Goodison, Steve

Abstract The interpretation of accumulating genomic data with respect to tumor evolution and cancer progression requires integrated models. We developed a computational approach that enables the construction of disease progression models using static sample data. Application to breast cancer data revealed a linear, branching evolutionary model with two distinct trajectories for malignant progression. Here, we used the progression model as a foundation to investigate the relationships between matched primary and metastasis breast tumor samples. Mapping paired data onto the model confirmed that molecular breast cancer subtypes can shift during progression and supported directional tumor evolution through luminal subtypes to increasingly malignant states. Cancer progression modeling through the analysis of available static samples represents a promising breakthrough. Further refinement of a roadmap of breast cancer progression will facilitate the development of improved cancer diagnostics, prognostics, and targeted therapeutics. Significance: Analysis of matched primary and metastatic tumor samples supports a unidirectional, linear cancer evolution process and sheds light on longstanding issues regarding the origins of molecular subtypes and their progression relationships.

Abstract Motivation Cancer subtype classification has the potential to significantly improve disease prognosis and develop individualized patient management. Existing methods are limited by their ability to handle extremely high-dimensional data and by the influence of misleading, irrelevant factors, resulting in ambiguous and overlapping subtypes. Results To address the above issues, we proposed a novel approach to disentangling and eliminating irrelevant factors by leveraging the power of deep learning. Specifically, we designed a deep-learning framework, referred to as DeepType, that performs joint supervised classification, unsupervised clustering and dimensionality reduction to learn cancer-relevant data representation with cluster structure. We applied DeepType to the METABRIC breast cancer dataset and compared its performance to state-of-the-art methods. DeepType significantly outperformed the existing methods, identifying more robust subtypes while using fewer genes. The new approach provides a framework for the derivation of more accurate and robust molecular cancer subtypes by using increasingly complex, multi-source data. Availability and implementation An open-source software package for the proposed method is freely available at http://www.acsu.buffalo.edu/~yijunsun/lab/DeepType.html. Supplementary information Supplementary data are available at Bioinformatics online.

471 Background: Urine based assays that can non-invasively detect bladder cancer (BCa) have the potential to reduce unnecessary and invasive procedures. The purpose of this study was to develop a multiplex immunoassay that can accurately and simultaneously monitor 10 diagnostic urinary protein biomarkers for application as a non-invasive test for BCa detection Methods: A custom electrochemiluminescent (ECL) multiplex assay was constructed (Meso Scale Diagnostics, LLC, Rockville, MD) to detect the following urinary proteins; IL8, MMP9, MMP10, ANG, APOE, SDC1, A1AT, PAI1, CA9 and VEGFA. Voided urine samples from two cohorts (cohort #1 n = 62 and cohort #2 n = 200) were collected prior to cystoscopy and samples were analyzed blinded to the clinical status of the participants. Means (±SD) and receiver operating characteristic (ROC) curve analysis were used to compare assay performance and to assess the diagnostic accuracy of the diagnostic signature. Results: Comparative diagnostic performance analyses revealed an AUROC value of 0.9258 for the multiplex assay and 0.9467 for the combination of the single-target ELISA assays (p = 0.625), so there was no loss of diagnostic utility for the MSD multiplex assay. Analysis of the independent 200-sample cohort using the multiplex assay achieved an overall diagnostic sensitivity of 0.85, specificity of 0.81, positive predictive value 0.82 and negative predictive value 0.84. Conclusions: It is technically feasible to simultaneously monitor complex urinary diagnostic signatures in a single assay without loss of performance. The described protein-based assay has the potential to be developed for the non-invasive detection of BCa.

Abstract Background: Improvements in the noninvasive clinical evaluation of patients at risk for bladder cancer would be of benefit both to individuals and to health care systems. We investigated the potential utility of a hybrid nomogram that combined key demographic features with the results of a multiplex urinary biomarker assay in hopes of identifying patients at risk of harboring bladder cancer. Methods: Logistic regression analysis was used to model the probability of bladder cancer burden in a cohort of 686 subjects (394 with bladder cancer) using key demographic features alone, biomarker data alone, and the combination of demographic features and key biomarker data. We examined discrimination, calibration, and decision curve analysis techniques to evaluate prediction model performance. Results: Area under the receiver operating characteristic curve (AUC) analyses revealed that demographic features alone predicted tumor burden with an accuracy of 0.806 [95% confidence interval (CI), 0.76–0.85], while biomarker data had an accuracy of 0.835 (95% CI, 0.80–0.87). The addition of molecular data into the nomogram improved the predictive performance to 0.891 (95% CI, 0.86–0.92). Decision curve analyses showed that the hybrid nomogram performed better than demographic or biomarker data alone. Conclusion: A nomogram construction strategy that combines key demographic features with biomarker data may facilitate the accurate, noninvasive evaluation of patients at risk of harboring bladder cancer. Further research is needed to evaluate the bladder cancer risk nomogram for potential clinical utility. Impact: The application of such a nomogram may better inform the decision to perform invasive diagnostic procedures. Cancer Epidemiol Biomarkers Prev; 25(9); 1361–6. ©2016 AACR.