Application of Bayesian modeling of autologous antibody responses against ovarian tumor-associated antigens to cancer detection

Bayesian multilevel model of micro RNA levels in ovarian-cancer and healthy subjects

In transcriptomics, micro RNAs (miRNAs) has gained much interest especially as potential disease indicators. However, apart from holding a great promise related to their clinical application, a lot of inconsistent results have been published. Our aim was to compare the miRNA expression levels in ovarian cancer and healthy subjects using the Bayesian multilevel model and to assess their potential usefulness in diagnosis. We have analyzed a case-control observational data on expression profiling of 49 preselected miRNA-based ovarian cancer indicators in 119 controls and 59 patients. A Bayesian multilevel model was used to characterize the effect of disease on miRNA levels controlling for differences in age and body weight. The difference between the miRNA level and health status of the patient on the scale of the data variability were discussed in the context of their potential usefulness in diagnosis. Additionally, the cross-validated area under the ROC curve (AUC) was used to assess the expected out-of-sample discrimination index of a different sets of miRNAs. The proposed model allowed us to describe the set of miRNA levels in patients and controls. Three highly correlated miRNAs: miR-101-3p, miR-142-5p, miR-148a-3p rank the highest with almost identical effect sizes that ranges from 0.45 to 1.0. For those miRNAs the credible interval for AUC ranged from 0.63 to 0.67 indicating their limited discrimination potential. A little benefit in adding information from other miRNAs was observed. There were several miRNAs in the dataset (miR-604, hsa-miR-221-5p) for which inferences were uncertain. For those miRNAs more experimental effort is needed to fully assess their effect in the context of new hits discovery and usefulness as disease indicators. The proposed multilevel Bayesian model can be used to characterize the panel of miRNA profile and to assess the difference in expression levels between healthy and cancer individuals.

Anti-nuclear antibodies in patients with breast cancer

To study the prevalence of anti-nuclear antibodies (ANA) in breast cancer patients and its association with tumour characteristics. Ninety-one patients with breast mass detected by image studies and assigned to conduct diagnostic biopsy and eventual surgical treatment were studied for demographical, tumour data and presence of ANA. Serum of positive ANA patients was screened for the extractable nuclear antigen (ENA) profile. As comparison, 91 healthy individuals matched for age and from the same geographical area were included. In this sample 72 of 91 (79·1%) had malignant lesions (83% ductal infiltrative carcinoma). ANA was positive in 44·4% of patients with malignant tumour and in 15·7% of those with benign lesions (malignant versus benign with P = 0·03). Controls had ANA positivity in 5·4%, and when compared with tumour samples showed P < 0·0001. The most common immunofluorescence pattern was a fine dense speckled pattern. In the ANA-positive patients with malignant lesions, seven had positivity for ENA profile (three for anti-RNP and anti-Sm, one for just anti-RNP, two for anti-Ro and anti-La e two for just anti-La). It was not possible to associate ANA positivity with tumour histological characteristics or staging or with patient's age. A negative association of ANA with hormonal (oestrogen or oestrogen plus progesterone) receptor status was found (P = 0·01). In this sample, there was a high prevalence of ANA positivity in breast cancer patients with a negative association with the presence of hormonal receptors. More studies are needed to understand the real value of this finding.

Selenium-Binding Protein 1 (SBP1) autoantibodies in ovarian disorders and ovarian cancer

