESULTS: We identified 167 significant probe sets that discriminate between the two classes, which were used to build gene predictor lists. Thus, 142 probe sets showed an accuracy of prediction ranging from 93% to 100% across all six prediction algorithms. The genes represented by these 142 probe sets were further classified into different functional networks that included cellular development, cellular movement, and cancer. CONCLUSIONS: The results presented herein offer encouraging preliminary data that may provide a basis for a more precise prognosis of HNSCC, as well as a molecular-based therapy decision for the management of these cancers.

Assessing the impact of tissue devitalization time on genome-wide gene expression analysis in ovarian tumor samples.

The utilization of genome-wide gene expression microarray technology in tumor stratification has proven a powerful tool to identify gene expression signatures associated with cancer prognosis and is currently under evaluation in clinical laboratories. Standardized protocols, including tumor tissue postoperatively handling guidelines are yet to be defined. We aimed at assessing a systematic effect of devitalization in ovarian tumors' gene expression profiling, using high-density oligonucleotide microarrays, under a standardized protocol following strict quality control criteria. Residual tissue from the surgical pathology specimen was divided into 5 samples. Half of each was immediately snap frozen in liquid nitrogen. The remaining halves were kept at room temperature for 0, 15, 30, 60, and 120 minutes, at which time the tissue was snap frozen in liquid nitrogen, and stored at -80 degrees C until RNA extraction. The entire process from RNA extraction through feature intensity distribution was rigorously monitored for quality. Identification of altered gene expression among each pair of snap frozen and devitalized samples per ovarian tumor specimen was assessed by using the Significance score (S-score) method. We identified only 4 probe sets that seemed to correlate with devitalization time in one of the ovarian tumor specimens, suggesting that they are not likely to have an impact on gene expression profiling tumor stratification. Our study suggests that with proper sample handling and rigorous quality control procedures for RNA extraction and microarray analysis, tumor classification based on global gene expression data will not be adversely affected if devitalization times are kept within a 120-minute window.

Human immunodeficiency virus genotype and hypertriglyceridemia.

Many HIV patients develop a progressive syndrome of abnormal body fat distribution accompanied by hypertriglyceridemia. Antiretroviral agents are thought to be etiologic in the syndrome, often termed "highly active antiretroviral therapy (HAART)-associated lipodystrophy." In the course of clinical HIV genotype testing, we observed that our HIV patients with hypertriglyceridemia had viral genotypes that were more highly mutated than those of our therapy-matched control patients. Hypertriglyceridemia was statistically associated with predicted resistance for three nucleoside reverse transcriptase inhibitors: zidovudine, abacavir, and stavudine. Statistical analysis of 51 patients in retrospect revealed a strong association of mutations at reverse transcriptase codons M41 and T215 with hypertriglyceridemia (chi-square (chi(2)) = 8.375, P=.0038; and chi(2)=7.445, P=.0064, respectively). This was in contrast to silent mutations, which occurred at equivalent rates in retroviral genotypes of patients with and without hypertriglyceridemia. The findings imply that the HIV genotype itself may be a significant etiologic factor in antiretroviral-associated lipodystrophy.

Evaluation of a linear amplification method for small samples used on high-density oligonucleotide microarray analysis.

High-density oligonucleotide microarray analysis has proven to be an excellent approach for gene expression profiling in human cancers. This technique assesses the expression of thousands of genes simultaneously, from at least 5 microg of total RNA per sample per experiment. This total RNA requirement poses a challenge when studying small, unique clinical samples, like biopsies. Recently, a new standardized protocol for small samples was released by Affymetrix, which includes a linear amplification step. To evaluate the impact of such amplification in the gene expression profiling of human ovarian cancer, we compared results obtained from 5 microg and 100 ng of total RNA from the same tumor sample, using the standard Affymetrix protocol and the new linear RNA amplification protocol, respectively. We identified a small bias in gene expression data caused by linear amplification, potentially due to shorter elongation products leading to misclassification of probe sets directed to the middle-5' region of the transcripts. Interestingly, the magnitude of the bias varied when different normalization and expression summary algorithms were used. However, this bias does not affect tumor gene expression profiling. Consequently, linear amplification may be of utility in cases of extremely low RNA recovery from critical and unique samples, such as small biopsies.

Analytical validation of a real-time reverse transcription-polymerase chain reaction quantitation of different transcripts of the Wilms' tumor suppressor gene (WT1).

Transcript variants of the same gene may play distinct functions in the tissue where they are expressed. Absolute quantitation of different transcript variants in malignant and normal tissues can address the specific role of each particular isoform in cancer development and progression. We have recently demonstrated differential expression of the wild-type Wilms' tumor transcript (wtWT1) and a novel truncated WT1 transcript (trWT1) which lacks the first five exons of wtWT1, among human prostate cancer, leukemia, and breast cancer cell lines. Here we report the analytical validation of a real-time RT-PCR assay for the absolute quantitation of these two different WT1 transcripts with specific primers and probes that ensure specificity for each WT1 variant. By cloning each WT1 transcript in a T3 promoter-containing plasmid, we obtained two WT1 transcript-specific in vitro-generated RNA calibrators for absolute quantitation. Serial dilution of each RNA calibrator demonstrated a 5 log linear dynamic range (5 x 10(1) to 5 x 10(6) copies/reaction, R(2)=0.9963 for wtWT1 and R(2)=0.9993 for trWT1). Dilution of the calibrators in total RNA from 1 x 10(3) non-WT1-expressing cells showed a decreased sensitivity without affecting the linear dynamic range. Precision studies for values within the linear dynamic range showed a coefficient of variation of less than 4% for both transcripts. The described method provides a sensitive and reliable technique for quantitating different WT1 mRNA transcripts.

