Dataset Information

Improving breast cancer survival analysis through competition-based multidimensional modeling.

ABSTRACT: Breast cancer is the most common malignancy in women and is responsible for hundreds of thousands of deaths annually. As with most cancers, it is a heterogeneous disease and different breast cancer subtypes are treated differently. Understanding the difference in prognosis for breast cancer based on its molecular and phenotypic features is one avenue for improving treatment by matching the proper treatment with molecular subtypes of the disease. In this work, we employed a competition-based approach to modeling breast cancer prognosis using large datasets containing genomic and clinical information and an online real-time leaderboard program used to speed feedback to the modeling team and to encourage each modeler to work towards achieving a higher ranked submission. We find that machine learning methods combined with molecular features selected based on expert prior knowledge can improve survival predictions compared to current best-in-class methodologies and that ensemble models trained across multiple user submissions systematically outperform individual models within the ensemble. We also find that model scores are highly consistent across multiple independent evaluations. This study serves as the pilot phase of a much larger competition open to the whole research community, with the goal of understanding general strategies for model optimization using clinical and molecular profiling data and providing an objective, transparent system for assessing prognostic models.

SUBMITTER: Bilal E

PROVIDER: S-EPMC3649990 | biostudies-literature | 2013

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Improving breast cancer survival analysis through competition-based multidimensional modeling.

Bilal Erhan E Dutkowski Janusz J Guinney Justin J Jang In Sock IS Logsdon Benjamin A BA Pandey Gaurav G Sauerwine Benjamin A BA Shimoni Yishai Y Moen Vollan Hans Kristian HK Mecham Brigham H BH Rueda Oscar M OM Tost Jorg J Curtis Christina C Alvarez Mariano J MJ Kristensen Vessela N VN Aparicio Samuel S Børresen-Dale Anne-Lise AL Caldas Carlos C Califano Andrea A Friend Stephen H SH Ideker Trey T Schadt Eric E EE Stolovitzky Gustavo A GA Margolin Adam A AA

PLoS computational biology 20130509 5

Breast cancer is the most common malignancy in women and is responsible for hundreds of thousands of deaths annually. As with most cancers, it is a heterogeneous disease and different breast cancer subtypes are treated differently. Understanding the difference in prognosis for breast cancer based on its molecular and phenotypic features is one avenue for improving treatment by matching the proper treatment with molecular subtypes of the disease. In this work, we employed a competition-based appr ...[more]

PMID: 23671412

Similar Datasets

Project description:BackgroundColon cancer has high morbidity and mortality rates among cancers. Existing clinical staging systems cannot accurately assess the prognostic risk of colon cancer patients. This study was aimed at improving the prognostic performance of the colon cancer clinical staging system through knowledge-based clinical-molecular integrated analysis.Methods374 samples from The Cancer Genome Atlas Colon Adenocarcinoma (TCGA-COAD) dataset were used as the discovery set. 98 samples from the Clinical Proteomic Tumor Analysis Consortium (CPTAC) dataset were used as the validation set. After converting gene expression data into pathway dysregulation scores (PDSs), the random survival forest and Cox model were used to identify the best prognostic supplementary factors. The corresponding clinical-molecular integrated prognostic model was built, and the improvement of prognostic performance was assessed by comparing with the clinical prognostic model.ResultsThe PDS of 14 pathways played important roles in prognostic prediction together with clinical prognostic factors through the random survival forest. Further screening with the Cox model revealed that the PDS of the pathway hsa00532 was the best clinical prognostic supplementary factor. The integrated prognostic model constructed with clinical factors and the identified molecular factor was superior to the clinical prognostic model in discriminative performance. Kaplan-Meier (KM) curves of patients grouped by PDS suggested that patients with a higher PDS had a poorer prognosis, and stage II patients could be distinctly distinguished.ConclusionsBased on the knowledge-based clinical-molecular integrated analysis, a clinical-molecular integrated prognostic model and corresponding nomogram for colon cancer overall survival prognosis was built, which showed better prognostic performance than the clinical prognostic model. The PDS of the pathway hsa00532 is a considerable clinical prognostic supplementary factor for colon cancer and may represent a potential prognostic marker for stage II colon cancer. The PDS calculation involves only 16 genes, which supports its potential for clinical application.

