Portal de Periódicos da CAPES

Development of a Novel Proteomic Approach for the Detection of Transitional Cell Carcinoma of the Bladder in Urine

2001; Elsevier BV; Volume: 158; Issue: 4 Linguagem: Inglês

10.1016/s0002-9440(10)64100-4

1525-2191

Antonia Vlahou, Paul F. Schellhammer, Savvas Mendrinos, Keyur Patel, Filippos I. Kondylis, Lei Gong, Suhail Nasim, George L. Wright,

Bladder and Urothelial Cancer Treatments

Development of noninvasive methods for the diagnosis of transitional cell carcinoma (TCC) of the bladder remains a challenge. A ProteinChip technology (surface enhanced laser desorption/ionization time of flight mass spectrometry) has recently been developed to facilitate protein profiling of biological mixtures. This report describes an exploratory study of this technology as a TCC diagnostic tool. Ninety-four urine samples from patients with TCC, patients with other urogenital diseases, and healthy donors were analyzed. Multiple protein changes were reproducibly detected in the TCC group, including five potential novel TCC biomarkers and seven protein clusters (mass range, 3.3 to 133 kd). One of the TCC biomarkers (3.4 kd) was also detected in bladder cancer cells procured from bladder barbotage and was identified as defensin. The TCC detection rates provided by the individual markers ranged from 43 to 70% and specificities from 70 to 86%. Combination of the protein biomarkers and clusters, increased significantly the sensitivity for detecting TCC to 87% with a specificity of 66%. Interestingly, this combinatorial approach provided sensitivity of 78% for detecting low-grade TCC compared to only 33% of voided urine or bladder-washing cytology. Collectively these results support the potential of this proteomic approach for the development of a highly sensitive urinary TCC diagnostic test. Development of noninvasive methods for the diagnosis of transitional cell carcinoma (TCC) of the bladder remains a challenge. A ProteinChip technology (surface enhanced laser desorption/ionization time of flight mass spectrometry) has recently been developed to facilitate protein profiling of biological mixtures. This report describes an exploratory study of this technology as a TCC diagnostic tool. Ninety-four urine samples from patients with TCC, patients with other urogenital diseases, and healthy donors were analyzed. Multiple protein changes were reproducibly detected in the TCC group, including five potential novel TCC biomarkers and seven protein clusters (mass range, 3.3 to 133 kd). One of the TCC biomarkers (3.4 kd) was also detected in bladder cancer cells procured from bladder barbotage and was identified as defensin. The TCC detection rates provided by the individual markers ranged from 43 to 70% and specificities from 70 to 86%. Combination of the protein biomarkers and clusters, increased significantly the sensitivity for detecting TCC to 87% with a specificity of 66%. Interestingly, this combinatorial approach provided sensitivity of 78% for detecting low-grade TCC compared to only 33% of voided urine or bladder-washing cytology. Collectively these results support the potential of this proteomic approach for the development of a highly sensitive urinary TCC diagnostic test. Bladder cancer is the second most common genitourinary malignancy accounting for ∼5% of all newly diagnosed cancers in the United States.1Klein A Zemer R Buchumensky V Klaper R Nissenkorn I Expression of cytokeratin 20 in urinary cytology of patients with bladder carcinoma.Cancer. 1998; 82: 349-354Crossref PubMed Scopus (80) Google Scholar More than 90% are of the transitional cell carcinoma (TCC) histology.2Stein JP Grossfeld GD Ginsberg DA Esrig D Freeman JA Figueroa AJ Skinner D Cote R Prognostic markers in bladder cancer: a contemporary review of the literature.J Urol. 1998; 160: 645-659Abstract Full Text Full Text PDF PubMed Scopus (256) Google Scholar At present, the most reliable way of diagnosis and surveillance of TCC is by cystoscopic examination and bladder biopsy for histological confirmation. The invasive and labor-intensive nature of this procedure presents a challenge to develop better, less costly, and noninvasive diagnostic tools. Urine cytology has for many years been the gold standard of the noninvasive approaches. It has high specificity and provides the advantage over biopsy of screening the entire urothelium.2Stein JP Grossfeld GD Ginsberg DA Esrig D Freeman JA Figueroa AJ Skinner D Cote R Prognostic markers in bladder cancer: a contemporary review of the literature.J Urol. 1998; 160: 645-659Abstract Full Text Full Text PDF PubMed Scopus (256) Google Scholar, 3Grossman HB Dinney CPN Markers of bladder cancer state of the art.Urol Oncology. 