Infertility is a risk factor for ovarian cancer (OvCa). The goal was to determine if antibodies to selenium-binding protein 1 (SBP1), an autoantibody we identified in patients with premature ovarian failure (POF), occurs in both infertility and OvCa patients, and thus could be associated with preneoplasia. Anti-SBP1 was measured by immunoassay against recombinant SBP1, in sera from OvCa (n = 41), infertility (n = 92) and control (n = 87) patients. Infertility causes were POF, unexplained, irregular ovulation or endometriosis. The percent of anti-SBP1-positive sera was higher in POF (P = 0.02), irregular ovulation (P = 0.001), unexplained causes (P = 0.02), late (III-IV)-stage OvCa (P = 0.02) but was not significant in endometriosis, benign ovarian tumors/cysts, early stage (I-II) OvCa or uterine cancer compared to healthy controls. Anti-SBP1 was significantly higher in women with serous (P = 0.04) but not non-serous (P = 0.33) OvCa compared to controls. Also, we determined if anti-SBP1 was associated with CA125 or anti-TP53, markers often studied in OvCa. Anti-TP53 and CA125 were measured by established immunoassays. The ability of anti-SBP1 alone to discriminate infertility or OvCa from controls or when combined with anti-TP53 and CA125, to identify OvCa was evaluated by comparing the area under the curve (AUC) in ROC analysis. Anti-SBP1 alone discriminated infertility (AUC = 0.7; P = 0.001) or OvCa (AUC = 0.67; P = 0.03) from controls. The sensitivity and specificity of OvCa identification was increased by combining CA125, anti-TP53 and anti-SBP1 (AUC = 0.96). Therefore, anti-SBP1 occurs in infertile women with POF, ovulatory disturbances or unexplained infertility and in serous OvCa. This suggests an autoimmune process is associated with the development of serous OvCa.

The Widening Sphere of Influence of HOXB7 in Solid Tumors

Strong lines of evidence have established a critical role for the homeodomain protein HOXB7 in cancer. Specifically, molecular and cellular studies have demonstrated that HOXB7 is a master regulatory gene, capable of orchestrating a variety of target molecules, resulting in the activation of several oncogenic pathways. HOXB7 overexpression correlates with clinical progression and poor outcome of cancer patients. Specific inhibition of HOXB7 is particularly relevant in cancers still lacking effective therapies, such as tamoxifen-resistant breast cancer and melanoma. Mechanistic studies are providing additional targets of therapy, and biomarker studies are further establishing its importance in early diagnosis and prognosis. Cancer Res; 76(10); 2857-62. ©2016 AACR.

Combination of Autoantibody Signature with PSA Level Enables a Highly Accurate Blood-Based Differentiation of Prostate Cancer Patients from Patients with Benign Prostatic Hyperplasia

Although an increased level of the prostate-specific antigen can be an indication for prostate cancer, other reasons often lead to a high rate of false positive results. Therefore, an additional serological screening of autoantibodies in patients’ sera could improve the detection of prostate cancer. We performed protein macroarray screening with sera from 49 prostate cancer patients, 70 patients with benign prostatic hyperplasia and 28 healthy controls and compared the autoimmune response in those groups. We were able to distinguish prostate cancer patients from normal controls with an accuracy of 83.2%, patients with benign prostatic hyperplasia from normal controls with an accuracy of 86.0% and prostate cancer patients from patients with benign prostatic hyperplasia with an accuracy of 70.3%. Combining seroreactivity pattern with a PSA level of higher than 4.0 ng/ml this classification could be improved to an accuracy of 84.1%. For selected proteins we were able to confirm the differential expression by using luminex on 84 samples. We provide a minimally invasive serological method to reduce false positive results in detection of prostate cancer and according to PSA screening to distinguish men with prostate cancer from men with benign prostatic hyperplasia.

An Antibody-based Blood Test Utilizing a Panel of Biomarkers as a New Method for Improved Breast Cancer Diagnosis

In order to develop a new tool for diagnosis of breast cancer based on autoantibodies against a panel of biomarkers, a clinical trial including blood samples from 507 subjects was conducted. All subjects showed a breast abnormality on exam or breast imaging and final biopsy pathology of either breast cancer patients or healthy controls. Using an enzyme-linked immunosorbent assay, the samples were tested for autoantibodies against a predetermined number of biomarkers in various models that were used to determine a diagnosis, which was compared to the clinical status. Our new assay achieved a sensitivity of 95.2% [CI = 92.8–96.8%] at a fixed specificity of 49.5%. Receiver-operator characteristic curve analysis showed an area under the curve of 80.1% [CI = 72.6–87.6%]. These results suggest that a blood test which is based on models comprising several autoantibodies to specific biomarkers may be a new and novel tool for improving the diagnostic evaluation of breast cancer.

Autoantibodies to mesothelin in infertility

BACKGROUND

According to extensive epidemiologic data, infertility is associated with increased ovarian cancer risk. Previous studies showed that both women with infertility and those with ovarian cancer have autoantibodies to ovarian antigens. The objective was to determine if women with infertility have antibodies to mesothelin, a well-characterized ovarian cancer antigen.

