Xuelin Huang - US grants

ABSTRACT The Biostatistics, Epidemiology and Bioinformatics Core (BEBiC) will provide professional expertise in biostatistics, epidemiology and bioinformatics for all research studies within the Partnership. The Core will be a comprehensive, multilateral resource for data acquisition and management, design of laboratory experiments, epidemiological studies and clinical trials, statistical analysis, and publishing translational research generated by the Partnership. The BEBiC will incorporate sound experimental design principles within all projects and programs; carry out data analyses using appropriate statistical methodology; and contribute to interpretation of results through written reports and frequent interaction with project and program co-leaders. The Core will provide an integrated data management system to facilitate communication among all projects and cores, which will be customized to meet the needs ofthe Partnership. This process includes prospective data collection, data quality control, data security, and patient confidentiality. Thus, from inception to reporting, projects and programs will benefit from collaboration with the BEBiC. In addition, the BEBiC will organize different workshops related with the statistical and bioinformatics needs for every research study, particularly with all types of regression modeling and variable selection, survival analysis and optimization of dynamic treatment regimes, clinical trials, sample size, power calculations and statistics and algorithms for bioinformatics. Also, the BEBiC's personnel will develop new statistical methodologies in cancer research according to the projects needs.

In manufacturing and industrial applications, it is critical to regularly inspect all the components of a working system, record data during the inspection, use these data to predict the next potential failure event, and take preventive actions. In medical practice and research, the health status of patients is measured repeatedly over time during post-treatment follow-up visits. During each visit, new information is obtained and physicians use that information to predict a patient's prognosis and design an appropriate treatment plan. These applications involve the use of current information to predict the time until the next failure event (such as disease progression). These continuously updated predictions, called dynamic predictions, are critical for patients with non-curable diseases such as cancer or AIDS. This project will develop and apply modern statistical techniques to extract useful features from massive data sets collected over time, and then use these features to conduct predictions as accurately as possible. When these statistical methods are built into a computer software program, they can be used online to conduct predictions. Patients and physicians can use such programs to evaluate disease progression and to make early decisions about treatment and prevention. Industrial engineers can use such programs to forecast a potential system failure and initiate maintenance. Commercial web sites can collect customers' reaction data online and then apply such methods to better predict customers' needs and improve sales and customer satisfaction.

Many statistical methods assume that longitudinal data trajectories follow parametric models, linear or nonlinear. However, the pattern of longitudinal data trajectories differs in each specific setting, making it difficult to identify a satisfactory parametric family that is suitable for all situations. Based on this consideration, a functional principal component analysis (FPCA) approach is used to capture the longitudinal data structures and functional patterns. The first goal of this project is to decompose biomarker trajectories into some feature functions, and then incorporate these features as covariates in the Cox proportional hazards model to make dynamic predictions. Given that the proportional hazards assumption may be too restrictive in some cases, the second goal of this project is to conduct dynamic prediction for the quantile functions of the residual event time under a flexible framework. The residual lifetime quantile regression model facilitates a meaningful interpretation and offers more direct answers than the Cox model. The third goal of this project is to develop analytic and visualization tools for identifying longitudinal data trajectory patterns prior to a failure event by looking at them backwards in time and aligning them with the failure events. Discerning these patterns can greatly facilitate dynamic prediction of the imminent failure event. The proposed methods are specially designed to handle the complications of censored data, irregular follow-up times and dynamically collected data to facilitate prediction over a range of time points.

Project Summary The major objective of the Biostatistics and Bioinformatics Core (Core 3) is to provide centralized biostatistics, bioinformatics, and database support for all Projects and Cores. Core 3 will provide guidance in the design and conduct of clinical trials and other experiments that arise from the ongoing research of the SPORE, facilitate prospective collection, entry, quality control, and integration of data for the basic science, pre-clinical, and clinical studies, and provide bioinformatics data analysis of high-throughput and high-dimensional genomics data. We will provide innovative and tailored statistical modeling, simulation techniques, and data analyses for the main projects, developmental research and career enhancement projects, and other cores to achieve their specific aims. We will conduct data analyses and prepare statistical reports for all experiments within all projects, ensure that the results of all projects are appropriately interpreted, and assist all project investigators in the publication of scientific results. Core 3 will also be a resource for intra- and inter-SPORE collaborations, including study design and developing databases for multi-center clinical trials.

