Updating risk prediction tools: a case study in prostate cancer

Online risk prediction tools for common cancers are now easily accessible and widely used by patients and doctors for informed decision-making concerning screening and diagnosis. A practical problem is as cancer research moves forward and new biomarkers and risk factors are discovered, there is a need to update the risk algorithms to include them. Typically, the new markers and risk factors cannot be retrospectively measured on the same study participants used to develop the original prediction tool, necessitating the merging of a separate study of different participants, which may be much smaller in sample size and of a different design. Validation of the updated tool on a third independent data set is warranted before the updated tool can go online. This article reports on the application of Bayes rule for updating risk prediction tools to include a set of biomarkers measured in an external study to the original study used to develop the risk prediction tool. The procedure is illustrated in the context of updating the online Prostate Cancer Prevention Trial Risk Calculator to incorporate the new markers %freePSA and [-2]proPSA measured on an external case-control study performed in Texas, U.S.. Recent state-of-the art methods in validation of risk prediction tools and evaluation of the improvement of updated to original tools are implemented using an external validation set provided by the U.S. Early Detection Research Network.     

Multiplex targeted proteomic assay for biomarker detection in plasma: a pancreatic cancer biomarker case study

Biomarkers are most frequently proteins that are measured in the blood. Their development largely relies on antibody creation to test the protein candidate performance in blood samples of diseased versus nondiseased patients. The creation of such antibody assays has been a bottleneck in biomarker progress due to the cost, extensive time, and effort required to complete the task. Targeted proteomics is an emerging technology that is playing an increasingly important role to facilitate disease biomarker development. In this study, we applied a SRM-based targeted proteomics platform to directly detect candidate biomarker proteins in plasma to evaluate their clinical utility for pancreatic cancer detection. The characterization of these protein candidates used a clinically well-characterized cohort that included plasma samples from patients with pancreatic cancer, chronic pancreatitis, and healthy age-matched controls. Three of the five candidate proteins, including gelsolin, lumican, and tissue inhibitor of metalloproteinase 1, demonstrated an AUC value greater than 0.75 in distinguishing pancreatic cancer from the controls. In addition, we provide an analysis of the reproducibility, accuracy, and robustness of the SRM-based proteomics platform. This information addresses important technical issues that could aid in the adoption of the targeted proteomics platform for practical clinical utility.     

Detecting microRNAs of high influence on protein functional interaction networks: a prostate cancer case study

ABSTRACT: BACKGROUND: The use of biological molecular network information for diagnostic and prognostic purposes and elucidation of molecular disease mechanism is a key objective in systems biomedicine. The network of regulatory miRNA-target and functional protein interactions is a rich source of information to elucidate the function and the prognostic value of miRNAs in cancer. The objective of this study is to identify miRNAs that have high influence on target protein complexes in prostate cancer as a case study. This could provide biomarkers or therapeutic targets relevant for prostate cancer treatment. RESULTS: Our findings demonstrate that a miRNA's functional role can be explained by its target protein connectivity within a physical and functional interaction network. To detect miRNAs with high influence on target protein modules, we integrated miRNA and mRNA expression profiles with a sequence based miRNA-target network and human functional and physical protein interactions (FPI). miRNAs with high influence on target protein complexes play a role in prostate cancer progression and are promising diagnostic or prognostic biomarkers. We uncovered several miRNA-regulated protein modules which were enriched in focal adhesion and prostate cancer genes. Several miRNAs such as miR-96, miR-182, and miR-143 demonstrated high influence on their target protein complexes and could explain most of the gene expression changes in our analyzed prostate cancer data set. CONCLUSIONS: We describe a novel method to identify active miRNA-target modules relevant to prostate cancer progression and outcome. miRNAs with high influence on protein networks are valuable biomarkers that can be used in clinical investigations for prostate cancer treatment.     

