Cancers of the urinary bladder result in aggressive and highly angiogenic tumors for which standard treatments have only limited success. Patients with advanced disease have a 5-year survival rate of less than 20%, and no new anticancer agent has been successfully introduced into the clinic armamentarium for the treatment of bladder cancer in more than 20 years. Investigations have identified plasminogen activator inhibitor-1 (PAI-1), a serine protease inhibitor, as being highly expressed in several malignancies, including bladder cancer, in which high expression is associated with a poor prognosis. In this study, we evaluated PAI-1 as a potential therapeutic target for bladder cancer. PAI-1 expression was manipulated in a panel of cell lines and functional inhibition was achieved using the small molecule tiplaxtinin. Reduction or inhibition of PAI-1 resulted in the reduction of cellular proliferation, cell adhesion, and colony formation, and the induction of apoptosis and anoikis in vitro. Treatment of T24 xenografts with tiplaxtinin resulted in inhibition of angiogenesis and induction of apoptosis, leading to a significant reduction in tumor growth. Similar results were obtained through evaluation of the human cervical cancer HeLa cell line, showing that PAI-1–mediated effects are not restricted to tumor cells of bladder origin. Collectively, these data show that targeting PAI-1 may be beneficial and support the notion that novel drugs such as tiplaxtinin could be investigated as anticancer agents.
Accurate urine assays for bladder cancer detection would benefit patients and health care systems. Through extensive genomic and proteomic profiling of urine components we previously identified a panel of 8 biomarkers that can facilitate the detection of bladder cancer in voided urine samples. In this study we confirmed this diagnostic molecular signature in a diverse multicenter cohort.
Materials and Methods
We performed a case-control, phase II study in which we analyzed voided urine from 102 subjects with bladder cancer and 206 with varying urological disorders. The urine concentration of 8 biomarkers (IL-8, MMP and 10, PAI-1, VEGF, ANG, CA9 and APOE) was assessed by enzyme-linked immunosorbent assay. Diagnostic performance of the panel of tested biomarkers was evaluated using ROCs and descriptive statistical values, eg sensitivity and specificity.
Seven of the 8 urine biomarkers were increased in subjects with bladder cancer relative to those without bladder cancer. The 7 biomarkers were assessed in a new model, which had an AUROC of 0.88 (95% CI 0.84–0.93), and 74% sensitivity and 90% specificity. In contrast, the sensitivity of voided urine cytology and the UroVysion® cytogenetic test in this cohort was 39% and 54%, respectively. Study limitations include analysis performed on banked urine samples and the lack of voided urine cytology and cytogenetic test data on controls.
The study provides further evidence that the reported panel of diagnostic biomarkers can reliably achieve the noninvasive detection of bladder cancer with higher sensitivity than currently available urine based assays.
urinary bladder neoplasms; urine; biological markers; molecular biology; diagnosis
To validate the expression of a urine-based bladder cancer associated diagnostic signature comprised of 10 targets; ANG, CA9, MMP9, MMP10, SERPINA1, APOE, SDC1, VEGFA, SERPINE1 and IL8 in bladder tumor tissues.
Immunohistochemical analyses were performed on tumor specimens from 213 bladder cancer patients (transitional cell carcinoma only) and 74 controls. Staining patterns were digitally captured and quantitated (Aperio, Vista, CA), and expression was correlated with tumor stage, tumor grade and outcome measures.
We revealed a positive association of 9 of the 10 proteins (excluding VEGF) in bladder cancer. Relative to control cases, a reduction in SDC1 and overexpression of MMP9, MMP10, SERPINE1, IL8, APOE, SERPINA1, ANG were associated with high stage bladder cancer. Reduced VEGF and increased SERPINA1 were associated with high-grade bladder cancer. Disease-specific survival was significantly reduced in tumors with high expression of SERPINE1 and/or IL8.
These findings confirm that the proteins in a urine-based diagnostic signature are aberrantly expressed in bladder tumor tissues, and support the potential additional utility of selected biomarkers for the clinicopathological evaluation of excised tissue or biopsy material.
