The impact of advanced proteomics in the search for markers and therapeutic targets of bladder cancer

Abstract

Bladder cancer is the most common cancer of the urinary tract and can be avoided through proper surveillance and monitoring. Several genetic factors are known to contribute to the progression of bladder cancer, many of which produce molecules that serve as cancer biomarkers. Blood, urine, and tissue are commonly analyzed for the presence of biomarkers, which can be derived from either the nucleus or the mitochondria. Recent advances in proteomics have facilitated the high-throughput profiling of data generated from bladder cancer–related proteins or peptides in parallel with high sensitivity and specificity, providing a wealth of information for biomarker discovery and validation. However, the transmission of screening results from one laboratory to another remains the main disadvantage of these methods, a fact that emphasizes the need for consistent and standardized procedures as suggested by the Human Proteome Organization. This review summarizes the latest discoveries and progress of biomarker identification for the early diagnosis, projected prognosis, and therapeutic response of bladder cancer, informs the readers of the current status of proteomic-based biomarker findings, and suggests avenues for future work.

Keywords

Bladder cancer proteomics diagnosis treatment prognosis

Introduction

Bladder cancer (BC) is the second most common tumor of the genitourinary malignancies after prostate cancer, with peak incidence occurring after the age of 70 years. The incidence of BC varies significantly between geographical regions, with the highest rates observed in more developed areas. Lower rates are observed in the less developed regions of Asia and most of Africa.^1,2 Much of this variability is thought to be artificial due to the lack of proper record-keeping as well as differences in the definitions of bladder tumors. Additionally, the most important risk factors for developing BC, occupational exposures and smoking, differ significantly between these regions.³ Approximately 70% of all newly diagnosed cases present as superficial BC (pTa–pT1) with a relatively high 5-year survival rate; however, 10%–20% of superficial BC will progress to muscle-invasive BC (MIBC: pT2–pT4) following resection, which is characterized by poor prognosis: 30%–50% of these patients die from their cancer.^4–7 Up to 70% of those patients will suffer from recurrence of the tumor, as the biological characteristics of BC render it prone to drug resistance and recurrences,⁸ despite the availability of various therapies, including radical cystectomy, transurethral resection, chemotherapy, and radiotherapy.⁹ A better understanding of the molecular basis of the tumorigenesis and progression of BC will lead to the identification of both biomarkers for early detection and potential targets for therapy.¹⁰

In 1994, the field of “proteomics” was first proposed and has been described as a characterization of the presence and activity of all proteins in a tissue, cell, or fluid.^11,12 Because gene expression is a highly regulated, multistep process, it is impossible to predict exact protein concentrations or their activity from the measurement of messenger RNA (mRNA) levels alone. In addition, from the transcription of DNA to post-translational modifications (PTMs), the expression or function of proteins may be modulated at many diverse points. For instance, numerous transcripts give rise to more than one protein through alternative splicing or alternative PTMs, including phosphorylation, glycosylation, and acetylation, that profoundly affect protein activity and lead to multiple protein products from the same gene.¹³ From microarray data, Shimwell et al.¹⁴ detected that the expression of HAI-1 and midkine mRNA was elevated in Ta BC but did not increase further with increasing stages. This suggests that the urinary protein levels and mRNA expression do not correlate across disease stages. Therefore, proteomics provides a different level of understanding than genomics. Providing quantitative and structural information about proteins, which are the major functional determinants of cells, has led to proteomics becoming a key tool in systems biology.^15,16 The proteome can be regarded as the protein complement of all actively expressed genes. In addition, alterations made through PTMs, as well as complexes and molecular machines formed and augmented through protein–protein – protein interactions (PPI), can be examined.^11,12 The extent and degree of this characterization vary depending on the assays used. Some of the more common proteomic approaches include two-dimensional gel electrophoresis (2D-GE), surface-enhanced laser desorption/ionization (SELDI), difference gel electrophoresis (DIGE), liquid chromatography coupled mass spectrometry (LC–MS), capillary electrophoresis (CE) coupled to MS (CE–MS), isobaric tag for relative and absolute quantification (iTRAQ), and protein arrays.^17–19

Several proteomics studies in this field of early detection of BC have been described in the literature, as outlined in a number of reviews from 2004 to 2010.^20–24 The technical advances have allowed the field of proteomics to become extremely productive in the identification of proteins as well as changes in specific protein concentrations in the blood, urine, or tissue of BC patients, many of which may have the potential to be utilized in the diagnosis of bladder tumors. Currently, an increasing number of studies^25–27 are being carried out with various goals aiming to improve the therapeutic treatment of BC and to better understand the potential prognostic role of proteomics in BC after follow-up. MS-based platforms such as CE–MS and multiple reaction monitoring (MRM) have already been demonstrated to be efficient screening tools to identify potential biomarkers. As the latest reviews have rarely provided a systematic overview of the role of proteomics in the identification of indicators for prognosis or targets for oncotherapy, we therefore seek to present the data from relevant literature to help bridge the gap between bench work results and bedside potential.

Methods for proteomic profiling

Because of the inherent complexity of a proteome, its analysis faces many limitations. Over the last few decades, proteomic techniques have become more and more precise, due to new instrumentation and a better understanding of the tremendous need for standard operating procedures in screening methods. There are many different proteomic technology platforms that can be widely applied. In the following section, we will briefly examine the more common of these methods.

MS-based techniques have long been the gold standards for total proteome analysis. The most commonly utilized MS methods for the analysis of peptides and proteins are electrospray ionization (ESI) and material-enhanced laser desorption/ionization (MALDI)-MS. When compared, MALDI can be much more easily automated than ESI and generates less complex spectra of mostly singly charged ions. A major disadvantage for the analysis of complex samples appears to be the pronounced signal suppression in MALDI, which is observed to a lesser degree in ESI.²⁸ Given the complexity of the proteome, an additional separation step before applying MS analysis (pre-MS separation) must be introduced to increase the overall analytical resolution.

The SELDI technology is widely accepted due to its ease of use and its high-throughput capabilities. SELDI utilizes selective interactions of polypeptides with different chip surfaces to achieve the fractionation of biological samples, thereby reducing their complexity for subsequent MALDI-MS analysis. Numerous reports on biomarkers for a variety of diseases have been published using this strategy.²⁹ However, the low-resolution results of the MS and an apparent lack of reproducibility in data are obstacles for its application.³⁰

LC provides a powerful protein fractionation method, which is the separation of proteins based on the charge, hydrophobicity, or other biophysical characteristics of the samples. This is performed prior to MS for specific protein identification.^31,32 LC–MS is looked upon as an excellent choice for proteome analysis due to its high sensitivity. The sequential separation of proteins using different principles in two independent steps provides a multidimensional fractionation that can generate vast amounts of information. Multidimensional protein identification technologies are well suited for the in-depth analysis of body fluids such as urine.³³ However, limitations include difficulties with comparative analysis, which are partly due to the variability within multidimensional separations, as well as the substantial time required for the analysis of a single sample.³⁴

CE–MS is another method that separates polypeptides in a single step with high resolution.³⁵ The separation is based on the migration through a gel in which an electrical field is applied, and proteins migrate differently based on their charge and molecular size. Following the electrophoretic separation, the polypeptides are analyzed and characterized by MS.³⁶ CE–MS provides fast separation and high resolution and is more robust and cheaper than LC-based techniques,³⁷ due to the use of inexpensive capillaries. Disadvantages of this technique are that it cannot be easily used in the separation and analysis of high-molecular-weight proteins. This is because large proteins tend to precipitate at low pH that is used in the running buffer. In addition, it has low sensitivity due to low loading capacities and small sample size.

