Search

Supplementary Data from Gene Expression Signature in Urine for Diagnosing and Assessing Aggressiveness of Bladder Urothelial Carcinoma

Purpose: To develop an accurate and noninvasive method for bladder cancer diagnosis and prediction of disease aggressiveness based on the gene expression patterns of urine samples.Experimental Design: Gene expression patterns of 341 urine samples from bladder urothelial cell carcinoma (UCC) patients and 235 controls were analyzed via TaqMan Arrays. In a first phase of the study, three consecutive gene selection steps were done to identify a gene set expression signature to detect and stratify UCC in urine. Subsequently, those genes more informative for UCC diagnosis and prediction of tumor aggressiveness were combined to obtain a classification system of bladder cancer samples. In a second phase, the obtained gene set signature was evaluated in a routine clinical scenario analyzing only voided urine samples.Results: We have identified a 12+2 gene expression signature for UCC diagnosis and prediction of tumor aggressiveness on urine samples. Overall, this gene set panel had 98% sensitivity (SN) and 99% specificity (SP) in discriminating between UCC and control samples and 79% SN and 92% SP in predicting tumor aggressiveness. The translation of the model to the clinically applicable format corroborates that the 12+2 gene set panel described maintains a high accuracy for UCC diagnosis (SN = 89% and SP = 95%) and tumor aggressiveness prediction (SN = 79% and SP = 91%) in voided urine samples.Conclusions: The 12+2 gene expression signature described in urine is able to identify patients suffering from UCC and predict tumor aggressiveness. We show that a panel of molecular markers may improve the schedule for diagnosis and follow-up in UCC patients. Clin Cancer Res; 16(9); 2624–33. ©2010 AACR.

Resumen Objetivo Analizar los perfiles de expresion genica del cancer de prostata (CaP) e identificar los genes diferencialmente expresados. Determinar si la expresion diferencial en tejido se mantiene en muestras de orina-posmasaje prostatico (PMP). Material y metodos Un total de 46 muestras de tejido prostatico (36 de pacientes con CaP y 10 controles) y 158 orinas-PMP (113 de pacientes con CaP y 45 controles) se recogieron entre diciembre de 2003 y mayo de 2007. Se utilizaron microarrays de ADN para identificar los genes diferencialmente expresados entre las muestras de tejido tumorales y las controles. Diez genes fueron seleccionados para la validacion tecnica de los microarrays en las mismas muestras tisulares mediante PCR cuantitativa (RT-qPCR). Se seleccionaron 42 genes para ser validados en muestras de orina-PMP mediante RT-qPCR. Resultados El grafico de escalado multidimensional mostro una clara separacion entre las muestras de tejido tumorales y las controles. Se han identificado 1.047 genes diferencialmente expresados (FDR ≤ 0,1) entre los 2 grupos. La correlacion entre los datos de microarrays y RT-qPCR fue alta (r = 0,928, p Conclusion Existe un perfil de expresion genica diferencial en el CaP. Aunque la extrapolacion de la expresion genica obtenida en tejido prostatico a orina-PMP se debe realizar con precaucion, el analisis del tejido prostatico permite la identificacion de nuevos biomarcadores para diagnostico no invasivo del CaP.

Background Routine histologic analysis of lymph nodes (LN) for detecting disseminated bladder urothelial carcinoma (BUC) lacks sensitivity. Objective To identify and test potential mRNA markers of BUC dissemination in LN that has been missed by histological analysis, and to compare the performance of selected markers with patients' clinical outcome. Design, setting, and participants Microarray data and a literature search were used to identify potential markers expressed in BUC but absent in LN. Five genes were finally selected to be studied by quantitative real-time RT-PCR (qRT-PCR) in 181 and 29 LN from 102 BUC patients and 29 controls, respectively, collected from 2002 to 2004 (median follow-up of 35 mo). Measurements The three most expressed genes plus two additional markers selected from the literature were finally evaluated by qRT-PCR. Gene expression values were statistically compared with histologic results and clinical outcome. Results and limitations A discriminant analysis showed that the combination of FXYD3 and KRT20 genes yielded a 100% sensitivity and specificity differentiating LN with BUC dissemination from controls. Combined, the expression of both genes allowed the identification of urothelial cells in LN in 20.5% of patients with previous histologically negative LN. These patients did not have a significantly worse survival than those who were negative by qRT-PCR. Conclusions Using molecular markers it was possible to improve the sensitivity of LN histologic analysis. However, since 20.5% of patients that reclassified as positive by qRT-PCR did not have a significantly worse survival, we assume lymphadenectomy was important to remove residual disease.