Perspective: a program to improve protein biomarker discovery for cancer

Biomarkers for cancer risk, early detection, prognosis, and therapeutic response promise to revolutionize cancer management. Protein biomarkers offer tremendous potential in this regard due to their great diversity and intimate involvement in physiology. An effective program to discover protein biomarkers using existing technology will require team science, an integrated informatics platform, identification and quantitation of candidate biomarkers in disease tissue, mouse models of disease, standardized reagents for analyzing candidate biomarkers in bodily fluids, and implementation of automation. Technology improvements for better fractionation of the proteome, selection of specific biomarkers from complex mixtures, and multiplexed assay of biomarkers would greatly enhance progress.     

Good practice guidelines for biomarker discovery from array data: a case study for breast cancer prognosis

Background

Biomarker discovery holds the promise for advancing personalized medicine as the biomarkers can help match patients to optimal treatment to improve patient outcomes. However, serious concerns have been raised because very few molecular biomarkers or signatures discovered from high dimensional array data can be successfully validated and applied to clinical use. We propose good practice guidelines as well as a novel tool for biomarker discovery and use breast cancer prognosis as a case study to illustrate the proposed approach.

Results

We applied the proposed approach to a publicly available breast cancer prognosis dataset and identified small numbers of predictive markers for patient subpopulations stratified by clinical variables. Results from an independent cross-platform validation set show that our model compares favorably to other gene signature and clinical variable based prognostic tools. About half of the discovered candidate markers can individually achieve very good performance, which further demonstrate the high quality of feature selection. These candidate markers perform extremely well for young patient with estrogen receptor-positive, lymph node-negative early stage breast cancers, suggesting a distinct subset of these patients identified by these markers is actually at high risk of recurrence and may benefit from more aggressive treatment than cur-rent practice.

Conclusion

The results show that by following good practice guidelines, we can identify highly predictive genes in high dimensional breast cancer array data. These predictive genes have been successfully validated using an independent cross-platform dataset.

     

Computational approaches to study protein unfolding: Hen egg white lysozyme as a case study

Four methods are compared to drive the unfolding of a protein: (1) high temperature (T-run), (2) high pressure (P-run), (3) by imposing a gradual increase in the mean radius of the protein using a penalty function added to the physical interaction function (F-run, radial force driven unfolding), and (4) by weak coupling of the difference between the temperature of the radially outward moving atoms and the radially inward moving atoms to an external temperature bath (K-run, kinetic energy driven unfolding). The characteristic features of the four unfolding pathways are analyzed in order to detect distortions due to the size or the type of the applied perturbation, as well as the features that are common to all of them. Hen egg white lysozyme is used as a test system. The simulations are analyzed and compared to experimental data like ¹H-NMR amide proton exchange-folding competition, heat capacity, and compressibility measurements. © 1995 Wiley-Liss, Inc.     

Proteomics in prostate cancer biomarker discovery

Despite advances in molecular medicine, genomics, proteomics and translational research, prostate cancer remains the second most common cause of cancer-related mortality for men in the Western world. Clearly, early detection, targeted treatment and post-treatment monitoring are vital tools to combat this disease. Tumor markers can be useful for diagnosis and early detection of cancer, assessment of prognosis, prediction of therapeutic effect and treatment monitoring. Such tumor markers include prostate-specific antigen (prostate), cancer antigen (CA)15.3 (breast), CA125 (ovarian), CA19.9 (gastrointestinal) and serum α-fetoprotein (testicular cancer). However, all of these biomarkers lack sensitivity and specificity and, therefore, there is a large drive towards proteomic biomarker discovery. Current research efforts are directed towards discovering biosignatures from biological samples using novel proteomic technologies that provide high-throughput, in-depth analysis and quantification of the proteome. Several of these studies have revealed promising biomarkers for use in diagnosis, assessment of prognosis, and targeting treatment of prostate cancer. This review focuses on prostate cancer proteomic biomarker discovery and its future potential.     

