SAS Innovate 2025: Observations Relevant to the Pharmaceutical Industry »» Bollinside Programming Services

SAS Innovate 2025: Observations Relevant to the Pharmaceutical Industry

SAS Innovate 2025, comprising the global event in Orlando (6^th-9^th May) and regional editions including London (3^rd-4^th June), featured extensive content addressing pharmaceutical industry applications. This document consolidates observations from sessions directly relevant to pharmaceutical research, development, regulatory affairs and commercial operations, illustrating platform evolution, regulatory compliance approaches and practical applications of advanced analytics in life sciences contexts.

Clinical Analytics Platform Evolution

SAS Clinical Acceleration: Next-Generation Clinical Data Management

The global event featured SAS Clinical Acceleration, positioned as a SAS Viya equivalent to SAS LSAF. The platform combines a secure clinical data repository with a statistical computing environment on SAS Viya, designed to modernise clinical trial data management, analysis and regulatory submission for life sciences organisations.

Key capabilities include:

Regulatory Compliance: Support for FDA Title 21 CFR Part 11 requirements through audit trails, electronic signatures, versioning and role-based privileges. Compliance with CDISC standards and initiatives including dataset-JSON and CDISC CORE ensures alignment with regulatory expectations for standardised data formats.

Multi-Language Programming: Support for SAS, R and Python within a governed and validated environment enables organisations to leverage diverse technical capabilities whilst maintaining compliance. This addresses the industry trend toward open-source analytics whilst preserving the validation and governance frameworks essential for regulatory submissions.

System Integration: Integration with electronic data capture systems, validation tools and metadata repositories enables end-to-end data flow from collection through analysis to submission, reducing manual data transfers and associated error risks.

Embedded AI Capabilities: AI functionality for protocol development and study planning extends platform utility beyond traditional execution-focused analytics into strategic planning activities upstream of trial initiation.

The platform addresses growing complexity in clinical trial designs, including hybrid and decentralised trials, whilst managing real-world data, biomarker information and digital protocol data. Performance improvements and additional capabilities represent meaningful enhancements over the predecessor platform, though much functionality appears familiar to existing LSAF users.

For pharmaceutical companies managing clinical trial portfolios, the platform offers:

Centralised global repository consolidating clinical information into single, secure environments
Validated computing environment supporting compliant analysis and reporting
Collaboration frameworks enabling sponsor-CRO partnerships with appropriate data governance
Scalability supporting growth from single studies to enterprise-wide deployments

AI-Powered Code Generation and Protocol Development

Presentations from Shionogi covered AI-powered SAS code generation for clinical studies and real-world evidence generation. The significant architectural detail involved the AI capability existing within SAS Viya rather than depending on external large language models, addressing pharmaceutical industry concerns about:

Data Governance: Clinical trial data subject to confidentiality agreements and competitive sensitivity cannot be transmitted to external AI services. Internal AI capabilities preserve data sovereignty whilst providing generative assistance.

Validation: External AI services update regularly without notification, creating validation challenges. Internal AI capabilities operate within controlled environments subject to change management and validation protocols.

Regulatory Compliance: Internal AI enables audit trails documenting how code was generated, what prompts were used and what outputs were produced, supporting regulatory requirements for reproducibility and traceability.

Additional sessions addressed interrogating and generating study protocols using SAS Viya, including functionality intended to support study planning in ways that could improve success probability. These capabilities represent directional shift from execution-focused tools to platforms supporting strategic planning:

Protocol Optimisation: AI analysis of historical protocol data to identify elements correlating with successful trial completion, informing protocol design decisions.

Feasibility Assessment: Automated analysis of proposed protocols against site capabilities, patient population availability and operational requirements.

Statistical Planning: AI assistance in determining appropriate sample sizes, randomisation schemes and statistical analysis approaches aligned with regulatory expectations.

This evolution extends analytics platforms beyond conventional clinical programming into activities sitting upstream of trial execution, potentially accelerating protocol development whilst improving trial design quality.

Platform Architecture for Pharmaceutical Data

Architectural Shifts: Bringing Analytics to Data

Sessions addressed fundamental shifts in data architecture particularly relevant to pharmaceutical organisations managing vast clinical and operational datasets across geographical boundaries and organisational entities. The traditional approach of moving massive datasets from various sources into a single centralised analytics engine is being challenged by bringing analytics to the data.

The integration of SAS Viya with SingleStore exemplifies this approach, where analytics processing occurs directly within the source database rather than requiring data extraction and loading. This architectural change can reduce infrastructure requirements for specific workloads by as much as 50 per cent, whilst eliminating complexity and cost associated with constant data movement and duplication.

For pharmaceutical companies, this architecture offers specific advantages:

Reduced Data Movement: Minimises security and compliance risks associated with transferring protected health information and commercially sensitive clinical data across systems and networks.