Two-dimensional polyacrylamide gel electrophoresis (2D-PAGE) followed by MS is the most commonly used method to separate and identify proteins >20 kDa. This approach separates proteins according to the isoelectric point and molecular mass of the proteins. Protein identification is routinely performed as a step-by-step approach, based on proteolytic in-gel digest, followed by gel extraction and subsequent MS analysis of the resultant peptide fragments. This approach also offers the possibility to detect PTMs of polypeptides.¹⁷ The 2D-GE provides high-resolution protein separation and direct information on the expression levels of intact proteins. However, this technique often will yield inconsistent or variable results and is time-consuming, leading to difficulties when trying to compare multiple datasets.³⁸

Protein array techniques are newly emerging approaches for proteomic analysis in recent years. These can be targeted approaches in which specific antibodies can be immobilized to form a protein microarray for the capture and quantification of specific protein concentrations.³⁹ Protein microarray analysis offers an additional method to investigate serum and urine proteomes in a targeted manner. It has been used in the separation and identification of proteins in cancer patients with high degrees of both sensitivity and specificity. However, the disadvantages of this method are the high costs of antibodies and the fact that PTMs of proteins are more difficult to analyze.^38,40

Markers for diagnosis

Currently, the principal methods of diagnosis include a voided urine cytology (VUC), cystoscopy with biopsy, and/or clinical presentation. First described in 1945 by Papanicolaou and Marshall, VUC has been used in both the diagnosis and follow-up of superficial BC and remains the primary noninvasive diagnostic method. Although it has a high specificity (>93%), its sensitivity is poor (only 25%–40%), particularly in well-differentiated (low-grade and low-stage) tumors.^41,42 While the procedure is uncomfortably invasive, costly, and can increase the likelihood of urinary infection, cystoscopy with biopsy is the current gold standard procedure for the identification of BC. At clinical presentation, approximately 85% of patients with BC and 40% with kidney cancer are found to have either microscopic or gross hematuria.⁴³ Because hematuria may be caused by a variety of benign urologic conditions, including benign prostatic hyperplasia, urinary tract infection, and kidney stones, the sensitivity and specificity of hematuria detection have a significantly reduced ability to diagnose BC.^44–46 Developing reliable diagnostic markers would be of great benefit to both patients and healthcare systems. The application of sensitive and specific proteomic analysis could potentially provide cost-effective, noninvasive monitoring for low-risk patients, as well increase the ability to identify the aggressiveness of high-risk refractory cancers before they metastasize. Thus, it is very critical to find markers of BC through the development of proteomic assays. Here, we will review the progress of proteomics in the analysis of the urine, blood, and tumor tissues of BC patients.

As urine can be sampled noninvasively, it not only is an ideal body fluid for the detection of potential biomarkers but also can serve as a source of proteins that represent the physiological or pathological state of the major urological tissues against a background of proteins and peptides that arise from the filtration of blood within the kidney. Moreover, proteomic analysis of urine has the potential to aid in the identification of key proteins in urine that are involved in the development and spread of BC in addition to having roles as biomarkers.^47,48 Aiming to establish a comprehensive analytical approach for the detection of both non-muscle invasive BC and MIBC in urine, Kumar et al.²⁶ used MS technology in combination with multiplex peptide stable isotope labeling to capture differentially secreted proteins in voided urine samples of patients. This approach identified five highly sensitive and specific biomarkers (Coronin-1A, Apolipoprotein A4, Semenogelin-2, Gamma-synuclein (SNCG) and DJ-1/PARK7), and its success suggests that quantitative proteomics can deliver accurate and unbiased information about the quantitative behavior of a wide variety of proteins within a urine sample. Kreunin et al.⁴⁹ developed a rapid, high-sensitivity technique that can be used to profile the N-linked glycoprotein component in naturally micturated human urine specimens. Of 186 proteins identified with high confidence via multiple analyses, 40% were secreted proteins, 18% were membrane proteins, and 14% were extracellular proteins. Several of the identified proteins appeared to be associated with the presence of BC, including alpha-1B-glycoprotein, which was detected in all tumor-bearing patient samples.

While protein separation by 2D-PAGE followed by proteome analysis by MS measurement is often utilized for the comprehensive identification of proteins in various types of biological samples, it has not yet been widely applied to the analysis of body fluids in clinical assessments. One of the reasons is that large amounts of proteins such as albumin and globulin in fluid can prevent the detection of small amounts of proteins with high clinical value. To improve the detection of proteins, Saito et al. combined proteomic analysis with gel-affinity purification of urine samples. The 2D-PAGE gels from cancer-bearing patients showed various amounts of matrix metalloproteinase-2 (MMP-2) and -9 (MMP-9) and fibronectin (FN), which were absent in the samples from normal individuals. This method proved to be useful not only for the diagnosis of BC but also for estimating the extent of tumor invasion.⁵⁰ Kageyama et al. examined 22 cancerous and 10 noncancerous surgical specimens from transurethral resection or radical cystectomy, focusing on the presence of calreticulin (CRT). Proteomic analysis revealed that CRT was present at increased levels in cancer tissue samples. They showed that urinary CRT could be a diagnostic indicator, with a sensitivity of 73% and a specificity of 86%.⁵¹ A large-scale proteomic analysis was performed by Iwaki et al., who identified SNCG and a soluble isoform of catechol-o-methyltransferase (s-COMT) as novel candidates for tumor biomarkers in urine samples. They indicated that the diagnostic capacity of these combined biomarkers was equal to or better than that of CRT alone when evaluating tumor characteristics such as size and outcome of urinary cytology. Furthermore, the concomitant use of CRT, SNCG, and s-COMT showed a higher sensitivity for the detection of BC than did the use of CRT alone.⁵²

Blood plasma/serum is another type of easily accessible body fluid that can comprehensively reflect the state of the body at the time of collection, and therefore, it can be used as a source for biomarker discovery.⁵³ This is based on the hypothesis of a complex interplay between tumor tissue and the surrounding microenvironment, resulting in an alteration to the serum protein profile.^38,54,55 Schwamborn et al.⁵⁶ identified the proteomic patterns of serum samples from 105 BC patients and 98 healthy controls with MALDI time-of-flight (TOF) MS. Multidimensional data analysis was performed to generate algorithms capable of distinguishing between cancer patients and healthy individuals based on a large number of differentially expressed proteins between 2 and 6 kDa. Their results showed a 96.4% sensitivity and 86.5% specificity. Based on this study, several combinations of markers could prove effective in the diagnosis of BC from serum testing. Lee et al.²⁷ have identified 50 plasma protein features that were significantly changed in protein expression levels with 2D-DIGE and combined this with MALDI-TOF MS/MS. Their results showed that the plasma proteins involved in inflammatory responses were up-regulated in samples, while the proteins responsible for cytoskeleton and cytoskeletal regulation were down-regulated. The newly defined marker (leucine-rich alpha-2-glycoprotein) may have potential value for BC diagnosis.

