GSK transforms clinical analytics and reporting with a modernized statistical compute environment (SCE)

About GSK

GSK, formerly known as GlaxoSmithKline, is a global biopharmaceutical leader that develops vaccines as well as speciality and general medicines to prevent and treat disease. With 37 manufacturing sites and 70,000 employees operating in over 75 countries, GSK is one of the ten largest pharmaceutical companies in the world. GSK strives to unite science, technology and talent to get ahead of disease and positively impact the health of 2.5 billion people by 2030.

The importance of a modern SCE

As pharmaceutical companies strive to discover, develop, and bring new drugs to market more efficiently, the need for a modern Statistical Computing Environment (SCE) has increasingly become essential to keep pace with emerging technologies, move at speed, and remain compliant. A modern SCE facilitates faster, more accurate reporting and analysis of clinical trial results, accelerates drug approval processes and enhances the ability to deliver life-saving treatments.

At GSK, modernizing and future-proofing for statistical computing and data science is critical to bring the best medicines and vaccines to patients faster, and at lower cost. They developed an agile implementation framework, known internally as the “Statistical & Programming Analytics Computing Environment,” or SPACE program, to deliver a modern SCE and associated modern toolset, and accelerate planning, analysis, publishing, and sharing of clinical trial results.

Bringing pharma and vaccines onto one SCE

As part of this initiative, GSK needed to integrate the Pharmaceuticals and Vaccines business units into a single Research and Development organization. But the two departments used different SCEs for clinical reporting: Pharmaceuticals had a home-grown, custom-built environment, while Vaccines leveraged a third party SaaS (Software as a Service) solution. Streamlining the two organizations onto one platform was critical for analyzing and reporting clinical trial results and preparing regulatory submissions in a timely, compliant manner. “Both of the legacy systems were traditional, siloed, monolithic, and inflexible,” said Eileen Ching, Senior Director of Biostatistics Technical Capability Delivery at GSK, “and now we're moving toward a single platform where we can do both GxP and non-GxP, exploratory and regulatory work all in a very flexible, robust environment.”

Moving to a unified platform means delivering business optimization and process standardization, and increasing efficiency and productivity.

Eileen Ching

Senior Director, Biostatistics Technical Capability Delivery, GSK

Modernizing statistical analytics with Domino

As a certified member of the TransCelerate BioPharma community, GSK adheres to the three tenets of the Modernization of Statistical Analytics Initiative: accuracy, reproducibility and traceability.

These key principles, along with the need for a scalable, cloud-native platform that was flexible enough to meet the needs of different users – including clinical statisticians, programmers, and research data scientists – led GSK to adopt Domino. As a modern, flexible, and compliant SCE, Domino offers a robust and friendly user interface with language and IDE flexibility, workflow automation and visualization, and full traceability and reproducibility. The centrally managed and audit-ready platform improves productivity, collaboration, and accuracy through end-to-end statistical analyses spanning both GxP and non-GxP work.

GSK uses Domino's Workspace environment to tailor to different user and compute needs based on factors like job size or hardware type. Statisticians, programmers, data scientists and other users can now leverage Domino containers to seamlessly work with both traditional and modern open source or commercial languages, in a single environment. This has led to increased collaboration between users and clinical data teams, and, as Eileen shared, created efficiencies for regulatory adherence as well. “Domino provides traceability from input, to data, to the code that was used to produce the output, and the output itself. Everything you need to show traceability, and enable full reproducibility, which is essential for GxP.”

We want to make leaps and bounds. We want to make sustainable improvements in efficiency and quality effectiveness by providing our teams an environment where they can truly innovate. They can enjoy the flexibility to think outside the box and find better ways of doing things.

Eileen Ching

Senior Director, Biostatistics Technical Capability Delivery, GSK

Increased flexibility, collaboration, and innovation

GSK delivered one unified SCE and increased speed, efficiency, productivity, innovation, and collaboration – without compromising quality or regulatory compliance. Working with technology partners and Domino Professional Services helped GSK deliver a fully qualified Domino install, a validated SCE workflow, and creation of a production-quality MVP in just the first eight months. After a 9-month migration effort, all of GSK’s clinical analysis and reporting work is now done in the new SCE.

Since implementing Domino, GSK has accelerated time to regulatory submission and increased productivity by planning more proactively, reducing volume of static outputs, streamlining data flows, and standardizing and automating clinical workflows. Statistical programmers and biostatisticians now have timely access to data, facilitating faster, high-quality decision making. From a regulatory perspective, agencies can easily access and review clinical study reports, ensuring GSK is compliant with rules and regulations. The flexibility of the Domino platform has reduced the learning curve for new users and enhanced collaboration across data teams, while improving overall user experience and satisfaction.

According to Saurin Mehta, Senior Product Director, RWD, Biostatistics, Analysis and Reporting at GSK, “Statistical programmers and Data science teams now have a choice of multiple languages and IDE’s to work with, an optimized scalable hardware tier to choose from to execute jobs, and we can further improve efficiency and user experience by automating various processes via available API’s. Our business partners are satisfied and the tech teams have gained efficiencies through a single platform that allows us to reallocate budget elsewhere.”

Looking to the future

As a founding member of the SCE Coalition, GSK is partnering with Domino and other top pharmaceutical organizations to bring together leaders and subject matter experts with the mission to better define, develop, and improve SCEs for clinical analytics. This includes exploring enhanced data ingestion and integration, workflow management, quality control, change management, application of AI/ML, traceability, and reproducibility. “Using AI/ML to enhance automated workflows and reduce routine manual tasks can help enable a faster cycle time,” shared Eileen. “At the end of the day it's about quality, but it’s also about speed, so delivering key milestones faster, at different stages of the development cycle is crucial.”

GSK is interested in the possibilities of AI automation and governance for quality control, and also exploring the possibilities of AI and machine learning for innovations that ultimately prevent and change the course of disease. The team is expanding its users beyond clinical statistics and programming to Real World Evidence (RWE) and chemical manufacturing and controls (CMC), which supports manufacturing. “With advancements in technology and science, the volume and variety of data from drug and vaccine trials has really skyrocketed, along with more novel data types like biomarkers and RNA sequencing,” said Eileen. “Data science, AI and ML are becoming key enablers for delivering more efficient, more effective trial design, and helping us make more informed decisions faster and ultimately get ahead of disease together.”

Being ambitious for patients is a core value of GSK. But we can't be ambitious for patients if we can’t be flexible and agile, and we can't innovate if we don't have the right tools. Domino gives us a full toolbox, not just different sized hammers, where we can easily pick the best tool for the job.

Eileen Ching

Senior Director, Biostatistics Technical Capability Delivery, GSK