METHODS

Sera were obtained from women with infertility (n = 109), ovarian cancer (n = 28), benign ovarian tumors or cysts (n = 24), and from healthy women (n = 152). Infertility included those with a risk for ovarian cancer; endometriosis (n = 23), ovulatory dysfunction (n = 17), premature ovarian failure (POF; n = 25) and unexplained infertility (n = 44). Sera were assayed for mesothelin antibodies and for circulating mesothelin antigen by immunoassay and compared with assay control sera (n = 16) to determine a positive result.

RESULTS

Mesothelin antibodies were significantly more frequent in women with prematurely reduced ovarian function including ovulatory dysfunction (59%), ovarian failure (44%) and unexplained infertility (25%) compared with controls. In contrast, women with endometriosis, who also have a high risk for ovarian cancer, did not have mesothelin antibodies. Serum levels of mesothelin were rarely elevated in women with infertility but were high in most patients with ovarian cancer.

CONCLUSIONS AND IMPACT

We show for the first time that antibodies to mesothelin, a well-characterized ovarian cancer antigen, occur in some women with epidemiologic risk for ovarian cancer. The results suggest it may be possible to identify which women with infertility have ovarian cancer risk.

Autoantibodies against stress-induced phosphoprotein-1 as a novel biomarker candidate for ovarian cancer

Detection of autoantibodies against tumor-associated antigens (TAA) has recently been shown to be a powerful tool for early detection of various cancers. The aim of this study was to investigate the possibility of using autoantibodies against TAA as novel biomarkers by a proteomics-based approach in patients with ovarian cancer. We used two-dimensional differential gel electrophoresis analysis of immuno-precipitated tumor antigens (2D-DITA) to compare the levels of autoantibodies in pretreatment and posttreatment sera of patients with ovarian cancers. The identified autoantibodies were validated by SYBR Green real-time polymerase chain reaction (PCR) and immunohistochemistry (IHC). We further evaluated the level of autoantibody in sera of 68 ovarian cancer patients by an enzyme-linked immunosorbent assay (ELISA). The autoantibody directed against stress-induced phosphoprotein-1 (STIP-1) emerged as a novel biomarker candidate for ovarian cancer. SYBR Green PCR and IHC confirmed that the STIP-1 mRNA and protein expression levels were significantly up-regulated in ovarian cancers compared with normal and benign tumors (P = 0.003 and P < 0.001, respectively). A preliminary ELISA study showed that the serum levels of anti-STIP-1 autoantibodies were significantly elevated in ovarian cancer patients compared with healthy controls (P = 0.03). The results suggest that 2D-DITA is a useful tool to detect autoantibodies and that STIP-1 is a potential biomarker candidate for ovarian cancers.

Improving seroreactivity-based detection of glioma

Seroreactivity profiling emerges as valuable technique for minimal invasive cancer detection. Recently, we provided first evidence for the applicability of serum profiling of glioma using a limited number of immunogenic antigens. Here, we screened 57 glioma and 60 healthy sera for autoantibodies against 1827 Escherichia coli expressed clones, including 509 in-frame peptide sequences. By a linear support vector machine approach, we calculated mean specificity, sensitivity, and accuracy of 100 repetitive classifications. We were able to differentiate glioma sera from sera of the healthy controls with a specificity of 90.28%, a sensitivity of 87.31% and an accuracy of 88.84%. We were also able to differentiate World Health Organization grade IV glioma sera from healthy sera with a specificity of 98.45%, a sensitivity of 80.93%, and an accuracy of 92.88%. To rank the antigens according to their information content, we computed the area under the receiver operator characteristic curve value for each clone. Altogether, we found 46 immunogenic clones including 16 in-frame clones that were informative for the classification of glioma sera versus healthy sera. For the separation of glioblastoma versus healthy sera, we found 91 informative clones including 26 in-frame clones. The best-suited in-frame clone for the classification glioma sera versus healthy sera corresponded to the vimentin gene (VIM) that was previously associated with glioma. In the future, autoantibody signatures in glioma not only may prove useful for diagnosis but also offer the prospect for a personalized immune-based therapy.