Both serum and urine have the potential to be used in novel biomarker discovery. Urine collection does not involve invasive procedures, is highly accessible, and can be measured easily. There are intrinsic factors detected in urine that could influence the pattern-based proteomic analysis and possibly influence the protein content and amount in urine. Comparatively, blood plasma/serum contains the most complex human-derived proteome, with a wide dynamic range in the abundance of proteins—a fact that results in challenges regarding the methodology. While both show promise, BC biomarker discovery studies still lack a consistent system to identify novel proteins to be introduced into clinical use. An overview of potential biomarkers identified by proteomic analysis is given in Table 1.

Table 1.

Overview of reported candidate diagnostic biomarkers.

Source of biomarker	Marker	Samples (n)	Method	Sensitivity (%)	Specificity (%)	Ref.
Urine	Coronin-1A Apolipoprotein A4 Semenogelin-2 Gamma-synuclein DJ-1/PARK7	451	LC–MS ELISA WB	79.2% 93.9%	100% 96.7%	Kumar et al.²⁶
	SAA4ProEGF	111	iTRAQLC–MSWB	NA	NA	Chen et al.⁵⁷
	Apo-A1	3	2D-MALDI-TOF-MSWB	NA	NA	Lei et al.⁵⁸
	41 Protein peaks	280	SELDI-MS	80%	100%	Majewski et al.⁵⁹
	PLK2	27	HPLC-ESI-MS/MSWB	80%	64%	Tan et al.⁶⁰
	22 Polypeptides	445	CE–MS	100	100	Theodorescu et al.⁶¹
Serum	Five protein peaks	496	MALDI-TOF-MS	96.4%	86.5%	Schwamborn et al.⁵⁶
	DynaminClusterin	311	PA	NA	NA	Orenes-Piñero et al.⁶²
	15 Peptides	106	MALDI-TOF MS	NA	NA	Villanueva et al.⁶³
	84 Plasma proteins	20	2D-DIGE-MALDI-TOF-MS	NA	NA	Lee et al.²⁷
Tissue	S100A8	24	SELDI-MSIHC	NA	NA	Tolson et al.⁶⁴
	58 Proteins	8	LC–MS	NA	NA	Niu et al.⁶⁵

LC–MS: liquid chromatography–mass spectrometry; ELISA: enzyme-linked immunosorbent assay; WB: western blot; SAA4: serum amyloid A 4; proEGF: precursor protein of epidermal growth factor; iTRAQ: isobaric tag for relative and absolute quantification MS; Apo-A1: Apolipoprotein-A1; 2D-MALDI-TOF-MS: two-dimensional material-enhanced laser desorption/ionization time-of-flight MS; SELDI-MS: surface-enhanced laser desorption/ionization-MS; PLK2: Polo-like kinase 2; HPLC–ESI-MS/MS: high-performance liquid chromatography–electrospray ionization tandem MS; CE–MS: capillary electrophoresis–MS; PA: protein arrays; 2D-DIGE: 2D-difference gel electrophoresis; S100A8: MRP-8, calgranulin A; IHC: immunohistochemistry.

Indicators for prognosis

High-quality detection methods are needed not only for initial diagnoses but also in surveillance for recurrent tumors. Based on histopathological analysis, BC is currently classified into a non-muscle-invasive type (75%), an invasive type (20%), or de novo metastasis type (5%).⁶⁶ Clinical and pathological factors such as the number of tumors, tumor size, prior-recurrence rate, T category, presence of carcinoma in situ (CIS), and tumor grade are used for the prediction of both recurrence and progression;⁶⁷ however, these parameters cannot predict with certainty the long-term outcome of the disease.⁶⁸ Although Lodde et al.⁶⁹ recently reported that positive urine cytology (odds ratio (OR): 6.8, 95% confidence interval (CI): 2.3–19.9, p = 0.01) is an independent prognostic factor of a residual tumor in a second transurethral resection of the bladder (TURB), thereby increasing the risk of the need for radical cystectomy, Horstmann et al.⁷⁰ found that the false-positive rates for cytology increased with age (p = 0.001). Therefore, the ability to identify aggressive tumors that are destined to recur and progress following initial treatment is extremely urgent.⁷¹

In the search for better biomarkers for BC prognosis, proteomic studies have utilized resected tumor tissue as a source for the discovery of a single marker or marker patterns. Hu et al.⁷² reported that nucleophosmin1 (NPM1), also called NO38, may be a potential biomarker. Using 2D-PAGE-based MALDI-TOF MS analysis, they found that NPM1 had the most prominent statistical significance among 19 up-regulated proteins in the BC cell lines PUMC-91 and PUMC-91/1.0ADM. They further examined the relationship between NPM1 and recurrent BC samples from tumors and showed that NPM1 was expressed at lower levels in late recurring (>2 years) BC tissue samples compared with samples that recurred at <6 months (p = 0.035). The higher expression of the NPM1 protein suggests that it could serve as a clinical biomarker associated with drug resistance and recurrence of BC. Additionally, Wu et al.⁷³ demonstrated that galectin-1 is up-regulated in high-grade BC compared to non-high-grade lesions. This study indicated that galectin-1 overexpression in tumors significantly predicted disease-specific survival at multivariate levels (p = 0.03, hazards ratio (HR): 2.438, 95% CI: 1.090–5.451), and immunohistochemical analysis confirmed the proteomics data and clarified the relevance of galectin-1 expression to urinary bladder urothelial carcinoma (UBUC) progression and prognosis. This suggests that galectin-1 may be a valid independent factor in BC prognosis. The proteins γ-glutamyl hydrolase (GGH), which has been associated with methotrexate resistance, and diazepam-binding inhibitor (DBI), previously identified as predictors of outcome following platinum-containing chemotherapeutic regimens have been identified as potential prognosis biomarkers. Pollard et al. reported that increased expression of GGH was observed across various tumor stages at 13.9-fold greater than normal levels in tumor tissue samples and was significantly correlated with the invasiveness of BC. Using the gene-weighted voting technique of Golub and Slonim to predict whether GGH, DBI, survivin, or emmprin expression correlated with clinical outcome, they found that DBI can successfully stratify patients treated with methotrexate, vinblastine, doxorubicin, and cisplatin (MVAC). When GGH was used with DBI in the same analysis, the prediction model had a greater statistical significance.⁷⁴

Due to compelling evidence that the heterogeneity of tumors is influenced at the genetic and epigenetic levels, multiplex biomarker panels are being advocated over single markers, which are unlikely to provide definitive stratification of patients. For example, comprehensive descriptions of the protein expression profiles within a much larger sample set consisting of 153 BC specimens were examined by Ohlsson et al. The results showed a striking down-regulation of adipocyte-type fatty acid–binding protein (A-FABP) in invasive lesions, suggesting that the aberrant level of A-FABP may play a role in BC progression.⁷⁵ Furthermore, Ohlsson et al. placed special emphasis on the combination of A-FABP with independent prognosticators, such as epidermal growth factor receptor (EGFR), E-cadherin, BC10, CK13, and gelsolin. Such combinations may be of value in predicting clinical outcomes.^76–81 Using proteomic techniques, Feldman et al. suggest that multiplexing cystatin B with cathepsins or other urinary biomarkers, such as CRT,⁸² NMP-22,^83,84 BLCA-4,^85,86 or MMPs,⁸⁷ may add significant predictive power. If these markers are involved in independent mechanisms of tumor pathogenesis, combining them may contribute additional sensitivity and specificity to urine cytology. To find new and desirable prognostic markers, SELDI-TOF MS was performed on 33 primary Ta tumors, and the data were compared to previously obtained mRNA expression profiles of 49 genes by Schultz et al. Protein peak 33331 and survivin identified three (17%) and eight (47%) patients with a recurrence-free period of at least 4 years, respectively, without generating false-negatives.⁸⁸

