While ICH E6(R3) provides the regulatory framework for modernized clinical trials, the real challenge lies in operationalizing RBQM across a global portfolio. At Roche, this transition wasn't just a process update—it was a cultural transformation. This session pulls back the curtain on the Roche RBQM Center of Excellence (CoE), sharing the successes, the "potholes," and the tangible lessons learned while embedding risk-based thinking from protocol concept through Quality by Design (QbD) to final reporting.

Traditional monitoring and risk management models often generate activity without insight, limiting effective decision-making. This session examines how integrating monitoring and risk management into a unified, quality-by-design-driven approach—anchored in critical-to-quality data and processes can enhance risk visibility, reduce operational noise, and enable outcome-focused decisions through cross-functional collaboration. 

• Moving from fragmented monitoring plans to integrated, decision-enabling frameworks
• Using critical-to-quality data for proactive risk identification and escalation
• Shifting focus from monitoring activities to meaningful outcomes and insights
• Enabling consistent decision-making through cross-functional data review

This talk will explore how centralized monitoring and site monitoring can be effectively combined in a risk-based manner to enhance study oversight at a major pharmaceutical company. We will showcase how at MSD we plan to achieve the following by having integrated centralized monitoring in a risk-based approach: Better patient safety and data quality, more efficient resource allocation, and faster issue detection and resolution.

Quality Tolerance Limits (QTLs, now examples of Acceptable Ranges according to ICH) are focused on monitoring the critical aspects of a trial. If they are deviated from by trial end then it indicates a serious quality problem and might, worst case, lead to inability to submit the data to a regulator. We will describe an approach being used to obtain an early signal that a QTL might be deviated from using a Composite Key Risk Indicator (KRI) built up from other KRIs. Trial teams are beginning to embrace this approach as it puts them back in control of the QTLs.

Risk-Based Quality Management generates increasing volumes of study-level signals, yet translating those signals into timely, consistent decisions remains a challenge. In the context of ICH E6(R3), this session examines why decisions stall and how to operationalize decision-making in practice. Practical approaches to defining thresholds, clarifying ownership, and enabling timely, traceable actions will support more effective, inspection-ready study oversight and continuous improvement.

Many RBQM programs underdeliver not due to lack of tools, but because early solutions evolve into one-off, study-specific implementations that struggle to scale or validate. This session shares a real-world journey from ad-hoc approaches to a validated RBQM foundation, highlighting practical enablers: disciplined data integration, strong engineering practices and cross-functional collaboration. It also shows how establishing a modern data platform such as Databricks supports standardized workflows today and creates a future-ready foundation for AI-enabled monitoring and decision support.

In this presentation, we explore how AI drives clinical quality improvements and inspection readiness. By offering quality signals, identifying risks, and enhancing decision-making, AI can boost operational efficiency. We'll examine the interplay between AI solutions and change management, sharing best practices for successful implementation. With real-world examples, attendees will learn how AI fosters continuous improvement and elevates performance. AI's role as a key enabler and effective change management will be central to our discussion.

This presentation introduces an AI-powered assistant designed to perform structured, protocol-driven risk identification in clinical trials. By deeply analyzing study design, population, endpoints, procedures, investigational product characteristics, and vendor dependencies, the tool systematically generates GCP-aligned risk statements mapped to Critical-to-Quality factors. The approach enhances consistency, traceability, and early detection of high-impact risks, supporting proactive RBQM implementation before study start-up.