Bristol Myers Squibb’s Journey: Adopting a Modern Data Infrastructure & Analytics Platform

Bristol Myers Squibb’s Journey: Adopting a Modern Data Infrastructure & Analytics Platform

David Weigand, Senior Director, Clinical Data Reporting & Analytics, Bristol Myers Squibb

Jayabarathi Rajasekaran, Director, Clinical Data Warehousing & Delivery, Bristol Myers Squibb

Sheila Rocchio, Chief Marketing Officer, eClinical Solutions

Lori Ellis, Editor, Citeline (Moderator)

KEY TAKEAWAYS

ƒ Key Industry Drivers

ƒ BMS Clinical Data Ecosystem

ƒ Phased Approach to Implementation

ƒ Lessons Learned

ƒ

What’s Next?

OVERVIEW

The amount and diversity of data incorporated into clinical trials are growing dramatically. In this environment, life sciences companies must optimize data flow at scale. Clinical data ecosystems must facilitate the exchange of information within organizations, as well as with outsourced partners.

To address these challenges, in 2018 Bristol Myers Squibb embarked on building a new clinical data infrastructure and analytics solution built on eClinical Solutions’ elluminate Clinical Data Cloud platform. This initiative has helped streamlined the clinical data lifecycle, reducing cycle times and time to insight.

CONTEXT

Sheila Rocchio described the eClinical Solutions’ elluminate Clinical Data Cloud. David Weigand and Jayabarathi Rajasekaran discussed lessons Bristol Myers Squibb (BMS) learned over the past four years adopting a new clinical data infrastructure and analytics platform. They also shared how BMS is taking their systems to the next level to improve clinical data efficiency.

KEY TAKEAWAYS

As the volume of data in clinical trials grows, life sciences companies need better ways to organize and access information.

The nature of clinical trials has changed significantly in recent years. Two key industry drivers are:

ƒ Clinical trials now include enormous volumes of data. Factors contributing to this trend are longitudinal studies, decentralized trials, and patient data from physiological monitoring devices. In 2022, a survey by the Tufts Center for the Study of Drug Development found the average Phase III trial now includes 3.6 million data points—which is three times more than a decade ago. If growth rates continue at the same pace, by the end of this decade the average trial will include 10 million data points or more. In a recent eClinical Solutions survey of 50 life sciences companies, 30% reported using more than 10 data sources per trial—double the average number of sources reported in a 2019 survey.

ƒ Large volumes of data generate richer patient insights but can be difficult for data consumers to access. Information for clinical trials comes in many formats and originates from a variety of sources. This can make it difficult for data consumers including data managers, medical reviewers, biostatisticians, and executive management to have relevant information for informed decision making.

In response to these market conditions, eClinical Solutions built the elluminate Clinical Data Cloud which serves as a bridge between data sources and data consumers. Core capabilities within elluminate include:

ƒ A robust data import capability, regardless of data format and source system. With elluminate, it’s possible to load data without first transforming it. The platform uses an ELT (extract, load, transform) model, rather than ETL (extract, transform, load).

ƒ Easy data transformation. Automation and a low-code environment allow for easy data transformation.

ƒ A complete set of data review tools. These include a data management workbench, analytics, AI/ ML enabled data review, a risk-based quality management application, and operational insights.

ƒ Flexible computing environment for faster data exploration and analysis.

elluminate gives data consumers self-service access to information for better, faster insights.

BMS Clinical Data Ecosystem

The BMS vision is to advance clinical development through data-based insights. BMS wanted to ingest, curate, and aggregate multiple clinical data sources so the company can rapidly generate insights that advance clinical and drug development efforts. As BMS worked towards realizing this vision, they focused on two business outcomes:

1. Reducing risk. The clinical data platform must improve response times to regulatory queries through faster and easier access to audit trails and data lineage information.

2. Improving efficiency. There are four areas where BMS wanted to improve efficiency: 1) Medidata Rave integration; 2) external non-eCRF data loading; 3) query management; and 4) data review and analysis.

We want traceability of the data lineage from the time that information enters the platform to the time it is sent to downstream systems. We want the ability to generate that information on-demand, so we are always inspection ready.

BMS adopted a phased approach to the implementation of its new data infrastructure and analytics program.

Lessons Learned

Building momentum for user adoption as early as possible is critically important. Converting a more significant amount of reporting content into elluminate earlier would have motivated more people within BMS to use the system sooner. In retrospect, the BMS team recommends eliminating or minimizing phased adoption, if business circumstances allow.