Because the urinary proteome reflects the urinary system microenvironment, it is an ideal noninvasive source of samples for the clinical diagnosis and prognosis of urinary system diseases. Urinary-soluble EGFR is a promising biomarker for rapidly identifying BC patients with the most aggressive forms of the disease. Bryan et al.⁸⁹ used shotgun proteomics to identify proteins released into the culture media by eight BC cell lines, and these data were compared with protein expression data from the Human Protein Atlas. As a result of this analysis, EGFR was identified as a candidate biomarker. Subsequently, via enzyme-linked immunosorbent assay (ELISA) analysis of urine from 436 BC patients, they demonstrated that EGFR is highly elevated in patients with high-grade BC. EGFR levels do appear to have prognostic value and can be predictive of BC-specific survival.

Many molecular alterations are associated with the pathology variables that predict the progression of superficial BC, such as CIS, grade 3, and T1 category. A wide variety of gene expression differences related to cell cycle regulators, proliferation antigens, cell adhesion molecules, and signaling proteins have been investigated for a better assessment of superficial BC prognosis.^10,90,91 Based on the urinary proteome database, Feng et al.⁹² used bioinformatics to construct a cancer-associated PPI network comprising 16 high-abundance urinary proteins. Following an analysis, it was shown that platelet-derived growth factor receptor beta (PDGFRB) was significantly increased in relapsed patients (n = 68) than in relapse-free patients (n = 117, p < 0.001), and it was significantly correlated with the increased risk of 3-year recurrence in superficial BC. Therefore, PDGFRB could serve as a prognostic biomarker for superficial BC recurrence. Chiang et al.⁹³ described the expression of SH3 domain–binding glutamic acid–rich protein-like 3 (SH3BGRL3), which could inhibit tumor necrosis factor (TNF)α-induced apoptosis and promote cell survival in the urine of patients with BC. They revealed that SH3BGRL3 expression is associated with an increased risk of progression in patients with superficial BC (p = 0.032), plays a role in cancer cell proliferation, epithelial–mesenchymal transition (EMT) and cell migration, and interacts with EGFR. It has also been shown to activate the Akt-associated signaling pathway. These results suggest that the SH3BGRL3 protein could serve as a potential prognostic biomarker for BC. An overview of potential prognostic biomarkers identified by proteomic analysis is given in Table 2.

Table 2.

Overview of reported candidate prognostic biomarkers.

Source of biomarker	Marker	Samples (n)	Method	Reported endpoints	Ref.
Urine	EGFREpCAM	496	ELISA	Cancer-specific survival	Bryan et al.⁸⁹
	PDGFRB	370	Bioinformatics analysisWBELISA	Superficial BC recurrence	Feng et al.⁹²
	SH3BGRL3	1190	LC–MS	Higher histologic gradingMuscle invasiveness	Chiang et al.⁹³
	Polypeptides from UROM, PGRC1, CO1A1, CO3A1	554	CE–MS	Muscle invasiveness	Schiffer et al.⁹⁴
Tissue	NPM 1	60	2D-MALDI-TOF-MS	Drug resistanceBC recurrence	Hu et al.⁷²
	GAL-1	185	IHCWB	Disease-specific survivalMuscle invasiveness	Wu et al.⁷³
	m/z 33331	33	SELDI-MS	Recurrence-free survival	Schultz et al.⁸⁸
	A-FABP	153	2D-PAGEWBIHC	Progression	Ohlsson et al.⁷⁵
	BLCAP	120	2D-PAGEWBIHC	Disease-specific survival	Moreira et al.⁹⁵

EGFR: epidermal growth factor receptor; EpCAM: epithelial cell adhesion molecule; ELISA: enzyme-linked immunosorbent assay; PDGFRB: platelet-derived growth factor receptor beta; WB: western blot; SH3BGRL3: SH3 domain–binding glutamic acid–rich protein-like 3; LC–MS: liquid chromatography–mass spectrometry; UROM: uromodulin; PGRC1: membrane associated progesterone receptor component 1; CO1A1: collagen α-1 (I) chain; CO3A1: collagen α-1 (III) chain; CE–MS: capillary electrophoresis–mass spectrometry; NPM1: nucleophosmin1; 2D-MALDI-TOF-MS: two-dimensional material-enhanced laser desorption/ionization time-of-flight MS; GAL-1: Galectin-1; IHC: immunohistochemistry; m/z 33331: mass-to-charge ratio 33331; SELDI–MS: surface-enhanced laser desorption/ionization–MS; A-FABP: adipocyte-type fatty acid–binding protein; 2D-PAGE: two-dimensional polyacrylamide gel electrophoresis; BLCAP: bladder cancer–associated protein.

Targets for oncotherapy

Proteomics provides a path for the identification of new drug targets, particularly when comparing the proteome of healthy versus diseased cells or tissues. With improved characterization of the molecular pathways driving BC, researchers have been able to identify disease biomarkers that suggest the following therapeutic angle: targeting for oncotherapy has powerful selectivity to kill tumor cells, therefore reducing the risk of injuring normal tissue. Peng et al.⁹⁶ identified 35 differentially expressed proteins by utilizing a 2D-PAGE and ESI-Q-TOF MS/MS-based proteomic method to compare and identify differentially expressed proteins in BC and adjacent normal tissues. Among these proteins, phosphoglycerate mutase 1 (PGAM1) had significantly higher expression levels in tumor samples. Furthermore, they found that attenuation of PGAM1 could up-regulate its substrate 3-PG and down-regulate the product 2-PG, thereby inhibiting the aerobic glycolytic and oxidative pentose phosphate pathways, which are essential to cancer cell proliferation. This indicated that the silencing of PGAM1 significantly induced antitumor effects, suggesting that it may serve as a potential therapeutic target for BC. While investigating the growth factor progranulin and its interacting proteins, Xu et al.⁹⁷ performed pull-down assays with recombinant progranulin and the protein extracts from 5637 BC cells. Proteomic analysis identified the F-actin-binding protein drebrin as a novel progranulin-binding partner. Furthermore, drebrin depletion in tumorigenic BC cells inhibited motility, anchorage-independent growth and tumor formation through the Akt and MAPK signaling pathways. This indicates that drebrin exerts an essential functional role in the regulation of progranulin action and may constitute a novel target for therapeutic intervention in BC. Chen et al.⁹⁸ placed special emphasis on the potential significance of NPM, a protein associated with cell proliferation, migration, and anti-apoptotic effects in bladder carcinogenesis. NPM was universally expressed in all uroepithelial cell lines examined, implying that NPM may play a role in human bladder carcinogenesis. Up-regulation of NPM tends to be dose- and time-dependent following treatment. As soy isoflavones could inhibit the expression of NPM in vitro, soybean-based foods may have potential in the suppression of As/NPM-related tumorigenesis.