Autoantibodies to tumor-associated antigens in epithelial ovarian carcinoma

This review will focus on recent knowledge related to circulating autoantibodies (AAbs) to tumor-associated antigens (TAAs) in epithelial ovarian carcinoma. So far, the following TAAs have been identified to elicit circulating AAbs in epithelial ovarian carcinoma: p53, homeobox proteins (HOXA7, HOXB7), heat shock proteins (HSP-27, HSP-90), cathepsin D, cancer-testis antigens (NY-ESO-1/LAGE-1), MUC1, GIPC-1, IL-8, Ep-CAM, and S100A7. Since AAbs to TAAs have been identified in the circulation of patients with early-stage cancer, it has been speculated that the assessment of a panel of AAbs specific for epithelial ovarian carcinoma TAAs might hold great potential as a novel tool for early diagnosis of epithelial ovarian carcinoma.

A 7 gene expression score predicts for radiation response in cancer cervix

BACKGROUND

Cervical cancer is the most common cancer among Indian women. The current recommendations are to treat the stage IIB, IIIA, IIIB and IVA with radical radiotherapy and weekly cisplatin based chemotherapy. However, Radiotherapy alone can help cure more than 60% of stage IIB and up to 40% of stage IIIB patients.

METHODS

Archival RNA samples from 15 patients who had achieved complete remission and stayed disease free for more than 36 months (No Evidence of Disease or NED group) and 10 patients who had failed radical radiotherapy (Failed group) were included in the study. The RNA were amplified, labelled and hybridized to Stanford microarray chips and analyzed using BRB Array Tools software and Significance Analysis of Microarray (SAM) analysis. 20 genes were selected for further validation using Relative Quantitation (RQ) Taqman assay in a Taqman Low-Density Array (TLDA) format. The RQ value was calculated, using each of the NED sample once as a calibrator. A scoring system was developed based on the RQ value for the genes.

RESULTS

Using a seven gene based scoring system, it was possible to distinguish between the tumours which were likely to respond to the radiotherapy and those likely to fail. The mean score +/- 2 SE (standard error of mean) was used and at a cut-off score of greater than 5.60, the sensitivity, specificity, Positive predictive value (PPV) and Negative predictive value (NPV) were 0.64, 1.0, 1.0, 0.67, respectively, for the low risk group.

CONCLUSION

We have identified a 7 gene signature which could help identify patients with cervical cancer who can be treated with radiotherapy alone. However, this needs to be validated in a larger patient population.

Development of a clinical decision model for thyroid nodules

Background

Thyroid nodules represent a common problem brought to medical attention. Four to seven percent of the United States adult population (10–18 million people) has a palpable thyroid nodule, however the majority (>95%) of thyroid nodules are benign. While, fine needle aspiration remains the most cost effective and accurate diagnostic tool for thyroid nodules in current practice, over 20% of patients undergoing FNA of a thyroid nodule have indeterminate cytology (follicular neoplasm) with associated malignancy risk prevalence of 20–30%. These patients require thyroid lobectomy/isthmusectomy purely for the purpose of attaining a definitive diagnosis. Given that the majority (70–80%) of these patients have benign surgical pathology, thyroidectomy in these patients is conducted principally with diagnostic intent. Clinical models predictive of malignancy risk are needed to support treatment decisions in patients with thyroid nodules in order to reduce morbidity associated with unnecessary diagnostic surgery.

Methods

Data were analyzed from a completed prospective cohort trial conducted over a 4-year period involving 216 patients with thyroid nodules undergoing ultrasound (US), electrical impedance scanning (EIS) and fine needle aspiration cytology (FNA) prior to thyroidectomy. A Bayesian model was designed to predict malignancy in thyroid nodules based on multivariate dependence relationships between independent covariates. Ten-fold cross-validation was performed to estimate classifier error wherein the data set was randomized into ten separate and unique train and test sets consisting of a training set (90% of records) and a test set (10% of records). A receiver-operating-characteristics (ROC) curve of these predictions and area under the curve (AUC) were calculated to determine model robustness for predicting malignancy in thyroid nodules.

