Wednesday, Aug 7: 8:30 AM - 10:20 AM
1607
Topic-Contributed Paper Session
Oregon Convention Center
Room: CC-E144
Open-source tools and platforms with various degrees of complexity and stakeholder base have gained traction in various industries, including the pharmaceutical industry, with some companies transitioning to full use of these tools. These tools offer cost-effective and collaborative solutions for data analysis, visualization, modelling, and other processes relevant to drug development. The elements in creating good open-source software business applications are like those for creating good software applications in general, e.g., well-documented codebase, sustainable/scalable development process, and security. This may involve things like setting up a continuous integration and deployment (CI/CD) pipeline and using issue tracking and project management tools. In addition, containerization platforms enable the creation of reproducible computing environments. These containers can encapsulate software dependencies and code, ensuring that analyses and processes can be replicated consistently, which is crucial for regulatory compliance.
The use of open-source software in submissions in support of NDAs is increasing concurrently, and it is likely to continue to grow in the future. This is because open-source software offers several advantages over proprietary software, such as increased transparency, reduced costs, improved collaboration, and accelerated innovation in addition to their powerful data visualization capabilities that support the presentation of data in regulatory submissions. Moreover, initiatives such as CDISC (Clinical Data Interchange Standards Consortium) provide open-source data standards that promote consistency in data collection, analysis, and submission. Using CDISC standards can facilitate the preparation of regulatory submissions. Open-source software can be used to format and convert data into required regulatory submission formats, such as the electronic Common Technical Document (eCTD) format.
In this session, we showcase examples and good practices of open-source applications development in pharma. It is recognized that while open-source software can offer many advantages, it is important to ensure that the tools and processes used adhere to regulatory requirements and validation standards. Open-source solutions should be validated and documented appropriately to meet regulatory expectations. It is also crucial to consider data privacy and security requirements to protect sensitive patient information and maintain compliance with regulations. When submitting a new drug application (NDA), it is important to ensure that the software and its associated technical specifications meet the regulatory requirements set forth by the relevant health authorities.
Applied
Yes
Main Sponsor
Section for Statistical Programmers and Analysts
Co Sponsors
Biopharmaceutical Section
ENAR
Presentations
Open-source tools have gained acceptance in the pharmaceutical industry and their use continues on an upward trajectory. This phenomenon is likely to continue as more companies get more familiar and comfortable in the routine use of open-source tools in planning, data exploration, reporting, and in submission work. Open-source tools offer many advantages relative to commercial software in terms of cost, cross-pharma collaboration, and innovation and often incorporate cutting age ideas and functionality, for example in data exploration and visualization. As one might expect, these tools come with varying degrees of complexity, functionality, accompanying documentation, and challenges. This presentation will discuss some considerations in leveraging open-source tools in the analysis and reporting of clinical trial data and some of the emerging trends and the landscape. The presentation will discuss some considerations regarding good practice in the development and use of open-source software for exploratory and reporting purposes in clinical trials and their use in regulatory submissions. Some open-source tools for meta-analysis will be highlighted for illustration purposes.
Innovation in clinical trial design may not be enabled by off-the shelf software. This talk will focus on group sequential design with variations such as testing for multiple hypotheses, design for possibly delayed treatment effects and stratified design with differing outcome distributions and treatment effects in different strata. Examples of trials with biomarker/histology subgroups and also multiple experimental treatment groups will be presented. We discuss software specification, testing, issue management and release strategies. Specific trials developed include gsDesign (group sequential design), gsDesign2 (group sequential design with more options, including non-proportional hazards), multiple testing in group sequential design (gMCPLite and WPGSD). Also, we discuss the use of Shiny to enable design without the need to program in R.
The advent of visual analytics signals a paradigm change in drug safety data exploration, evaluation, and distribution from conventional tabular representations to more dynamic and interactive techniques. This shift strategically combines data visualization, statistical analysis, and data mining tools to boost stakeholder participation while also increasing the effectiveness of data review. Such advanced visualization modalities are instrumental in distilling complex safety data into coherent narratives, thereby facilitating a more intuitive understanding of drug safety findings. The development of sophisticated tools for visual analytics in drug safety, which incorporate structured assessments tailored to specific safety inquiries, is imperative. CVARS is designed to generate interactive forest and volcano plots, offering a compelling alternative to static outputs. These interactive plots play a critical role in elucidating adverse events and analyses based on FDA Medical Queries (FMQs), thereby providing invaluable insights for regulatory submissions. In addition, 3D visualization techniques offer unparalleled depth and clarity, enabling stakeholders to navigate through complex
{teal} is an innovative open-source and scalable R-shiny based framework that is transforming the way clinical trial data is being analyzed and visualized. It enables data scientists to streamline the creation of web applications, bringing data closer and faster to stakeholders, resulting in quicker insights and better-informed decisions. The framework's key features, such as dynamic data filtering, code reproducibility and report generation, promote transparency and push the boundaries of interactive analysis in the data exploration process. With over 50 analysis templates and the ability to easily integrate customized modules for different analyses or data types, {teal} offers a comprehensive and extendable solution for clinical trial data exploration. In this talk, we will introduce the {teal} framework, highlight its key features, share how this has been adopted by hundreds of data scientists inside our organization, and discuss how collaborative development with external partners are fostering even more values and impacts across the pharmaceutical data science community. For more information, visit https://insightsengineering.github.io/teal/