Mass profiling-directed isolation and identification of a stage-specific serologic protein biomarker of advanced prostate cancer

Carcinoma of the prostate (CaP) is the second leading cause of cancer-related mortality among American men. While high cure rates are associated with localized CaP, no cure exists for advanced recurrent disease. At present there are no known serologic biomarkers specific to this stage of the disease. Several groups have used mass spectrometry (MS) based mass profiling (MP) combined with multivariate analysis to identify diagnostically predictive protein peaks for CaP in serum and tissues. Nevertheless, an appreciable level of skepticism exists for MP attributed primarily to a lack of definitive protein characterization. To address this problem, we have applied an approach that combines MP with a whole-protein based top-down separation strategy for the identification of a stage-specific marker in a group comprising 16 patients with CaP (metastatic and localized disease) and 15 healthy individuals. MP, combined with multivariate analysis, yielded 17 serum proteins specific to metastatic disease. A single protein detected at m/z 7771 was found to be significantly decreased in the sera of all the metastatic CaP patients but not in localized CaP or healthy individuals. This protein was therefore chosen as the primary candidate for further analysis. The complex nature of the serologic proteome necessitated an isolation strategy that included a C18 prefractionation, followed by multidimensional liquid chromatography and, finally, two-dimensional gel electrophoresis. The separation process was monitored by UV-Vis and matrix-assisted laser desorption/ionization-time of flight MS analysis. This strategy was found to greatly facilitate subsequent MS characterization of the unknown protein, which was identified as platelet factor 4, a chemokine with prothrombolytic and antiangiogenic activities. Confirmation was achieved using both Western blot analysis and enzyme-linked immunosorbent assay. With the growing interest in using MP for patient classification and diagnosis, our approach and its variations should be powerful in the separation and characterization of proteins following MP.     

Advances in proteomic prostate cancer biomarker discovery

Prostate cancer is the most common non-cutaneous cancer in men in the United States. For reasons largely unknown, the incidence of prostate cancer has increased in the last two decades, in spite or perhaps because of a concomitant increase in serum prostate-specific antigen (PSA) screening. While PSA is acknowledged not to be an ideal biomarker for prostate cancer detection, it is however widely used by physicians due to lack of an alternative. Thus, the identification of a biomarker(s) that can complement or replace PSA represents a major goal for prostate cancer research. Screening complex biological specimens such as blood, urine, and tissue to identify protein biomarkers has become increasingly popular over the last decade thanks to advances in proteomic discovery methods. The completion of human genome sequence together with new development in mass spectrometry instrumentation and bioinformatics has been a major driving force in biomarker discovery research. Here we review the current state of proteomic applications as applied to various sample sources including blood, urine, tissue, and “secretome” for the purpose of prostate cancer biomarker discovery. Additionally, we review recent developments in validation of putative markers, efforts at systems biology approach, and current challenges of proteomics in biomarker discovery.     

Statistical issues and methods for meta-analysis of microarray data: a case study in prostate cancer

With the proliferation of related microarray studies by independent groups, a natural step in the analysis of these gene expression data is to combine the results across these studies. However, this raises a variety of issues in the analysis of such data. In this article, we discuss the statistical issues of combining data from multiple gene expression studies. This leads to more complications than those in standard meta-analyses, including different experimental platforms, duplicate spots and complex data structures. We illustrate these ideas using data from four prostate cancer profiling studies. In addition, we develop a simple approach for assessing differential expression using the LASSO method. A combination of the results and the pathway databases are then used to generate candidate biological pathways for cancer.     