The virtual slide(s) for this article can be found here: http://www.diagnosticpathology.diagnomx.eu/vs/13000_2014_200
Bladder cancer; Diagnosis; Grade; Signature; Stage
As molecular profiling data continue to accumulate, the design of integrative computational analyses that can provide insights into the dynamic aspects of cancer progression becomes feasible. Here, we present a novel computational method for the construction of cancer progression models based on the analysis of static tumor samples. We demonstrate the reliability of the method with simulated data, and describe the application to breast cancer data. Our findings support a linear, branching model for breast cancer progression. An interactive model facilitates the identification of key molecular events in the advance of disease to malignancy.
Electronic supplementary material
The online version of this article (doi:10.1186/s13059-014-0440-0) contains supplementary material, which is available to authorized users.
Intravesical Bacillus Calmette-Guérin (BCG) has been shown to induce a specific immunologic response (i.e., activation of IL-2 and effector T-cells), while preclinical studies using ALT-803 (mutated IL-15 analogue combined with IL-15Rα-Fc fusion) have shown promising results by prolonging the agent's half-life and stimulating CD8+ T-cells. Based on these results, we hypothesized that the intravesical administration of ALT-803 along with BCG will generate an immunologic response leading to significant bladder tumor burden reduction. Using a well-established carcinogen induced rat non-muscle invasive bladder cancer (NMIBC) model, we studied the effects of intravesical ALT-803 with and without BCG. Rat tissues were evaluated to document treatment response. Intravesical ALT-803 was safe and well tolerated alone and in combination with BCG. As a single treatment agent, ALT-803 reduced tumor burden by 35% compared to control whereas BCG alone only reduced tumor burden by 15%. However, the combination of ALT-803 plus BCG reduced tumor burden by 46% compared to control. Immune monitoring suggested that the antitumor response was linked to the production and secretion of IL-1α, IL-1β and RANTES, which in turn, induced the proliferation and activation of NK cells. Lastly, tumoral responses of the combinational treatment were associated with 76% reduction in angiogenesis, which is significantly higher than when assessed with either agent alone. The enhanced therapeutic index seen with this duplet provides justification for the development of this regimen for future clinical trials.
The ability to reliably diagnose bladder cancer in voided urine samples would be a major advance. Using high throughput technologies, we identified a panel of bladder cancer associated biomarkers with potential clinical usefulness. In this study we tested 4 potential biomarkers for the noninvasive detection of bladder cancer.
Materials and Methods
We examined voided urine specimens from 124 patients, including 63 newly diagnosed with bladder cancer and 61 controls. Concentrations of proteins were assessed by enzyme-linked immunosorbent assay, including α1-antitrypsin, apolipoprotein E, osteopontin and pentraxin 3. Data were compared to the results of urinary cytology and the BTA Trak® enzyme-linked immunosorbent assay based bladder cancer detection assay. We used the AUC of ROC curves to compare the usefulness of each biomarker to detect bladder cancer.
Urinary levels of α1-antitrypsin, apolipoprotein E and bladder tumor antigen were significantly increased in subjects with bladder cancer. α1-Antitrypsin (AUC 0.9087, 95% CI 0.8555–0.9619) and apolipoprotein E (AUC 0.8987, 95% CI 0.8449–0.9525) were the most accurate biomarkers. The combination of α1-antitrypsin and apolipoprotein E (AUC 0.9399) achieved 91% sensitivity, 89% specificity, and a positive and negative predictive value of 89% and 90%, respectively. Multivariate regression analysis highlighted only apolipoprotein E as an independent predictor of bladder cancer (OR 24.9, 95% CI 4.22–146.7, p = 0.0004).
Alone or in combination, α1-antitrypsin and apolipoprotein E show promise for the noninvasive detection of bladder cancer (OR 24.9, 95% CI 4.22– 146.7, p = 0.0004). Larger, prospective studies including more low grade, low stage tumors are needed to confirm these results.
urinary bladder; urinary bladder neoplasms; alpha 1-antitrypsin; apolipoproteins A; diagnosis
Cancer invasion and metastasis develops through a series of steps that involve the loss of cell to cell and cell to matrix adhesion, degradation of extracellular matrix and induction of angiogenesis. Different protease systems (e.g., matrix metalloproteinases, MMPs) are involved in these steps. MMP-10, one of the lesser studied MMPs, is limited to epithelial cells and can facilitate tumor cell invasion by targeting collagen, elastin and laminin. Enhanced MMP-10 expression has been linked to poor clinical prognosis in some cancers, however, mechanisms underlying a role for MMP-10 in tumorigenesis and progression remain largely unknown. Here, we report that MMP-10 expression is positively correlated with the invasiveness of human cervical and bladder cancers.
Using commercial tissue microarray (TMA) of cervical and bladder tissues, MMP-10 immunohistochemical staining was performed. Furthermore using a panel of human cells (HeLa and UROtsa), in vitro and in vivo experiments were performed in which MMP-10 was overexpressed or silenced and we noted phenotypic and genotypic changes.
Experimentally, we showed that MMP-10 can regulate tumor cell migration and invasion, and endothelial cell tube formation, and that MMP-10 effects are associated with a resistance to apoptosis. Further investigation revealed that increasing MMP-10 expression stimulates the expression of HIF-1α and MMP-2 (pro-angiogenic factors) and PAI-1 and CXCR2 (pro-metastatic factors), and accordingly, targeting MMP-10 with siRNA in vivo resulted in diminution of xenograft tumor growth with a concomitant reduction of angiogenesis and a stimulation of apoptosis.
Taken together, our findings show that MMP-10 can play a significant role in tumor growth and progression, and that MMP-10 perturbation may represent a rational strategy for cancer treatment.
Angiogenesis; Apoptosis; Cancer; Invasion; MMP-10
The ability to accurately measure multiple proteins simultaneously in a single assay has the potential to markedly improve the efficiency of a myriad of clinical assays. Here, we tested the performance of a new, multiplex protein array platform to quantitate three bladder cancer-associated proteins in urine samples. The following analytes, interleukin 8 (IL8), matrix metallopeptidase 9 (MMP9), and vascular endothelial growth factor A (VEGFA) were monitored using Q-plex, a customized multiplex ELISA system from Quansys Biosciences, and individual target commercial ELISA kits. The performance of the two approaches was compared by evaluating the diagnostic accuracy of the biomarker assays in samples from a cohort of 73 subjects of known bladder cancer status.
The combination biomarker panel analyses revealed an AUROC value of 0.9476 for the Q-plex assay, and 0.9119 for the combination of the single-target ELISA assays. The Q-plex assay achieved an overall diagnostic sensitivity of 0.93 and specificity of 0.81, and the individual target ELISA assays achieved an overall sensitivity of 0.77 and specificity of 0.91.
Based on these encouraging preliminary data, we believe that the Q-Plex technology is a viable platform that can be exploited as an efficient, highly accurate tool to quantitate multiplex panels of diagnostic proteins in biologic specimens.
Biomarkers; Bladder cancer; Q-plex; Protein; Urine
To determine the diagnostic and prognostic capability of urinary and tumoral syndecan-1 (SDC-1) levels in patients with cancer of the urinary bladder.
SDC-1 levels were quantitated by enzyme-linked immunosorbent assay (ELISA) in 308 subjects (102 cancer subjects and 206 non-cancer subjects) to assess its diagnostic capabilities in voided urine. The performance of SDC-1 was evaluated using the area under the curve of a receiver operating characteristic curve. In addition, immunohistochemical (IHC) staining assessed SDC-1 protein expression in 193 bladder specimens (185 cancer subjects and 8 non-cancer subjects). Outcomes were correlated to SDC-1 levels.
Mean urinary levels of SDC-1 did not differ between the cancer subjects and the non-cancer subjects, however, the mean urinary levels of SDC-1 were reduced in high-grade compared to low-grade disease (p < 0.0001), and in muscle invasive bladder cancer (MIBC) compared to non-muscle invasive bladder cancer (NMIBC) (p = 0.005). Correspondingly, preliminary data note a shift from a membranous cellular localization of SDC-1 in normal tissue, low-grade tumors and NMIBC, to a distinctly cytoplasmic localization in high-grade tumors and MIBC was observed in tissue specimens.
Alone urinary SDC-1 may not be a diagnostic biomarker for bladder cancer, but its urinary levels and cellular localization were associated with the differentiation status of patients with bladder tumors. Further studies are warranted to define the potential role for SDC-1 in bladder cancer progression.