Results

Thyroid nodule size, FNA cytology, US and EIS characteristics were highly predictive of malignancy. Cross validation of the model created with Bayesian Network Analysis effectively predicted malignancy [AUC = 0.88 (95%CI: 0.82–0.94)] in thyroid nodules. The positive and negative predictive values of the model are 83% (95%CI: 76%–91%) and 79% (95%CI: 72%–86%), respectively.

Conclusion

An integrated predictive decision model using Bayesian inference incorporating readily obtainable thyroid nodule measures is clinically relevant, as it effectively predicts malignancy in thyroid nodules. This model warrants further validation testing in prospective clinical trials.

Antigen-specific immunotherapy of cervical and ovarian cancer

We contrast the efforts to treat ovarian cancer and cervical cancer through vaccination because of their different pathobiology. A plethora of approaches have been developed for therapeutic vaccination against cancer, many of which target defined tumor-associated antigens (TAAs). Persistent infection with oncogenic human papillomavirus (HPV) types causes cervical cancer. Furthermore, cervical cancer patients frequently mount both humoral and T-cell immune responses to the HPV E6 and E7 oncoproteins, whose expression is required for the transformed phenotype. Numerous vaccine studies target these viral TAAs, including recent trials that may enhance clearance of pre-malignant disease. By contrast, little is known about the etiology of epithelial ovarian cancer. Although it is clear that p53 mutation or loss is a critical early event in the development of epithelial ovarian cancer, no precursor lesion has been described for the most common serous histotype, and even the location of its origin is debated. These issues have complicated the selection of appropriate ovarian TAAs and the design of vaccines. Here we focus on mesothelin as a promising ovarian TAA, because it is overexpressed and immunogenic at high frequency in patients, is displayed on the cell surface, and potentially contributes to ovarian cancer biology.

Pattern of serum autoantibodies allows accurate distinction between a tumor and pathologies of the same organ

PURPOSE

Recent studies impressively showed the diagnostic potential of seroreactivity patterns for different tumor types, offering the prospect for low-cost screening of numerous tumor types simultaneously. One of the major challenges toward this goal is to prove that seroreactivity profiles do not only allow for identifying a tumor but also allow for distinguishing tumors from other pathologies of the same organ.

EXPERIMENTAL DESIGN

We chose glioma as a model system and tested 325 sera (88 glioma, 95 intracranial tumors, 60 other brain pathologies, and 82 healthy controls) for seroreactivity on a panel of 35 antigens.

RESULTS

We were able to discriminate between glioma and all other sera with cross-validated specificity of 86.1%, sensitivity of 85.2%, and accuracy of 85.8%. We obtained comparably good results for the separation of glioma versus nontumor brain pathologies and glioma versus other intracranial tumors.

CONCLUSION

Our study provides first evidence that seroreactivity patterns allow for an accurate discrimination between a tumor and pathologies of the same organ even between different tumor types of the same organ.

Use of a combination of approaches to identify and validate relevant tumor-associated antigens and their corresponding autoantibodies in ovarian cancer patients

PURPOSE

Novel biomarkers are urgently needed to increase the sensitivity of CA125 for the early detection of ovarian cancer. Indeed, it has been shown that as much as 20% of early-stage patients do not express significant levels of this biomarker. Therefore, the possibility of using autoantibodies directed against tumor-associated antigens as putative cancer markers is being more examined. Indeed, many autoantibodies have recently been shown to correlate with cancer patient prognosis or to be suitable for detection of the disease.

EXPERIMENTAL DESIGN

In this study, we have used a new approach involving the use of proteomics, immunology, and ELISA methods to identify relevant autoantibodies in the plasma of ovarian cancer patients. To do so, we developed an innovative technique called two-dimensional differential gel electrophoresis analysis of immunoprecipitated tumor antigens.

RESULTS

This strategy allowed us to successfully identify novel circulating autoantibodies directed against the S100A7 protein in the plasma of ovarian cancer patients. Further real-time reverse transcription-PCR and immunohistochemical studies confirmed that the S100A7 mRNA and protein were highly expressed in ovarian tumors but absent in normal and benign tissues. Moreover, a preliminary study involving 138 patients confirmed that the plasma levels of anti-S100A7 antibodies are significantly elevated in early- and late-stage ovarian cancer patients compared with healthy controls and with patients with benign gynecologic diseases.

