Mass spectrometry assays of plasma biomarkers to predict radiographic progression of knee osteoarthritis

Introduction

Biomarkers to identify osteoarthritis (OA) patients at risk for disease progression are needed. As part of a proteomic analysis of knee synovial fluid from normal and OA patients, differentially expressed proteins were identified that could represent potential biomarkers for OA. This study aimed to use mass spectrometry assays to identify representative peptides from several proteins in synovial fluid and peripheral blood, and assess their levels as biomarkers of OA progression.

Methods

Multiplexed high throughput selected reaction monitoring (SRM) assays were developed to measure tryptic peptides representative of 23 proteins in matched serum and synovial fluid samples from late OA subjects at the time of joint replacement. Subsequently plasma samples from the baseline visit of 173 subjects in an observational OA cohort were tested by SRM for peptides from nine of these proteins: afamin, clusterin, cartilage oligomeric matrix protein, hepatocyte growth factor, kallistatin, insulin-like growth factor binding protein, acid labile subunit, lubricin, lumican, and pigment epithelium-derived factor. Linear regression was used to determine the association between the peptide biomarker level at baseline and change in joint space width (ΔJSW) from baseline to 30 months, adjusting for age and sex.

Results

In the matched cohort, 17 proteins could be identified in synovial fluid and 16 proteins were detected in serum. For the progression cohort, the average age was 62 and average ΔJSW over 30 months was 0.68 mm. A high correlation between different peptides from individual proteins was observed, indicating our assays correctly measured their target proteins. Peptides representative of clusterin, lumican and lubricin showed statistically significant associations with joint space narrowing after adjustment for age and sex. Partial R² values showed clusterin FMETVAEK and lubricin LVEVNPK peptide biomarkers explains about 2 to 3% of the variability of ΔJSW, similar to that explained by age. A biomarker score combining normalized data for both lubricin and clusterin peptides increased the model R² to 0.079.

Conclusions

Our results suggest that when combined, levels of peptides representative of clusterin and lubricin in plasma are as predictive of OA progression as age. Replication of these findings in other prospective OA cohorts is planned.

Electronic supplementary material

The online version of this article (doi:10.1186/s13075-014-0456-6) contains supplementary material, which is available to authorized users.     

Identification of treatment efficacy-related host factors in chronic hepatitis C by ProteinChip serum analysis

Recent development of proteomic array technology, including protein profiling coupling ProteinChip array with surface-enhanced laser desorption ionization time-of-flight mass spectrometry (SELDI-TOF/MS), provides a potentially powerful tool for discovery of new biomarkers by comparison of its profiles according to patient phenotypes. We used this approach to identify the host factors associated with treatment response in patients with chronic hepatitis C (CHC) receiving a 48-wk course of pegylated interferon (PEG-IFN) alpha 2b plus ribavirin (RBV). Protein profiles of pretreatment serum samples from 32 patients with genotype 1b and high viral load were conducted by SELDI-TOF/MS by using the three different ProteinChip arrays (CM10, Q10, IMAC30). Proteins showed significantly different peak intensities between sustained virological responders (SVRs), and non-SVRs were identified by chromatography, SDS-PAGE, TOF/MS and tandem mass spectrometry (MS/MS) assay. Eleven peak intensities were significantly different between SVRs and non-SVRs. The three SVR-increased peaks could be identified as two apolipoprotein (Apo) fragments and albumin and, among the eight non-SVR-increased proteins, four peaks identified as two iron-related and two fibrogenesis-related protein fragments, respectively. Multivariate analysis showed that the serum ferritin and three peak intensity values (Apo A1, hemopexin and transferrin) were independent variables associated with SVRs, and the area under the receiver operating characteristic (ROC) curves for SVR prediction by using the Apo A1/hemopexin and hemopexin/transferrin were 0.964 and 0.936. In conclusion, pretreatment serum protein profiling by SELDI-TOF/MS is variable for identification of response-related host factors, which are useful for treatment efficacy prediction in CHC receiving PEG-IFN plus RBV. Our data also may help us understand the mechanism for treatment resistance and development of more effective antiviral therapy targeted toward the modulation of lipogenesis or iron homeostasis in CHC patients.     