Genome-wide detection of LOH in prostate cancer using human SNP microarray technology.

Loss of heterozygosity (LOH) of chromosomal regions is crucial in tumor progression. In this study we assessed the potential of the Affymetrix GeneChip HuSNP mapping assay for detecting genome-wide LOH in prostate tumors. We analyzed two human prostate cell lines, P69SV40Tag (P69) and its tumorigenic subline, M12, and 11 prostate cancer cases. The M12 cells showed LOH in chromosomes 3p12.1-p22.1, 11q22.1-q24.2, 19p13.12, and 19q13.42. All of the prostate cases with informative single-nucleotide polymorphism (SNP) markers showed LOH in 1p31.2, 10q11.21, 12p13.1, 16q23.1-q23.2, 17p13.3, 17q21.31, and 21q21.2. Additionally, a high percentage of cases showed LOH at 6p25.1-p25.3 (75%), 8p22-p23.2, and 10q22.1 (70%). Several tumor suppressor genes (TSGs) have been mapped in these loci. These results demonstrate that the HuSNP mapping assay can serve as an alternative to comparative genomic hybridization for assessing genome-wide LOH and can identify chromosomal regions harboring candidate TSGs implicated in prostate cancer.

Evaluation of quality-control criteria for microarray gene expression analysis.

BACKGROUND: Development of quality-control criteria to ensure reproducibility of microarray results for potential clinical application is still in its infancy. METHODS: In the present studies we developed quality-control criteria and evaluated their effect in microarray data analysis using total RNA from cell lines, frozen tumors, and a commercially available reference RNA. Quality-control criteria such as A(260)/A(280) ratios, percentage of rRNA, and median size of cDNA and cRNA synthesis products were evaluated for robustness in microarray analysis. Furthermore, precision studies using a reference material were performed on the Affymetrix HG-U133A high-density oligonucleotide microarrays. The same reference RNA sample was examined in 16 different chips run on 2 different days in the four different modules of the Affymetrix fluidics workstation. Fresh and frozen fragmented cRNAs were also compared. An ANOVA model was fit to identify the main sources of variation. RESULTS: Good-quality samples showed >30% rRNA in the electropherograms and cDNA and cRNA synthesis products with median sizes of 2.0 and 3.0 kb, respectively. Precision studies showed that the main source of variation was the day-to-day variability, minimally affecting hybridization exogenous control genes. Altogether, the results showed that the Affymetrix Genechip system is highly reproducible when RNA that meet the quality-control criteria are used (overall P >0.01). CONCLUSIONS: These results confirm the need to establish defined quality-control criteria for sample quality to distinguish between analytical and biological variability.

Comparison of three commercially available assays for HCV RNA using the international unit standard: implications for management of patients with chronic hepatitis C virus infection in clinical practice.

OBJECTIVES: The present study was performed to evaluate the impact of the international unit standard for measuring HCV RNA in the management of patients with chronic hepatitis C virus (HCV) infection. METHODS: The three assays used were Amplicor Monitor PCR, the National Genetics Institute PCR assay, and branched chain DNA. HCV RNA was measured at four time points (baseline, 3 months after the start of therapy, at the end of treatment, and 6 months after discontinuation of therapy) in 106 consecutive patients who received interferon and ribavirin for chronic HCV. RESULTS: The mean age of the patients was 44 yr. Of the patients, 62% were male, 24% were African American, 38% had bridging fibrosis or cirrhosis, and 75% were HCV genotype 1. Of the 424 samples analyzed, 82-89% of values were within 1 log unit and 85-92% were within 2 log units by the various assays. This variability was not dependent upon HCV genotype. HCV RNA was undetectable in 1.4-6.8% of samples when virus was detected by another assay. The mean HCV RNA in these discordant samples was 1.47-6.33 log IU/ml (30-2,100,000 IU/ml). CONCLUSIONS: These data demonstrate that approximately 90% of serum values for HCV RNA were within 1 log unit by the international unit standard regardless of which virological assay was used. However, false positive and false negative results as well as variations in the HCV RNA level of more than 1 to 2 log units can occur with any of the assays, and these results may have an impact upon the management of patients receiving interferon therapy. It is therefore unwise in clinical practice to base important treatment decisions upon a single HCV RNA determination.

Effects of stress management on PNI-based outcomes in persons with HIV disease.

A pretest-posttest, repeated-measures design was used to evaluate the effects of two stress management interventions on a battery of outcomes derived from a psychoneuroimmunological (PNI) framework. The effects of cognitive-behavioral relaxation training groups (CBSM) and social support groups (SSG) were compared with a WAIT-listed control group on the outcomes of psychosocial functioning, quality of life, neuroendocrine mediation, and somatic health. Participants were 148 individuals (119 men, 29 women), diagnosed with HIV disease; 112 (76%) completing the study groups. Using analysis of covariance, the CBSM group was found to have significantly higher postintervention emotional well-being and total quality-of-life scores than did either the SSG or WAIT groups. SSG participants had significantly lower social/family well-being scores immediately postintervention and lower social support scores after 6 months. The findings point to a pressing need for further, well-controlled research with these common intervention modalities.