Computational prediction for the protein interactions of tyrosinase: Protein experimental interactome MAP

Tyrosinase (EC 1.14.18.1) is a diversely distributed enzyme in various organisms with physiological roles related to pigment production. Tyrosinase has gained the attention of researchers due to its biological functions and potential applications. In this regard, studies on the partner proteins of tyrosinase are important. In this study, we predicted the 3D structure of human tyrosinase and simulated the protein–protein interactions between tyrosinase and binding partners by using the PEIMAP algorithm. As a result, we found that tyrosinase is predicted to bind with G protein-related proteins, potassium voltage-gated channel-related proteins, and vesicle/sorting-related proteins. In particular, GIPC1, GIPC2, GIPC3, TYRP1, and DCT were predicted to primarily bind with tyrosinase. Interacting proteins (103) were secondarily bound to these 5 interacting proteins in the PEIMAP network of tyrosinase. An involvement in melanogenesis was also newly predicted by associating the predicted binding proteins. We simulated the protein–protein bindings by probing the residues of each complex facing toward the predicted site of interaction with tyrosinase. Among the interacting residues, some were predicted to dock to the active site of tyrosinase, which could affect its activity directly. Our computational predictions will be useful for elucidating the protein–protein interactions of tyrosinase and for studying its binding mechanisms and the melanin biosynthesis pathway.     

MALDI mass spectrometry in prostate cancer biomarker discovery

Matrix-assisted laser desorption/ionisation (MALDI) mass spectrometry (MS) is a highly versatile and sensitive analytical technique, which is known for its soft ionisation of biomolecules such as peptides and proteins. Generally, MALDI MS analysis requires little sample preparation, and in some cases like MS profiling it can be automated through the use of robotic liquid-handling systems. For more than a decade now, MALDI MS has been extensively utilised in the search for biomarkers that could aid clinicians in diagnosis, prognosis, and treatment decision making. This review examines the various MALDI-based MS techniques like MS imaging, MS profiling and proteomics in-depth analysis where MALDI MS follows fractionation and separation methods such as gel electrophoresis, and how these have contributed to prostate cancer biomarker research. This article is part of a Special Issue entitled: Biomarkers: A Proteomic Challenge.     

Periostin identified as a potential biomarker of prostate cancer by iTRAQ-proteomics analysis of prostate biopsy

Background  

Proteomics may help us better understand the changes of multiple proteins involved in oncogenesis and progression of prostate cancer(PCa) and identify more diagnostic and prognostic biomarkers. The aim of this study was to screen biomarkers of PCa by the proteomics analysis using isobaric tags for relative and absolute quantification(iTRAQ).     

Sequence- and interactome-based prediction of viral protein hotspots targeting host proteins: a case study for HIV Nef

Virus proteins alter protein pathways of the host toward the synthesis of viral particles by breaking and making edges via binding to host proteins. In this study, we developed a computational approach to predict viral sequence hotspots for binding to host proteins based on sequences of viral and host proteins and literature-curated virus-host protein interactome data. We use a motif discovery algorithm repeatedly on collections of sequences of viral proteins and immediate binding partners of their host targets and choose only those motifs that are conserved on viral sequences and highly statistically enriched among binding partners of virus protein targeted host proteins. Our results match experimental data on binding sites of Nef to host proteins such as MAPK1, VAV1, LCK, HCK, HLA-A, CD4, FYN, and GNB2L1 with high statistical significance but is a poor predictor of Nef binding sites on highly flexible, hoop-like regions. Predicted hotspots recapture CD8 cell epitopes of HIV Nef highlighting their importance in modulating virus-host interactions. Host proteins potentially targeted or outcompeted by Nef appear crowding the T cell receptor, natural killer cell mediated cytotoxicity, and neurotrophin signaling pathways. Scanning of HIV Nef motifs on multiple alignments of hepatitis C protein NS5A produces results consistent with literature, indicating the potential value of the hotspot discovery in advancing our understanding of virus-host crosstalk.     

Computational method for prediction of protein functional sites using specificity determinants

The current available data on protein sequences largely exceeds the experimental capabilities to annotate their function. So annotation in silico, i.e. using computational methods becomes increasingly important. This annotation is inevitably a prediction, but it can be an important starting point for further experimental studies. Here we present a method for prediction of protein functional sites, SDPsite, based on the identification of protein specificity determinants. Taking as an input a protein sequence alignment and a phylogenetic tree, the algorithm predicts conserved positions and specificity determinants, maps them onto the protein's 3D structure, and searches for clusters of the predicted positions. Comparison of the obtained predictions with experimental data and data on performance of several other methods for prediction of functional sites reveals that SDPsite agrees well with the experiment and outperforms most of the previously available methods. SDPsite is publicly available under http://bioinf.fbb.msu.ru/SDPsite.     