Identification of probable early-onset biomarkers for tuberculosis disease progression

Determining what constitutes protective immunity to TB is critical for the development of improved diagnostics and vaccines. The comparison of the immune system between contacts of TB patients, who later develop TB disease (progressors), versus contacts who remain healthy (non-progressors), allows for identification of predictive markers of TB disease. This study provides the first comprehensive analysis of the immune system of progressors and non-progressors using a well-characterised TB case-contact (TBCC) platform in The Gambia, West Africa. 22 progressors and 31 non-progressors were analysed at recruitment, 3 months and 18 months (time to progression: median[IQR] of 507[187-714] days). Immunophenotyping of PBMC, plasma cytokine levels and RT-MLPA analysis of whole blood-derived RNA was performed to capture key immune system parameters. At recruitment, progressors had lower PBMC proportions of CD4+ T cells, NKT cells and B cells relative to non-progressors. Analysis of the plasma showed higher levels of IL-18 in progressors compared to non-progressors and analysis of the RNA showed significantly lower gene expression of Bcl2 but higher CCR7 in progressors compared to non-progressors. This study shows several markers that may predict the onset of active TB at a very early stage after infection. Once these markers have been validated in larger studies, they provide avenues to prospectively identify people at risk of developing TB, a key issue in the testing of new TB vaccines.     

Comparison of proteomes in various human plasma preparations by two-dimensional gel electrophoresis

Serum or plasma can be utilized in a variety of studies targeted toward the discovery of disease biomarkers. In this study, the proteome profiles of plasma samples prepared using various anticoagulants (EDTA, heparin or citrate), were compared with those of serum using two-dimensional electrophoresis (2-DE). Proteins which evidenced different levels in the plasma and serum were screened and identified using ESI-Q-TOF MS/MS. The proteins which became detectable after the removal of fibrinogen from serum were identified as pigment epithelial differentiating factor (four spots), fetuin-like protein, and the hemopexin precursor. In particular, three proteins, pre-serum amyloid P component, plasma glutathione peroxidase precursor, and tetranectin, evidenced increased volume intensity only in the plasma samples prepared with EDTA.     

Plasma and synovial fluid inflammatory cytokine profiles in primary knee osteoarthritis

Objective: The objective of this study is to compare inflammatory cytokine levels in primary knee osteoarthritis (OA) patients and healthy controls.

Methods: A total of 32 knee OA patients and 14 healthy controls were enrolled. A multiplex immunoassay was utilized for 10 cytokines in plasma and synovial fluid.

Results: Plasma IL-2, IL-4, and IL-6 concentrations were significantly greater in knee OA patients than controls. Moreover, both plasma IL-4 and IL-6 were positively correlated with the radiographic severity of knee OA.

Conclusions: Plasma IL-4 and IL-6 may serve as biomarkers reflecting the severity of OA.     

Association of plasma acylcarnitines with insulin sensitivity, insulin secretion,and prediabetes in a biracial cohort

