Neural network using combined urine nuclear matrix protein-22, monocyte chemoattractant protein-1 and urinary intercellular adhesion molecule-1 to detect bladder cancer

Examining longitudinal markers of bladder cancer recurrence through a semiautonomous machine learning system for quantifying specimen atypia from urine cytology

BACKGROUND

Urine cytology is generally considered the primary approach for screening for recurrence of bladder cancer. However, it is currently unclear how best to use cytological examinations for assessment and early detection of recurrence, beyond identifying a positive finding that requires more invasive methods to confirm recurrence and decide on therapeutic options. Because screening programs are frequent, and can be burdensome, finding quantitative means to reduce this burden for patients, cytopathologists, and urologists is an important endeavor and can improve both the efficiency and reliability of findings. Additionally, identifying ways to risk-stratify patients is crucial for improving quality of life while reducing the risk of future recurrence or progression of the cancer.

METHODS

In this study, a computational machine learning tool, AutoParis-X, was leveraged to extract imaging features from urine cytology examinations longitudinally to study the predictive potential of urine cytology for assessing recurrence risk. This study examined how the significance of imaging predictors changes over time before and after surgery to determine which predictors and time periods are most relevant for assessing recurrence risk.

RESULTS

Results indicate that imaging predictors extracted using AutoParis-X can predict recurrence as well or better than traditional cytological/histological assessments alone and that the predictiveness of these features is variable across time, with key differences in overall specimen atypia identified immediately before tumor recurrence.

CONCLUSIONS

Further research will clarify how computational methods can be effectively used in high-volume screening programs to improve recurrence detection and complement traditional modes of assessment.

Diagnostic performance of nuclear matrix protein 22 and urine cytology for bladder cancer: A meta-analysis

PURPOSE

To compare and analyze the diagnostic efficacy of nuclear matrix protein 22 (NMP22) and urine cytology (UC) in the diagnosis of bladder cancer.

METHODS

Search the Chinese and English studies on NMP22 and urinary cytology in the diagnosis of bladder tumors published between 1999 and June, and conduct quality evaluation, data extraction and analysis.

RESULTS

A total of 397 related articles were retrieved, and 12 articles were finally included after screening, including 2456 subjects. The heterogeneity test shows that there is no discernible threshold effect. Perform meta-analysis according to the random effects model. The results showed that the total sensitivity of NMP22 and UC were 0.79 (95% CI [0.73, 0.84]) (CI: Confidence interval), 0.55 (95% CI [0.41, 0.69]), and the total specificity 0.59 (95% CI [0.46], respectively, 0.71), 0.91 (95% CI (0.81, 0.96]), +LR 1.9 (95% CI [1.4, 2.6]) (+LR: positive likelihood ration), 5.9 (95% CI [3.3, 10.6]), -LR 0.35 (-LR: negative likelihood ration), respectively (95% CI [0.27, 0.47]), 0.49 (95% CI [0.38, 0.64]), diagnostic odds ratios 5 (95% CI [3, 9]), 12 (95% CI [7, 21]). The area under the summary receiver operating characteristics curve (AUC) was 0.79 (95% CI [0.75, 0.82]) and 0.81 (95% CI [0.77, 0.84]), respectively.

CONCLUSIONS

NMP22 has moderate diagnostic efficiency for bladder cancer. Its sensitivity is greater than UC, but its specificity is significantly lower than that of UC. At present, it cannot replace traditional cystoscopy and UC, but it can be combined to detect bladder tumors. It plays a major role in screening, postoperative monitoring and follow-up.

A systematic review of the applications of Expert Systems (ES) and machine learning (ML) in clinical urology

BACKGROUND

Testing a hypothesis for 'factors-outcome effect' is a common quest, but standard statistical regression analysis tools are rendered ineffective by data contaminated with too many noisy variables. Expert Systems (ES) can provide an alternative methodology in analysing data to identify variables with the highest correlation to the outcome. By applying their effective machine learning (ML) abilities, significant research time and costs can be saved. The study aims to systematically review the applications of ES in urological research and their methodological models for effective multi-variate analysis. Their domains, development and validity will be identified.