Project description:BackgroundThe relation between breast cancer molecular subtype and survival has been studied in several jurisdictions, but limited information is available for Ontario. The aim of this study was to determine breast cancer survival by molecular subtype and to assess the effect on survival of selected demographic and tumour-based characteristics.MethodsWe extracted 29 833 breast cancer cases (in 26 538 girls and women aged ≥ 15 yr) diagnosed between 2010 and 2012 from the Ontario Cancer Registry. Cancers were categorized into 4 molecular subtypes: 1) luminal A (estrogen-receptor-positive and/or progesterone-receptor-positive [ER+ and/or PR+] and negative for human epidermal growth factor receptor 2 [HER2-]), 2) luminal B (ER+ and/or PR+/HER2+), 3) HER2-enriched (ER- and PR-/HER2+) and 4) triple-negative (ER- and PR-/HER2-). We estimated associations with predictor variables (age, stage at diagnosis, histologic type, comorbidity and place of residence [urban or rural]) using a multivariate Cox proportional hazards model. Likelihood ratio testing was used to evaluate differences in risk of death.ResultsLuminal A was the most commonly diagnosed subtype (59.0%) and had the greatest survival, whereas triple-negative had the poorest survival. For all subtypes, a dose-response effect was observed between the hazard of death and age and stage at diagnosis, with the greatest effect found for the HER2-enriched subtype (age: hazard ratio [HR] 7.87 [95% confidence interval (CI) 3.68-11.81]; stage at diagnosis: HR 37.71 [95% CI 34.64-41.27]). Moderate comorbidity (Charlson Comorbidity Index score 1 or 2) was associated with increased risk of death for triple-negative cancers (HR 2.42 [95% CI 1.36-4.31]), and severe comorbidity (Charlson Comorbidity Index score ≥ 3) increased the risk for all molecular subtypes.InterpretationThe results indicate the importance of including molecular subtype, along with age, stage at diagnosis and comorbidity, in assessing breast cancer survival. They highlight the need to address outcomes related to hormone-receptor-negative cancers, for which survival lags behind that for hormone-receptor-positive cancers.

Project description:Background: Although inflammatory breast cancer (IBC) has poor overall survival (OS), there is little information about using imaging features for predicting the prognosis. Computed tomography (CT)-based texture analysis, a non-invasive technique to quantify tumor heterogeneity, could be a potentially useful imaging biomarker. The aim of the article was to investigate the usefulness of chest CT-based texture analysis to predict OS in IBC patients. Methods: Of the 3,130 patients with primary breast cancers between 2006 and 2016, 104 patients (3.3%) with IBC were identified. Among them, 98 patients who underwent pre-treatment contrast-enhanced chest CT scans, got treatment in our institution, and had a follow-up period of more than 2 years were finally included for CT-based texture analysis. Texture analysis was performed on CT images of 98 patients, using commercially available software by two breast radiologists. Histogram-based textural features, such as quantification of variation in CT attenuation (mean, standard deviation, mean of positive pixels [MPP], entropy, skewness, and kurtosis), were recorded. To dichotomize textural features for survival analysis, receiver operating characteristic curve analysis was used to determine cutoff points. Clinicopathologic variables, such as age, node stage, metastasis stage at the time of diagnosis, hormonal receptor positivity, human epidermal growth factor receptor 2 positivity, and molecular subtype, were assessed. A Cox proportional hazards model was used to determine the association of textural features and clinicopathologic variables with OS. Results: During a mean follow-up period of 47.9 months, 41 of 98 patients (41.8%) died, with a median OS of 20.0 months. The textural features of lower mean attenuation, standard deviation, MPP, and entropy on CT images were significantly associated with worse OS, as was the M1 stage among clinicopathologic variables (all P-values < 0.05). In multivariate analysis, lower mean attenuation (hazard ratio [HR], 3.26; P = 0.003), lower MPP (HR, 3.03; P = 0.002), and lower entropy (HR, 2.70; P = 0.009) on chest CT images were significant factors independent from the M1 stage for predicting worse OS. Conclusions: Lower mean attenuation, MPP, and entropy on chest CT images predicted worse OS in patients with IBC, suggesting that CT-based texture analysis provides additional predictors for OS.

Dataset Information

Improving breast cancer survival analysis through competition-based multidimensional modeling.

Publications

Improving breast cancer survival analysis through competition-based multidimensional modeling.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets