Proteomic classification of pancreatic adenocarcinoma tissue using protein chip technology

SELDI-TOF MS Analysis of Hepatocellular Carcinoma in an Australian Cohort

BACKGROUND

Hepatocellular carcinoma (HCC) is a common cause of cancer death worldwide. Resection offers the best chance of long-term survival, but a consistent adverse prognostic factor is the presence of microvascular invasion (MVI). In this study, surface-enhanced laser desorption/ionization time-of-flight mass spectrometry (SELDI-TOF MS), a high throughput method of analyzing complex samples, was used to explore differentially expressed proteins between HCC and adjacent nontumour liver tissue (ANLT). These findings were correlated with clinical outcomes.

MATERIALS AND METHODS

From 2002 to 2011, tumor and ANLT were collected from patients who underwent liver resection and these samples were later prepared for SELDI-TOF MS. Output data were then used to identify proteins capable of discriminating HCC from ANLT. Proteins from the multivariate analysis were then analyzed to determine prognostic factors and the m/z ratios of these proteins were entered into the ExPASy database to infer potential candidates.

RESULTS

During the study period, 30 patients had SELDI-TOF MS performed on their HCC and ANLT samples. On multivariate analysis, a panel of four proteins-m/z 5840, m/z 8921, m/z 9961, and m/z 25,872-discriminated HCC from ANLT with an area under the ROC curve of 0.954 (P < 0.001). On prognostic factor assessment, decreased m/z 9961 was significantly associated with the presence of MVI (P = 0.025) and shorter disease-free survival (P = 0.045) in our patients. A potential candidate for this protein was coxsackievirus and adenovirus receptor, isoform 3 (CAR 3/7), which helps maintain tight junction integrity.

CONCLUSIONS

Using SELDI TOF-MS, we identified a panel of four proteins with excellent discriminative capacity between HCC and ANLT. Of these, m/z 9961 was the only protein significantly associated with a known poor prognostic factor (presence of MVI) and survival (shorter disease-free survival). While loss of CAR 3/7 could lead to MVI, further research is warranted to validate the identity of protein m/z 9961.

Potential prognostic significance of a new proteomic profile in patients with advanced pancreatic adenocarcinoma

OBJECTIVES

Seven-signal proteomic approach has recently been developed as a new proteomic profile measured by matrix-assisted laser desorption/ionization mass spectrometry. The aim of this study was to evaluate prognostic significance of this proteomic value in patients with pancreatic adenocarcinoma.

METHODS

Blood samples from the patients with pancreatic adenocarcinoma were prospectively collected before treatments including surgical resection and systemic chemotherapies. The seven-signal proteomic profiles of the samples were measured, and the prognostic significance of the proteomic value was evaluated through comparison with other existing prognostic markers.

RESULTS

Cut-off value of the proteomic profiles at 52 stratified overall prognosis of the patients (6.5 months vs. 10.9 months with the values ≥52 vs. <52, p = 0.020). In subgroup analyses of inoperable cases with carcinoembryonic antigen level of <5 ng/ml or performance status of 0-1, the proteomic value at 52 stratified their prognosis (p = 0.002 and p = 0.006, respectively).

CONCLUSIONS

The new seven-signal proteomics showed useful prognostic significance for patients with pancreatic adenocarcinoma. Further studies with a large sample size would be required to evaluate whether this proteomic approach possibly complements the existing parameters, such as carcinoembryonic antigen and performance status.

Phytochemical properties and anti-proliferative activity of Olea europaea L. leaf extracts against pancreatic cancer cells

Olea europaea L. leaves are an agricultural waste product with a high concentration of phenolic compounds; especially oleuropein. Oleuropein has been shown to exhibit anti-proliferative activity against a number of cancer types. However, they have not been tested against pancreatic cancer, the fifth leading cause of cancer related death in Western countries. Therefore, water, 50% ethanol and 50% methanol extracts of Corregiola and Frantoio variety Olea europaea L. leaves were investigated for their total phenolic compounds, total flavonoids and oleuropein content, antioxidant capacity and anti-proliferative activity against MiaPaCa-2 pancreatic cancer cells. The extracts only had slight differences in their phytochemical properties, and at 100 and 200 μg/mL, all decreased the viability of the pancreatic cancer cells relative to controls. At 50 μg/mL, the water extract from the Corregiola leaves exhibited the highest anti-proliferative activity with the effect possibly due to early eluting HPLC peaks. For this reason, olive leaf extracts warrant further investigation into their potential anti-pancreatic cancer benefits.