Syndecan; Bladder; Cancer biomarker; Specificity
The development of accurate and reliable molecular assays that could diagnose bladder cancer would be of significant benefit to both patients and the healthcare system. Non-invasive assays that have utility not only for diagnosis, but also for monitoring disease recurrence and response to treatment, are needed. Current urinary tests lack sufficient sensitivity or specificity, often because of a reliance on single biomarkers, but high-throughput technologies are enabling the derivation of more accurate panels of biomarkers. In this article, we review some of the promising investigational studies that are revealing multiplex biomarker signatures that may augment current bladder cancer detection strategies.
bladder cancer; biomarkers; non-invasive; urinalysis; diagnosis
Details of metastasis, the deadliest aspect of cancer, are unclear. Cell surface proteins play central roles in adhesive contacts between the tumor cell and the stroma during metastasis. We optimized a fast, small-scale isolation of biotinylated cell surface proteins to reveal novel metastasis-associated players froman isogenic pair of human MDA-MB-435 cancer cells with opposite metastatic phenotypes. Isolated proteins were trypsin digested and analyzed using LC–MS/MS followed by quantitation with the Progenesis LC–MS software. Sixteen proteins displayed over twofold expression differences between the metastatic and non-metastatic cells. Interestingly, overexpression of most of them (14/16) in the metastatic cells indicates a gain of novel surface protein profile as compared to the non-metastatic one. All five validated, differentially expressed proteins showed higher expression in the metastatic cells in culture, and four of these were further validated in vivo. Moreover, we analyzed the expression of two of the identified proteins, CD109 and ITGA6 in 3-dimensional cultures of six melanoma cell lines. Both proteins marked the surface of cells derived from melanoma metastasis over cells derived from primary melanoma. These unbiased identification and validation of both known and novel metastasis-associated proteins indicate a reliable approach for the identification of differentially expressed surface proteins.
Cell surface; Biotinylation; Metastasis; CD109; ITGA6; LC–MS/MS
In this study, we investigated the influence of hematuria on the performance of the BTA tests in a clinical cohort and in an experimental model.
Materials and Methods
Analysis of urine samples from a cohort of 126 subjects (64 with BCa and 62 controls) were analyzed by ELISA for hemoglobin and BTA. The experimental model involved the spiking of urine with blood from the same subject, and hemoglobin, red blood cell count, and BTA levels (BTA stat© and BTA-TRAK©). BTA-TRAK© analyses were also performed on serum samples obtained from 40 subjects (20 with confirmed with BCa).
In the 126-subject cohort, correlation between hemoglobin and BTA was 0.732. Of the 64 BCa samples, 72% had a positive BTA assay, but 47% of controls were also positive. The sensitivity and specificity of BTA to detect BCa was 72% and 53%, respectively. Hematuria, measured by urinary hemoglobin, was a better indicator of BCa with 75% sensitivity and 90% specificity. Spiking of BTA-negative urine samples with as little as 1μl/10ml was enough to produce positive a BTA test. High levels of BTA were found equally in the serum of subjects with, or without BCa (mean BTA levels 355,159 U/ml vs. 332,329 U/ml, respectively).
Rather than detecting a bladder tumor antigen, urinary BTA assays may be measuring serum cFH introduced by bleeding, a common presenting factor in BCa subjects. The presence of hematuria in subjects without malignant disease can result in false-positive BTA assays.
complement factor H; hematuria; BTA; bladder cancer; diagnosis
Bladder cancer (BCa) is among the five most common malignancies world-wide, and due to high rates of recurrence, one of the most prevalent. Improvements in non-invasive urine-based assays to detect BCa would benefit both patients and healthcare systems. In this study, the goal was to identify urothelial cell transcriptomic signatures associated with BCa.
Gene expression profiling (Affymetrix U133 Plus 2.0 arrays) was applied to exfoliated urothelia obtained from a cohort of 92 subjects with known bladder disease status. Computational analyses identified candidate biomarkers of BCa and an optimal predictive model was derived. Selected targets from the profiling analyses were monitored in an independent cohort of 81 subjects using quantitative real-time PCR (RT-PCR),
Transcriptome profiling data analysis identified 52 genes associated with BCa (p≤0.001), and gene models that optimally predicted class label were derived. RT-PCR analysis of 48 selected targets in an independent cohort identified a 14-gene diagnostic signature that predicted the presence of BCa with high accuracy.
Exfoliated urothelia sampling provides a robust analyte for the evaluation of patients with suspected BCa. The refinement and validation of the multi-gene urothelial cell signatures identified in this preliminary study may lead to accurate, non-invasive assays for the detection of BCa.
The development of an accurate, non-invasive BCa detection assay would benefit both the patient and healthcare systems through better detection, monitoring and control of disease.