Another method of treatment includes the targeting of pivotal proteins utilizing specific antibodies or coalescent ligands. This is suggested to increase the concentration of proteins related to suppressing BC formation or to lower the proportion of proteins associated with promoting BC progression. Given that cell-trafficking targets may exist in various forms, often with completely different functions within multiple cellular compartments, careful interpretation of proteomics data is needed for an accurate understanding of gene function. Not unexpectedly, proteomics could reveal novel and unforeseen biological processes with important ramifications for target validation in drug discovery.¹⁵ Green tea, one of the most widely consumed beverages worldwide, has shown promising anticancer effects on various cancers including BC.^99,100 At the laboratory level, green tea extract (GTE) and some of its major catechin components, such as epigallocatechin-3-gallate (EGCG), can produce a broad range of biological activities in various cell models, including antiproliferation, antioxidation, antiangiogenesis, apoptosis induction, and the inhibition of DNA methyltransferase.^101,102 However, there have been no specific markers identified from most of these trials. To further examine this, Xiao et al.¹⁰³ compared the proteomic profile of MC-T11 cells treated with or without GTE. Among the identified 20 GTE-induced proteins, 3 actin-binding proteins (ABPs), tropomodulin, cofilin, and annexin-I, were identified, but only annexin-I showed dose- and time-dependent expression. Immunohistochemistry of a BC tissue array showed that there is a decrease in annexin-I expression in CIS and low-grade papillary carcinoma compared to nontumor urothelium; however, it was shown to be increased in some high-grade tumors. The increased annexin-I correlated with actin remodeling and was the result of up-regulation at the transcriptional level. Furthermore, 5-Azacytidine, a DNA methylation inhibitor, exhibited no effect on annexin-I expression when used alone but had an additive effect for GTE-induced annexin-I expression.

A better understanding of the molecular mechanisms of effective treatment can lead to both the identification of biomarkers and new targets for therapy. The development of advanced technologies, such as the quantitative proteomics approach especially, offers a systematic approach for identifying active targets for tailored therapeutics in BC. Since the work of Morales et al. in 1976, intravesical Bacillus Calmette–Guérin (BCG) instillation has been the standard of care for superficial BC. Although it is known to trigger an innate immune response, the mechanism of intravesical BCG therapy is not fully understood. To further explore it, Bisiaux et al.¹⁰⁴ characterized the innate immune response to intravesical BCG therapy by analyzing 36 statistically significant changes in blood and urine derived from patients during the third instillation compared to the initial treatment. The identified proteins were classified into three categories: (1) plasma proteins that leaked into the urine, (2) cytokines/chemokines produced locally during the first hours of inflammation, and (3) other innate molecules that modulate the bladder microenvironment. These changes in the proteome were reflected in the enhanced cellular infiltrate at week 3 following a typical prime/boost pattern, suggesting that administration of multiple doses of BCG resulted in a boosted innate response. Therefore, establishing a framework of proteomic screens could improve vaccination strategies while limiting adverse events. Herpes simplex virus thymidine kinase (HSV-TK)–mediated suicide gene therapy in BC can convert the nontoxic nucleoside analog ganciclovir (GCV) into a toxic triphosphorylated form, subsequently leading to the death of rapidly dividing cells. Researchers found that Bifidobacterium infantis (BI), which selectively localizes and proliferates within the hypoxic regions of tumors as a non-pathogenic and anaerobic bacterium and TK/GCV (BI-TK/GCV) system, exhibited a sustainable antitumor growth activity in a rodent BC model in vivo, which involved both extrinsic and intrinsic apoptotic pathways. To understand the molecular mechanisms and identify the potential target protein for BI-TK/GCV treatment systems, Jiang et al.¹⁰⁵ resorted to MS-based iTRAQ to obtain comprehensive differential protein profiles following treatment with BI-TK/GCV in Sprague-Dawley rats. They found 192 down-regulated proteins, including proliferating cell nuclear antigen (PCNA), pyruvate kinase isozymes M2 (PKM2), hexokinase 1 (HXK-1), 6-phosphofructokinase (PFK-B), and cell surface glycoprotein (CD146), which indicates a decrease in cancer proliferation, metabolism, and invasion. Furthermore, Jiang et al.¹⁰⁶ confirmed that Peroxiredoxin-I (Prx-I), which had rarely been linked directly with BC, was significantly down-regulated in BC following BI-TK/GCV treatment. Silencing Prx-I could significantly suppress growth, promote apoptosis, and regulate the cell cycle in T24 cells. In addition, phospho-NF-kB, p50, and p65 protein expression has revealed links between Prx-I and NF-kB pathway, which have been indicated by Ingenuity pathway analysis (IPA). These findings yield new insights into the therapy of BC, revealing Prx-I as a new therapeutic target and identifying the BI-TK/GCV system as a prospective therapy by down-regulation of Prx-I through the NF-kB signaling pathway.

Although chemotherapy is an important tool for BC treatment, multidrug resistance (MDR) may render treatment ineffective and remains a major obstacle to overcome. Adriamycin is an important drug used for systemic and intravesical chemotherapy against advanced and superficial BC to prevent recurrence. However, due to the high rate of MDR, treatment with adriamycin or other agents often fails.^9,107 As MDR is a phenomenon whereby resistance to one anticancer drug is accompanied by resistance to drugs whose structures and mechanisms of action may be completely different, Meng et al.¹⁰⁸ established a human BC cell line that is resistant to adriamycin (pumc-91/ADM) and screened the differentially expressed proteins within the pumc-91/ADM and pumc-91 cell lines by MALDI-TOF/TOF MS. Thirty proteins were found to be differentially expressed and were further classified into 13 functional categories: transferase, oxidoreductase, transporter, transcription factor, calcium ion binding, nucleic acid binding, enzyme modulator, cytoskeletal protein, signaling molecule, ligase, hydrolase, lyase, and storage protein. Among them, annexin A2 (ANXA2), which is associated with the degradation of the extracellular matrix to facilitate cell invasion and migration, and NPM, which is highly up-regulated in proliferative cells and promotes cell growth by inhibiting tumor suppressors, may take part in the mechanism of MDR in BC. The strategy for the analysis of the urinary proteome is given in Figure 1.

Figure 1.

Pathways to personalized cancer management by proteomic profiling. This process improves the predictive ability of clinicopathologic information or survival prognosis. Meanwhile, it offers systematic approaches to identify biomarkers expressed in different issues, which may help screen for active targets for drug discovery and tailored therapeutics.

Conclusion and outlook

The Human Proteome Organization (HUPO) was founded in 2001 with a goal to organize data from multiple laboratories and to facilitate scientific collaborations. In 2007, HUPO initiated the Human Kidney and Urine Proteome Project (HKUPP) to better understand kidney function and disease. In addition, it has worked to establish standard protocols and guidelines for the proteomic analysis of urine. With the era of proteomics being ushered in, the door is wide open for the development of methods related to the prevention, diagnosis, treatment, and improved prognosis of BC. A series of questions must yet be resolved through ongoing proteomic analyses to pave the way for future research of BC: how best to detect the various types of proteins which may be extremely acidic, highly alkaline, slightly soluble, of low abundance or of low copy number in any cell or tissue? How best to develop an assay to identify one or a group of specific proteins for prevention screening of seemingly healthy individuals who may fall within high-risk groups for BC? How best to increase the accuracy and specificity of proteomic indicators with respect to both diagnosis and prognosis, to convert basic laboratorial research into clinical applications, both practicably and expediently? How best to implement the idea of targeted therapy, making use of proteomic indicators, and practicably using that information for the development of molecular therapeutics?

Current proteomic-based approaches have successfully identified potential biomarkers for prostate cancer, BC, and renal cell carcinoma that reflect a diverse number of key cellular processes, including modification of the extracellular environment, invasion and metastasis, chemotaxis, differentiation, metabolite transport, and apoptosis.

It is known that reliable diagnosis and patient stratification requires the use of a multiple biomarker panel to enable the correct classification of patients, especially given the multiple cancer subtypes and different underlying and evolving genetic changes and epigenetic influences, which is certainly a work in progress. There are many challenges involved in the translation of potential biomarkers to clinical use, including assay development, the need for robust evidence of clinical benefit, health economic studies, and patient acceptability. Multicenter evaluation of biomarkers in clinical patient populations is becoming more common for urological cancers, and some markers (annexin A3, BTA-Stat, and NMP22) have completed the initial confirmatory stages of the biomarker pipeline.

In the future, urinary biomarkers can be used to complement the emerging findings of genomic studies, aiding in the overall improvement of the management of urological cancers, ranging from earlier diagnosis to the selection of optimal treatment. The systematic approach to the discovery of proteomic profiling will allow identification of patterns of molecules predictive of prognosis as well as response to chemotherapy or BI-TK/GCV treatment. Additional refinement and subsequent validation of novel biomarkers by independent cohorts will provide target molecules for assays that more accurately predict prognosis and tumor response to specific chemotherapeutic agents, further contributing to the development of “personalized medicine” for BC patients.

Footnotes

Acknowledgements

H.Z. and Y.F. contributed equally to this work.

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

This work was supported by the National Natural Science Foundation of China (81472746).

References

Hemelt

Yamamoto

Cheng

. The effect of smoking on the male excess of bladder cancer: a meta-analysis and geographical analyses. Int J Cancer 2009; 124: 412–419.

Zargar

Espiritu

Fairey

. Multicenter assessment of neoadjuvant chemotherapy for muscle-invasive bladder cancer. Eur Urol 2015; 67: 241–249.

Bassett

Gore

Kwan

. Knowledge of the harms of tobacco use among patients with bladder cancer. Cancer 2014; 120: 3914–3922.

Cancer Genome Atlas Research Network. Comprehensive molecular characterization of urothelial bladder carcinoma. Nature 2014; 507: 315–322.

Eifler

Scarpato

Clark

PE.

Management of noninvasive bladder cancers. Curr Opin Oncol 2015; 27: 185–190.

Kaufman

Shipley

Feldman

AS.

Bladder cancer. Lancet 2009; 374: 239–249.

Svatek

Hollenbeck

Holmäng

. The economics of bladder cancer: costs and considerations of caring for this disease. Eur Urol 2014; 66: 253–262.

Zargar

Aning

Ischia

. Optimizing intravesical mitomycin C therapy in non-muscle-invasive bladder cancer. Nat Rev Urol 2014; 11: 220–230.

Gingrich

JR.

Bladder cancer: chemohyperthermia for bladder cancer–clinically effective. Nat Rev Urol 2011; 8: 414–416.

10.

Izumi

Ito

Miyamoto

. Expression of androgen receptor in non-muscle-invasive bladder cancer predicts the preventive effect of androgen deprivation therapy on tumor recurrence. Oncotarget 2016; 7(12): 14153–14160.

11.

Hachey

Chaurand

Proteomics in reproductive medicine: the technology for separation and identification of proteins. J Reprod Immunol 2004; 63: 61–73.

12.

Aebersold

Mann

Mass spectrometry-based proteomics. Nature 2003; 422: 198–207.

13.

Davalieva

Maleva

Dwork

AJ.

Proteomics research in schizophrenia. Front Cell Neurosci 2016; 10: 18.

14.

Shimwell

Bryan

Wei

. Combined proteome and transcriptome analyses for the discovery of urinary biomarkers for urothelial carcinoma. Br J Cancer 2013; 108: 1854–1861.

15.

Gstaiger

Aebersold

Applying mass spectrometry-based proteomics to genetics, genomics and network biology. Nat Rev Genet 2009; 10: 617–627.

16.

Kabbage

Trimeche

Ben

. Expression of the molecular chaperone αB-crystallin in infiltrating ductal breast carcinomas and the significance thereof: an immunohistochemical and proteomics-based strategy. Tumour Biol 2012; 33: 2279–2288.

17.

Schiffer

Mischak

Zimmerli

LU.

Proteomics in gerontology: current applications and future aspects – a mini-review. Gerontology 2009; 55: 123–137.

18.

Bhavsar

Auweter

Finlay

BB.

Proteomics as a probe of microbial pathogenesis and its molecular boundaries. Future Microbiol 2010; 5: 253–265.

19.

Deng

Zhou

. iTRAQ-based quantitative proteomic analysis of esophageal squamous cell carcinoma. Tumour Biol 2015; 37(2): 1909–1918.

20.

Celis

Gromova

Moreira

. Impact of proteomics on bladder cancer research. Pharmacogenomics 2004; 5: 381–394.

21.

Liu

Ehrlich

JR.

Proteomics approaches to urologic diseases. Expert Rev Proteomics 2006; 3: 283–296.

22.

Schiffer

Mischak

Theodorescu

. Challenges of using mass spectrometry as a bladder cancer biomarker discovery platform. World J Urol 2008; 26: 67–74.

23.

Shirodkar

Lokeshwar

VB.

Potential new urinary markers in the early detection of bladder cancer. Curr Opin Urol 2009; 19: 488–493.

24.

Gaston

Grossman

HB.

Proteomic assays for the detection of urothelial cancer. Methods Mol Biol 2010; 641: 303–323.

25.

Peng

Gong

Chen

. Proteomics identification of PGAM1 as a potential therapeutic target for urothelial bladder cancer. J Proteomics 2016; 132: 85–92.

26.

Kumar

Nandi

Tan

. Highly sensitive and specific novel biomarkers for the diagnosis of transitional bladder carcinoma. Oncotarget 2015; 6: 13539–13549.

27.

Lee

Chen

Tsai

. Redox- and expression-proteomic analysis of plasma biomarkers in bladder transitional cell carcinoma. Mol Biosyst 2012; 8: 3314–3324.

28.

Zurbig

Renfrow

Schiffer

. Biomarker discovery by CE-MS enables sequence analysis via MS/MS with platform-independent separation. Electrophoresis 2006; 27: 2111–2125.

29.

Schaub

Wilkins

Weiler

. Urine protein profiling with surface-enhanced laser-desorption/ionization time-of-flight mass spectrometry. Kidney Int 2004; 65: 323–332.

30.

Mischak

Coon

Novak

. Capillary electrophoresis-mass spectrometry as a powerful tool in biomarker discovery and clinical diagnosis: an update of recent developments. Mass Spectrom Rev 2009; 28: 703–724.

31.

Gygi

Aebersold

Mass spectrometry and proteomics. Curr Opin Chem Biol 2000; 4: 489–494.

32.

Issaq

HJ.