DATA AND OMICS SCIENCES CORE (DATAOmics) PROJECT SUMMARY/ABSTRACT The primary objective of the Data and Omics Sciences Core (DATAOmics) is to assist the work of the Infection-Driven Malignancies Program for Advancing Careers and Translational Sciences (IMPACT), a cancer research partnership between the University of Puerto Rico (UPR) and the MD Anderson Cancer Center (MDACC), through an enhanced infrastructure to support IMPACT's research, education, and outreach activities, in quantitative analyses and scientific reporting. We propose to expand and strengthen the relationships between IMPACT and the UPR Comprehensive Cancer Center by establishing a formal unit to support their researchers and participants (undergraduate and graduate students) in areas such as biostatistics, epidemiology, bioinformatics and Omics sciences. DATAOmics capabilities will be enhanced, with respect to the previous Biostatistics, Epidemiology and Bioinformatics Core (BEBiC) in at least two ways: 1) the incorporation of a new investigator with experimental and analytical expertise in Omics sciences, particularly in next generation sequencing and mass spectrometry, and with wet lab experience, a dimension which is now of paramount importance in modern cancer research, particularly for Full Project A and Pilot Project A; and 2) by adding a small and fully dedicated computer server to DATAOmics, which will enhance our computational ability to undertake realistic bioinformatics and Omics problems and will enable participants to be trained in bioinformatics analyses. The overall goal of DATAOmics is to enhance the productivity of IMPACT's research projects through the provision of high-quality and timely biostatistics, data sciences, bioinformatics and Omics advice and consultancy at all stages of research. The specific aims of DATAOmics are: 1) Provide support and consulting on study design, data gathering and management, and statistical analysis to all research projects; 2) Increase cancer research capacity in design and statistical analyses in biostatistics, data sciences, bioinformatics and Omics through educational activities for our researchers and participants; and 3) Develop and apply innovative biostatistics and bioinformatics methodologies to the cancer research projects initiated by our researchers. DATAOmics will include two components: 1) Biostatistics and Data Science, and 2) Bioinformatics and Omics Sciences. The Biostatistics and Data Science component at UPR will have two statisticians and one epidemiologist dedicated to the methodological and statistical aspects of the research projects. MDACC will engage one senior biostatistician and one senior data analyst. The Bioinformatics and Omics Sciences components at UPR will have two bioinformaticians engaged in the bioinformatics facets of the projects in a broad sense (including: genomics, transcriptomics, pharmacogenomics, proteomics, epigenomics, metabolomics, lipidomics, and microbiomics), and MDACC will allocate one senior bioinformatician.

ABSTRACT The CISNET Breast Working Group (BWG) proposes innovative modeling research focused on new precision oncology paradigms that are expected to re-define breast cancer control best practices. We selected significant topics where modeling is suited to fill evidence gaps and facilitate clinical and policy translation. Unique components of our approach include modeling of absolute risk of disease accounting for multiple risk factors, addressing important comorbidities?specifically type 2 diabetes?that affect both disease risk and survival, exploring emerging biomarker-based approaches for screening, providing guidance regarding precision systemic treatments and their impact on quality of life in survivors, and investigating race disparities. The specific aims are to: 1) Evaluate the impact of novel precision screening approaches; 2) Evaluate the impact of precision treatment paradigms in the adjuvant, neo-adjuvant, and metastatic setting; 3) Synthesize Aims 1 and 2 to quantify the contributions of precision screening and precision treatment to US breast cancer mortality reductions; and 4) Provide evidence to guide interventions to reduce race disparities by quantifying multiple risk, screening, treatment, and survival factors that impact disparities. This scope of work would not be feasible without the availability of six distinctive BWG models: Dana Farber (D), Erasmus (E), Georgetown- Einstein (GE), MD Anderson (M), Stanford (S) and Wisconsin-Harvard (W). The aims encompass multiple RFA priority areas, and we have set aside Rapid Response funds to address remaining priority areas, support cross-cancer CISNET collaborations, and foster junior career enhancement. Each aim includes three or more model groups selected for their unique structure and includes outside collaborators and junior investigators. The models will share common inputs and provide a standard set of outcomes for benefits (e.g., distant recurrences and deaths avoided, mortality reductions, distant disease-free survival, and life years and quality- adjusted life years), harms (e.g., false positives and benign biopsies, interval cancers, advanced stage diagnoses, overdiagnosis and treatment impact on quality of life), and costs. Continuously funded for the past 19 years, the modeling teams have published 204 research papers informing public health policy decisions and trained 13 junior investigators. For this proposal, the BWG will partner with the American Cancer Society, the American College of Radiology, the American Society of Clinical Oncology, the Breast Cancer Surveillance Consortium, and others. An experienced Coordinating Center provides the infrastructure to support the project goals including resource sharing and model accessibility. The exceptional environment provides unprecedented synergy and leveraging of resources to address new research questions and support career development that would not otherwise be possible. Overall, this research will advance modeling research and guide breast cancer control policy.