Geographic Distribution: Enables analytics processing to occur where data resides, addressing data sovereignty requirements in jurisdictions restricting cross-border data transfers.

CRO Integration: Facilitates analytics on data residing in contract research organisation systems without requiring data transfer to sponsor environments, accelerating interim analyses and oversight activities.

Cost Efficiency: Reduces infrastructure investments through efficient resource utilisation, eliminating need for duplicate storage and processing environments.

Performance: Accelerates time-to-insight by eliminating extract, transform and load bottlenecks that delay analysis commencement.

For organisations operating global clinical trial programmes with data distributed across sponsors, CRO’s, central laboratories and imaging core facilities, this architectural approach addresses practical challenges of distributed data analytics whilst maintaining governance and compliance frameworks.

Cloud Migration and Managed Services

Parexel’s Transition to Managed Cloud

The closing session featured Parexel’s CIO discussing their transition to SAS managed cloud services. Characterised as a modernisation initiative, reported outcomes included reduced outage frequency. This observation aligns with experiences from other multi-tenant systems, where maintaining stability and availability represents fundamental requirements that often prove more complex than external perspectives might suggest.

For pharmaceutical companies and CRO’s considering cloud migration, the presentation addressed:

Validation in Cloud Environments: Approaches to qualifying cloud-based analytics platforms to meet regulatory requirements, including infrastructure qualification, application validation and ongoing compliance monitoring.

Regulatory Compliance: Maintaining compliance with FDA Title 21 CFR Part 11 and equivalent regulations in cloud deployments, addressing electronic signature, audit trail and data integrity requirements.

Stability and Availability: Service level agreements, disaster recovery capabilities and business continuity planning for mission-critical analytics supporting regulatory submissions.

Managed Service Models: Comparing managed services where vendors handle infrastructure and application management against self-hosted cloud deployments where organisations retain operational responsibility.

The reported reduction in outage frequency represents significant value for organisations where analytics infrastructure supports time-sensitive regulatory submissions and business-critical decisions. Unplanned downtime during critical analysis periods—such as database locks for regulatory submissions—can delay submissions and impact market access timelines.

Sessions on implementing SAS Viya on-premises under restrictive security requirements described solutions requiring sustained collaboration with SAS over multiple years to achieve necessary modifications. This illustrated that certain pharmaceutical deployments, particularly those handling extremely sensitive data or operating under heightened security requirements, remain defined primarily by governance, controls and assurance requirements rather than by standard product features.

Natural Language Interfaces and Data Democratisation

SAS Viya Copilot: Accessible Analytics

Content addressing SAS Viya Copilot demonstrated natural language capabilities enabling users to interact with analytics through conversational queries rather than requiring technical syntax. Built on Microsoft Azure OpenAI Service, the Copilot functionality supports code generation, model development assistance and natural language explanations of analytical outputs.

For pharmaceutical organisations, this democratisation of data access enables:

Clinical Operations: Clinical trial managers and medical monitors can query trial data for safety signals, enrolment trends and protocol deviations without requiring programming expertise, accelerating decision-making during study conduct.

Medical Affairs: Medical science liaisons and medical information specialists can analyse real-world evidence and published literature using natural language queries, supporting healthcare provider interactions and medical information responses.

Regulatory Affairs: Regulatory professionals can extract analyses and summaries for regulatory submissions without depending on programming resources, particularly valuable for ad hoc regulatory authority questions requiring rapid response.

Commercial Teams: Brand managers and market access professionals can access market data, prescription trends and competitive intelligence through conversational interfaces, enabling data-driven commercial strategy without technical barriers.

Pharmacovigilance: Safety professionals can query adverse event databases using natural language to identify signal patterns, assess case series and generate safety reports.

The capability maintains governance and validation requirements essential for regulated environments:

All queries and results are logged for audit purposes
Access controls ensure users only reach authorised data
Generated code can be reviewed and validated before execution
Results include explanations of analytical methods employed

This approach addresses the pharmaceutical industry challenge of data accessibility: whilst organisations invest heavily in data collection and management, the value remains locked behind technical barriers requiring specialised programming skills. Natural language interfaces democratise access whilst maintaining the rigour and compliance requirements the industry demands.

Trustworthy AI and Regulatory Alignment

AI as Organisational Mirror

Keynote presentations addressed the relationship between AI systems and organisational practices. SAS Vice President of Data Ethics Practice Reggie Townsend articulated that when AI produces biased results, “it’s not a technical failure… What it’s doing is it’s showing us the biases and the beneficiaries that we’ve embedded in our cultural and our organisational practices.”

For pharmaceutical companies, this perspective proves particularly relevant:

Clinical Trial Diversity: AI models trained on historical trial data may perpetuate historical under-representation of particular demographic groups, making patient recruitment algorithms potentially discriminatory unless organisations consciously address these biases in both data and processes.