The role of separation science in proteomics research. Electrophoresis 2001; 22: 3629–3638.

33.

Chen

Hewel

Felding-Habermann

. Large scale protein profiling by combination of protein fractionation and multidimensional protein identification technology (MudPIT). Mol Cell Proteomics 2006; 5: 53–56.

34.

Theodorescu

Mischak

Mass spectrometry based proteomics in urine biomarker discovery. World J Urol 2007; 25: 435–443.

35.

Fliser

Novak

Thongboonkerd

. Advances in urinary proteome analysis and biomarker discovery. J Am Soc Nephrol 2007; 18: 1057–1071.

36.

Johannesson

Wetterhall

Markides

. Monomer surface modifications for rapid peptide analysis by capillary electrophoresis and capillary electrochromatography coupled to electrospray ionization-mass spectrometry. Electrophoresis 2004; 25: 809–816.

37.

Hernandez-Borges

Neususs

Cifuentes

. On-line capillary electrophoresis-mass spectrometry for the analysis of biomolecules. Electrophoresis 2004; 25: 2257–2281.

38.

Schwamborn

Gaisa

Henkel

Tissue and serum proteomic profiling for diagnostic and prognostic bladder cancer biomarkers. Expert Rev Proteomics 2010; 7: 897–906.

39.

Haab

Dunham

Brown

PO.

Protein microarrays for highly parallel detection and quantitation of specific proteins and antibodies in complex solutions. Genome Biol 2001; 2: RESEARCH0004.

40.

Kozarova

Petrinac

Ali

. Array of informatics: applications in modern research. J Proteome Res 2006; 5: 1051–1059.

41.

Brown

FM.

Urine cytology. It is still the gold standard for screening. Urol Clin North Am 2000; 27: 25–37.

42.

Cajulis

Haines

3rd Frias-Hidvegi

. Cytology, flow cytometry, image analysis, and interphase cytogenetics by fluorescence in situ hybridization in the diagnosis of transitional cell carcinoma in bladder washes: a comparative study. Diagn Cytopathol 1995; 13: 214–223; discussion 224.

43.

Wakui

Shiigai

Urinary tract cancer screening through analysis of urinary red blood cell volume distribution. Int J Urol 2000; 7: 248–253.

44.

Ramakumar

Bhuiyan

Besse

. Comparison of screening methods in the detection of bladder cancer. J Urol 1999; 161: 388–394.

45.

Halling

King

Sokolova

. A comparison of BTA stat, hemoglobin dipstick, telomerase and Vysis UroVysion assays for the detection of urothelial carcinoma in urine. J Urol 2002; 167: 2001–2006.

46.

Schroeder

Lorenzo-Gomez

Hautmann

. A side by side comparison of cytology and biomarkers for bladder cancer detection. J Urol 2004; 172: 1123–1126.

47.

Wood

Knowles

Thompson

. Proteomic studies of urinary biomarkers for prostate, bladder and kidney cancers. Nat Rev Urol 2013; 10: 206–218.

48.

Strong

Osicka

Comper

WD.

Urinary-peptide excretion by patients with and volunteers without diabetes. J Lab Clin Med 2005; 145: 239–246.

49.

Kreunin

Zhao

Rosser

. Bladder cancer associated glycoprotein signatures revealed by urinary proteomic profiling. J Proteome Res 2007; 6: 2631–2639.

50.

Saito

Kimoto

Araki

. Proteome analysis of gelatin-bound urinary proteins from patients with bladder cancers. Eur Urol 2005; 48: 865–871.

51.

Kageyama

Isono

Iwaki

. Identification by proteomic analysis of calreticulin as a marker for bladder cancer and evaluation of the diagnostic accuracy of its detection in urine. Clin Chem 2004; 50: 857–866.

52.

Iwaki

Kageyama

Isono

. Diagnostic potential in bladder cancer of a panel of tumor markers (calreticulin, gamma -synuclein, and catechol-o-methyltransferase) identified by proteomic analysis. Cancer Sci 2004; 95: 955–961.

53.

Anderson

NG.

The human plasma proteome: history, character, and diagnostic prospects. Mol Cell Proteomics 2002; 1: 845–867.

54.

Liotta

Kohn

EC.

The microenvironment of the tumour-host interface. Nature 2001; 411: 375–379.

55.

Norden

Rodriguez-Cutillas

Unwin

RJ.

Clinical urinary peptidomics: learning to walk before we can run. Clin Chem 2007; 53: 375–376.

56.

Schwamborn

Krieg

Grosse

. Serum proteomic profiling in patients with bladder cancer. Eur Urol 2009; 56: 989–996.

57.

Chen

Lin

Tsai

. Identification of potential bladder cancer markers in urine by abundant-protein depletion coupled with quantitative proteomics. J Proteomics 2013; 85: 28–43.

58.

Lei

Zhao

Jin

. Discovery of potential bladder cancer biomarkers by comparative urine proteomics and analysis. Clin Genitourin Cancer 2013; 11: 56–62.

59.

Majewski

Spiess

Bondaruk

. Detection of bladder cancer using proteomic profiling of urine sediments. PLoS ONE 2012; 7: e42452.

60.

Tan

Chen

Yuan

. Identification of urine PLK2 as a marker of bladder tumors by proteomic analysis. World J Urol 2010; 28: 117–122.

61.

Theodorescu

Wittke

Ross

. Discovery and validation of new protein biomarkers for urothelial cancer: a prospective analysis. Lancet Oncol 2006; 7: 230–240.

62.

Orenes-Piñero

Barderas

Rico

. Serum and tissue profiling in bladder cancer combining protein and tissue arrays. J Proteome Res 2009; 9(1): 164–173.

63.

Villanueva

Shaffer

Philip

. Differential exoprotease activities confer tumor-specific serum peptidome patterns. J Clin Invest 2006; 116: 271–284.

64.

Tolson

Flad

Gnau

. Differential detection of S100A8 in transitional cell carcinoma of the bladder by pair wise tissue proteomic and immunohistochemical analysis. Proteomics 2006; 6: 697–708.

65.

Niu

Zhang

Jiang

. Differences in shotgun protein expression profile between superficial bladder transitional cell carcinoma and normal urothelium. Urol Oncol 2009; 27: 400–406.

66.

Donat

SM.

Evaluation and follow-up strategies for superficial bladder cancer. Urol Clin North Am 2003; 30: 765–776.

67.

Sylvester

RJ.

Natural history, recurrence, and progression in superficial bladder cancer. Sci World J 2006; 6: 2617–2625.

68.

Sidransky

Messing

Molecular genetics and biochemical mechanisms in bladder cancer. Oncogenes, tumor suppressor genes, and growth factors. Urol Clin North Am 1992; 19: 629–639.

69.

Lodde

Mayr

Martini

. Positive urine cytology and carcinoma in situ prior to second transurethral resection of the bladder correlate with positive second resection histology and the need for subsequent cystectomy. World J Urol 2012; 30: 841–846.

70.

Horstmann

Todenhöfer

Hennenlotter

. Influence of age on false positive rates of urine-based tumor markers. World J Urol 2013; 31: 935–940.

71.

Stein

Lieskovsky

Cote

. Radical cystectomy in the treatment of invasive bladder cancer: long-term results in 1,054 patients. J Clin Oncol 2001; 19: 666–675.

72.