METHODS

The PRISMA methodology was applied to formulate an effective method for data gathering and analysis. This study search included seven most relevant information sources: WEB OF SCIENCE, EMBASE, BIOSIS CITATION INDEX, SCOPUS, PUBMED, Google Scholar and MEDLINE. Eligible articles were included if they applied one of the known ML models for a clear urological research question involving multivariate analysis. Only articles with pertinent research methods in ES models were included. The analysed data included the system model, applications, input/output variables, target user, validation, and outcomes. Both ML models and the variable analysis were comparatively reported for each system.

RESULTS

The search identified n = 1087 articles from all databases and n = 712 were eligible for examination against inclusion criteria. A total of 168 systems were finally included and systematically analysed demonstrating a recent increase in uptake of ES in academic urology in particular artificial neural networks with 31 systems. Most of the systems were applied in urological oncology (prostate cancer = 15, bladder cancer = 13) where diagnostic, prognostic and survival predictor markers were investigated. Due to the heterogeneity of models and their statistical tests, a meta-analysis was not feasible.

CONCLUSION

ES utility offers an effective ML potential and their applications in research have demonstrated a valid model for multi-variate analysis. The complexity of their development can challenge their uptake in urological clinics whilst the limitation of the statistical tools in this domain has created a gap for further research studies. Integration of computer scientists in academic units has promoted the use of ES in clinical urological research.

Economic evaluations of big data analytics for clinical decision-making: a scoping review

OBJECTIVE

Much has been invested in big data analytics to improve health and reduce costs. However, it is unknown whether these investments have achieved the desired goals. We performed a scoping review to determine the health and economic impact of big data analytics for clinical decision-making.

MATERIALS AND METHODS

We searched Medline, Embase, Web of Science and the National Health Services Economic Evaluations Database for relevant articles. We included peer-reviewed papers that report the health economic impact of analytics that assist clinical decision-making. We extracted the economic methods and estimated impact and also assessed the quality of the methods used. In addition, we estimated how many studies assessed "big data analytics" based on a broad definition of this term.

RESULTS

The search yielded 12 133 papers but only 71 studies fulfilled all eligibility criteria. Only a few papers were full economic evaluations; many were performed during development. Papers frequently reported savings for healthcare payers but only 20% also included costs of analytics. Twenty studies examined "big data analytics" and only 7 reported both cost-savings and better outcomes.

DISCUSSION

The promised potential of big data is not yet reflected in the literature, partly since only a few full and properly performed economic evaluations have been published. This and the lack of a clear definition of "big data" limit policy makers and healthcare professionals from determining which big data initiatives are worth implementing.

Bladder cancer in the time of machine learning: Intelligent tools for diagnosis and management

Machine learning (ML) is the subfield of artificial intelligence (AI), born from the marriage between statistics and computer science, with the unique purpose of building prediction algorithms able to improve their performances by automatically learning from massive data sets. The availability of ever-growing computational power and highly evolved pattern recognition software has led to the spread of ML-based systems able to perform complex tasks in bioinformatics, medical imaging, and diagnostics. These intelligent tools could be the answer to the unmet need for non-invasive and patient-tailored instruments for the diagnosis and management of bladder cancer (BC), which are still based on old technologies and unchanged nomograms. We reviewed the most significant evidence on ML in the diagnosis, prognosis, and management of bladder cancer, to find out if these intelligent technologies are ready to be introduced into the daily clinical practice of the urologist.

Using multi-layer perceptron with Laplacian edge detector for bladder cancer diagnosis

In this paper, the urinary bladder cancer diagnostic method which is based on Multi-Layer Perceptron and Laplacian edge detector is presented. The aim of this paper is to investigate the implementation possibility of a simpler method (Multi-Layer Perceptron) alongside commonly used methods, such as Deep Learning Convolutional Neural Networks, for the urinary bladder cancer detection. The dataset used for this research consisted of 1997 images of bladder cancer and 986 images of non-cancer tissue. The results of the conducted research showed that using Multi-Layer Perceptron trained and tested with images pre-processed with Laplacian edge detector are achieving AUC value up to 0.99. When different image sizes are compared it can be seen that the best results are achieved if 50×50 and 100×100 images were used.