2000; 5: 3-10Abstract Full Text Full Text PDF PubMed Scopus (9) Google Scholar However, its high false-negative rate, particularly for low-grade tumors, has limited its use as an adjunct to cystoscopy. Many noninvasive molecular diagnostic tests have been developed based on an ever-increasing knowledge about the molecular alterations associated with bladder cancer pathogenesis. The bladder tumor antigen,4Schamhart DHJ Reijke TM Poel HG Witjes JA Boer EC Kurth K-H Schalken JA The Bard BTA test: its mode of action, sensitivity, and specificity compared to cytology of voided urine, in the diagnosis of superficial bladder cancer.Eur Urol. 1998; 34: 99-106Crossref PubMed Scopus (36) Google Scholar the bladder tumor antigen stat,5Sarosdy MF Hudson MA Ellis WJ Soloway MS de Vere White R Sheinfeld J Jarowenko MV Schellhammer PF Schervish EW Patel JV Chodak GW Lamm DL Johnson RD Henderson M Adams G Blumenstein BA Thoelke KR Pfalzgraf RD Murchison HA Brunelle SL Improved detection of recurrent bladder cancer using the bard BTA stat test.Urology. 1997; 50: 349-353Abstract Full Text PDF PubMed Scopus (165) Google Scholar the fibrinogen/fibrin degradation products,6Schmetter BS Habicht KK Lamm DL Morales A Bander NH Grossman HB Hanna Jr, MG Silberamn SR Butman BT A multicenter trial evaluation of the fibrin/fibrinogen degradation products test for detection and monitoring bladder cancer.J Urol. 1997; 158: 801-805Abstract Full Text Full Text PDF PubMed Scopus (112) Google Scholar and the nuclear matrix protein-22 tests,3Grossman HB Dinney CPN Markers of bladder cancer state of the art.Urol Oncology. 2000; 5: 3-10Abstract Full Text Full Text PDF PubMed Scopus (9) Google Scholar, 7Soloway MS Briggman JV Caprinito GA Chodak GW Church PA Lamm DL Lange P Messing E Pasciak RM Resevitz GB Rukstalis DB Sarosdy MF Stadler WM Thiel RP Hayden CL Use of a new tumor marker, urinary NMP22, in the detection of occult or rapidly recurring transitional cell carcinoma of the urinary tract following surgical treatment.J Urol. 1996; 156: 363-367Abstract Full Text Full Text PDF PubMed Scopus (292) Google Scholar have been approved by the Food and Drug Administration to be used in conjunction with cystoscopy. Additional molecular assays currently being evaluated for their diagnostic/prognostic utility2Stein JP Grossfeld GD Ginsberg DA Esrig D Freeman JA Figueroa AJ Skinner D Cote R Prognostic markers in bladder cancer: a contemporary review of the literature.J Urol. 1998; 160: 645-659Abstract Full Text Full Text PDF PubMed Scopus (256) Google Scholar, 3Grossman HB Dinney CPN Markers of bladder cancer state of the art.Urol Oncology. 2000; 5: 3-10Abstract Full Text Full Text PDF PubMed Scopus (9) Google Scholar, 8Orntoft TF Wolf H Molecular alteration in bladder cancer.Urol Res. 1998; 26: 223-233Crossref PubMed Scopus (95) Google Scholar, 9Halachmi S Linn JF Amiel GE Moskovitz B Nativ O Urine cytology, tumour markers and bladder cancer.Br J Urol. 1998; 82: 647-654Crossref PubMed Scopus (37) Google Scholar are the Telomerase,10Hoshi S Takahashi T Satoh M Numahata K Suzuki K-I Ohyama C Mori M Mituoka T Nakagawara K-I Orikasa S Telomerase activity. Simplification of assay and detection in bladder tumor and urinary exfoliated cells.Urol Oncol. 