Building Trust in Real-World Evidence With AI and High-Quality Data

As artificial intelligence (AI) rapidly transforms oncology research, generating reliable real-world evidence (RWE) depends on high-quality, transparent, and rigorous clinical data. In this episode of Exploring AI in Oncology, expert faculty explain how AI is accelerating complex biostatistical workflows without compromising scientific rigor. They explore Ontada’s three-pillar framework for real-world data quality (relevance, reliability, and external validity), the role of generative AI in trial emulation, and why human oversight and regulatory transparency remain essential. Read the full accompanying article on Oncology Data Advisor: https://bit.ly/3SpSv9S What You’ll Learn in This Episode 🔹 Data Quality Before AI: Understanding why complete, traceable, and structured real-world data (RWD) is a prerequisite for generating trustworthy AI analyses. 🔹 Three Pillars of Reliable RWE: Analyzing relevance, reliability, and external validity using Ontada’s dataset of 4 million patients across 700 community oncology clinics. 🔹 Accelerating Trial Emulation: How generative AI assists biostatisticians in defining patient cohorts, writing statistical code, and building analytic frameworks. 🔹 Essential Human Oversight: Why progressive summaries, milestone checkpoints, and human expertise prevent AI "black box" decisions and manage missing EHR data. 🔹 Regulatory Transparency: Meeting FDA and EMA expectations through structured prompting, audit trails, and early bias detection to ensure reproducible findings. Chapter Timestamps About the Speakers Waqas Haque, MD, MPH, is a second-year Hematology/Oncology Fellow at the University of Chicago. He completed his Internal Medicine Residency at New York University (NYU) in a Clinical Investigator Track, with research focusing on innovative clinical trial design, value-based care delivery, and clinical investigation. Saamir Pasha, MPH, is a Senior Biostatistician and Data Scientist at Ontada.