Evaluation of urinary and serum level of chemokine (C-C motif) ligand 2 as a potential biomarker in canine urothelial tumours

Chemokine (C-C motif) ligand 2 (CCL2) is a chemotactic cytokine recruiting monocytes, releasing growth factors and promoting adhesion in vascular endothelium. Elevated serum and urinary CCL2 levels and expression of its receptor (CCR2) have been associated with tumorigenesis in human urinary malignancies. CCL2 implication has not been investigated in canine urothelial carcinoma. The aim of this study was to evaluate CCL2 serum and urine levels (measured by ELISA) in dogs with urothelial carcinoma or non-neoplastic urinary tract disease. CCL2 serum and urine levels were significantly higher in diseased dogs compared with healthy dogs (P < 0.001). Dogs with carcinoma had significantly higher serum and urine CCL2 levels (P = 0.001) than healthy dogs. Dogs with metastases showed significantly lower serum and urine CCL2 levels compared with the non-metastasised tumour group (P = 0.007). CCL2 as a diagnostic marker for urothelial carcinoma held a sensitivity of 95.2% and a specificity of 38.2% in the urine. As a staging marker, sensitivity was 85.7% and specificity was 57.1% with a positive predictive value of 75.7% and a negative predictive value of 71.9%. Further investigation is needed to define the role of CCL2 as a prognostic marker in canine urothelial carcinoma.

The negative predictive value of a negative repeat urinalysis in patients presenting with haematuria: A review of 1138 patients

Objective:

Haematuria may be transient for a number of benign conditions, particularly a urinary-tract infection (UTI). We set out to determine if a negative repeat urinalysis at the time of urological assessment for patients with haematuria could predict negative investigations and whether investigations could be tailored by this test.

Methods:

This was a retrospective analysis of records for all patients attending a haematuria clinic between 16 September 2013 and 12 September 2014. This included patients with visible and non-visible (microscopic) haematuria.

Results:

There were 1138 patients, 599 with visible haematuria (VH) and 460 with non-visible haematuria (NVH). Seventy-two patients were excluded. A total of 546 patients had a positive repeat urinalysis for blood; 438 patients had a negative repeat urinalysis when tested at the haematuria clinic, 298/599 for VH and 140/460 NVH. For those who had negative repeat urinalysis, urothelial cancer was found in 15/298 VH and 1/140 NVH. The one patient with negative repeat urinalysis and NVH was found to have a grade 2 (high grade) bladder tumour. The negative predictive value for a negative repeat urinalysis in transient haematuria was 0.95 for VH and 0.99 for NVH. Twenty-nine patients with VH and repeat negative urinalysis on assessment had a positive urine culture suggesting a UTI as a cause. None of these patients was found to have urothelial cancer ( p = 0.0413).

Conclusion:

Patients who experience transient VH and subsequent repeat negative urinalysis in the absence of infection have a 5% chance of urothelial cancer and should still be investigated. For those with transient NVH, the probability of finding a urothelial cancer is <1%, although we did find a high-grade bladder tumour in this group. If patients have a positive urine culture and a negative repeat urinalysis following treatment, they could be spared haematuria investigations.

Medical follow-up for workers exposed to bladder carcinogens: the French evidence-based and pragmatic statement

BACKGROUND

The aim of this work was to establish recommendations for the medical follow-up of workers currently or previously exposed to carcinogenic substances for the bladder.

METHODS

A critical synthesis of the literature was conducted. Sectors of activity where workers are or were exposed to carcinogenic substances for the bladder were listed and classified according to the level of bladder cancer risk. Performances of techniques available for the targeted screening of bladder cancer were analysed, including a simulation of results among high-risk populations in France.

RESULTS

The risk level for the professional group and the latency period between the start of exposure and the natural history of the disease were selected to define a targeted screening protocol. The NMP22BC test, exclusive haematuria testing, and combinations of urine cytology with, respectively, the NMP22BC test and haematuria test, generated an extremely high proportion of false positive results.

CONCLUSION

Urine cytology is the test that offers the best specificity. Although poor for all bladder cancer stages and grades combined, its sensitivity is better for high grades, which require early diagnosis since late-stage cancers are of very poor prognosis. These results suggest that urine cytology is currently the only technique suitable for proposal within the context of a first line targeted screening strategy for occupational bladder cancer. An algorithm summarising the recommended medical follow-up for workers currently or previously exposed to carcinogenic substances for the bladder is proposed, based on the level of risk of bladder cancer.