Genomic profiling; Bladder cancer; Urinalysis; Non-invasive detection
In this study, we further investigated the association of two biomarkers, CCL18 and A1AT, with bladder cancer (BCa) and evaluated the influence of potentially confounding factors in an experimental model.
In a cohort of 308 subjects (102 with BCa), urinary concentrations of CCL18 and A1AT were assessed by enzyme-linked immunosorbent assay (ELISA). In an experimental model, benign or cancerous cells, in addition to blood, were added to urines from healthy controls and analyzed by ELISA. Lastly, immunohistochemical staining for CCL18 and A1AT in human bladder tumors was performed.
Median urinary protein concentrations of CCL18 (52.84 pg/ml vs. 11.13 pg/ml, p < 0.0001) and A1AT (606.4 ng/ml vs. 120.0 ng/ml, p < 0.0001) were significantly elevated in BCa subjects compared to controls. Furthermore, the addition of whole blood to pooled normal urine resulted in a significant increase in both CCL18 and A1AT. IHC staining of bladder tumors revealed CCL18 immunoreactivity in inflammatory cells only, and there was no significant increase in these immunoreactive cells within benign and cancerous tissue and no association with BCa grade nor stage was noted. A1AT immunoreactivity was observed in the cytoplasm of epithelia cells and intensity of immunostaining increased with tumor grade, but not tumor stage.
Further development of A1AT as a diagnostic biomarker for BCa is warranted.
Biomarkers; Bladder cancer; Specificity; Urine
Erythropoietin (EPO) provides an alternative to transfusion for increasing red blood cell mass and treating anemia in cancer patients. However, recent studies have reported increased adverse events and/or reduced survival in patients receiving both EPO and chemotherapy, potentially related to EPO-induced cancer progression. Additional preclinical studies that elucidate the possible mechanism underlying EPO cellular growth stimulation are needed.
Using commercial tissue microarray (TMA) of a variety of cancers and benign tissues, EPO and EPO receptor immunohistochemical staining was performed. Furthermore using a panel of human renal cells (Caki-1, 786-O, 769-P, RPTEC), in vitro and in vivo experiments were performed with the addition of EPO in normoxic and hypoxic states to note phenotypic and genotypic changes.
EPO expression score was significantly elevated in lung cancer and lymphoma (compared to benign tissues), while EPOR expression score was significantly elevated in lymphoma, thyroid, uterine, lung and prostate cancers (compared to benign tissues). EPO and EPOR expression scores in RCC and benign renal tissue were not significantly different. Experimentally, we show that exposure of human renal cells to recombinant EPO (rhEPO) induces cellular proliferation, which we report for the first time, is further enhanced in a hypoxic state. Mechanistic investigations revealed that EPO stimulates the expression of cyclin D1 while inhibiting the expression of p21cip1 and p27kip1 through the phosphorylation of JAK2 and ERK1/2, leading to a more rapid progression through the cell cycle. We also demonstrate an increase in the growth of renal cell carcinoma xenograft tumors when systemic rhEPO is administered.
In summary, we elucidated a previously unidentified mechanism by which EPO administration regulates progression through the cell cycle, and show that EPO effects are significantly enhanced under hypoxic conditions.
Cancer; ERK1/2; Erythropoietin; Hypoxia; JAK2; Renal
Chemokines, including chemokine (C-X-C motif) ligand 1 (CXCL1), may regulate tumor epithelial-stromal interactions that facilitate tumor growth and invasion. Studies have linked CXCL1 expression to gastric, colon and skin cancers, but limited studies to date have described CXCL1 protein expression in human bladder cancer (BCa).
CXCL1 protein expression was examined in 152 bladder tissue specimens (142 BCa) by immunohistochemical staining. The expression of CXCL1 was scored by assigning a combined score based on the proportion of cells staining and intensity of staining. CXCL1 expression patterns were correlated with clinicopathological features and follow-up data.
CXCL1 protein expression was present in cancerous tissues, but was entirely absent in benign tissue. CXCL1 combined immunostaining score was significantly higher in high-grade tumors relative to low-grade tumors (p = 0.012). Similarly, CXCL1 combined immunostaining score was higher in high stage tumors (T2-T4) than in low stage tumors (Ta-T1) (p < 0.0001). An increase in the combined immunostaining score of CXCL1 was also associated with reduced disease-specific survival.