Objective To evaluate the use of a multiprobe fluorescent in situ hybridization (FISH) assay for determining the response of patients with high-risk superficial bladder tumour (HRSBT) to bacillus Calmette-Guerin (BCG) therapy. Methods Bladder washing specimens from 65 HRSBT patients collected before and after BCG therapy were analyzed by FISH. Labelled probes for chromosomes 3, 7, 9, and 17 were used to assess chromosomal abnormalities indicative of malignancy. Results Fifty-five of 65 (85%) patients had a positive pre-BCG FISH result; 29 (45%) patients had a positive and 36 (55%) had a negative post-BCG FISH result. Patients with a positive post-BCG FISH status had a 2.7 times higher risk for tumour recurrence than patients with a negative post-BCG FISH status ( p =0.017; 95% CI: 1.18–6.15). In addition, patients who maintained a positive FISH status before and after BCG therapy had a risk for tumour recurrence 2.96 times higher than patients whose FISH result changed from positive to negative after BCG ( p =0.02; 95% CI: 1.17–7.54). On the other hand, there was no significant difference between the risk for tumour progression in patients with a positive versus a negative post-BCG FISH result ( p =0.49). Conclusions The high percentage of positive pre-BCG FISH results suggests the need for adjuvant therapy in patients with HRSBT after the initial transurethral resection. In addition, patients with a positive post-BCG FISH result were more likely to relapse after therapy. Thus, FISH appears to be useful for the surveillance of patients with HRSBT following BCG therapy.

Objectives To determine the impact of different levels of RNA degradation on gene expression measurements and to ascertain if the gene expression profile obtained from bladder washing (BW) correlates to that obtained from the related bladder tumour (BT). Methods BT and BW RNAs from the same patient were heat shocked to obtain three RNA degradation states, which were compared with intact RNAs from healthy bladders by using complementary DNA (cDNA) microarrays. All samples were amplified by means of a T3N9-based transcription method. In addition, four of the differentially expressed genes in microarrays related to bladder cancer ( KRT20 , IGF2 , GSN , and CCL2 ) were analyzed in 36 tumoural and 14 control BW samples by quantitative reverse transcriptase-polymerase chain reaction (qRT-PCR). Results A high percentage of overlapping differentially expressed genes were detected between BT arrays (85–91%) and between BW arrays (78–93%). Furthermore, the similarity between BW and BT arrays was relatively high and independent of the RNA degradation state (52–60%). Finally, expression differences for the four selected genes were confirmed in the vast majority of extended BW samples tested by qRT-PCR. Conclusions Our results showed that partially degraded RNA samples analyzed by cDNA microarrays yielded gene expression profiles comparable to those obtained using intact RNA. Moreover, BW RNA exhibited gene expression patterns similar to those identified in the BT, indicating that BW is an appropriate sample for studying gene expression profiles of BT using cDNA microarrays. In addition, qRT-PCR results further support the suitability of BW for gene expression profiling and its potential use for routine diagnostics.

UEV proteins are enzymatically inactive variants of the E2 ubiquitin-conjugating enzymes that regulate noncanonical elongation of ubiquitin chains. In Saccharomyces cerevisiae, UEV is part of the RAD6-mediated error-free DNA repair pathway. In mammalian cells, UEV proteins can modulate c-FOS transcription and the G2-M transition of the cell cycle. Here we show that the UEV genes from phylogenetically distant organisms present a remarkable conservation in their exon–intron structure. We also show that the human UEV1 gene is fused with the previously unknown gene Kua. In Caenorhabditis elegans and Drosophila melanogaster, Kua and UEV are in separated loci, and are expressed as independent transcripts and proteins. In humans, Kua and UEV1 are adjacent genes, expressed either as separate transcripts encoding independent Kua and UEV1 proteins, or as a hybrid Kua–UEV transcript, encoding a two-domain protein. Kua proteins represent a novel class of conserved proteins with juxtamembrane histidine-rich motifs. Experiments with epitope-tagged proteins show that UEV1A is a nuclear protein, whereas both Kua and Kua–UEV localize to cytoplasmic structures, indicating that the Kua domain determines the cytoplasmic localization of Kua–UEV. Therefore, the addition of a Kua domain to UEV in the fused Kua–UEV protein confers new biological properties to this regulator of variant polyubiquitination./n/n[Kua cDNAs isolated by RT-PCR and described in this paper have been deposited in the GenBank data library under accession nos. AF1155120 (H. sapiens) and AF152361 (D. melanogaster). Genomic clones containing UEV genes: S. cerevisiae, YGL087c (accession no. Z72609); S. pombe, c338 (accession no. AL023781); P. falciparum, MAL3P2 (accession no. AL034558); A. thaliana, F26F24 (accession no. AC005292); C. elegans, F39B2 (accession no. Z92834); D. melanogaster, AC014908; and H. sapiens, 1185N5 (accession no. AL034423). Accession numbers for Kua cDNAs in GenBank dbEST: M. musculus, AA7853; T. cruzi, AI612534. Other Kua-containing sequences: A. thaliana genomic clones F10M23 (accession no. AL035440), F19K23 (accession no. AC000375), and T20K9 (accession no. AC004786).

One of the first useful products from the human genome will be a set of predicted genes. Besides its intrinsic scientific interest, the accuracy and completeness of this data set is of considerable importance for human health and medicine. Though progress has been made on computational gene identification in terms of both methods and accuracy evaluation measures, most of the sequence sets in which the programs are tested are short genomic sequences, and there is concern that these accuracy measures may not extrapolate well to larger, more challenging data sets. Given the absence of experimentally verified large genomic data sets, we constructed a semiartificial test set comprising a number of short single-gene genomic sequences with randomly generated intergenic regions. This test set, which should still present an easier problem than real human genomic sequence, mimics the ∼200kb long BACs being sequenced. In our experiments with these longer genomic sequences, the accuracy ofGENSCAN, one of the most accurate ab initio gene prediction programs, dropped significantly, although its sensitivity remained high. Conversely, the accuracy of similarity-based programs, such as GENEWISE,PROCRUSTES, andBLASTX, was not affected significantly by the presence of random intergenic sequence, but depended on the strength of the similarity to the protein homolog. As expected, the accuracy dropped if the models were built using more distant homologs, and we were able to quantitatively estimate this decline. However, the specificities of these techniques are still rather good even when the similarity is weak, which is a desirable characteristic for driving expensive follow-up experiments. Our experiments suggest that though gene prediction will improve with every new protein that is discovered and through improvements in the current set of tools, we still have a long way to go before we can decipher the precise exonic structure of every gene in the human genome using purely computational methodology.

We validated the performance of our previously reported test for bladder cancer based on urine gene expression patterns using an independent cohort. We also ascertained whether alternative models could achieve better accuracy.Gene expression patterns of the previously reported 48 genes, including the 12 + 2 genes of the signature, were analyzed by TaqMan® arrays in an independent set of 207 urine samples. We pooled all samples analyzed to date to obtain a larger training set of 404 and used it to search for putative improved new models.Our 12 + 2 gene expression signature had overall 80% sensitivity with 86% specificity (AUC 0.914) to discriminate between bladder cancer and control samples. It had 75% sensitivity and 75% specificity (AUC 0.83) to predict tumor aggressiveness in the validation set of urine samples. After grouping all samples 3 new signatures for diagnosis containing 2, 5 and 10 genes, respectively, and 1 containing 6 genes for prognosis were designed. Diagnostic performance of the 2, 5, 10 and 12-gene signatures was maintained or improved in the enlarged sample set (AUC 0.913, 0.941, 0.949 and 0.944, respectively). Performance to predict aggressiveness was also improved in the 14 and 6-gene signatures (AUC 0.855 and 0.906, respectively).This validation study confirms the accuracy of the 12 + 2 gene signature as a noninvasive tool for assessing bladder cancer. We present improved models with fewer genes that must be validated in future studies.

Objective: To analyze gene expression profiles of prostate cancer (PCa) with the aim of determining the relevant differentially expressed genes and subsequently ascertain whether this differential expression is maintained in post-prostatic massage (PPM) urine samples. Material and methods: Forty-six tissue specimens (36 from PCa patients and 10 controls) and 158 urine PPM-urines (113 from PCa patients and 45 controls) were collected between December 2003 and May 2007. DNA microarrays were used to identify genes differentially expressed between tumour and control samples. Ten genes were technically validated in the same tissue samples by quantitative RT-PCR (RT-qPCR). Forty two selected differentially expressed genes were validated in an independent set of PPM-urines by qRT-PCR. Results: Multidimensional scaling plot according to the expression of all the microarray genes showed a clear distinction between control and tumour samples. A total of 1047 differentially expressed genes (FDR

Table 2: Bladder Cancers Detected Placebo Vitamin E Selenium Combination Years of Follow-up (median, IQR) 7.0 (6.1, 8.0) 7.0 (6.0, 8.0) 7.0 (6.0, 8.0) 7.0 (6.0, 8.0) No. of Bladder Cancers 53 56 60 55 Stage CIS 3 4 4 3 TA 32 26 30 36 T1 10 18 19 7 T1 6 6 6 7 Unknown 2 2 1 2 Grade Well differentiated 21 18 23 29 Moderately differentiated 9 7 10 5 Poorly differentiated 22 29 26 19 Unknown 1 2 1 2 N-stage N0 3 2 3 5 N1 3 2 1 0 NX 47 52 56 50 M-stage M0 2 0 3 3 M1 3 3 0 1 MX 48 53 57 51 Histology Urothelial cell 51 53 59 50 Squamous cell 1 0 1 0 Adenocarcinoma 0 0 0 1 Small cell 0 1 0 0 Other/Unknown 1 2 0 4 Deaths (cause) Bladder cancer 6 6 3 3 Other cause 5 3 4 5 Source of Funding: This work was supported in part by Public Health Service Cooperative Agreement grant CA37429 awarded by the National Cancer Institute, National Institutes of Health, Department of Health and Human Services, and in part by the National Center for Complementary and Alternative Medicine (National Institutes of Health). Study agents and packaging were provided by Perrigo Company (Allegan, Michigan), Sabinsa Corporation (Piscataway, New Jersey), Tishcon Corporation (Westbury, New York), and DSM Nutritional Products Inc. (Parsipanny, New Jersey).

Purpose: To develop an accurate and noninvasive method for bladder cancer diagnosis and prediction of disease aggressiveness based on the gene expression patterns of urine samples. Experimental Design: Gene expression patterns of 341 urine samples from bladder urothelial cell carcinoma (UCC) patients and 235 controls were analyzed via TaqMan Arrays. In a first phase of the study, three consecutive gene selection steps were done to identify a gene set expression signature to detect and stratify UCC in urine. Subsequently, those genes more informative for UCC diagnosis and prediction of tumor aggressiveness were combined to obtain a classification system of bladder cancer samples. In a second phase, the obtained gene set signature was evaluated in a routine clinical scenario analyzing only voided urine samples. Results: We have identified a 12+2 gene expression signature for UCC diagnosis and prediction of tumor aggressiveness on urine samples. Overall, this gene set panel had 98% sensitivity (SN) and 99% specificity (SP) in discriminating between UCC and control samples and 79% SN and 92% SP in predicting tumor aggressiveness. The translation of the model to the clinically applicable format corroborates that the 12+2 gene set panel described maintains a high accuracy for UCC diagnosis (SN = 89% and SP = 95%) and tumor aggressiveness prediction (SN = 79% and SP = 91%) in voided urine samples. Conclusions: The 12+2 gene expression signature described in urine is able to identify patients suffering from UCC and predict tumor aggressiveness. We show that a panel of molecular markers may improve the schedule for diagnosis and follow-up in UCC patients. Clin Cancer Res; 16(9); 2624–33. ©2010 AACR.

Purpose: There is a need in urological practice to identify new bladder cancer molecular markers to further develop noninvasive diagnostic tests. We analyzed bladder cancer gene expression profiles to determine the relevant differentially expressed genes and whether this differential expression is maintained in urine samples. Materials and Methods: We collected 55 tissue specimens from a total of 43 patients with bladder cancer and 12 controls, and 49 urine samples from bladder washings from a total of 36 patients with bladder cancer and 13 controls between September 2003 and December 2004. DNA microarrays (GeneChip (R) Human Genome U133 Plus 2.0 Array) were used to identify differentially expressed genes at 3 bladder cancer stages. Selected differentially expressed genes were validated in an independent set of bladder washings by quantitative reverse transcriptase-polymerase chain reaction. Results: Unsupervised cluster analysis of DNA microarray data showed a clear distinction in control vs tumor samples and low vs high grade tumors. Genes with at least 2-fold differential expression in controls vs tumors (2,937 probe sets or 2,295 genes) and in low vs high grade tumors (674 probe sets or 530 genes) were identified and ranked. Gene expression measurements in bladder washings of the 6 most differentially expressed genes in controls vs tumors were confirmed for the 2 over expressed genes tested by quantitative reverse transcriptase-polymerase chain reaction. All 8 selected differentially expressed genes in low vs high grade tumors were confirmed in bladder washing samples. Conclusions: Bladder cancer analysis by DNA microarrays provides new putative mRNA markers for bladder cancer diagnosis and/or prognosis that can be extrapolated to bladder fluids.

Background An accurate gene expression quantification using TaqMan Arrays (TA) could be limited by the low RNA quantity obtained from some clinical samples. The novel cDNA preamplification system, the TaqMan PreAmp Master Mix kit (TPAMMK), enables a multiplex preamplification of cDNA targets and therefore, could provide a sufficient amount of specific amplicons for their posterior analysis on TA. Findings A multiplex preamplification of 47 genes was performed in 22 samples prior to their analysis by TA, and relative gene expression levels of non-preamplified (NPA) and preamplified (PA) samples were compared. Overall, the mean cycle threshold (CT) decrement in the PA genes was 3.85 (ranging from 2.07 to 5.01). A high correlation (r) between the gene expression measurements of NPA and PA samples was found (mean r = 0.970, ranging from 0.937 to 0.994; p < 0.001 in all selected cases). High correlation coefficients between NPA and PA samples were also obtained in the analysis of genes from degraded RNA samples and/or low abundance expressed genes. Conclusion We demonstrate that cDNA preamplification using the TPAMMK before TA analysis is a reliable approach to simultaneously measure gene expression of multiple targets in a single sample. Moreover, this procedure was validated in genes from degraded RNA samples and low abundance expressed genes. This combined methodology could have wide applications in clinical research, where scarce amounts of degraded RNA are usually obtained and several genes need to be quantified in each sample.