Fruit-derived phenolic compounds and pancreatic cancer: perspectives from Australian native fruits

ETHNOPHARMACOLOGICAL RELEVANCE

Pancreatic cancer is a devastating cancer that presents late, is rapidly progressive and has current therapeutics with only limited efficacy. Bioactive compounds are ubiquitously present in fruits and numerous studies in vitro are addressing the activity of these compounds against pancreatic cancer, thus studies of specific bioactive compounds could lead to new anti-pancreatic cancer strategies. Australian native fruits have been used as foods and medicines by Australian Aboriginals for thousands of years, and preliminary studies have found these fruits to contain rich and diversified bioactive components with high antioxidant activity. Thus, Australian native fruits may possess key components for preventing or delaying the onset of tumorigenesis, or for the treatment of existing cancers, including pancreatic cancer.

MATERIALS AND METHODS

Numerous databases including PubMed, SciFinder, Web of Knowledge, Scopus, and Sciencedirect were analysed for correlations between bioactive components from fruits and pancreatic cancer, as well as studies concerning Australian native fruits.

RESULTS

In this review, we comprehensively highlight the proposed mechanisms of action of fruit bioactives as anti-cancer agents, update the potential anti-pancreatic cancer activity of various major classes of bioactive compounds derived from fruits, and discuss the existence of bioactive compounds identified from a selection Australian native fruits for future studies.

CONCLUSION

Bioactive compounds derived from fruits possess the potential for the discovery of new anti-pancreatic cancer strategies. Further, Australian native fruits are rich in polyphenols including some flora that contain unique phenolic compounds, thereby warranting further investigations into their anti-cancer properties.

Tissue biomarkers of breast cancer and their association with conventional pathologic features

Background:

Tissue protein expression profiling has the potential to detect new biomarkers to improve breast cancer (BC) diagnosis, staging, and prognostication. This study aimed to identify tissue proteins that differentiate breast cancer tissue from healthy breast tissue using protein chip mass spectrometry and to examine associations with conventional pathological features.

Methods:

To develop a training model, 82 BC and 82 adjacent unaffected tissue (AT) samples were analysed on cation-exchange protein chips by time-of-flight mass spectrometry. For validation, 89 independent BC and AT sample pairs were analysed.

Results:

From the protein peaks that were differentially expressed between BC and AT by univariate analysis, binary logistic regression yielded two peaks that together classified BC and AT with a ROC area under the curve of 0.92. Two proteins, ubiquitin and S100P (in a novel truncated form), were identified by liquid chromatography/tandem mass spectrometry and validated by immunoblotting and reactive-surface protein chip immunocapture. The combined marker panel was positively associated with high histologic grade, larger tumour size, lymphovascular invasion, ER and PR positivity, and HER2 overexpression, suggesting that it may be associated with a HER2-enriched molecular subtype of breast cancer.

Conclusion:

This independently validated protein panel may be valuable in the classification and prognostication of breast cancer patients.

Evaluation of serum diagnosis of pancreatic cancer by using surface-enhanced laser desorption/ionization time-of-flight mass spectrometry

Proteomic methods have been widely used in disease marker discovery research. The aim of this study was to discover potential biomarkers for pancreatic cancer (PCa) using surface-enhanced laser desorption/ionization time-of-flight mass spectrometry (SELDI-TOF-MS). Crude serum samples from 132 patients with PCa and 67 healthy controls (HCs) were analyzed in duplicate using SELDI. Support vector machine (SVM) analysis of the spectra was used to generate a predictive algorithm based on proteins that were maximally differentially expressed between patients with PCa and the HCs in the training cohort. This algorithm was tested using leave-one-out cross-validation in the test cohort. From the 4 significant peaks in the training cohort, a classifier for separating patients with PCa from HCs was developed. The classifier was challenged with all samples achieving 96.67% sensitivity and 100% specificity in the training cohort and 93.1% sensitivity and 78.57% specificity in the test cohort. Additionally, the classifier correctly classified 12/12 stage Ia and 13/16 stage IIa PCa cases. The combination of the SELDI panel and CA19-9 was superior to CA19-9 alone in distinguishing individuals with PCa from the healthy subject group. These results suggest that high-throughput proteomic profiling has the capacity to provide new biomarkers for the early detection and diagnosis of PCa.

Proteomic assessment of markers for malignancy in the mucus of intraductal papillary mucinous neoplasms of the pancreas

OBJECTIVES

Intraductal papillary mucinous neoplasms (IPMN) of the pancreas evolve from dysplasia to invasive adenocarcinoma. The aims of this study were to look for candidate protein profiles in IPMN mucus according to histological grade, using a differential proteomic technique, and to highlight protein peaks associated with malignant transformation.

METHODS

Forty-three mucus samples obtained from surgically resected IPMN and categorized as benign (low/moderate dysplasia) or malignant (severe dysplasia/invasive adenocarcinoma) in 21 and 22 patients, respectively. A surface-enhanced laser desorption ionization time-of-flight mass spectrometry was used to determine candidate protein expression profiles. Protein peaks that significantly differed between benign/malignant IPMN (area under curve > 0.88; P < 10; high intensity) were identified using adapted software.