CONCLUSIONS

This shows that our approach is a valuable tool to successfully identify autoantibodies and tumor-associated antigens in cancer patients and that future research assessing their putative clinical usefulness would be worthwhile.

Proteomic profiling identifies afamin as a potential biomarker for ovarian cancer

PURPOSE

To discover and validate serum glycoprotein biomarkers in ovarian cancer using proteomic-based approaches.

EXPERIMENTAL DESIGN

Serum samples from a "discovery set" of 20 patients with ovarian cancer or benign ovarian cysts or healthy volunteers were compared by fluorescence two-dimensional differential in-gel electrophoresis and parallel lectin-based two-dimensional profiling. Validation of a candidate biomarker was carried out with Western blotting and immunoassay (n = 424).

RESULTS

Twenty-six proteins that changed significantly were identified by mass spectrometric sequencing. One of these, confirmed by Western blotting, was afamin, a vitamin E binding protein, with two isoforms decreasing in patients with ovarian cancer. Validation using cross-sectional samples from 303 individuals (healthy controls and patients with benign, borderline, or malignant ovarian conditions and other cancers) assayed by ELISA showed significantly decreased total afamin concentrations in patients with ovarian cancer compared with healthy controls (P = 0.002) and patients with benign disease (P = 0.046). However, the receiver operating characteristic areas for total afamin for the comparison of ovarian cancer with healthy controls or benign controls were only 0.67 and 0.60, respectively, with comparable figures for CA-125 being 0.92 and 0.88 although corresponding figures for a subgroup of samples analyzed by isoelectric focusing for afamin isoform 2 were 0.85 and 0.79. Analysis of a further 121 samples collected prospectively from 9 patients pretreatment through to relapse indicated complementarity of afamin with CA-125, including two cases in whom CA-125 was noninformative.

CONCLUSIONS

Afamin shows potential complementarity with CA-125 in longitudinal monitoring of patients with ovarian cancer, justifying prospective larger-scale investigation. Changes in specific isoforms may provide further information.

Endometriosis markers: immunologic alterations as diagnostic indicators for endometriosis

Because endometriosis, a chronic disease affecting 7% to 10% of women, is associated with immunologic aberrations, the authors hypothesize that the presence of specific immune alterations may be diagnostic. Autoantibodies were assayed by Western immunoblotting using antigens derived from the plasma membrane, cytosol, and nucleus from endometrial and ovarian cells. Natural killer (NK) activity was defined by levels of signaling protein zeta and induction of interferon (IFN)-gamma following exposure to patients' sera. Patients with endometriosis exhibited autoantibodies reactive with cellular proteins; endometrial membrane proteins exhibited the greatest reactivity, followed by nuclear antigens. In all subcellular fractions, patients with stage 3 endometriosis exhibited significantly more immunoreactivity than did stage 2 patients, which was greater than that observed in stage 1 patients. The stage-associated increased reactivity resulted from both recognition of additional proteins and enhanced reactivity with shared proteins. Patient sera suppressed NK zeta expression, which resulted in suppression of NK IFN induction. Alterations in autoreactivity and NK activity are observed in endometriosis and may be useful as diagnostic markers, even in early stage disease.

SePaCS--a web-based application for classification of seroreactivity profiles

Immunogenic antigen sets possess high potential for minimally invasive disease detection and monitoring. For various diseases, including cancer, appropriate antigen sets have already been detected in blood sera of patients. Typically, a large number of sera from diseased and unaffected persons is screened for the antigens of interest. Sophisticated statistical learning approaches are trained on the resulting data set to classify sera as either tumor or normal sera. We developed a web-based application, called ‘Seroreactivity Profile Classification Service’ (SePaCS) that enables clinical groups to carry out analyzes of training sets and predictions of unclassified seroreactivity profiles with minimal effort. SePaCS provides a broad range of classification methods: four versions of a Naïve Bayes Classifier, Support Vector Machines with a radial basis function kernel, Linear Discriminant Analysis, and Diagonal Discriminant Analysis. The computed results are summarized in a PDF file. We demonstrate the functionality of SePaCS exemplarily for meningioma, a generally benign intracranial tumor. As a second example, we evaluated SePaCS on glioma, a malignant brain tumor. SePaCS is freely available at http://www.bioinf.uni-sb.de/sepacs.