The ability to predict prediabetes, which affects ∼90 million adults in the US and ∼400 million adults worldwide, would be valuable to public health. Acylcarnitines, fatty acid metabolites, have been associated with type 2 diabetes risk in cross-sectional studies of mostly Caucasian subjects, but prospective studies on their link to prediabetes in diverse populations are lacking. Here, we determined the association of plasma acylcarnitines with incident prediabetes in African Americans and European Americans enrolled in a prospective study. We analyzed 45 acylcarnitines in baseline plasma samples from 70 adults (35 African-American, 35 European-American) with incident prediabetes (progressors) and 70 matched controls (non-progressors) during 5.5-year (mean 2.6 years) follow-up in the Pathobiology of Prediabetes in a Biracial Cohort (POP-ABC) study. Incident prediabetes (impaired fasting glucose/impaired glucose tolerance) was confirmed with OGTT. We measured acylcarnitines using tandem mass spectrometry, insulin sensitivity by hyperinsulinemic euglycemic clamp, and insulin secretion using intravenous glucose tolerance test. The results showed that progressors and non-progressors during POP-ABC study follow-up were concordant for 36 acylcarnitines and discordant for nine others. In logistic regression models, beta-hydroxy butyryl carnitine (C4-OH), 3-hydroxy-isovaleryl carnitine/malonyl carnitine (C5-OH/C3-DC), and octenoyl carnitine (C8:1) were the only significant predictors of incident prediabetes. The combined cut-off plasma levels of <0.03 micromol/L for C4-OH, <0.03 micromol/L for C5-OH/C3-DC, and >0.25 micromol/L for C8:1 acylcarnitines predicted incident prediabetes with 81.9% sensitivity and 65.2% specificity. Thus, circulating levels of one medium-chain and two short-chain acylcarnitines may be sensitive biomarkers for the risk of incident prediabetes among initially normoglycemic individuals with parental history of type 2 diabetes.     

The association of a distinct plasma proteomic profile with the cervical high-grade squamous intraepithelial lesion of Uyghur women: a 2D liquid-phase chromatography/mass spectrometry study

Objective: To identify plasma protein biomarkers of cervical high-grade squamous intraepithelial lesion (HSIL) of Uyghur women by proteomics approach.

Methods: Plasma protein samples of Uyghur women with HSIL and chronic cervicitis were analyzed with 2D HPLC followed by detection of target proteins with Linear Trap Quadrupole Mass Spectrometer (LTQ MS/MS).

Results: We detected three upregulated and one downregulated protein peaks representing protein constituents distinguishing HSIL from controls by 2D HPLC, identified 31 target proteins by LTQ MS/MS. Further confirmed analysis with online software IPA^® 8.7 and ELISA assay showed APOA1 and mTOR as potential biomarkers.

Conclusions: A distinct plasma proteomic profile may be associated with HSIL of Uyghur women.     

Differential carbonylation of proteins as a function of in vivo oxidative stress

This study reports for the first time qualitative and quantitative differences in carbonylated proteins shed into blood as a function of increasing levels of OS. Carbonylated proteins in freshly drawn blood from pairs of diabetic and lean rats were derivatized with biotin hydrazide, dialyzed, and enriched with avidin affinity chromatography. Proteins thus selected were used in several ways. Differences between control and diabetic subjects in relative concentration of proteins was achieved by differential labeling of tryptic digests with iTRAQ reagents followed by reversed phase chromatography (RPC) and tandem mass spectrometry (MS/MS). Identification and characterization of OS induced post-translational modification sites in contrast was achieved by fractionation of affinity selected proteins before proteolysis and RPC-MS/MS. Relative quantification of peptides bearing oxidative modifications was achieved for the first time by selective reaction monitoring (SRM). Approximately 1.7% of the proteins in Zucker diabetic rat plasma were selected by the avidin affinity column as compared to 0.98% in lean animal plasma. Among the 35 proteins identified and quantified, Apo AII, clusterin, hemopexin precursor, and potassium voltage-gated channel subfamily H member 7 showed the most dramatic changes in concentration. Seventeen carbonylation sites were identified and quantified, 11 of which changed more than 2-fold in oxidation state. Three types of carbonylation were identified at these sites: direct oxidative cleavage from reactive oxygen species, glycation and addition of advanced glycation end products, and addition of lipid peroxidation products. Direct oxidation was the dominant form of carbonylation observed while hemoglobin and murinoglobulin 1 homologue were the most heavily oxidized proteins.     