RESULTS

Among 952 protein peaks, 31 were differentially expressed in benign/malignant IPMN (P < 0.001). Among them, 5 candidate proteins of interest (mass-to-charge ratio [m/z]: 5217, 6326, 6719, 10,453, and 10,849 d) were selected by their high diagnostic accuracy and ability to distinguish between malignant and benign tumors. No correlation was found between peak profiles and duct involvement.

CONCLUSIONS

Carcinogenic process in IPMN is associated with changes in mucus proteome with characteristic peaks that could be potential candidate biomarkers of malignancy.

ABBREVIATIONS

IPMN - intraductal papillary mucinous neoplasm, EPC - extrapancreatic cancer, MRI - magnetic resonance imaging, ERCP - endoscopic retrograde cholangiopancreatography.

Tumor classification of six common cancer types based on proteomic profiling by MALDI imaging

In clinical diagnostics, it is of outmost importance to correctly identify the source of a metastatic tumor, especially if no apparent primary tumor is present. Tissue-based proteomics might allow correct tumor classification. As a result, we performed MALDI imaging to generate proteomic signatures for different tumors. These signatures were used to classify common cancer types. At first, a cohort comprised of tissue samples from six adenocarcinoma entities located at different organ sites (esophagus, breast, colon, liver, stomach, thyroid gland, n = 171) was classified using two algorithms for a training and test set. For the test set, Support Vector Machine and Random Forest yielded overall accuracies of 82.74 and 81.18%, respectively. Then, colon cancer liver metastasis samples (n = 19) were introduced into the classification. The liver metastasis samples could be discriminated with high accuracy from primary tumors of colon cancer and hepatocellular carcinoma. Additionally, colon cancer liver metastasis samples could be successfully classified by using colon cancer primary tumor samples for the training of the classifier. These findings demonstrate that MALDI imaging-derived proteomic classifiers can discriminate between different tumor types at different organ sites and in the same site.

Proteomics portrait of archival lesions of chronic pancreatitis

Chronic pancreatitis is a chronic inflammatory disorder of the pancreas. The etiology is multi-fold, but all lead to progressive scarring and loss of pancreatic function. Early diagnosis is difficult; and the understanding of the molecular events that underlie this progressive disease is limited. In this study, we investigated differential proteins associated with mild and severe chronic pancreatitis in comparison with normal pancreas and pancreatic cancer. Paraffin-embedded formalin-fixed tissues from five well-characterized specimens each of normal pancreas (NL), mild chronic pancreatitis (MCP), severe chronic pancreatitis (SCP) and pancreatic ductal adenocarcinoma (PDAC) were subjected to proteomic analysis using a “label-free” comparative approach. Our results show that the numbers of differential proteins increase substantially with the disease severity, from mild to severe chronic pancreatitis, while the number of dysregulated proteins is highest in pancreatic adenocarcinoma. Important functional groups and biological processes associated with chronic pancreatitis and cancer include acinar cell secretory proteins, pancreatic fibrosis/stellate cell activation, glycoproteins, and inflammatory proteins. Three differential proteins were selected for verification by immunohistochemistry, including collagen 14A1, lumican and versican. Further canonical pathway analysis revealed that acute phase response signal, prothrombin activation pathway, and pancreatic fibrosis/pancreatic stellate cell activation pathway were the most significant pathways involved in chronic pancreatitis, while pathways relating to metabolism were the most significant pathways in pancreatic adenocarcinoma. Our study reveals a group of differentially expressed proteins and the related pathways that may shed light on the pathogenesis of chronic pancreatitis and the common molecular events associated with chronic pancreatitis and pancreatic adenocarcinoma.

A proteomic analysis of Curcuma comosa Roxb. rhizomes

BACKGROUND

The similarly in plant physiology and the difficulty of plant classification, in some medicinal plant species, especially plants of the Zingiberaceae family, are a major problem for pharmacologists, leading to mistaken use. To overcome this problem, the proteomic base method was used to study protein profiles of the plant model, Curcuma comosa Roxb., which is a member of the Zingiberaceae and has been used in traditional Thai medicine as an anti-inflammatory agent for the treatment of postpartum uterine bleeding.

RESULTS

Due to the complexity of protein extraction from this plant, microscale solution-phase isoelectric focusing (MicroSol-IEF) was used to enrich and improve the separation of Curcuma comosa rhizomes phenol-soluble proteins, prior to resolving and analyzing by two-dimensional polyacrylamide gel electrophoresis and identification by tandem mass spectrometry. The protein patterns showed a high abundance of protein spots in the acidic range, including three lectin proteins. The metabolic and defense enzymes, such as superoxide dismutase (SOD) and ascorbate peroxidase, that are associated with antioxidant activity, were mainly found in the basic region. Furthermore, cysteine protease was found in this plant, as had been previously reported in other Zingiberaceae plants.