Impacts of ICAM-1 gene polymorphisms on urothelial cell carcinoma susceptibility and clinicopathologic characteristics in Taiwan

Intercellular adhesion molecule (ICAM)-1, a cell adhesion molecule, is reportedly overexpressed in several cancers and may contribute to tumorgenesis and metastasis. The current study explored the effect of ICAM-1 gene polymorphisms on the susceptibility of developing urothelial cell carcinoma (UCC) and the clinicopathological status. A total of 558 participants, including 279 healthy people and 279 patients with UCC, were recruited for this study. Four single-nucleotide polymorphisms of the ICAM-1 gene were assessed by a real-time polymerase chain reaction with the TaqMan assay. After adjusting for other covariants, the individuals carrying at least one G allele at ICAM-1 rs5498 had a 1.603-fold risk of developing UCC than did wild-type (AA) carriers. Furthermore, UCC patients who carried at least one G allele at rs5498 had a higher invasive stage risk (p < 0.05) than did patients carrying the wild-type allele. In conclusion, the rs5498 polymorphic genotypes of ICAM-1 might contribute to the prediction of susceptibility to and pathological development of UCC. This is the first study to provide insight into risk factors associated with ICAM-1 variants in carcinogenesis of UCC in Taiwan.

Urinary markers in the everyday diagnosis of bladder cancer

Bladder cancer (BC) represents the fourth most common neoplasia in men and the ninth most common cancer in women, with a significant morbidity and mortality. Cystoscopy and voided urine cytology (involving the examination of cells in voided urine to detect the presence of cancerous cells) are currently the routine initial investigations in patients with hematuria or other symptoms suggestive of BC. Around 75-85% of the patients are diagnosed as having non-muscle-invasive bladder cancer (NMIBC). Despite the treatment, these patients have a probability of recurrence at 5 years ranging from 50 to 70% and of progression to muscle invasive disease of 10-15%. Patients with NMIBC must undergo life-long surveillance, consisting of serial cystoscopies, possibly urine cytology and ultrasonography. Cystoscopy is unsuitable for screening because of its invasiveness and costs; serial cystoscopies may cause discomfort and distress to patients. Furthermore, cystoscopy may be inconclusive, falsely positive or negative. Although urine cytology has a reasonable sensitivity for the detection of high-grade BC, it lacks sensitivity to detect low-grade tumors (sensitivity ranging from 4 to 31%). The overall sensitivity and specificity of urine cytology range from 7 to 100 and from 30 to 70%, respectively. There is a need for new urine biomarkers that may help in BC diagnosis and surveillance. A lot of urinary biomarkers with high sensitivity and/or specificity have been investigated. Although none of these markers have proven to be powerful enough to replace standard cystoscopy, some of them may represent accurate predictors of BC. A review of recent studies is presented.

Diagnostic biomarkers differentiating metastatic melanoma patients from healthy controls identified by an integrated MALDI-TOF mass spectrometry/bioinformatic approach

The prognosis of advanced metastatic melanoma (American Joint Committee on Cancer (AJCC) stage IV) remains dismal with a 5-year survival rate of 6-18%. In the present study, an integrated MALDI mass spectrometric approach combined with artificial neural networks (ANNs) analysis and modeling has been used for the identification of biomarker ions in serum from stage IV melanoma patients allowing the discrimination of metastatic disease from healthy status with high specificities of 92% for protein ions and 100% for peptide biomarkers. Our ANNs model also correctly classified 98% of a blind validation set of AJCC stage I melanoma samples as nonstage IV samples, emphasizing the power of the newly defined biomarkers to identify patients with late-stage metastatic melanoma. Sequence analysis identified peptides derived from metastasis-associated proteins; alpha 1-acid glycoprotein precursor-1/2 (AAG-1/2) and complement C3 component precursor-1 (CCCP-1). Furthermore, quantitation of serum AAG by an immunoassay showed a significant (p<0.001) increase in AAG serum concentration in stage IV patients in comparison with healthy volunteers; moreover; the quantity of AAG plotted against MALDI-MS peak intensity classified the groups into two distinct clusters. Ongoing studies of other disease stages will provide evidence whether our strategy is sufficiently robust to give rise to stage-specific protein/peptide signatures in melanoma.