To date, this is the largest study describing increased CXCL1 protein expression in more aggressive phenotypes in human BCa. Further studies are warranted to define the role CXCL1 plays in bladder carcinogenesis and progression.
Bladder Cancer; Chemokine Ligand 1 (CXCL1); Tumor Grade; Tumor Stage
To investigate whether elevated urinary levels of vascular endothelial growth factor (VEGF), carbonic anhydrase 9 (CA9) and angiogenin are associated with BCa.
This is a case-control study in which voided urines from 127 patients: control subjects (n = 63) and tumor bearing subjects (n = 64) were analyzed. The urinary concentrations of VEGF, CA9, angiogenin and BTA were assessed by enzyme-linked immunosorbent assay (ELISA). We used the area under the curve (AUC) of receiver operating characteristic curves to determine the ability of VEGF, CA9, and angiogenin to detect BCa in voided urine samples. Data were also compared to a commercial ELISA-based BCa detection assay (BTA-Trak©). Sensitivity, specificity, positive and negative predictive values were calculated.
Urinary concentrations of VEGF, CA9, angiogenin and BTA were significantly elevated in BCa. VEGF was the most accurate urinary biomarker (AUC: 0.886; 95% confidence interval [CI]: 0.8301–0.9418). Furthermore, multivariate regression analysis highlighted VEGF (OR: 5.90; 95% CI: 2.60–13.40, p < 0.0001) as an independent variable. The sensitivities and specificities for VEGF (sensitivity, 83% and specificity, 87%) outperformed BTA (sensitivity, 80% and specificity, 84%).
VEGF may be a valuable addition to voided urine sample analysis for the detection of BCa. Larger, prospective studies are needed to determine the clinical utility of urinary VEGF and angiogenin as biomarkers in the non-invasive evaluation of BCa patients.
angiogenin; bladder cancer; biomarkers; diagnosis; VEGF
Bladder cancer is one of the most prevalent cancers worldwide, but the treatment and management of this disease can be very successful if the disease is detected early. The development of molecular assays that could diagnose bladder cancer accurately, and at an early stage, would be a significant advance. Ideally, such molecular assays would be applicable to non-invasively obtained body fluids, and be designed not only for diagnosis but also for monitoring disease recurrence and response to treatment. In this article, we assess the performance of current diagnostic assays for bladder cancer and discuss some of the emerging biomarkers that could be developed to augment current bladder cancer detection strategies.
Mycoplasma hyorhinis is a eubacterium belonging to the Mollicutes class and is responsible for porcine respiratory and arthritic diseases. It is also the major contaminant of mammalian tissue cultures in laboratories worldwide. Here, we report the complete genome sequence of M. hyorhinis strain SK76.
In the United States, breast cancer affects one in eight women, with mortality that is second only to lung cancer. Although chemotherapy is widely used in breast cancer treatment, its side effects remain a challenge. One way to address this problem is through drug delivery by the internalization of cell type-specific probes. Although nucleic acid aptamers are excellent probes for molecular recognition, only a few reports have demonstrated that aptamers can be internalized into living cells. Therefore, in this work, we report the development of a cancer cell-specific DNA aptamer probe, KMF2-1a. Using the cell-SELEX method, this aptamer was selected against breast cancer cell line MCF-10AT1. Our results showed that KMF2-1a was internalized efficiently and specifically to the endosome of target breast cancer cells. These results indicate that KMF2-1a is a promising agent for cell type-specific intracellular delivery with both diagnostic and therapeutic implications.
aptamer; breast cancer; internalization; MCF-10AT1
Accurate urinary assays for bladder cancer (BCa) detection would benefit both patients and healthcare systems. Through genomic and proteomic profiling of urine components, we have previously identified a panel of biomarkers that can outperform current urine-based biomarkers for the non-invasive detection of BCa. Herein, we report the diagnostic utility of various multivariate combinations of these biomarkers. We performed a case-controlled validation study in which voided urines from 127 patients (64 tumor bearing subjects) were analyzed. The urinary concentrations of 14 biomarkers (IL-8, MMP-9, MMP-10, SDC1, CCL18, PAI-1, CD44, VEGF, ANG, CA9, A1AT, OPN, PTX3, and APOE) were assessed by enzyme-linked immunosorbent assay (ELISA). Diagnostic performance of each biomarker and multivariate models were compared using receiver operating characteristic curves and the chi-square test. An 8-biomarker model achieved the most accurate BCa diagnosis (sensitivity 92%, specificity 97%), but a combination of 3 of the 8 biomarkers (IL-8, VEGF, and APOE) was also highly accurate (sensitivity 90%, specificity 97%). For comparison, the commercial BTA-Trak ELISA test achieved a sensitivity of 79% and a specificity of 83%, and voided urine cytology detected only 33% of BCa cases in the same cohort. These datashow that a multivariate urine-based assay can markedly improve the accuracy of non-invasive BCa detection. Further validation studies are under way to investigate the clinical utility of this panel of biomarkers for BCa diagnosis and disease monitoring.