CONCLUSION

This report presents the protein profiles of the ginger plant, Curcuma comosa. Several interesting proteins were identified in this plant that may be used as a protein marker and aid in identifying plants of the Zingiberaceae family.

Proteomics in pancreatic cancer research

In this review, we give an overview of the actual role of proteomic technologies in the study of pancreatic cancers (PCs). We describe PC proteomics on the basis of sample origins, i.e. tissues, body fluids, and PC cell lines. As regards PC tissues, we report the identification of a number of candidate biomarkers of precursor lesions that may allow early diagnosis of this neoplasia. Moreover, we describe cytoskeletal and hypoxia-regulated proteins that confirm the involvement of cytoskeleton modifications and metabolism adaptations in carcinogenesis. We also discuss the most important biomarkers identified by proteomic analysis involved in local invasion and distant metastasis, and in the cross-talk between pancreatic tumor and the surrounding stroma. Furthermore, we report novel candidate biomarkers identified in serum, plasma, and pancreatic juice of cancer patients compared with cancer-free controls. Proteomic alterations in PC cell line models as compared to normal controls and studies on cell lines treated with drugs or new agents to understand their mechanism of pharmacological action or the onset of drug resistance are also presented. Finally, we discuss the recent improvements obtained in classical 2-DE and high-throughput proteomic strategies able to allow the overcoming of relevant proteomic drawbacks.

Discovery of serum biomarkers for pancreatic adenocarcinoma using proteomic analysis

Background and aims:

The serum/plasma proteome was explored for biomarkers to improve the diagnostic ability of CA19-9 in pancreatic adenocarcinoma (PC).

Methods:

A Training Set of serum samples from 20 resectable and 18 stage IV PC patients, 54 disease controls (DCs) and 68 healthy volunteers (HVs) were analysed by surface-enhanced laser desorption and ionisation time-of-flight mass spectrometry (SELDI-TOF MS). The resulting protein panel was validated on 40 resectable PC, 21 DC and 19 HV plasma samples (Validation-1 Set) and further by ELISA on 33 resectable PC, 28 DC and 18 HV serum samples (Validation-2 Set). Diagnostic panels were derived using binary logistic regression incorporating internal cross-validation followed by receiver operating characteristic (ROC) analysis.

Results:

A seven-protein panel from the training set PC vs DC and from PC vs HV samples gave the ROC area under the curve (AUC) of 0.90 and 0.90 compared with 0.87 and 0.91 for CA19-9. The AUC was greater (0.97 and 0.99, P<0.05) when CA19-9 was added to the panels and confirmed on the validation-1 samples. A simplified panel of apolipoprotein C-I (ApoC-I), apolipoprotein A-II (ApoA-II) and CA19-9 was tested on the validation-2 set by ELISA, in which the ROC AUC was greater than that of CA19-9 alone for PC vs DC (0.90 vs 0.84) and for PC vs HV (0.96 vs 0.90).

Conclusions:

A simplified diagnostic panel of CA19-9, ApoC-I and ApoA-II improves the diagnostic ability of CA19-9 alone and may have clinical utility.

Elevated levels of the acute-phase serum amyloid are associated with heightened lung cancer risk

BACKGROUND

The authors investigated whether early stage lung cancer could be identified by proteomic analyses of plasma.

METHODS

For the first case-control study, plasma samples from 52 patients with lung cancer and from a group of 51 controls were analyzed by surface-enhanced laser desorption/ionization time-of-flight mass spectrometry. In a second case-control study, a classifier of 4 markers (mass-to-charge ratio, 11,681, 6843, 5607, and 8762) also was tested for validation on plasma from 16 consecutive patients with screen-detected cancer versus 406 healthy individuals. The most relevant marker was identified, and an enzyme-linked immunosorbent assay-based analysis revealed that signal intensity was correlated with concentration.

RESULTS

The classifier had a sensitivity of 94.23% and a specificity of 76.47% in the first study but lost predictive value in the second study. Nevertheless, the 11,681 cluster, which was identified as serum amyloid protein A (SAA), resulted in a multiple logistic regression model that indicated a strong association with lung cancer. When both studies were considered as a together, the odds ratio (OR) for an SAA intensity > or =0.5 was 10.27 (95% confidence interval [CI], 4.64-22.74), whereas an analysis restricted to stage I cancers (TNM classification) revealed an OR of 8.45 (95% CI, 2.76-25.83) for T1 lung cancer and 21.22 (95% CI, 5.62-80.14) for T2 lung cancer.

CONCLUSIONS

SAA levels were predictive of an elevated risk of lung cancer, supporting the general view that inflammation is implicated in lung cancer development.

Simpler evaluation of predictions and signature stability for gene expression data

Scientific advances are raising expectations that patient-tailored treatment will soon be available. The development of resulting clinical approaches needs to be based on well-designed experimental and observational procedures that provide data to which proper biostatistical analyses are applied. Gene expression microarray and related technology are rapidly evolving. It is providing extremely large gene expression profiles containing many thousands of measurements. Choosing a subset from these gene expression measurements to include in a gene expression signature is one of the many challenges needing to be met. Choice of this signature depends on many factors, including the selection of patients in the training set. So the reliability and reproducibility of the resultant prognostic gene signature needs to be evaluated, in such a way as to be relevant to the clinical setting. A relatively straightforward approach is based on cross validation, with separate selection of genes at each iteration to avoid selection bias. Within this approach we developed two different methods, one based on forward selection, the other on genes that were statistically significant in all training blocks of data. We demonstrate our approach to gene signature evaluation with a well-known breast cancer data set.

Advances in pancreatic cancer detection

Pancreatic cancer represents a major challenge for research studies and clinical management. No specific tumor marker for the diagnosis of pancreatic cancer exists. Therefore, extensive genomic, transcriptomic, and proteomic studies are being developed to identify candidate markers for use in high-throughput systems capable of large cohort screening. Understandably, the complex pathophysiology of pancreatic cancer requires sensitive and specific biomarkers that can improve both early diagnosis and therapeutic monitoring. The lack of a single diagnostic marker makes it likely that only a panel of biomarkers is capable of providing the appropriate combination of high sensitivity and specificity. Biomarker discovery using novel technology can improve prognostic upgrading and pinpoint new molecular targets for innovative therapy.

Pancreatic cancer biomarkers discovery by surface-enhanced laser desorption and ionization time-of-flight mass spectrometry

BACKGROUND

Surface-enhanced laser desorption and ionization time-of-flight mass spectrometry (SELDI-TOF/MS), a laboratory-friendly technique, is used to identify biomarkers for cancer. The aim of the present study was to explore the application of SELDI proteomic patterns in serum for distinguishing between cases of pancreatic cancer, chronic pancreatitis, type 2 diabetes mellitus and healthy controls.

METHODS

Sera from 12 healthy controls, 24 patients with type 2 diabetes mellitus, 126 with pancreatic cancer, including 84 with diabetes, and 61 with chronic pancreatitis, 32 of which were diabetics, were analyzed using SELDI-TOF/MS. Spectra (IMAC-30) were clustered and classified using Biomarker Wizard and Biomarker Pattern software.

RESULTS

Two decision tree classification algorithms, one with and one without CA 19-9, were constructed. In the absence of CA 19-9, the splitting protein peaks were: m/z 1526, 1211, and 3519; when CA 19-9 was used in the analysis, it replaced the m/z 3519 splitter. The two algorithms performed equally for classifying patients. A classification tree that considered diabetic patients only was constructed; the main splitters were: 1211, CA 19-9, 7903, 3359, 1802. With this algorithm, 100% of patients with type 2 diabetes mellitus, 97% with chronic pancreatitis and 77% of patients with pancreatic cancer were correctly classified. SELDI-TOF/MS features improved the diagnostic accuracy of CA 19-9 (AUC = 0.883 for CA 19-9; AUC = 0.935 for CA 19-9 and SELDI-TOF/MS features combined).

CONCLUSIONS

SELDI-TOF/MS allows identification of new peptides which, in addition to CA 19-9, enable the correct classification of the vast majority of patients with pancreatic cancer, which can be distinguished from patients with chronic pancreatitis or type 2 diabetes mellitus.

The current state of proteomics in GI oncology

Proteomics refers to the study of the entire set of proteins in a given cell or tissue. With the extensive development of protein separation, mass spectrometry, and bioinformatics technologies, clinical proteomics has shown its potential as a powerful approach for biomarker discovery, particularly in the area of oncology. More than 130 exploratory studies have defined candidate markers in serum, gastrointestinal (GI) fluids, or cancer tissue. In this article, we introduce the commonly adopted proteomic technologies and describe results of a comprehensive review of studies that have applied these technologies to GI oncology, with a particular emphasis on developments in the last 3 years. We discuss reasons why the more than 130 studies to date have had little discernible clinical impact, and we outline steps that may allow proteomics to realize its promise for early detection of disease, monitoring of disease recurrence, and identification of targets for individualized therapy.

Diagnosis of pancreatic adenocarcinoma using protein chip technology

BACKGROUND

To develop a serum-specific protein fingerprint which is capable of differentiating samples from patients with pancreatic cancer and those with other pancreatic conditions.

METHODS

We used SELDI-TOF-MS coupled with CM10 chips and bioinformatics tools to analyze a total of 118 serum samples in this study; 78 serum samples were analyzed to establish the diagnostic models and the other 40 samples were analyzed on the second day as an independent test set.

RESULTS

The analysis of this independent test set yielded a specificity of 91.6% and a sensitivity of 91.6% for pattern 1, which distinguished pancreatic adenocarcinoma (PC) from healthy individuals and a specificity of 80.0% and a sensitivity of 90.9% for pattern 2, which distinguished PC from chronic pancreatitis.

CONCLUSION

This study indicated that the SELDI-TOF-MS technique can facilitate the discovery of better serum tumor biomarkers and a combination of specific models is more accurate than a single model in diagnosis of PC.

Differential expression of serum proteomic spectra in rat model of pancreatic intraepithelial neoplasia and dimethylbenzanthracene-induced pancreatic carcinoma

AIM: To investigate relationship between differential expression of serum proteomic spectra in rat model of pancreatic intraepithelial neoplasia (PanIN) and dimethylbenzanthracene-induced pancreatic carcinoma (PC) using surface-enhanced laser desorption/ionization time-of-flight mass spectrometry (SELDI-TOF MS) technology.

METHODS: Forty male SD rats were implanted with DMBA into the pancreas to induce rat model of PanIN and PC. Histopathology was evaluated according to PanIN classification system. And normal control group of twenty-six male SD rats was established. The serum protein spectra were detected using IMAC-Cu²⁺ proteinchip and SELDI-TOF MS. The data were analyzed using Biomarker Wizard 3.0 Software of Ciphergen Biosystem Co.

RESULTS: DMBA was implanted into pancreas of rats in PC group (n = 11) and PanIN group (n = 18). Compared with the normal control group, there were significant differences (P < 0.001) of 30 protein peaks in PanIN and PC of which 19 protein peaks were up-regulated and 11 down-regulated. The expression of 9 protein peaks, with a ratio of mass to charge (M/Z) of 5835.2, 4087.3, 4786.5, 4800.5, 3932.2, 5765.9, 5924.8, 5001.9, 3913.7 gradually increased from normal to PanIN and PC group, and 4 protein peaks with a M/Z ratio of 1096.9, 1478.9, 8572.9, 1007.1 gradually decreased.

CONCLUSION: Serum proteomic spectra were differentially expressed in rat model of PanIN and PC. Identification and function of these differentially expressed proteins necessitate further investigation.

Classification of premalignant pancreatic cancer mass-spectrometry data using decision tree ensembles

Background

Pancreatic cancer is the fourth leading cause of cancer death in the United States. Consequently, identification of clinically relevant biomarkers for the early detection of this cancer type is urgently needed. In recent years, proteomics profiling techniques combined with various data analysis methods have been successfully used to gain critical insights into processes and mechanisms underlying pathologic conditions, particularly as they relate to cancer. However, the high dimensionality of proteomics data combined with their relatively small sample sizes poses a significant challenge to current data mining methodology where many of the standard methods cannot be applied directly. Here, we propose a novel methodological framework using machine learning method, in which decision tree based classifier ensembles coupled with feature selection methods, is applied to proteomics data generated from premalignant pancreatic cancer.

Results

This study explores the utility of three different feature selection schemas (Student t test, Wilcoxon rank sum test and genetic algorithm) to reduce the high dimensionality of a pancreatic cancer proteomic dataset. Using the top features selected from each method, we compared the prediction performances of a single decision tree algorithm C4.5 with six different decision-tree based classifier ensembles (Random forest, Stacked generalization, Bagging, Adaboost, Logitboost and Multiboost). We show that ensemble classifiers always outperform single decision tree classifier in having greater accuracies and smaller prediction errors when applied to a pancreatic cancer proteomics dataset.

Conclusion

In our cross validation framework, classifier ensembles generally have better classification accuracies compared to that of a single decision tree when applied to a pancreatic cancer proteomic dataset, thus suggesting its utility in future proteomics data analysis. Additionally, the use of feature selection method allows us to select biomarkers with potentially important roles in cancer development, therefore highlighting the validity of this method.

Proteomics of pancreatic cancer

Pancreatic cancer is a devastating disease, with a mortality rate almost identical with its incidence. Late diagnosis and limited therapeutic options make early detection of pancreatic cancer a pressing clinical problem. In this context, the investigation of the pancreatic cancer proteome has recently gained considerable attention because profiles of proteins may be able to more accurately identify disease states, such as cancer. Recent pancreatic cancer proteome studies may be categorized into basic studies cataloguing the pancreatic proteome, studies investigating differential protein expression patterns, and studies searching for proteome-based biomarkers for early cancer detection and differentiation. Although these studies clearly demonstrate that a range of biological samples are suitable for proteomic analyses, comparison of different studies is problematic due to the diversity of methodologies, sample sources, and characterization of patient populations. Reproducibility between studies has rarely been investigated, and no investigation has compared the different methods of proteomic research. The results of this review have shown that more stringent requirements concerning the design and the analysis of future studies should be implemented. These include an adequate patient number, obligatory histological examination of tissues, appropriate control groups, identification of proteins and peaks, validation of differential expression using independent cohorts and/or a second methodology, and, finally, demonstration of result reproducibility. This will hopefully lead to the discovery of prognostic and predictive biomarkers that help to improve prognosis of pancreatic cancer patients.

