Annual and Interim Progress Report Summaries
Principal Investigator: Steve Self
Project: Vaccine Immunology Statistical Center
Submitted March 31, 2011
The Vaccine Immunology Statistical Center is organized within the CAVD to provide services in the areas of statistical design and data analysis, a central repository for key CAVD studies, and support for laboratory data management. Our diverse staff has expertise in the areas of biostatistics, immunologic assays, bioinformatics, data management, and information systems. We ensure that CAVD studies are efficient by providing state-of-the-art statistical methods and, when required, develop novel methods.
Areas of methodological development and collaboration by the VISC include work on analytic methods for assessing immunological correlates of risk and of protection, the design of repeated low-dose challenge non-human primate studies, signal processing and normalization of flow cytometry-based assays, signal processing, normalization and analysis of antigen array data, signal processing and modeling of data from the luminex assay, and the analysis of high-dimensional immunologic checkerboard data.
Collaborations in progress include a project with the Comprehensive Antibody Vaccine Immune Monitoring Consortium (CA‐VIMC) and Monogram to analyze Tier 2 neutralization responses of vaccine and placebo recipients in Vax004, which sets a bar for what constitutes non-protective neutralizing antibody responses for a rgp120 HIV-1 vaccine.
In collaboration with the Comprehensive T Cell Vaccine Immune Monitoring Consortium (CTC-VIMC), analyses of data from the CAVD-VIMC ICS assay qualification study were completed and statistical analytical plans were developed for the CAVD-VIMC ICS assay head-to-head comparison study and the ICS assay transfer study. Data pipelines were also developed to receive data for those studies.
Data from the initial cohort of the DCVax-001 study were analyzed in collaboration with both the CA- and CTC- VIMCs.
In collaboration with the Letvin VDC, data from four vaccine studies in non-human primates were analyzed to assess the vaccine effect on per-challenge infection risk, the vaccine effect on viral load progression, and host and immunologic correlates of SIV and SHIV infection risk. Additional work is underway for a sieve analysis of these with emphasis on machine learning methods to detect amino acid sites/patterns that distinguish transmitting sequences of vaccinated infected animals from control infected animals.
In collaboration with Louis Picker (McElrath VDC) data were analyzed from a multi-stage adaptive design vaccine trial in non-human primates to identify immune correlates of protection.
Data models and pipelines were developed and implemented in the past year for the Luminex multiplex cytokine assay, peptide microarray, flow cytometry-based assays (intra-cellular cytokine staining) and a high-throughput Neutralizing Antibody assay.
The VISC data system, with its customized web portal, is the nexus for data sharing and collaboration among investigators within the CAVD. Over the past year, we have developed a plan for creating a considerably enhanced and expanded facility for data sharing using the concept from the computer science discipline of “DataSpace”. We expect to meet the future needs for data sharing within the CAVD by evolving the current VISC data systems to embrace the new model of DataSpace.
Submitted August 13, 2010