2000; 5: 25-30Abstract Full Text Full Text PDF PubMed Scopus (4) Google Scholar Immunocyt,11Fradet Y Lochart C Performance characteristics of a new monoclonal antibody test for bladder cancer: immunoCytTM.Can J Urol. 1997; 4: 400-405PubMed Google Scholar and hyaluronic acid/hyaluronidase12Pham HT Block NL Lokeshwar VB Tumor-derived hyaluronidase: a diagnostic urine marker for high grade bladder cancer.Cancer Res. 1997; 57: 778-783PubMed Google Scholar, 13Lokeshwar VB Obek C Soloway MS Block NL Tumor-associated hyaluronic acid: a new sensitive and specific urine marker for bladder cancer.Cancer Res. 1997; 57: 773-777PubMed Google Scholar tests, microsatellite analysis,14Steiner G Schoenberg MP Linn JF Mao L Sidransky D Detection of bladder cancer recurrence by microsatellite analysis of urine.Nat Med. 1997; 6: 621-624Crossref Scopus (245) Google Scholar as well as assays detecting blood group antigens,15Golijanin D Sherman Y Shapiro A Pode D Detection of bladder tumors by immunostaining of the Lewis X antigen in cells from voided urine.Urology. 1995; 46: 173-177Abstract Full Text PDF PubMed Scopus (67) Google Scholar carcinoembryonic antigen,16Liu BC Neuwirth H Wei Zhu L Stock LM DeKernion JB Fahey JL Detection of onco-fetal bladder antigen in urine of patients with transitional cell carcinoma.J Urol. 1987; 137: 1258-1261PubMed Google Scholar p53 and retinoblastoma proteins,3Grossman HB Dinney CPN Markers of bladder cancer state of the art.Urol Oncology. 2000; 5: 3-10Abstract Full Text Full Text PDF PubMed Scopus (9) Google Scholar E cadherin,17Banks RE Porter WH Whelam P Smith PH Seldy PJ Soluble forms of E-cadherin in urine.J Clin Pathol. 1995; 48: 179-180Crossref PubMed Scopus (67) Google Scholar, 18Protheroe AS Banks RE Mzimba M Southgate J Singh PN Bosomworth M Harnden P Smith PH Whelan P Selby PJ Urinary concentrations of the soluble adhesion molecule E-cadherin and total protein in patients with bladder cancer.Br J Cancer. 1999; 80: 273-278Crossref PubMed Scopus (34) Google Scholar and various growth factors.9Halachmi S Linn JF Amiel GE Moskovitz B Nativ O Urine cytology, tumour markers and bladder cancer.Br J Urol. 1998; 82: 647-654Crossref PubMed Scopus (37) Google Scholar Because of the molecular heterogeneity of these tumors, it is likely that there will be no single molecular assay that will replace cystoscopy. The identification and simultaneous analysis of a panel of biomarkers, representative of the various biological characteristics of the cancer, has greater potential for improving the early detection/diagnosis of TCC. For many years, two-dimensional (2D) gel electrophoresis has been the principal tool for the separation and analysis of multiple proteins.19O' Farrell PH High resolution two-dimensional gel electrophoresis of proteins.J Biol Chem. 1975; 250: 4007-4021Abstract Full Text PDF PubMed Google Scholar This methodology, which is able to resolve thousands of proteins in one experiment, provides the highest resolution in protein separation. However, it is labor intensive, requires large quantities of starting material, lacks interlab reproducibility, and is not practical for clinical application. Although development of image analysis software for the comparison of 2D gel-protein maps and automation of protein spot excision20Patterson SD From electrophoretically separated proteins to identification: strategies for sequence and mass analysis.Anal Biochem. 1994; 221: 1-15Crossref PubMed Scopus (109) Google Scholar have facilitated the analysis of the separated proteins, most of the major technical difficulties of 2D gel electrophoresis remain. Significant technological advances in protein chemistry in the last 2 decades have established mass spectrometry as an indispensable tool for protein study.21Carr SA Hemling ME Bean MF Roberts GD Integration of mass spectrometry in analytical biotechnology.Anal Chem. 1991; 63: 2802-2824Crossref PubMed Scopus (240) Google Scholar, 22Carr SA Annan RS Overview of peptide and protein analysis by mass spectrometry. Current Protocols in Molecular Biology.in: Ausubel FM Brent R Kingston RE Moore DD Seidman JG Smith JA Struhl K John Wiley & Sons Inc., New York1998: 10.21.1-10.21.27Google Scholar, 23Patterson SD Protein identification and characterization by mass spectrometry. Current Protocols in Molecular Biology.in: Ausubel FM Brent R Kingston RE Moore DD Seidman JG Smith JA Struhl K John Wiley & Sons Inc., New York1998: 10.22.1-10.22.24Google Scholar Although the resolving power of 2D gels remains unchallenged, the high sensitivity, speed, and reproducibility of mass spectrometry have boosted its application in all aspects of protein analysis, including discovery, identification (ie, peptide mapping, sequencing), and structural characterization. Analogous to the DNA chip technologies that allow the study of gene expression profiles, Ciphergen Biosystems, Inc. (Fremont, CA) has recently developed the ProteinChip technology coupled with SELDI-TOF-MS (surface-enhanced laser desorption/ionization time of flight mass spectrometry) to facilitate protein profiling of complex biological mixtures.24Hutchens TW Yip TT New desorption strategies for the mass spectrometric analysis of macromolecules.Rapid Commun Mass Spectrom. 1993; 7: 576-580Crossref Scopus (592) Google Scholar, 25Kuwata H Yip T-T Yip CL Tomita M Hutchens WT Bactericidal domain of lactofferin: detection, quantitation and characterization of lactoferricin in serum by SELDI affinity mass spectrometry.Biochem Biophys Res Commun. 1998; 245: 764-773Crossref PubMed Scopus (119) Google Scholar This technology utilizes patented chip arrays to capture individual proteins from complex mixtures that are subsequently resolved by mass spectrometry. This innovative technology has numerous advantages over 2D-polyacrylamide gel electrophoresis: it is much faster, has a high-throughput capability, requires orders of magnitude lower amounts of the protein sample, has a sensitivity for detecting proteins in the picomole to attamole range, can effectively resolve low mass proteins (2,000 to 20,000 Da), and is directly applicable for clinical assay development. The efficacy of the SELDI technology for discovery of prostate cancer protein markers in serum, seminal plasma, and cell extracts, as well as the development of immunoassays for the detection of known prostate cancer markers has recently been demonstrated by our laboratory.26Wright Jr, GL Cazares LH Leung S-M Nasim S Adam B-L Yip T-T Schellhammer PF Gong L Vlahou A ProteinChip surface enhanced laser desorption/ionization mass spectrometry: a novel protein biochip technology for detection of prostate cancer biomarkers in complex protein mixtures.Prostate Cancer Prostate Dis. 1999; 2: 264-276Crossref PubMed Scopus (255) Google Scholar, 27Paweletz CP Gillespie JW Ornstein DK Simone NL Brown MR Cole KA Wang Q-H Huang J Hu N Yip T-T Rich WE Kohn EC Linehan WM Weber T Taylor P Emmert-Buck MR Liotta LA Petricoin III, EF Rapid protein profiling of cancer progression directly from human tissue using a protein biochip.Drug Dev Res. 2000; 49: 34-42Crossref Scopus (139) Google Scholar This report describes our initial evaluation using the ProteinChip SELDI-MS system to detect potential TCC biomarkers in urine, and to assess these biomarkers for the diagnosis of TCC. Multiple protein changes were reproducibly found in the urine of TCC patients, including five potentially novel urinary TCC biomarkers, and seven protein cluster regions consisting of different numbers of proteins observed in the cancer versus the control groups. One of these potential urinary TCC-associated protein biomarkers was identified as belonging to the defensin family of peptides. Urine samples were collected throughout a period of several months from patients seen in the department of Urology, Eastern Virginia Medical School. The urine samples were immediately aliquoted and stored at −80°C in the Tissue and Body Fluid Bank of the Virginia Prostate Center, until assayed. A total of 94 urine specimens were collected. The demographics of the TCC patient and control groups are provided in Table 1. Healthy controls (n = 34) included volunteers with no evidence of disease, and healthy individuals (ie, no history or evidence of urological cancer) participating in the prostate cancer screening program at Eastern Virginia Medical School. TCC (n = 30 patients) was histologically or cytologically confirmed at the time of specimen collection. In the case of recurrences none of the patients had received chemo- or immunotherapy within 3 months before specimen collection. Grading was assessed using the World Health Organization system. Tumor stage and grade of patients with TCC are shown in Table 2. Other urogenital diseases (n = 30 patients) included clinical or pathologically confirmed prostatitis (n = 6), prostatism (n = 9), urinary tract infections (n = 1), benign