Drug Development Priorities: AI systems analysing market opportunities and development candidates may reflect existing organisational biases toward particular therapeutic areas, patient populations or commercial markets rather than unmet medical needs.

Safety Signal Detection: Adverse event detection algorithms may exhibit differential sensitivity across patient subgroups, reflecting biases in historical pharmacovigilance data collection and reporting patterns.

Market Access: Pricing and reimbursement models incorporating AI may inadvertently reflect historical access disparities unless organisations deliberately audit for such patterns.

The framing positions AI not as source of new problems but as diagnostic revealing existing organisational issues requiring attention. For pharmaceutical companies committed to health equity and patient-centred drug development, this perspective suggests that AI deployment should include deliberate bias assessment and mitigation as core components rather than afterthoughts.

Healthcare Professional Perspectives

The London event featured presentations from healthcare professionals providing context regarding the operational environment within which pharmaceutical companies function. Dr. Michel van Genderen, an Internist-Intensivist at Erasmus Medical Centre, discussed AI governance in healthcare, stating: “I will only use AI when I’m confident it is safe, explainable and trustworthy. Because in my line of work, decisions can be the difference between life and death.”

His work with the World Health Organisation, leading European hospitals and SAS focuses on:

AI model inventory and governance frameworks
Bias assessment methodologies for clinical AI applications
Maturity assessment for organisational trust and accountability
Integration of AI into clinical decision workflows

For pharmaceutical companies developing AI-enabled products or deploying AI in clinical development, these perspectives reinforce several imperatives:

Explainability: Healthcare providers require understanding of how AI systems reach conclusions, necessitating interpretable models rather than black-box algorithms.

Validation: Clinical AI must demonstrate safety and effectiveness through rigorous validation analogous to clinical trial evidence standards.

Governance: Organisations must establish frameworks ensuring AI systems remain current, monitoring for model drift and updating as clinical practice evolves.

Human Oversight: AI should augment rather than replace clinical judgement, with appropriate human-in-the-loop safeguards for consequential decisions.

SAS CTO Bryan Harris expressed appreciation for pharmaceutical research and development work during the London event, an acknowledgement recognising the industry’s role in advancing human health through data-driven innovation.

Development Practices and Quality Assurance

Continuous Integration for Clinical Programming

Sessions included content on using Bitbucket with SAS Viya to support continuous integration and continuous deployment pipelines for SAS code. Git formed the foundation of the approach, with supporting tools such as JQ.

For pharmaceutical clinical programming, this addresses genuine operational challenges:

Macro Validation: Current manual validation processes for SAS macros used across clinical portfolios can require several weeks. Automated validation through CI/CD pipelines offers:

Consistent, repeatable validation procedures
Reduced validation timelines from weeks to hours or days
Version control enabling complete audit trails
Automated testing ensuring code quality before deployment
Simplified collaboration across programming teams and CRO’s

Code Review: Automated workflows enable systematic code review processes, ensuring programming standards compliance and error detection before validation.

Deployment Management: Controlled deployment processes ensure validated code versions reach production environments without unauthorised modifications.

Change Control: Integration with change control systems provides traceability from change requests through implementation, testing, validation and deployment.

For organisations managing hundreds or thousands of clinical programmes and macros supporting analyses across therapeutic areas and regions, automation of validation and deployment processes represents substantial efficiency gains whilst enhancing quality and compliance.

Migration Assessment and Data Lineage

Sessions on migrating from SAS 9 to SAS Viya focused on assessment methods for determining what requires migration and techniques for locating existing assets. The content reflected the reality that the discovery phase often constitutes the primary work effort rather than a preliminary step.

For pharmaceutical organisations with extensive legacy SAS environments accumulated over decades of clinical development, systematic assessment proves essential:

Discovery: Identifying all SAS code, data, macros and dependencies across clinical development, commercial analytics, manufacturing and quality systems.

Prioritisation: Determining which components require migration based on business criticality, frequency of use and strategic importance.

Dependency Mapping: Understanding relationships between programmes, macros, data sources and downstream processes to avoid breaking critical workflows.

Validation Planning: Assessing validation requirements for migrated code, determining which components require full revalidation versus lighter validation approaches.

Parallel Operations: Planning extended periods where SAS 9 and Viya operate concurrently, enabling gradual migration without disrupting ongoing studies and submissions.

Content on data lineage reframed this capability from purely technical concern to strategic tool for transformation planning and regulatory compliance. For pharmaceutical companies, comprehensive mapping of data flows, transformations and dependencies provides:

Regulatory Audit Trails: Complete documentation of data flow from source systems through transformations to regulatory submissions, supporting regulatory authority inspections and queries.