Projections of lung cancer mortality in West Germany: a case study in Bayesian prediction

We apply a generalized Bayesian age-period-cohort (APC) model to a data-set on lung cancer mortality in West Germany, in the period 1952-1996. Our goal is to predict future death rates until the year 2010, separately for males and females. Since age and period are not measured on the same grid, we propose a generalized APC model where consecutive cohort parameters represent strongly overlapping birth cohorts. This approach results in a rather large number of parameters, where standard algorithms for statistical inference by Markov chain Monte Carlo methods turn out to be computationally intensive. We propose a more efficient implementation based on ideas of block sampling from the time series literature. We entertain two different formulations, penalizing either first or second differences of age, period and cohort parameters. To assess the predictive quality of both formulations, we first forecast the rates for the period 1987-1996 based on data until 1986. A comparison with the actual observed rates is made based on a predictive deviance criterion. Predictions of lung cancer mortality until 2010 are then reported and a modification of the formulation in order to include information on cigarette consumption is finally described.To whom correspondence should be addressed. Currently at Imperial College School of Medicine, Department of Epidemiology and Public Health, Norfolk Place, London W2 1PG, UK.     

11C]Choline as a PET biomarker for assessment of prostate cancer tumor models

[(11)C]Choline has been evaluated as a positron emission tomography (PET) biomarker for assessment of established human prostate cancer tumor models. [(11)C]Choline was prepared by the reaction of [(11)C]methyl triflate with 2-dimethylaminoethanol (DMAE) and isolated and purified by solid-phase extraction (SPE) method in 60-85% yield based on [(11)C]CO(2), 15-20 min overall synthesis time from end of bombardment (EOB), 95-99% radiochemical purity and specific activity >0.8 Ci/micromol at end of synthesis (EOS). The biodistribution of [(11)C]choline was determined at 30 min post iv injection in prostate cancer tumor models C4-2, PC-3, CWR22rv, and LNCaP tumor-bearing athymic mice. The results showed the accumulation of [(11)C]choline in these tumors was 1.0% dose/g in C4-2 mouse, 0.4% dose/g in PC-3 mice, 3.2% dose/g in CWR22rv mice, and 1.4% dose/g in LNCaP mice; the ratios of tumor/muscle (T/M) and tumor/blood (T/B) were 2.3 (T/M, C4-2), 1.4 (T/M, PC-3), 2.5 (T/M, CWR22rv), 1.2 (T/M, LNCaP) and 2.6 (T/B, C4-2), 2.6 (T/B, PC-3), 7.8 (T/B, CWR22rv), 3.2 (T/B, LNCaP), respectively. The micro-PET imaging of [(11)C]choline in prostate cancer tumor models was acquired from a C4-2, PC-3, CWR22rv, or LNCaP implanted mouse at 30 min post iv injection of 1 mCi of the tracer using a dedicated high resolution (<3 mm full-width at half-maximum) small FOV (field-of-view) PET imaging system, IndyPET-II scanner, developed in our laboratory, which showed the accumulation of [(11)C]choline in C4-2, PC-3, CWR22rv, or LNCaP tumor implanted in a nude athymic mouse. The initial dynamic micro-PET imaging data indicated the average T/M ratios were approximately 3.0 (C4-2), 2.1 (PC-3), 3.5 (CWR22rv), and 3.3 (LNCaP), respectively, which showed the tumor accumulation of [(11)C]choline in all four tumor models is high. These results suggest that there are significant differences in [(11)C]choline accumulation between these different tumor types, and these differences might offer some useful measure of tumor biological process.