Prognostic significance of growth factors and the urokinase-type plasminogen activator system in pancreatic ductal adenocarcinoma

OBJECTIVES

To determine the diagnostic and prognostic significance of growth factors and the urokinase-type plasminogen activator (uPA) system in pancreatic ductal adenocarcinoma (PDAC) using a multigene assay.

METHODS

Messenger RNA (mRNA) expression of 15 genes from epidermal growth factor receptor, insulin-like growth factor (IGF), and uPA families were measured in 46 PDAC tissue samples using quantitative real-time reverse transcription-polymerase chain reaction. These results were compared with those of the uninvolved adjacent (AP) tissue and benign mucinous cystadenomas (BMC). The mRNA expression was evaluated using logistic regression and receiver operating characteristic area under the curve (ROC AUC) analyses. Their relationship with prognosis was tested by Cox regression multivariate analysis.

RESULTS

All genes were overexpressed in most of the PDAC tissue. When compared with AP tissue, the median expression values for IGF-binding protein 3 (IGFBP-3) and uPA receptor (uPAR) was 9.8- and 9.6-fold, respectively. Expression levels of uPA, uPAR, IGF-I, and IGFBP-3 mRNA were significantly greater in PDAC than in BMC. The IGFBP-3 mRNA expression demonstrated greatest ROC AUC values for PDAC versus AP tissue (ROC AUC, 0.745; 95% confidence interval [CI], 0.65-0.86); whereas ROC AUC values were greatest for uPAR when PDAC was compared with BMC (ROC AUC, 0.846; 95% CI, 0.76-0.94). The combination of uPA, uPAR, and IGF-I significantly improved discriminatory power (ROC AUC, 0.965; 95% CI, 0.93-1.00). The IGFBP-3, uPA, plasminogen activator inhibitor-2, and International Union Against Cancer stage had a significant influence on survival, but the effect of IGFBP-3 was lost after multivariate stepwise analysis.

CONCLUSIONS

These results indicate that there is an influence of IGF system in tumor progression from BMC to PDAC, whereas the uPA/uPAR system has the greater influence on survival in PDAC.

Classification of pancreatic cystic lesions using SELDI-TOF mass spectrometry

BACKGROUND

The diagnosis of pancreatic cystic lesions is problematical with difficulties arising in the differentiation between malignant, premalignant or benign lesions. This preliminary study aimed to analyse pancreatic cyst fluid, using a proteomic approach, to generate reproducible protein profiles to assist in the classification of malignant and non-carcinoma samples.

METHODS

Pancreatic cyst fluid samples from patients with pancreatic adenocarcinoma and non-carcinoma cystic lesions were analysed on hydrophobic protein chip arrays by surface-enhanced laser desorption/ionization time-of-flight mass spectrometry (SELDI-TOF MS).

RESULTS

Differential protein expression profiles were observed between pancreatic adenocarcinoma and non-carcinoma cyst fluid samples using SELDI-TOF MS, with 12 protein peaks differentially expressed between pancreatic adenocarcinoma and non-carcinoma. Additionally, unique patterns were observed between the different subtypes of non-carcinoma samples as well as malignant adenocarcinoma.

CONCLUSIONS

In this preliminary study we used SELDI-TOF MS to identify protein expression profiles of pancreatic cyst fluid, showing a potential to aid in the differential diagnosis of pancreatic cystic lesions.

Application of surface-enhanced laser desorption/ionization time-of-flight mass spectrometry for early detection of pancreatic cancer

Pancreatic cancer is a devastating and lethal disease. Early detection continues to be a serious, unsolved problem. However, proteomics is emerging as a powerful new tool for the diagnosis of pancreatic cancer. Surface-enhanced laser desorption/ionization time-of-flight mass spectrometry (SELDI-TOF MS) is a new technique that allows for rapid high-throughput screening of protein expression in clinical samples. The progress and challenges in applying SELDI-TOF MS to protein biomarker discovery in pancreatic cancer are reviewed in this paper.

Progress on molecular markers of pancreatic cancer

PURPOSE OF REVIEW

To describe advances in the development of biomarkers for pancreatic cancer over the past year.