Specific peptide patterns of follicular fluids at different growth stages analyzed by matrix-assisted laser desorption/ionization time-of-flight mass spectrometry

Human follicular fluid (HFF) has been suggested to influence oocyte development potential, and some of HFF proteins may be potential markers for oocyte maturation during follicular development. Using matrix-assisted laser desorption /ionization-time of flight mass spectrometry (MALDI-TOF MS), the presence of specific peptide peaks in HFF which could represent the follicle development potential was evaluated. HFF from different developmental stages were first digested and the resultant peptide mixtures were directly analyzed by MALDI-TOF MS. It was shown that the frequencies of specific peaks demonstrated higher reproducibility than peak intensities after multiple measurements (>or=6 times) per sample. Using this approach, a reliable peak list for each different sample could be generated by combining the information from multiple measurements. By comparing the peak lists from different samples at different growth stages, we found that 5 specific peaks appeared in the 100% frequency category of 6 replicates in all the HFF samples containing mature oocyte. Similarly, such 25 peptide peaks were also identified for HFF containing immature oocyte. These specific peaks could be used to distinguish HFF from different stages as biomarkers related to follicle development and maturation. After searching the protein database, some proteins that are known to be involved in the development and maturation of oocyte were identified, such as apolipoprotein A-I, collagen type IV, integrin, et al. Identification of such proteins in our experiment further proved that the direct analysis of tryptic digests could be of practical value.     

Hemopexin is up-regulated in plasma from type 1 diabetes mellitus patients: Role of glucose-induced ROS

Type 1 diabetes mellitus (T1DM) is an insulin-dependent metabolic disease in the world and often occurs in children and adolescents. Recent advances in quantitative proteomics offer potential for the discovery of plasma proteins as biomarkers for tracking disease progression and for understanding the molecular mechanisms of diabetes. Comparative proteomic analysis of the plasma proteomes from T1DM cases and healthy donors with lysine- and cysteine-labeling 2D-DIGE combining MALDI-TOF/TOF mass spectrometry revealed that 39 identified T1DM-associated plasma proteins showed significant changes in protein expression including hemopexin, and 41 in thiol reactivity. Further study showed that hemopexin can be induced in numerous cell lines by increasing the glucose concentration in the medium. Interestingly, glucose-induced hemopexin expression can be reduced by reactive oxygen species (ROS) scavengers such as glutathione, implying that hemopexin expression is linked to glucose-induced oxidative stress. In conclusion, the current work has identified potential T1DM biomarkers and one of these, hemopexin, can be modulated by glucose through a ROS-dependent mechanism.     

Proteomic identification of plasma biomarkers in uterine leiomyoma

Recent progresses in quantitative proteomics have offered opportunities to discover plasma proteins as biomarkers for tracking the progression and for understanding the molecular mechanisms of uterine leiomyomas. In the present study, plasma samples were analyzed by fluorescence two-dimensional differential gel electrophoresis (2D-DIGE) and differentially expressed proteins were identified by matrix assisted laser desorption ionization-time of flight mass spectrometry (MALDI-TOF MS). In total, 20 proteins have been firmly identified representing 13 unique gene products. These proteins mainly functioned in transportation (such as apolipoprotein A-I) and coagulation (such as fibrinogen gamma chain). Additionally, our quantitative proteomic approach has identified numerous previous reported plasma markers of uterine leiomyomas such as alpha-1-antitrypsin. On the contrary, we have presented several putative uterine leiomyomas biomarkers including afamin, apolipoprotein A-I, carbonic anhydrase 1, fibrinogen beta chain, fibrinogen gamma chain, gelsolin, hemopexin, leucine-rich alpha-2-glycoprotein, serotransferrin and vitamin D-binding protein which have not been reported and may be associated with the progression and development of the disease. In summary, we report a comprehensive patient-based proteomic approach for the identification of potential plasma biomarkers for uterine leiomyomas. The potential of utilizing these markers for screening and treating uterine leiomyomas warrants further investigations.     