Artificial neural networks trained to detect viral and phage structural proteins

Phages play critical roles in the survival and pathogenicity of their hosts, via lysogenic conversion factors, and in nutrient redistribution, via cell lysis. Analyses of phage- and viral-encoded genes in environmental samples provide insights into the physiological impact of viruses on microbial communities and human health. However, phage ORFs are extremely diverse of which over 70% of them are dissimilar to any genes with annotated functions in GenBank. Better identification of viruses would also aid in better detection and diagnosis of disease, in vaccine development, and generally in better understanding the physiological potential of any environment. In contrast to enzymes, viral structural protein function can be much more challenging to detect from sequence data because of low sequence conservation, few known conserved catalytic sites or sequence domains, and relatively limited experimental data. We have designed a method of predicting phage structural protein sequences that uses Artificial Neural Networks (ANNs). First, we trained ANNs to classify viral structural proteins using amino acid frequency; these correctly classify a large fraction of test cases with a high degree of specificity and sensitivity. Subsequently, we added estimates of protein isoelectric points as a feature to ANNs that classify specialized families of proteins, namely major capsid and tail proteins. As expected, these more specialized ANNs are more accurate than the structural ANNs. To experimentally validate the ANN predictions, several ORFs with no significant similarities to known sequences that are ANN-predicted structural proteins were examined by transmission electron microscopy. Some of these self-assembled into structures strongly resembling virion structures. Thus, our ANNs are new tools for identifying phage and potential prophage structural proteins that are difficult or impossible to detect by other bioinformatic analysis. The networks will be valuable when sequence is available but in vitro propagation of the phage may not be practical or possible.

Bacteriophages are extremely abundant and diverse biological entities. All phage particles are comprised of nucleic acids and structural proteins, with few other packaged proteins. Despite their simplicity and abundance, more than 70% of phage sequences in the viral Reference Sequence database encode proteins with unknown function based on FASTA annotations. As a result, the use of sequence similarity is often insufficient for detecting virus structural proteins among unknown viral sequences. Viral structural protein function is challenging to detect from sequence data because structural proteins possess few known conserved catalytic motifs and sequence domains. To address these issues we investigated the use of Artificial Neural Networks as an alternative means of predicting function. Here, we trained thousands of networks using the amino acid frequency of structural protein sequences and identified the optimal architectures with the highest accuracies. Some hypothetical protein sequences detected by our networks were expressed and visualized by TEM, and produced images that strongly resemble virion structures. Our results support the utility of our neural networks in predicting the functions of unknown viral sequences.

Biomarkers for detection and surveillance of bladder cancer

INTRODUCTION

Bladder cancer is the fourth most common cancer in men and the ninth most common cancer in women in Canada. Early detection of tumours is essential for improved prognosis and long-term survival. The standard method for detection and surveillance is cystoscopy together with urine cytology. Cystoscopy is relatively sensitive but is expensive and invasive. Urinary cytology is a noninvasive method that has poor sensitivity but high specificity; it is relied on for the detection of carcinoma in situ. Currently, several urinary-based bladder tumour biomarkers with USFDA/Health Canada approval are available commercially, but none have been widely adopted by urologists despite their offering high sensitivity and/or specificity. We present here a review of recent studies evaluating 7 commercial biomarker assays for the detection and/or surveillance of bladder cancer.

RESULTS

SENSITIVITY AND SPECIFICITY RANGES, RESPECTIVELY, FOR EACH MARKER WERE REPORTED AS FOLLOWS: BTA Stat (Polymedco), 52.5%-78.0% and 69.0%-87.1%; BTA Trak (Polymedco), 51%-100% and 73%-92.5%; cytology, 12.1%-84.6% and 78.0%-100%; hematuria dipstick, 47.0%-92.6% and 51.0%-84.0%; NMP22 Bladder Cancer Test (Matritech), 34.6%-100% and 60.0%-95.0%; NMP22 BladderChek (Matritech), 49.5%-65.0% and 40.0%-89.8%; ImmunoCyt/uCyt+ (DiagnoCure), 63.3%-84.9% and 62.0%-78.1%; ImmunoCyt/uCyt+ and cytology, 81.0%-89.3% and 61.0%-77.7%; and UroVysion (Abbott Molecular)/florescence in situ hybridization, 68.6%-100% and 65.0%-96.0%.