RECENT FINDINGS

Several new approaches were taken in the search for biomarkers for pancreatic cancer. Studies of CA19-9 revealed new prognostic abilities of the already well known biomarker. New blood biomarkers were investigated and CEACAM1 and MIC-1 were found to be superior to CA19-9 at distinguishing cancer from normal but, unfortunately, not from chronic pancreatitis. MUC1 was reported to be superior to CA19-9 based on the use of a novel immunoassay. The superiority of the concept of a panel of biomarkers as opposed to single biomarkers was supported by several studies, but no such panel was identified. RNA levels in blood and DNA methylation in pancreatic juice yielded some promising findings. Advancements were also made in the area of tissue biomarkers, which can improve the diagnostic accuracy of fine-needle aspirations and provide prognostic information. A new source of potential biomarkers, microRNAs, also made its debut in the past year.

SUMMARY

The tools to identify pancreatic-cancer biomarkers and sources of samples needed in this search are expanding. The field has not yet achieved its aims, but several encouraging breakthroughs have been made.

Technology insight: the application of proteomics in gastrointestinal disease

Analysis of the human genome has increased our knowledge of the genes that are associated with disease. At the same time, however, it has become clear that having complete DNA sequences alone is not sufficient to elucidate the biological functions of the proteins that they encode. For this reason, proteomics-the analysis of proteins-has become increasingly attractive, because the proteome reflects both the intrinsic genetic programming of a cell and the impact of its immediate environment. The principal goals of clinical proteomics are to identify biomarkers for the early diagnosis of disease and potential targets for therapeutic intervention. Other goals include the identification of biomarkers for the early detection of disease recurrence (relapse) and how they might be combined with diagnostic imaging techniques to improve the sensitivity for detecting disease. This Review describes conventional proteomic technologies, their strengths and limitations, and demonstrates their application to clinical practice, with specific reference to their use in the gastroenterology field.

Early diagnostic potential for hepatocellular carcinoma using the SELDI ProteinChip system

Early detection of HCC increases the potential for curative treatment and improves survival. To facilitate early detection of HCC, this study sought to identify novel diagnostic markers of HCC using surface-enhanced laser desorption ionization time-of-flight mass spectrometry (SELDI-TOF/MS) ProteinChip technology. Serum samples were obtained from 153 patients with or without HCC, all of whom had been diagnosed with HCV-associated chronic liver disease. To identify proteins associated with HCC, serum samples were analyzed using SELDI-TOF/MS. We constructed an initial decision tree for the correct diagnosis of HCC using serum samples from patients with (n = 35) and without (n = 44) HCC. Six protein peaks were selected to construct a decision tree using this first group. The efficacy of the decision tree was then assessed using a second group of patients with (n = 29) and without (n = 33) HCC. The sensitivity and specificity of this decision tree for the diagnosis of HCC were 83% and 76%, respectively. For a third group, we analyzed sera from seven patients with HCC obtained before the diagnosis of HCC by ultrasonography (US) and from five patients free of HCC for the past 3 years. Use of these diagnostic markers predicted the diagnosis of HCC in six of these seven patients before HCC was clinically apparent without any false positives.

CONCLUSION

Serum profiling using the SELDI ProteinChip system is useful for the early detection and prediction of HCC in patients with chronic HCV infection.

Proteomic profiling of cholangiocarcinoma: diagnostic potential of SELDI-TOF MS in malignant bile duct stricture

Proteomic techniques promise to improve the diagnosis of cholangiocarcinoma (CC) in both tissue and serum as histological diagnosis and existing serum markers exhibit poor sensitivities. We explored the use of surface-enhanced laser desorption/ionization time-of-flight mass spectrometry (SELDI-TOF MS) to identify potential protein biomarkers of CC. Twenty-two resected CC samples were compared with adjacent noninvolved bile duct tissue. Serum from patients with CC (n=20) was compared with patients with benign disease (n=20), and healthy volunteers (n=25). Samples were analyzed on hydrophobic protein chips via SELDI-TOF MS, and classification models were developed using logistic regression and cross-validation analysis. Univariate analysis revealed 14 individual peaks differentially expressed between CC and bile duct tissue, 4 peaks between CC and benign disease, and 12 peaks between CC and sera of healthy volunteers. The 4,462 mass-to-charge serum peak had superior discriminatory ability to carbohydrate antigen 19.9 (CA19.9) and carcinoembryonic antigen (CEA) (P=.004; receiver operating characteristic [ROC] area under the curve [AUC]=0.76, 0.73, and 0.70, respectively). The training models developed panels of peaks that distinguished CC from bile duct tissue (92.5% sensitivity, 92.3% specificity; ROC AUC=0.96), CC from benign serum (65.0% sensitivity, 70.0% specificity; ROC AUC=0.83), and CC from sera of healthy volunteers (75.0% sensitivity, 100% specificity; ROC AUC=0.92). Serum results were further improved with the inclusion of CA19.9 and CEA (ROC AUC=0.86 and 0.99 for CC vs benign and healthy volunteer serum, respectively). In conclusion, biomarker panels are capable of distinguishing CC from nonmalignant tissue; serum markers have important diagnostic implications for unknown bile duct stricture.