Mechanical biomarkers of medial compartment knee osteoarthritis diagnosis and severity grading: Discovery phase

ObjectiveTo investigate, as a discovery phase, if 3D knee kinematics assessment parameters can serve as mechanical biomarkers, more specifically as diagnostic biomarker and burden of disease biomarkers, as defined in the Burden of Disease, Investigative, Prognostic, Efficacy of Intervention and Diagnostic classification scheme for osteoarthritis (OA) (Altman et al., 1986). These biomarkers consist of a set of biomechanical parameters discerned from 3D knee kinematic patterns, namely, flexion/extension, abduction/adduction, and tibial internal/external rotation measurements, during gait recording.Methods100 medial compartment knee OA patients and 40 asymptomatic control subjects participated in this study. OA patients were categorized according to disease severity, by the Kellgren and Lawrence grading system. The proposed biomarkers were identified by incremental parameter selection in a regression tree of cross-sectional data. Biomarker effectiveness was evaluated by receiver operating characteristic curve analysis, namely, the area under the curve (AUC), sensitivity and specificity.ResultsDiagnostic biomarkers were defined by a set of 3 abduction/adduction kinematics parameters. The performance of these biomarkers reached 85% for the AUC, 80% for sensitivity and 90% for specificity; the likelihood ratio was 8%. Burden of disease biomarkers were defined by a 3-decision tree, with sets of kinematics parameters selected from all 3 movement planes.ConclusionThe results demonstrate, as part of a discovery phase, that sets of 3D knee kinematic parameters have the potential to serve as diagnostic and burden of disease biomarkers of medial compartment knee OA.     

Application of proteomics in the discovery of candidate protein biomarkers in a diabetes autoantibody standardization program sample subset

Novel biomarkers of type 1 diabetes must be identified and validated in initial, exploratory studies before they can be assessed in proficiency evaluations. Currently, untargeted "-omics" approaches are underutilized in profiling studies of clinical samples. This report describes the evaluation of capillary liquid chromatography (LC) coupled with mass spectrometry (MS) in a pilot proteomic analysis of human plasma and serum from a subset of control and type 1 diabetic individuals enrolled in the Diabetes Autoantibody Standardization Program, with the goal of identifying candidate biomarkers of type 1 diabetes. Initial high-resolution capillary LC-MS/MS experiments were performed to augment an existing plasma peptide database, while subsequent LC-FTICR studies identified quantitative differences in the abundance of plasma proteins. Analysis of LC-FTICR proteomic data identified five candidate protein biomarkers of type 1 diabetes. alpha-2-Glycoprotein 1 (zinc), corticosteroid-binding globulin, and lumican were 2-fold up-regulated in type 1 diabetic samples relative to control samples, whereas clusterin and serotransferrin were 2-fold up-regulated in control samples relative to type 1 diabetic samples. Observed perturbations in the levels of all five proteins are consistent with the metabolic aberrations found in type 1 diabetes. While the discovery of these candidate protein biomarkers of type 1 diabetes is encouraging, follow up studies are required for validation in a larger population of individuals and for determination of laboratory-defined sensitivity and specificity values using blinded samples.     

Serum protein profile in systemic-onset juvenile idiopathic arthritis differentiates response versus nonresponse to therapy

Systemic-onset juvenile idiopathic arthritis (SJIA) is a disease of unknown etiology with an unpredictable response to treatment. We examined two groups of patients to determine whether there are serum protein profiles reflective of active disease and predictive of response to therapy. The first group (n = 8) responded to conventional therapy. The second group (n = 15) responded to an experimental antibody to the IL-6 receptor (MRA). Paired sera from each patient were analyzed before and after treatment, using surface-enhanced laser desorption/ionization time-of-flight mass spectrometry (SELDI-TOF MS). Despite the small number of patients, highly significant and consistent differences were observed before and after response to therapy in all patients. Of 282 spectral peaks identified, 23 had mean signal intensities significantly different (P < 0.001) before treatment and after response to treatment. The majority of these differences were observed regardless of whether patients responded to conventional therapy or to MRA. These peaks represent potential biomarkers of active disease. One such peak was identified as serum amyloid A, a known acute-phase reactant in SJIA, validating the SELDI-TOF MS platform as a useful technology in this context. Finally, profiles from serum samples obtained at the time of active disease were compared between the two patient groups. Nine peaks had mean signal intensities significantly different (P < 0.001) between active disease in patients who responded to conventional therapy and in patients who failed to respond, suggesting a possible profile predictive of response. Collectively, these data demonstrate the presence of serum proteomic profiles in SJIA that are reflective of active disease and suggest the feasibility of using the SELDI-TOF MS platform used as a tool for proteomic profiling and discovery of novel biomarkers in autoimmune diseases.     