CONCLUSION

We find that no currently available bladder cancer urinary marker is sensitive enough to eliminate the need for cystoscopy. In addition, cytology remains integral to the detection of occult cancer. However, owing to their relatively high sensitivities, these markers may be used to extend the period between cystoscopies in the surveillance of patients with transitional cell carcinoma. Further study is required to determine which markers, alone or in panel, would best accomplish this.

[Influence of the delay in urinary NMP22 preanalytical processing]

OBJECTIVES

To evaluate the influence of the delay in urinary NMP22 preanalytical processing.

MATERIAL AND METHODS

Twenty-eight voided urine samples were taken: bladder cancer (14), urine tract infections (4), lithiasis (4), healthy volunteers (1), and with other no malignant bladder diseases (5). All samples, were maintained at environment temperature, and were processed according to the stabilization of parts of urine collected at 0, 30, 90 and 150 minutes. Samples were stored at 4 degrees C until its determination. NMP22 was determined with the IMMULITE One analyzer.

RESULTS

There were no significant differences for NMP22 levels between each different point of time studied.

CONCLUSIONS

Delay up to 2 hours and a half when we add stabilization solution to urine samples no affects NMP22 results. That thing, might provide more confidence and flexibility on quantitative immunoassays that required urine stabilization.

Application of artificial intelligence to the management of urological cancer

PURPOSE

Artificial intelligence techniques, such as artificial neural networks, Bayesian belief networks and neuro-fuzzy modeling systems, are complex mathematical models based on the human neuronal structure and thinking. Such tools are capable of generating data driven models of biological systems without making assumptions based on statistical distributions. A large amount of study has been reported of the use of artificial intelligence in urology. We reviewed the basic concepts behind artificial intelligence techniques and explored the applications of this new dynamic technology in various aspects of urological cancer management.

MATERIALS AND METHODS

A detailed and systematic review of the literature was performed using the MEDLINE and Inspec databases to discover reports using artificial intelligence in urological cancer.

RESULTS

The characteristics of machine learning and their implementation were described and reports of artificial intelligence use in urological cancer were reviewed. While most researchers in this field were found to focus on artificial neural networks to improve the diagnosis, staging and prognostic prediction of urological cancers, some groups are exploring other techniques, such as expert systems and neuro-fuzzy modeling systems.

CONCLUSIONS

Compared to traditional regression statistics artificial intelligence methods appear to be accurate and more explorative for analyzing large data cohorts. Furthermore, they allow individualized prediction of disease behavior. Each artificial intelligence method has characteristics that make it suitable for different tasks. The lack of transparency of artificial neural networks hinders global scientific community acceptance of this method but this can be overcome by neuro-fuzzy modeling systems.

Partially Degraded RNA from Bladder Washing is a Suitable Sample for Studying Gene Expression Profiles in Bladder Cancer

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.

Molecular markers in bladder cancer: A critical appraisal

The diagnosis of both primary and recurrent bladder tumors currently relies upon the urine cytology and cystoscopy. Neither of these diagnostic tools is completely accurate. Prognostication of bladder cancer is largely based on pathologic tumor grade and stage. Over the past 2 decades, there is accumulating evidence that like many other cancers, bladder cancer, too, has a distinct molecular signature that separates it from other cancers and normal bladder tissue. Bladder tumors of different grades and stages even possess unique, and specific genotypic and phenotypic characteristics. Although recognition of several of these molecular alterations is possible by analyzing tumor tissue, urine, and serum samples, few if any of these "molecular markers" for bladder cancer are widely used in clinical practice. These markers include some that can be applied during the diagnostic work-up of symptoms (e.g., hematuria), those under surveillance for recurrence of superficial disease and forecasting long-term prognosis, or response to chemotherapy. In this review of molecular markers for bladder cancer, effectiveness of markers in each of these categories that are identifiable in the urine of patients with bladder cancer was examined. Many of the diagnostic markers appear to hold an advantage over urine cytology in terms of sensitivity, especially for the detection of low-grade superficial tumors. However, most markers tend to be less specific than cytology, yielding more false-positives. This result is more commonly observed in patients with concurrent bladder inflammation or other benign bladder conditions. Although there are several candidate markers for assessing prognosis or response to chemotherapy, studies of large patient populations are lacking. Further studies involving larger numbers of patients are required to determine their accuracy and widespread applicability in guiding treatment of bladder cancer.

