Day 1 – 9 April 
Day 1 Recording
Presentation TitleSpeaker(s)Abstract

How AI Is Transforming Unstructured Healthcare Data into Actionable Insights

Aracelis Torres, Verana HealthOver 80% of healthcare data is in an unstructured format that is not easily analyzable. These notes contain rich clinical information that has historically been underused in the absence of reliable AI approaches. While manual review may be thorough, findings are likely to grow stale, given the effort and resources it requires. Verana Health’s state-of-the-art AI enables insight into over 90 million patients at high velocity without sacrificing quality. Come learn, via an oncology case study cemented in the urology setting, how Verana helps accelerate the drug development life cycle. With approximately 70% of bladder and prostate cancer patients diagnosed by urologists at early stages of disease, AI provides a way to unlock a critical part of the patient journey at scale. This allows for important insights into key subpopulations and real-world endpoints to be extracted to inform key decisions that impact on patient care.

Leveraging Real World Data for Safety Assessments in the Drug Development Life Cycle

Harini Vadlamudi, GSK

Incorporating real-world data (RWD) into the drug development process offers unprecedented opportunities for safety teams to refine and enhance safety assessments. Firstly, RWD enables proactive monitoring of possible adverse events by providing insights into a wide variety of patient demographics, including those underrepresented in clinical trials. Through continuous data collection from diverse healthcare settings, safety teams can identify potential risks tied to specific subpopulations or pre-existing conditions. Secondly, RWD facilitates the detection of rare adverse effects that may not surface in smaller or shorter-duration trials, ensuring a more comprehensive safety profile. Moreover, it allows for the evaluation of drug-drug interactions and off-label usage patterns, to inform risk mitigation strategies. Safety teams acquire comprehensive information from real-world insights, thereby supporting regulatory decision-making throughout the drug development life cycle. This presentation will explore the multifaceted role of RWD in early- and/or late-stage drug development.

The Patient-Mediated Approach To Collecting Real World Data

Ashley Pitcher, IQVIAIndividual rights over personal data are growing around the world, from HIPAA in the US to GDPR in the European Union, and similar legislation elsewhere. Many health systems are also working on making health data more accessible to patients, and digitalisation of health records continues to grow. What this means is that there is a new avenue for accessing clinical data for health research – via the patient. The traditional way of collecting real-world data not already collected in a registry or database is to engage sites who will recruit patients and collect the data. However, there are limitations with this approach, for example when patients are not typically managed at research sites. In addition, focusing research on patients connected to research sites may limit patient diversity and create unequal access to research participation. A patient-mediated approach may address some of these limitations.
Leveraging RWD To Streamline Clinical Trial Recruitment and Pragmatic Study ExecutionJoanna Guild & Abhishek Dabral, Flatiron Health

Flatiron Health is transforming clinical research through our expertise in RWD and RWE, coupled with a deep understanding of the clinical trial landscape. We are uniquely positioned to address persistent challenges in prospective clinical studies, and bridge the gap between clinical care and clinical research. We will describe the methods by which Flatiron uses RWD and ML-/NLP-enabled abstraction processes to convert unstructured data at scale (e.g. patient charts, physician notes, lab reports) and efficiently identify and pre-screen eligible subjects. We will also explain how our technology surfaces potential candidates to sites within their EHR workflows, to facilitate faster enrolment and increased patient diversity and participation. Once subjects enrol, we streamline data acquisition by automating the transfer of both structured (HL7 FHIR) and unstructured EHR RWD directly to study EDC (using Clinical Pipe™, Flatiron’s EHR2EDC product), to significantly reduce resourcing, cost, and timelines for sites.




Day 2 – 10 April
Day 2 Recording
Presentation TitleSpeaker(s)Abstract

Towards Standardised and Regulated RWD-Based Treatment Comparisons

Tamas Soroncz-Szabo, Takeda

A New CDISC Standard for Reliable Real World Data (RWD)

Tasha Nagamine, Droice Labs & Anita Umesh, Genentech/Roche

FDA guidance indicates RWD submissions must be in a format included in the FDA’s Data Standards Catalog, which currently means CDISC SDTM. To meet FDA RWD reliability requirements, the SDTMs housing RWD need additional lineage/traceability to audit source data and quantify the performance of data transformations. RWD collected for clinical care does not easily map to SDTM, which is designed for clinical research, thereby resulting in data loss during the complex processing required to transform RWD to SDTM. This data loss causes unquantifiable biases that render the RWD unreliable, precluding regulatory use. CDISC’s RWD Lineage project aims to provide standardised and comprehensive data lineage for each RWD data point, represented in a CDISC standard metadata model. This standardisation will allow lineage to be combined across disparate data sources and enable the unification of tools, workflows, and analytics for validation and audit activities.

Emulating RCTs with Real World Data

James Joseph, EDA ClinicalReal-world evidence (RWE) studies can achieve conclusions comparable to randomised clinical trials (RCTs) when their design and measurements are closely aligned. This presentation will introduce a structured approach to defining and quantifying agreement metrics essential for successful emulation. We will simulate the accrual of RWE sources and showcase the step-by-step process of SPIFD2 to design fit-for-purpose RWE studies, select appropriate RWE data, and document decisions. Originally intended for comparative research using observational designs –such as single-arm trials with external comparators – SPIFD2 can also support single-group studies, like natural history studies, by framing a hypothetical single-arm target trial. The framework tackles key challenges, including missing data, confounding and bias, to generate robust and reliable evidence. Aligning real-world data with RCT methodologies improves the credibility of RWE, making it more applicable for regulatory and clinical decision-making.

Verifying the Quality and Reliability of Data Through Cryptography

Ali Dootson, EQTY Life SciencesThis presentation will provide visibility into an emerging technology currently tested by sponsors, which cryptographically certifies data at each step across organisations. By leveraging open data standards and open-source tools, integrity and compliance are embedded into every data point, enhancing the reliability of real-world data (RWD) used as evidence. This approach adds a new layer of quality and governance to existing systems and SOPs, enabling data provenance and process traceability back to the source – without requiring auditors to access datasets or systems beyond those submitted by the sponsor. Using a public prototype will demonstrate the foundation of the deployed solution, which provides data and process traceability from SDTM to FHIR. Guided by governance controls that protect patient privacy, reduce bias, and provide explainability for system and human actions, we show how mathematics provides data reliability, strengthens its consideration as credible evidence and enables new payment models for data providers based on quality.



Sponsorship 

Hosting the Real World Data Spring Event digitally means that no matter where you are in the world you can participate. It provides the industry with a broader opportunity to share knowledge on a global scale, connecting through the virtual event platform. The sponsor options offer a range of benefits with ample company exposure. See the prospectus for more details.



Real World Evidence Working Group Leads