This paper considers feature selection for data classification in the presence of a huge number of irrelevant features. We propose a new feature-selection algorithm that addresses several major issues with prior work, including problems with algorithm implementation, computational complexity, and solution accuracy. The key idea is to decompose an arbitrarily complex nonlinear problem into a set of locally linear ones through local learning, and then learn feature relevance globally within the large margin framework. The proposed algorithm is based on well-established machine learning and numerical analysis techniques, without making any assumptions about the underlying data distribution. It is capable of processing many thousands of features within minutes on a personal computer while maintaining a very high accuracy that is nearly insensitive to a growing number of irrelevant features. Theoretical analyses of the algorithm’s sample complexity suggest that the algorithm has a logarithmical sample complexity with respect to the number of features. Experiments on 11 synthetic and real-world data sets demonstrate the viability of our formulation of the feature-selection problem for supervised learning and the effectiveness of our algorithm.
Feature selection; local learning; logistical regression; ℓ1 regularization; sample complexity
Hepatocellular carcinoma (HCC) has a tendency for intravascular dissemination leading to a poor prognosis. The importance of the sinusoidal structure of the tumor vasculature in HCC has been implicated in the metastasis formation. To clarify the role of tumor angiogenesis in HCC metastasis, we morphologically investigated the interaction of HCC cells with blood vessels during the sequential process of metastasis. Autopsy specimens of 80 patients with HCC were examined with immunohistochemistry using a specific antibody against CD31, a marker for endothelial cells. The most frequent sites of metastasis were the liver (82.5%) and lung (43.8%). In most cases, the metastatic process was initiated by vascular involvement where tumor nests surrounded by sinusoidal vessels extend into the portal and hepatic veins. Subsequently, these endothelial-coated tumor emboli enter the circulation, embolize at distant organs, proliferate within the blood vessel and ultimately form metastatic foci. These steps are indicative of an invasion-independent pathway. Our findings in animal models and now in human cases suggest that sinusoidal angiogenesis may represent a novel target for therapeutic strategies to limit HCC metastasis. In combination with primary tumor treatment, perturbation of tumor emboli may reduce dissemination of disease.
Hepatocellular carcinoma; Metastasis; Angiogenesis; Invasion-independent pathway; Autopsy
Accurate prognosis of breast cancer can spare a significant number of breast cancer patients from receiving unnecessary adjuvant systemic treatment and its related expensive medical costs. Recent studies have demonstrated the potential value of gene expression signatures in assessing the risk of post-surgical disease recurrence. However, these studies all attempt to develop genetic marker-based prognostic systems to replace the existing clinical criteria, while ignoring the rich information contained in established clinical markers. Given the complexity of breast cancer prognosis, a more practical strategy would be to utilize both clinical and genetic marker information that may be complementary.
A computational study is performed on publicly available microarray data, which has spawned a 70-gene prognostic signature. The recently proposed I-RELIEF algorithm is used to identify a hybrid signature through the combination of both genetic and clinical markers. A rigorous experimental protocol is used to estimate the prognostic performance of the hybrid signature and other prognostic approaches. Survival data analyses is performed to compare different prognostic approaches.
The hybrid signature performs significantly better than other methods, including the 70-gene signature, clinical makers alone and the St. Gallen consensus criterion. At the 90% sensitivity level, the hybrid signature achieves 67% specificity, as compared to 47% for the 70-gene signature and 48% for the clinical makers. The odds ratio of the hybrid signature for developing distant meta-stases within five years between the patients with a good prognosis signature and the patients with a bad prognosis is 21.0 (95% CI: 6.5–68.3), far higher than either genetic or clinical markers alone.