Simultaneous and exact interval estimates for the contrast of two groups based on an extremely high dimensional variable: application to mass spec data

MOTIVATION: Analysis of high-throughput proteomic/genomic data, in particular, surface-enhanced laser desorption/ionization time-of-flight mass spectrometry (SELDI-TOF MS) data and microarray data, has led to a multitude of techniques aimed at identifying potential biomarkers. Most of the statistical techniques for comparing two groups are based on qualitative measures such as P-value. A quantitative way such as interval estimation for the contrasts of two groups is more appealing. RESULTS: We have devised a simultaneous confidence bands method capable of detecting potential biomarkers, while controlling for overall confidence coverage level, in high-dimensional datasets that discriminate two treatment groups using a permutation scheme. For example, for the SELDI-TOF MS data, we deal with the entire spectrum simultaneously and construct (1 - alpha) confidence bands for the mean differences between groups. Furthermore, peaks were identified based on the maximal differences between the groups as determined by the confidence bands. The analysis method herein described gives both qualitative (P-value) and quantitative data (magnitude of difference). The Clinical Proteomics Programs Databank's ovarian cancer dataset and data from in-house samples containing known spiked-in proteins were analyzed. We were able to identify potential biomarkers similar to those described in previous analysis of the ovarian cancer data, however, while these markers are highly significant between cancer and normal groups, our analysis indicated the absolute difference between the two groups was minimal. In addition, we found additional markers than those previously described with greater differences in average intensities. The proposed confidence bands method successfully detected the spiked-in peaks, as well as, secondary peaks generated by adducts and double-charged species. We also illustrate our method utilizing paired gene expression data from a prostate cancer microarray experiment by constructing confidence bands for the fold changes between cancer and normal samples. AVAILABILITY: R-package, 'seie.zip' (license: GNU GPL), is publiclly available at http://research2.dfci.harvard.edu/dfci/MS_spike-in_data/     

Comparison of proteomes in various human plasma preparations by two-dimensional gel electrophoresis

Serum or plasma can be utilized in a variety of studies targeted toward the discovery of disease biomarkers. In this study, the proteome profiles of plasma samples prepared using various anticoagulants (EDTA, heparin or citrate), were compared with those of serum using two-dimensional electrophoresis (2-DE). Proteins which evidenced different levels in the plasma and serum were screened and identified using ESI-Q-TOF MS/MS. The proteins which became detectable after the removal of fibrinogen from serum were identified as pigment epithelial differentiating factor (four spots), fetuin-like protein, and the hemopexin precursor. In particular, three proteins, pre-serum amyloid P component, plasma glutathione peroxidase precursor, and tetranectin, evidenced increased volume intensity only in the plasma samples prepared with EDTA.     

A wide range of protein isoforms in serum and plasma uncovered by a quantitative intact protein analysis system

We have implemented an orthogonal 3-D intact protein analysis system (IPAS) to quantitatively profile protein differences between human serum and plasma. Reference specimens consisting of pooled Caucasian-American serum, citrate-anticoagulated plasma, and EDTA-anticoagulated plasma were each depleted of six highly abundant proteins, concentrated, and labeled with a different Cy dye (Cy5, Cy3, or Cy2). A mixture consisting of each of the labeled samples was subjected to three dimensions of separation based on charge, hydrophobicity, and molecular mass. Differences in the abundance of proteins between each of the three samples were determined. More than 5000 bands were found to have greater than two-fold difference in intensity between any pair of labeled specimens by quantitative imaging. As expected, some of the differences in band intensities between serum and plasma were attributable to proteins related to coagulation. Interestingly, many proteins were identified in multiple fractions, each exhibiting different pI, hydrophobicity, or molecular mass. This is likely reflective of the expression of different protein isoforms or specific protein cleavage products, as illustrated by complement component 3 precursor and clusterin. IPAS provides a high resolution, high sensitivity, and quantitative approach for the analysis of serum and plasma proteins, and allows assessment of PTMs as a potential source of biomarkers.     

Large-scale gene expression profiles, differentially represented in osteoarthritic synovium of the knee joint using cDNA microarray technology

Osteoarthritis (OA) is one of the most common age-related chronic disorders of articular cartilage, joints and bone tissue. Diagnosis of OA commonly depends on clinical and radiographic findings. However, changes in cartilage associated with the early stage of OA cannot be detected using radiographs, because significant cartilage degeneration must occur before radiographic findings show alterations of the appearance of cartilage. To identify new biomarkers of OA, we analysed gene expression profiles of synovium from 43 patients with OA, ten patients with rheumatoid arthritis (RA), and eight non-OA/non-RA patients using a novel cDNA microarray chip. We identified 21 genes with simultaneous significant differences in expression between OA and non-OA/non-RA groups and between OA and RA groups. Linear discriminant analysis showed that the three groups could be well separated using those 21 genes. Statistical analysis also revealed that several of the 21 genes were associated with disease progression and clinical presentation. The graphical modelling method indicated that some of the 21 genes are significantly associated with a particular clinical presentation, suggesting biological relationships among those genes. This is the first report of the use of cDNA microarray technology to create large-scale gene expression profiles differentially expressed in situ in OA synovium of the knee joint.     

Evaluation of changes in serum protein profiles during neoadjuvant chemotherapy in HER2‐positive breast cancer using an LC‐MALDI‐TOF/MS procedure

Comparison of protein profiles of sera acquired before and after preoperative chemotherapy for breast cancer may reveal tumor markers that could be used to monitor tumor response. In this study, we analyzed pre‐ and post‐chemotherapy protein profiles of sera from 39 HER2‐postive breast cancer patients (n=78 samples) who received 6 months of preoperative chemotherapy using LC‐MALDI‐TOF/MS technology. We detected qualitative and quantitative differences in pair‐wise comparison of pre‐ and post chemotherapy samples that were different in patients who achieved pathological complete response (pCR, n=21) compared with those with residual disease (n=18). We identified 2329 and 3152 peaks as differentially expressed in the pre‐chemotherapy samples of the responders and non‐responders. Comparison of matching pre‐ and post‐chemotherapy samples identified 34 (32 decreased, two increased) and 304 peaks (157 decreased, 147 increased) that significantly changed (p<0.01, false discovery rate ≤20%) after treatment in responders and non‐responders, respectively. The top 11 most significantly altered peptide peaks with the greatest change in intensity were positively identified. These corresponded to eight proteins including α‐2‐macroglobulin, complement 3, hemopexin, and serum amyloid P in the responder group and chains C and A of apolipoprotein A‐I, hemopexin precursor, complement C, and amyloid P component in the non‐responding groups. All proteins decreased after therapy, except chain C apolipoprotein A and hemopexin precursor that increased. These results suggest that changes in serum protein levels occur in response to chemotherapy and these changes partly appear different in patients who are highly sensitive to chemotherapy compared with those with lesser response.     

The wisdom of the commons: ensemble tree classifiers for prostate cancer prognosis

MOTIVATION: Classification and regression trees have long been used for cancer diagnosis and prognosis. Nevertheless, instability and variable selection bias, as well as overfitting, are well-known problems of tree-based methods. In this article, we investigate whether ensemble tree classifiers can ameliorate these difficulties, using data from two recent studies of radical prostatectomy in prostate cancer. RESULTS: Using time to progression following prostatectomy as the relevant clinical endpoint, we found that ensemble tree classifiers robustly and reproducibly identified three subgroups of patients in the two clinical datasets: non-progressors, early progressors and late progressors. Moreover, the consensus classifications were independent predictors of time to progression compared to known clinical prognostic factors.