Background Alport syndrome (AS) is a genetically heterogeneous hereditary renal disease. X-Linked AS (XLAS) is responsible for 80% to 85% of familial cases and is caused by mutations in the COL4A5 collagen gene. To date, indirect molecular diagnosis for XLAS is not well defined, and mutation screening of the COL4A5 gene is time consuming and complicated because of its large size and high allelic heterogeneity. Our aim is to facilitate XLAS genetic testing. Methods For linkage analysis, we tested the applicability of 4 microsatellite markers defining a 1.2-megabase region flanking the COL4A5 gene. For mutation screening of the COL4A5 gene, we describe a new strategy based on direct sequencing of hair root COL4A5 messenger RNA (mRNA). Results Three microsatellite markers proved accurate ( DXS1120 , DXS6802 , and DXS1210 ) and 1 was discarded ( DXS6797 ) because it was difficult to interpret. The mutation screening method provides results in 4 days, and when applied to 29 patients suspected of having XLAS, it identified mutations in 76% (22 of 29 patients). This study correlates COL4A5 mutations with effects at the mRNA level and suggests that mutations affecting mRNA splicing of the COL4A5 gene (41%; 9 of 22 patients) are more common than previously described. Many splicing mutations did not alter the canonical 5′ and 3′ splice sites. Conclusions A more reliable linkage analysis and a simple, fast, and efficient mutation screening are now available for the genetic testing of patients with XLAS.

We evaluated the performance of a multiprobe FISH (fluorescence in situ hybridization) assay for noninvasive detection of superficial urothelial carcinoma (UC) in the bladder, in comparison to urinary cytology. Voided urine samples from 74 patients with superficial UC were analyzed by both techniques. Urine samples from 19 patients with muscle-invasive tumors and from 19 healthy control subjects were also studied. For FISH analysis, labeled probes for chromosomes 3, 7, 9, and 17 were used to assess chromosomal abnormalities indicative of malignancy. We found a significant difference between the overall sensitivity of FISH and cytology in superficial UC detection (70.3 versus 35.1%, respectively; P0.0001). This significant difference was maintained when superficial UCs were broken down into low grade (52.8 versus 13.9%, respectively; P0.0005) and high grade (86.8 versus 55.3%, respectively; P0.0015) tumors. Overall specificity was 100% for cytology and 94.7% for FISH (difference not significant). Of patients with suspicious cytology, 69% were positive by FISH. Together, these findings suggest that FISH assay for chromosomes 3, 7, 9, and 17 has a higher sensitivity than cytology and a similar specificity in the detection of superficial UC--which could be useful for reducing some cystoscopies in the accurate follow-up usually performed in these patients.

We evaluate a number of computer programs designed to predict the structure of protein coding genes in genomic DNA sequences. Computational gene identification is set to play an increasingly important role in the development of the genome projects, as emphasis turns from mapping to large-scale sequencing. The evaluation presented here serves both to assess the current status of the problem and to identify the most promising approaches to ensure further progress. The programs analyzed were uniformly tested on a large set of vertebrate sequences with simple gene structure, and several measures of predictive accuracy were computed at the nucleotide, exon, and protein product levels. The results indicated that the predictive accuracy of the programs analyzed was lower than originally found. The accuracy was even lower when considering only those sequences that had recently been entered and that did not show any similarity to previously entered sequences. This indicates that the programs are overly dependent on the particularities of the examples they learn from. For most of the programs, accuracy in this test set ranged from 0.60 to 0.70 as measured by the Correlation Coefficient (where 1.0 corresponds to a perfect prediction and 0.0 is the value expected for a random prediction), and the average percentage of exons exactly identified was less than 50%. Only those programs including protein sequence database searches showed substantially greater accuracy. The accuracy of the programs was severely affected by relatively high rates of sequence errors. Since the set on which the programs were tested included only relatively short sequences with simple gene structure, the accuracy of the programs is likely to be even lower when used for large uncharacterized genomic sequences with complex structure. While in such cases, programs currently available may still be of great use in pinpointing the regions likely to contain exons, they are far from being powerful enough to elucidate its genomic structure completely.