Design of a bio-mathematical prediction model using serum tumor markers and immunohistochemistry in peritoneal carcinomatosis with ovarian involvement: a pilot study

The diagnosis of disseminated intra-abdominal malignancy in women with ovarian involvement can be problematic. Whilst both blood tumor markers and use of immunohistochemical staining on tissue can help decide the origin of the tumor, this is done separately. This study looked at the blood and tissue marker profiles of 198 cases of disseminated malignancy to construct a model, which may help to determine tumor origin. The original histology material from 198 cases of disseminated intra-abdominal epithelial malignancy were reviewed, blind, and reassessed as to the likely site of origin. These cases had immunohistochemical (IHC) staining for cytokeratins (CK) 7 and 20, carcinoembryonic antigen (CEA) and CA125. Blood values for CEA and CA125 were also known at diagnosis. The histologic types of the tumors in this pilot study were of ovarian type morphologically in 130 cases (65.7%), nonovarian in 32 (16.1%), and not assigned in 36 cases (18.2%). The majority of the nonovarian cases were of mucinous type or too poorly differentiated to classify. Analysis showed an overall sensitivity and specificity of 93% and 69%, and positive predictive and negative predictive value of 92% and 71%, respectively, for a diagnosis of ovarian vs nonovarian origin using histology alone vs histology and IHC. Use of an ordinal regression developed a model which uses tissue staining for CK 7 and CEA along with blood levels of CEA to help determine the site of tumor origin.

Targeting serum antibody for cancer diagnosis: a focus on colorectal cancer

The ability of the immune system to magnify the appearance of disease by generating relatively large amounts of antibody in response to small amounts of disease makes it a natural biosensor, and serum antibodies have emerged as promising biomarkers for the detection of cancer. This review summarizes recent progress in targeting serum antibodies for cancer diagnosis, with a particular focus on colorectal cancer (CRC). Several serum antibodies have been detected at increased levels in CRC patients, including p53, carcinoembryonic antigen, Ras, topoisomerase II-alpha, histone deacetylase 3 and 5, ubiquitin C-terminal hydrolase L3, tropomyosin and cyclin B1. As each antibody is only present in a limited proportion of patients (usually < 40%), a combination of serum antibodies that defines the 'immunological signature' of cancer needs to be developed. High-throughput methods to identify new serum antibodies for cancer diagnosis are also reviewed.

A minimally invasive multiple marker approach allows highly efficient detection of meningioma tumors

Background

The development of effective frameworks that permit an accurate diagnosis of tumors, especially in their early stages, remains a grand challenge in the field of bioinformatics. Our approach uses statistical learning techniques applied to multiple antigen tumor antigen markers utilizing the immune system as a very sensitive marker of molecular pathological processes. For validation purposes we choose the intracranial meningioma tumors as model system since they occur very frequently, are mostly benign, and are genetically stable.

Results

A total of 183 blood samples from 93 meningioma patients (WHO stages I-III) and 90 healthy controls were screened for seroreactivity with a set of 57 meningioma-associated antigens. We tested several established statistical learning methods on the resulting reactivity patterns using 10-fold cross validation. The best performance was achieved by Naïve Bayes Classifiers. With this classification method, our framework, called Minimally Invasive Multiple Marker (MIMM) approach, yielded a specificity of 96.2%, a sensitivity of 84.5%, and an accuracy of 90.3%, the respective area under the ROC curve was 0.957. Detailed analysis revealed that prediction performs particularly well on low-grade (WHO I) tumors, consistent with our goal of early stage tumor detection. For these tumors the best classification result with a specificity of 97.5%, a sensitivity of 91.3%, an accuracy of 95.6%, and an area under the ROC curve of 0.971 was achieved using a set of 12 antigen markers only. This antigen set was detected by a subset selection method based on Mutual Information. Remarkably, our study proves that the inclusion of non-specific antigens, detected not only in tumor but also in normal sera, increases the performance significantly, since non-specific antigens contribute additional diagnostic information.

Conclusion

Our approach offers the possibility to screen members of risk groups as a matter of routine such that tumors hopefully can be diagnosed immediately after their genesis. The early detection will finally result in a higher cure- and lower morbidity-rate.