Additionally, the team also identified these lessons learned:

1. Early internal alignment is important. This requires end-to-end standardization.

2. It’s crucial to set CRO expectations. Although BMS shared standards with CRO partners, vendors didn’t interpret them correctly. Timely communication, training, and test studies could have prevented this.

3. Sponsor oversight is required at all stages of a study. Before studies go live and throughout the study lifecycle, there must be a robust process for standardization checks.

With our new BMS outsourced studies, the main issue was that the CROs didn’t interpret our standards correctly. We learned that we must provide proper training before starting a study and also do a couple of test studies before bringing data into production. Timely communication throughout the study contract is critical.

What’s Next?

Work is underway in three key areas to further build out an optimized data architecture including:

1. Programming and transformation. Expanded capabilities in the recently available elluminate statistical computing environment are being considered as they will enable the team to promote global programming across all studies, as well as deliver data to end users more rapidly.

2. Reporting and analytic capabilities. This work is focused on enhanced variance reporting, visualization and dashboarding, and pooled study capabilities in elluminate Data Central.

3. AI/ML opportunities. BMS would leverage AI/ML capabilities to identify discrepancies and anomalies and is evaluating elluminate IQ. This would reduce customized programming and minimize the burden on data reviewers. Another area of interest is metadata mapping and code snippets.

CONCLUSION

In today’s dynamic business environment, life sciences companies need a flexible, scalable clinical data infrastructure and analytics platform that will reduce business risk and improve operational efficiency. Clinical trials now incorporate information in different data formats, from diverse source systems. From a strategic perspective, systems and processes must be extensible to accommodate mergers and acquisitions.

Through its partnership with eClinical Solutions, BMS has created a strong foundation levaraging capabilities in the elluminate Clinical Data Cloud. The team at BMS appreciates elluminate’s existing capabilities, as well as its ability to handle future expansion and innovation. When deploying a clinical data infrastructure and analytics solution, a key lesson is that processes are as important as technology. Early internal alignment is critical, as is clear expectation-setting with outsourced partners like CROs. In addition, building momentum for user adoption as early as possible is essential. Learn more about the elluminate Clinical Data Cloud and request a demo at eclinicalsol.com

BIOGRAPHIES

David Weigand

Senior Director, Clinical Data Reporting & Analytics, Bristol Myers Squibb

David Weigand brings over 20 years of clinical research experience in various clinical programming, data management, and project management roles across multiple therapeutic areas. In his role as Senior Director, Clinical Data Reporting and Analytics at Bristol Myers Squibb, Weigand provides operational leadership of the clinical data reporting and analytics function enabling the presentation of clinical trial datasets, data reports, listings, and visualizations to various stakeholder groups for data review, patient safety oversight, protocol deviation and data trend detection, internal decision making, and statistical analysis. He holds a BS in Nursing from Penn State University.

As the Director, Clinical Data Warehousing & Delivery, Jayabarathi Rajasekaran, M.S. Computer Science, is responsible for the receipt, ingestion, transformation, warehousing, and delivery of clinical data to internal stakeholders and external partners. As subject matter expert, she provides knowledgeable consultation on the technology and tools to ensure robust and compliant data receipt, modeling, transformation, warehousing, and best-fit reporting and analytics delivery. Rajasekaran has 20+ years of experience in Healthcare and Pharmaceutical industry specialized in Data warehousing and Clinical Programming.

Sheila Rocchio has more than 20 years of industry experience including marketing, product management, and strategy roles in life sciences technology companies. She spent 15 years at PHT Corporation (now ERT) in a variety of marketing and executive leadership roles helping to drive the company’s growth and the industry adoption of electronic clinical outcome assessments (eCOA). At eClinical Solutions, Rocchio manages all aspects of marketing including branding, communications, lead generation and product marketing. She enjoys finding creative ways to tell customer stories, and building products and services that help clinicians, data scientists, and technologists do the challenging and important work of bringing life-saving new therapies to market.

Lori Ellis is a content writer within Pharma Intelligence’s Custom Content team, providing clients with engaging content to support their marketing objectives. In addition to delivering written content, she provides award-winning voiceover, hosting, and camera skills for clients. She specializes in producing, moderating, and hosting podcasts, webinars, and roundtables. Areas she has covered span the drug development process, from pre-clinical to commercial stages. Her expertise focuses on technology’s ever-evolving impact on the pharmaceutical industry.