Meng

Lei

. Nucleophosmin1 associated with drug resistance and recurrence of bladder cancer. Clin Exp Med 2015; 15: 361–369.

73.

Chien

. Galectin-1 dysregulation independently predicts disease specific survival in bladder urothelial carcinoma. J Urol 2015; 193: 1002–1008.

74.

Pollard

Nitz

Baras

. Genoproteomic mining of urothelial cancer suggests {gamma}-glutamyl hydrolase and diazepam-binding inhibitor as putative urinary markers of outcome after chemotherapy. Am J Pathol 2009; 175: 1824–1830.

75.

Ohlsson

Moreira

Gromov

. Loss of expression of the adipocyte-type fatty acid-binding protein (A-FABP) is associated with progression of human urothelial carcinomas. Mol Cell Proteomics 2005; 4: 570–581.

76.

Celis

Ostergaard

Basse

. Loss of adipocyte-type fatty acid binding protein and other protein biomarkers is associated with progression of human bladder transitional cell carcinomas. Cancer Res 1996; 56: 4782–4790.

77.

Gromova

Gromov

Celis

JE.

bc10: a novel human bladder cancer-associated protein with a conserved genomic structure downregulated in invasive cancer. Int J Cancer 2002; 98: 539–546.

78.

Cordon-Cardo

Molecular alterations in bladder cancer. Cancer Surv 1998; 32: 115–131.

79.

Byrne

Shariat

Brown

. E-cadherin immunostaining of bladder transitional cell carcinoma, carcinoma in situ and lymph node metastases with long-term followup. J Urol 2001; 165: 1473–1479.

80.

Bellmunt

Hussain

Dinney

CP.

Novel approaches with targeted therapies in bladder cancer. Therapy of bladder cancer by blockade of the epidermal growth factor receptor family. Crit Rev Oncol Hematol 2003; 46(Suppl.): S85–S104.

81.

Rao

Seligson

Visapaa

. Tissue microarray analysis of cytoskeletal actin-associated biomarkers gelsolin and E-cadherin in urothelial carcinoma. Cancer 2002; 95: 1247–1257.

82.

Dube

Bertozzi

CR.

Glycans in cancer and inflammation – potential for therapeutics and diagnostics. Nat Rev Drug Discov 2005; 4: 477–488.

83.

Prakash

Robbins

PW.

Glycotyping of prostate specific antigen. Glycobiology 2000; 10: 173–176.

84.

Peracaula

Tabares

Royle

. Altered glycosylation pattern allows the distinction between prostate-specific antigen (PSA) from normal and tumor origins. Glycobiology 2003; 13: 457–470.

85.

Ishioka

Takahashi

Putnam

FW.

Amino acid sequence of human plasma alpha 1B-glycoprotein: homology to the immunoglobulin supergene family. Proc Natl Acad Sci U S A 1986; 83: 2363–2367.

86.

Catanese

Kress

LF.

Isolation from opossum serum of a metalloproteinase inhibitor homologous to human alpha 1B-glycoprotein. Biochemistry 1992; 31: 410–418.

87.

Chirco

Liu

Jung

. Novel functions of TIMPs in cell signaling. Cancer Metastasis Rev 2006; 25: 99–113.

88.

Schultz

De Kok

Witjes

. Simultaneous proteomic and genomic analysis of primary Ta urothelial cell carcinomas for the prediction of tumor recurrence. Anticancer Res 2007; 27: 1051–1058.

89.

Bryan

Regan

Pirrie

. Protein shedding in urothelial bladder cancer: prognostic implications of soluble urinary EGFR and EpCAM. Br J Cancer 2015; 112: 1052–1058.

90.

Chen

Chang

. The role of PD-L1 in the radiation response and clinical outcome for bladder cancer. Sci Rep 2016; 6: 19740.

91.

Lee

Roh

Kim

. Activation of EZH2 and SUZ12 regulated by E2F1 predicts the disease progression and aggressive characteristics of bladder cancer. Clin Cancer Res 2015; 21: 5391–5403.

92.

Feng

Song

. Platelet-derived growth factor receptor beta: a novel urinary biomarker for recurrence of non-muscle-invasive bladder cancer. PLoS ONE 2014; 9: e96671.

93.

Chiang

Pan

Chang

. SH3BGRL3 protein as a potential prognostic biomarker for urothelial carcinoma: a novel binding partner of epidermal growth factor receptor. Clin Cancer Res 2015; 21: 5601–5611.

94.

Schiffer

Vlahou

Petrolekas

. Prediction of muscle-invasive bladder cancer using urinary proteomics. Clin Cancer Res 2009; 15: 4935–4943.

95.

Moreira

Ohlsson

Gromov

. Bladder cancer-associated protein, a potential prognostic biomarker in human bladder cancer. Mol Cell Proteomics 2010; 9: 161–177.

96.

Peng

Gong

Chen

. Proteomics identification of PGAM1 as a potential therapeutic target for urothelial bladder cancer. J Proteomics 2016; 132: 85–92.

97.

Buraschi

Morcavallo

. A novel role for drebrin in regulating progranulin bioactivity in bladder cancer. Oncotarget 2015; 6: 10825–10839.

98.

Chen

Wang

Hsu

. Nucleophosmin in the pathogenesis of arsenic-related bladder carcinogenesis revealed by quantitative proteomics. Toxicol Appl Pharmacol 2010; 242: 126–135.

99.

Kuroda

Hara

Antimutagenic and anticarcinogenic activity of tea polyphenols. Mutat Res 1999; 436: 69–97.

100.

Kohlmeier

Weterings

Steck

. Tea and cancer prevention: an evaluation of the epidemiologic literature. Nutr Cancer 1997; 27: 1–13.

101.

Moyers

Kumar

NB.

Green tea polyphenols and cancer chemoprevention: multiple mechanisms and endpoints for phase II trials. Nutr Rev 2004; 62: 204–211.

102.

Fang

Wang

. Tea polyphenol (-)-epigallocatechin-3-gallate inhibits DNA methyltransferase and reactivates methylation-silenced genes in cancer cell lines. Cancer Res 2003; 63: 7563–7570.

103.

Xiao

Jin

. Annexin-I as a potential target for green tea extract induced actin remodeling. Int J Cancer 2007; 120: 111–120.

104.

Bisiaux

Thiounn

Timsit

. Molecular analyte profiling of the early events and tissue conditioning following intravesical bacillus calmette-guerin therapy in patients with superficial bladder cancer. J Urol 2009; 181: 1571–1580.

105.

Jiang

Ren

Xiao

. Proteomic analysis of bladder cancer by iTRAQ after Bifidobacterium infantis-mediated HSV-TK/GCV suicide gene treatment. Biol Chem 2013; 394: 1333–1342.

106.

Jiang

Xiao

Ren

. Proteomic analysis of bladder cancer indicates Prx-I as a key molecule in BI-TK/GCV treatment system. PLoS ONE 2014; 9: e98764.

107.

Baguley

BC.

Multidrug resistance in cancer. Methods Mol Biol 2010; 596: 1–14.

108.

Meng

Lei

Zhang

. Identification of proteins differentially expressed in adriamycin-resistant (pumc-91/ADM) and parental (pumc-91) human bladder cancer cell lines by proteome analysis. J Cancer Res Clin Oncol 2013; 139: 509–519.