Bladder tumor markers beyond cytology: International Consensus Panel on bladder tumor markers

This is the first of 2 articles that summarize the findings of the International Consensus Panel on cytology and bladder tumor markers. The objectives of our panel were to reach a consensus on the areas where markers are needed, to define the attributes of an ideal tumor marker, and to identify which marker(s) would be suitable for diagnosis and/or surveillance of bladder cancer. Our panel consisted of urologists and researchers from Europe, Asia, and the United States who reviewed original articles, reviews, and book chapters on individual bladder tumor markers published in the English language mainly using the PubMed search engine. Panel members also met during 3 international meetings to write recommendations regarding bladder tumor markers. The panel found that the most practical use of noninvasive tests is to monitor bladder cancer recurrence, thereby reducing the number of surveillance cystoscopies performed each year. Markers also may be useful in the screening of high-risk individuals for early detection of bladder cancer. However, more prospective studies are needed to strengthen this argument. Case-control and cohort studies show that several markers have a higher sensitivity to detect bladder cancer. However, cytology is the superior marker in terms of specificity, although some markers in limited numbers of studies have shown specificity equivalent to that of cytology. Our panel believes that several bladder tumor markers are more accurate in detecting bladder cancer than prostate-specific antigen (PSA) is in detecting prostate cancer. However, bladder tumor markers are held to a higher standard than PSA. Therefore, use of bladder tumor markers in the management of patients with bladder cancer will require the willingness of both urologists and clinicians to accept them.

The use of artificial neural networks in decision support in cancer: a systematic review

Artificial neural networks have featured in a wide range of medical journals, often with promising results. This paper reports on a systematic review that was conducted to assess the benefit of artificial neural networks (ANNs) as decision making tools in the field of cancer. The number of clinical trials (CTs) and randomised controlled trials (RCTs) involving the use of ANNs in diagnosis and prognosis increased from 1 to 38 in the last decade. However, out of 396 studies involving the use of ANNs in cancer, only 27 were either CTs or RCTs. Out of these trials, 21 showed an increase in benefit to healthcare provision and 6 did not. None of these studies however showed a decrease in benefit. This paper reviews the clinical fields where neural network methods figure most prominently, the main algorithms featured, methodologies for model selection and the need for rigorous evaluation of results.

Development and use of biomarkers in oncology drug development

Successful development and use of biomarkers will improve the productivity of oncology drug development. Recognition of the importance of biomarkers for speeding drug development is reflected in the precise definitions and concepts proposed by an NIH Working Group to standardize terminology and promote a more coherent and systematic approach to the development and use of biomarkers. Potential clinical biomarkers of drug efficacy are often identified through pre-clinical studies or basic research. Identification of potential biomarkers for use in oncology is moving rapidly forward through continuing advances in clinical imaging technologies, especially molecular and functional imaging. Other rapid advances are a product of the growing availability of new scientific reagents for established technologies and of high-throughput genomic and proteomic technologies that can generate hundreds of potential biomarkers for further evaluation. In certain cases, conventional clinical diagnostic techniques or assays can be adapted for use in pre-clinical models to evaluate their ability to serve as biomarkers for predicting clinical responses to new drug candidates. Evaluation (pre-clinical and clinical) of a potential biomarker is often the longest stage of biomarker development, and standards for evaluation or validation depend on the intended use and stage of clinical development. Biomarkers verified for use in preclinical studies can be used to help select appropriate animal models and lead compounds. Biomarkers verified for use in clinical trials can confirm a drug's pharmacological or biological mechanism of action, guide protocol design, aid patient and dose selection, and help to minimize safety risks. Oncology drug development can be optimized by using a tiered set of clinical biomarkers that predict compound efficacy and safety with increasing confidence at each rise in tier thereby aiding corporate decision-making about advancing compounds. In oncology, a special class of extensively evaluated biomarkers of efficacy (surrogate endpoints) that generally correlate with desired clinical outcomes can be used as a basis for corporate decisions as well as for gaining accelerated provisional regulatory approval of a drug.