Impact Assessment: Understanding downstream effects of system changes, ensuring modifications do not inadvertently affect validated processes supporting regulatory submissions.

ALCOA+ Compliance: Supporting data integrity principles (Attributable, Legible, Contemporaneous, Original, Accurate, plus Complete, Consistent, Enduring and Available) through comprehensive documentation of data lineage.

Transformation Planning: Foundation for accurate effort estimation, budgetary planning and risk assessment for major system migrations.

Business Continuity: Visibility into critical data flows enabling rapid restoration in disaster recovery scenarios.

Life Sciences Stream Content

The Life Sciences stream at the London event focused heavily on AI, with presentations from AWS, AstraZeneca and IQVIA addressing the subject, followed by panel discussions. The scale of technological change represents tangible shift affecting all parts of the pharmaceutical ecosystem.

Presentations from AWS addressed cloud infrastructure approaches for life sciences workloads, including compliance frameworks, data security and scalable computing for computational chemistry and genomics analyses.

AstraZeneca content likely addressed practical implementations of advanced analytics in pharmaceutical development, though specific details were not accessible. AstraZeneca has partnered with SAS to use SAS Viya and SAS Life Science Analytics Framework in delivery and approval processes, and to leverage synthetic data capabilities through SAS Data Maker for decision-making in drug discovery and development.

IQVIA presentations presumably addressed real-world data analytics, clinical trial optimisation and commercial analytics applications, reflecting IQVIA’s role as major provider of healthcare data and clinical research services to the pharmaceutical industry.

Panel discussions enabled dialogue across pharmaceutical sponsors, technology providers and service organisations, facilitating knowledge sharing regarding practical implementations, challenges encountered and solutions developed. The emphasis on trustworthiness, responsibility and governance emerged as prominent themes throughout the stream, reflecting pharmaceutical industry priorities around patient safety, regulatory compliance and ethical innovation.

Organisational Change Management

The Human Element of Technology Projects

Presentations on organisational change management accompanying technical migrations emphasised that successful technology projects require attention to human factors alongside technical implementation. Strategies discussed included:

Formal Transition Events: Launch events marking migration milestones, creating shared sense of new beginnings and organisational commitment to change.

Structured Support: Office hours or drop-in sessions providing venues for technical questions, preventing individuals from struggling alone with implementation challenges.

Community Building: Social events designed to foster relationships and maintain engagement during periods of change, recognising that technology adoption succeeds or fails based partly on interpersonal dynamics and organisational culture.

For pharmaceutical companies undertaking major analytics platform migrations, these strategies address practical realities:

Resistance to Change: Long-tenured programmers and biostatisticians may resist learning new platforms, particularly when existing tools meet current needs. Community-building and peer support reduce isolation and create positive associations with new technology.

Knowledge Transfer: Formal and informal knowledge-sharing mechanisms accelerate learning, enabling experienced users to assist colleagues and preventing knowledge hoarding.

Stakeholder Engagement: Involving users in migration planning and execution creates ownership and investment in success rather than passive resistance.

Sustained Momentum: Extended migrations risk losing organisational energy and commitment. Regular events and communications maintain focus and celebrate progress.

The recognition that “fostering community and personal relationships is as critical to the success of a technology project as the code itself” reflects mature understanding that technology projects are fundamentally human endeavours, succeeding or failing based on people’s willingness to adopt change rather than purely on technical merit.

Conclusion

SAS Innovate 2025 demonstrated substantial advancement in analytics capabilities addressing pharmaceutical industry requirements. From SAS Clinical Acceleration’s comprehensive clinical trial data management to natural language interfaces democratising data access, the platform evolution addresses industry-specific challenges whilst maintaining governance, validation and transparency requirements essential for regulated environments.

The emphasis on trustworthy AI, with particular focus on bias detection, explainability and human oversight, aligns with pharmaceutical industry imperatives around patient safety and regulatory compliance. The integration of healthcare professional perspectives into technology discussions grounds innovation in clinical reality, ensuring technology serves rather than dictates medical and scientific judgement.

The architectural shifts toward cloud-native platforms, analytics-at-the-data-source and integrated development environments promise to accelerate pharmaceutical innovation whilst maintaining rigorous standards the industry requires. The recognition that successful technology implementation depends equally on human factors and technical capabilities reflects mature understanding of organisational change dynamics.

As pharmaceutical companies navigate increasing clinical development complexity, growing regulatory expectations and mounting pressure to accelerate therapeutic innovation, the capabilities and approaches demonstrated at SAS Innovate 2025 offer practical pathways forward. The evolution from experimental tools to production-grade infrastructure supporting life-saving work continues, with platforms increasingly designed not merely for technical sophistication but for human usability, regulatory compliance and meaningful contribution to pharmaceutical companies’ core mission of improving patient outcomes through data-driven innovation.

See more Trends & Insights