prostatic hyperplasia (n = 12), amyloidosis (n= 1), inflammation of prostate and bladder (n = 1), bladder outlet obstruction (n = 1), and prostate cancer (n = 1). One patient with benign prostatic hyperplasia and one with prostatism had concomitant prostatitis.Table 1Demographics of the Study (TCC) and Control (Normal, Other Diseases) GroupsTCCNormalOtherNo. of samples303430Age range42–8623–7141–82Mean age69.45568.5 Open table in a new tab Table 2Stage/Grade of Bladder TumorsStageNo. of samplesGradeNo. of samplesTa10I4Ta-CIS4II5T17III21T1-CIS1T24T2-CIS1T31CIS2CIS, carcinoma in situ. Open table in a new tab CIS, carcinoma in situ. Urine samples were thawed and briefly centrifuged (1 minute, 10,000 rpm) for the removal of cellular material. Protein concentration of the supernatants was estimated using the bicinchoninic acid kit (Pierce, Rockford, IL). Samples were diluted with binding buffer (20 mmol/L Tris, pH 9, 0.4 mol/L NaCl, 0.1% Triton X 100) to equal protein concentration (2 mg/ml) and subjected to protein size fractionation using a K30 microspin column (Ciphergen Biosystems, Inc.). After a 30-minute incubation on ice, diluted urine samples were applied to the spin columns and centrifuged for 3 minutes at 720 × g. The ProteinChip SELDI analysis was performed similar to that described in an earlier report.26Wright Jr, GL Cazares LH Leung S-M Nasim S Adam B-L Yip T-T Schellhammer PF Gong L Vlahou A ProteinChip surface enhanced laser desorption/ionization mass spectrometry: a novel protein biochip technology for detection of prostate cancer biomarkers in complex protein mixtures.Prostate Cancer Prostate Dis. 1999; 2: 264-276Crossref PubMed Scopus (255) Google Scholar Briefly, 5-μl aliquots of the flow-through (fraction) and the unfractionated sample diluted in 20 mmol/L Tris, pH 9.0, 0.1% Triton X-100, were directly applied onto different arrays of a SAX2 chip that consists of a strong anion exchanger chemistry. After a brief wash with H2O, 0.5 μl of saturated matrix solution (α-cyano-4-hydroxycinnamic acid in 50% acetonitrile and 0.5% trifluoroacetic acid) was applied on the array and allowed to air dry. The chips were then placed in the PBS-I mass reader, where nanosecond laser pulses are generated from a nitrogen laser (337 nm). Spectra were generated using an average of 60 laser shots at each of the following laser intensities (L): 15 (filter in), 30 (filter in), and 55 (filter out) and manually compared for the detection of protein differences between the various groups. A protein or protein cluster was considered to be differentially expressed in the TCC group, if statistically significant differences in its frequency, compared to the normal and/or other diseases group, were observed. For the calculation of protein peak numbers resolved at low laser intensities, spectra collected at L15 and L30 were combined using the SELDI software (0.5% variation). External calibration was performed using bovine insulin (5,733.6 Da), bovine cytochrome C (12,230.9 Da), and bovine serum albumin (66,410 Da) as standards (Ciphergen Biosystems, Inc.). Bladder washings were centrifuged at 1,500 rpm for 5 minutes for the collection of cellular material. Supernatants were discarded with the exception of 1 to 2 ml that were used for resuspending the cell pellet. Cytospin preparations of 50 to 100 μl of the resuspended cell pellet were then made, the slides immediately placed in 100% EtOH, and stained with hematoxylin and eosin. The stained slides were examined by a pathologist (SN) to identify the cancer cells, and the individual cancer cells or clusters were procured using the Pixcell 100 Laser Capture Microdissection Microscope (Arcturus Engineering, Mountain View, CA), as previously described.26Wright Jr, GL Cazares LH Leung S-M Nasim S Adam B-L Yip T-T Schellhammer PF Gong L Vlahou A ProteinChip surface enhanced laser desorption/ionization mass spectrometry: a novel protein biochip technology for detection of prostate cancer biomarkers in complex protein mixtures.Prostate Cancer Prostate Dis. 1999; 2: 264-276Crossref PubMed Scopus (255) Google Scholar, 28Emmert-Buck MR Bonner RF Smith PD Chuagui FR Zhuang Z Goldstein SR Weiss RA Liotta LA Laser capture microdissection.Science. 1996; 274: 998-1001Crossref PubMed Scopus (2091) Google Scholar Protein extracts were prepared from 500 to 1,000 microdissected cells by resuspending the cells in 3 to 5 μl of 20 mmol/L Hepes containing 0.1% Nonidet P-40, vortexing for 5 minutes, and then centrifugation at 14,000 rpm for 1 minute. The entire lysate was applied onto a nickel IMAC3 (Immobilized Metal Affinity) chip array, and incubated for 1 hour. The chips were washed with 20 mmol/L Tris, pH 7.5, 0.1% Triton X-100, 0.5 mol/L NaCl (5 times), and HPLC-H20 (5 times). Mass analysis was performed as described for urine, using either α-cyano-4-hydroxycinnamic acid or sinapinic acid as the energy absorbing molecules. Sensitivity is defined as the ratio of the TCC patients that contained the biomarker to the total number of TCC patients included in the study. Specificity is defined as the ratio of the individuals that do not have the protein peak and do not have TCC, to the total number of individuals without TCC. Positive predictive value is defined as the probability that an individual with the biomarker has TCC. Negative predictive value is defined as the probability that an individual without the biomarker does not have TCC. Statistics were performed using the chi-square test, after organizing the data in two-dimensional contingency tables and testing for independence of variables. Comparison of peak numbers between the various groups was performed using Student’s t-test. In all cases, P < 0.05 was considered statistically significant. The SELDI immunoassay was performed similar to that described in a previous report.26Wright Jr, GL Cazares LH Leung S-M Nasim S Adam B-L Yip T-T Schellhammer PF Gong L Vlahou A ProteinChip surface enhanced laser desorption/ionization mass spectrometry: a novel protein biochip technology for detection of prostate cancer biomarkers in complex protein mixtures.Prostate Cancer Prostate Dis. 1999; 2: 264-276Crossref PubMed Scopus (255) Google Scholar Briefly, the arrays of a preactivated chip (PS1, Ciphergen Biosystems, Inc.), were coated with 4 μl of Protein G (0.5 mg/ml in 50 mmol/L sodium bicarbonate, pH.8: Sigma Chemical Co., St. Louis, MO) for 2 to 4 hours at room temperature with shaking. Residual active sites were subsequently blocked with 1 mol/L ethanolamine (30 minutes, room temperature), followed by sequential washes in 15-ml conical tubes with phosphate-buffered saline (PBS) and 0.5% Triton X (3×) and PBS (4×). Two μl of defensins-1, -2, and -3 (HNP-1, -2, and -3) monoclonal antibody (mAb) (IgG1, 0.2 mg/ml; Serotec), prostate-specific membrane antigen (PSMA) 7E11C5.3 mAb (IgG1, 0.2 mg/ml; kindly provided by Cytogen Corporation, Princeton, NJ) or mouse IgG1 (30 μg/ml) were applied on the chip and allowed to bind at 4°C, overnight with shaking. Unbound Abs were removed by sequential washes in 15-ml conical tubes with PBS and 0.5% Triton X (1×), PBS and 0.1% Triton X (3×), and PBS (4 ×). Urine samples were diluted in 100 μl of PBS-0.1% CETAB29Panyutich AV Voitenok NN Lehrer RI Ganz T An enzyme immunoassay for human defensins.J Immunol Methods. 1991; 141: 149-155Crossref PubMed Scopus (58) Google Scholar (Sigma Chemical Co., St. Louis, MO) at a total protein concentration of 0.055 mg/ml, and after a 20-minute incubation in ice, were applied onto the arrays using a bioprocessor (Ciphergen). After a 3-hour incubation at 4°C, the unbound urinary proteins were washed away by five washes with PBS-0.1% CETAB (5 minutes each, room temperature) followed by five washes with HPLC-H20, α-cyano-4-hydroxycinnamic acid added, and the chip subjected to mass analysis. The spectra were generated using signal averaging of 90 laser shots. Ninety-four urine samples were assayed by SELDI mass spectrometry. Processing on a strong anion exchanger chip surface resolved up to 70 protein peaks. Figure 1a is a representative protein spectrum showing the protein masses between 2,000 to 150,000 Da of a single urine specimen. Generation of spectra was performed at laser intensities 15, 30, and 55, so as to better resolve low- and high-molecular mass proteins, respectively. As shown, the SELDI technology was particularly effective in resolving the low molecular weight (<10 kd) proteins and polypeptides. Interestingly, urine samples from TCC patients appeared to contain higher numbers of protein peaks. Collection of data at laser intensities 15 and 30, generated an average of 33 protein peaks from the TCC urine samples versus an average of 21 and 22 for the normal and other urogenital diseases, respectively (P < 0.001). Similarly, at higher laser intensities (ie, 55-filter out), TCC samples had an average of 34 protein peaks, versus 27 and 20 in the normal and other urogenital diseases groups (P< 0.001 for the normals and P = 0.008 for the other diseases). All samples were processed in either duplicate or triplicate to confirm reproducibility in resolving the urinary proteins. Figure 1, b and c, shows that reproducibility was quite acceptable. The mean, standard deviation (SD) and coefficient of variation (CV) were determined for four prominent peaks, designated as proteins 1 to 4. The intraassay reproducibility, ie, the mean mass, SD (%CV) for protein 1 was 6,440.6 ± 0.92 Da (0.014%); for protein 2, 7,914 ± 3.32 Da (0.042%); for protein 3, 13,262 ± 0.78 Da (0.006%); and for protein 4, 66,288 ± 69.3 Da (0.1%) (Figure 1, b and c; spectra 1 and 2). The interassay reproducibility was determined to be 6,443.3 ± 3.85 Da (0.06%) for protein 1, 7,918.3 ± 6.12 Da (0.08%) for protein 2, 13,267 ± 8.42 Da (0.06%) for protein 3, and 66,277 ± 15.56 Da (0.023%) for protein 4 (Figure 1, b and c; spectra 1 and 2 versus 3). Analysis of urine specimens from patients with TCC, patients with other diseases of the urogenital tract, and normal individuals, revealed that five prominent protein peaks were preferentially expressed in TCC. Representative mass spectra and gel views of these proteins are shown in Figure 2. One of the proteins was observed as a doublet or occasionally as a triplet protein peak (Figure 2a) having an average mass of 3.353 (SD: 21 Da), 3.432 (SD: 24.4 Da), and 3.470 kd (SD: 6.32 Da), respectively. This protein will be referred to as marker urinary bladder cancer 1 or UBC1. The average SELDI mass associated with the other four TCC-associated proteins are UBC2: 9.495 kd (SD: 46.5 Da); UBC3: 44.6 kd (SD: 372.8 Da); UBC4: 100.120 kd (SD: 866.8 Da); and UBC5: 133.190 kd (SD: 772.9 Da) (Figure 2; b, c, and d). Of the TCC patient urine samples evaluated, 47% (14 of 30) were positive for UBC1, 53% (16 of 30) for UBC2, 70% (21 of 30) for UBC3, 43% (13 of 30) for UBC4, and 63% (19 of 30) for UBC5 (Table 3A). Frequency of almost all markers was observed to increase with progression from low-grade (I to II) to high-grade (III) and low-stage (Ta) to higher stage (T1–3) carcinomas (data not shown). Nevertheless, larger numbers of samples will need to be analyzed to confirm these initial observations.Table 3Summary of TCC-Associated Protein DataA. Detection of the 5 TCC-associated proteins in the study and control groupsNumber positive/number testedMarkerSensitivity %Specificity N%Specificity O%Specificity All %PPV%NPV%P*P**UBC147 (14/30)85 (5/34)87 (4/30)86 (9/64)61760.01 < P < 0.0250.01 < P < 0.025UBC253 (16/30)91 (3/34)70 (9/30)81 (12/64)5779P < 0.0010.1 < P < 0.25UBC370 (21/30)88 (4/34)70 (9/30)80 (13/64)6285P < 0.0010.001 < P < 0.005UBC443 (13/30)85 (5/34)87 (4/30)86 (9/64)59760.01 < P < 0.0250.01 < P < 0.025UBC563 (19/30)79 (7/34)60 (12/30)70 (19/64)50800.001 < P < 0.0050.1 < P < 0.25B. Frequency of the TCC-associated proteins according to ageMarkerSpecificity N% (>50)Specificity N% ( 50, normal individuals older than 50 years old (range 50–71, mean 61.95); 50, normal individuals older than 50 years old (range 50–71, mean 61.95); <50, normal individuals younger than 50 years old (range 23–49, mean 42.8). The percent positive samples for the five biomarkers in the normal population were 15 (5 of 34) for UBC1, 9 (3 of 34) for UBC2, 12 (4 of 34) for UBC3, 15 (5 of 34) for UBC4, and 21 (7 of 34) for UBC5, corresponding to a specificity of 85, 91, 88, 85, and 79%, respectively (Table 3A). The frequency of the markers in this control group is significantly different from their frequency in the TCC urine samples (Table 3A), and does not appear to change significantly when aged-matched normal individuals (ie, older than 50 years old)