Comparison of nuclear matrix proteins between gastric cancer and normal gastric tissue

AIM: To study the alteration of nuclear matrix proteins (NMPs) in gastric cancer.

METHODS: The NMPs extracted from 22 cases of gastric cancer and normal gastric tissues were investigated by SDS-PAGE technique and the data were analyzed using Genetools analysis software.

RESULTS: Compared with normal gastric tissue, the expression of 30 ku and 28 ku NMPs in gastric cancer decreased significantly (P = 0.002, P = 0.001, P < 0.05). No significant difference was found in the expression of the two NMPs between the various differentiated grades (P = 0.947, P = 0.356) and clinical stages of gastric cancer (P = 0.920, P = 0.243, P > 0.05).

CONCLUSION: The results suggested that the alteration of NMPs in gastric cancer occurred at the early stage of gastric cancer development.

AN ARTIFICIAL NEURAL NETWORK FOR PREDICTING THE PRESENCE OF SPERMATOZOA IN THE TESTES OF MEN WITH NONOBSTRUCTIVE AZOOSPERMIA

PURPOSE

We developed an artificial neural network (ANN) for predicting spermatozoa prior to testicular biopsy in men with nonobstructive azoospermia. The performance of this ANN was compared to that of the standard logistic regression (LR) model.

MATERIALS AND METHODS

Data were retrospectively collected from the physical examination and laboratory records of 303 patients who presented with infertility due to nonobstructive azoospermia. Input factors were patient age, duration of infertility, serum follicle-stimulating hormone, luteinizing hormone, total testosterone and prolactin, and left and right testicular volume. The ANN and LR models were constructed based on data on cases in which spermatozoa were and were not detected on testicular sperm extraction. The ANN was trained and validated with the data in the training (200 cases) and validation (30 cases) sets, and the model was then used to predict findings in the test set (73 cases). The LR model was constructed using the same data on the 230 training and validation cases. The same 73 patients served as the test set.

RESULTS

The sensitivity of the ANN model was significantly higher than that of the LR model (68% vs 28%, p < 0.0001). The neural network correctly predicted the outcome in 59 of the 73 test set patients (80.8%), whereas LR correctly predicted the outcome in 48 (65.7%, p = 0.07).

CONCLUSIONS

This ANN model, which is based on age, duration of infertility, serum hormone levels and testicular volumes, has clinically acceptable sensitivity. It may be of value for predicting spermatozoa in men with nonobstructive azoospermia.

Urinary markers of bladder carcinoma

PURPOSE

Transitional cell carcinoma of bladder is one of the most common tumors of genitourinary tract. Cystoscopy along with cytology is the mainstay for diagnosing bladder cancer. Cytology is specific for diagnosis of bladder carcinomas but less sensitive particularly in low-grade disease. Cystoscopy on the other hand is invasive and relatively costly technique and may also be inconclusive at times particularly in case of cystitis. A simple noninvasive marker for detecting bladder cancer would be helpful. A clinically useful urinary marker should be easy to perform, should have minimum requirements for sample preparations and should have high sensitivity and specificity in diagnosis. In this review we discussed the various urinary markers and their role in detection of bladder cancer.

MATERIALS AND METHODS

We reviewed the literature on urinary markers and tests that may have clinical usefulness.

RESULTS

Several urinary markers and tests such as BTA Stat, BTA TRAK, NMP22, telomerase, HA and HAse tests, Immunocyt, Quanticyt, FDP, BLCA-4, FISH, CYFRA-21-1 have enough potential for future clinical use. BTA stat, NMP22 (bladder check)and AccuDX (FDP) tests are presently point of care tests. The rest of the tests are laboratory-based and may need trained technicians. Majorities of the urinary markers have higher sensitivity and specificity than cytology. However, voided urinary cytology has the highest specificity.

CONCLUSIONS

Till now there is no urinary marker or test that can replace the need of cystoscopy. However, some markers have the potential for future clinical use.