Utilizing Fit for Purpose Real World Data to Accelerate Clinical Trial Success

Clinical trials are becoming harder to execute; patient populations are more fragmented, and eligibility criteria are more precise. Trials are expected to move faster while generating stronger evidence for regulators, payers, and clinicians alike. Yet the operational constructs that underpin trial design and execution often remain rooted in a narrower view of patient data; one confined largely to what happens inside the four walls of the trial.

At the Indegene Digital Summit, a conversation between Mayank Raizada, Senior Director at Indegene, and Melissa Wissner, Principal for Clinical Development Solutions at Datavant, explored a question that sits at the heart of this challenge:

What would it take to make Real World Data work - not as an abstract concept, but as a practical, fit for purpose accelerator for clinical trial success?

The answers surface a fundamental shift in mindset, away from viewing Real World Evidence as an optional add-on and toward treating it as a foundational layer of modern clinical development.

Clinical trial data, by design, is highly controlled and deliberately siloed. Patients are assigned subject IDs to protect privacy; data collection is protocol-driven and external context is minimized to preserve internal validity. These safeguards are essential; however they create blind spots.

A trial captures only a narrow slice of a patient's health journey. What happened before screening? Why did some patients fail eligibility? What treatments were tried, and abandoned, before enrolment? How do outcomes evolve long after the final study visit? These are not peripheral questions. They sit at the core of smarter trial design, more effective patient recruitment, and stronger downstream evidence generation strategies. Yet traditional clinical data structures, built for control, not continuity, make them difficult, and often impossible, to answer. This is where fit for purpose Real World Data enters the conversation, not as a replacement for trial data, but as a way to restore context and close critical evidence gaps.

Much of the conversation centred on tokenization; a term that is increasingly familiar in life sciences, but not always fully understood across R&D teams. At its core, tokenization is the process of de-identifying patient data to create a secure, irreversible patient key. Personally Identifiable Information (PII) is removed and encrypted, allowing data from different sources to be linked back to the same individual without exposing identity. The real value of tokenization lies in what it enables within an integrated evidence strategy:

Tokenization doesn't just solve a data problem. It reshapes how clinical teams think about evidence continuity across the patient lifecycle, connecting Real World Evidence to trial outcomes in ways that were previously unachievable.

The most immediate application of fit for purpose Real World Data is patient recruitment, especially in difficult-to-recruit populations. Before a patient ever enters a trial, they carry years of history embedded in electronic health records, claims data, lab results, and specialty care notes. When appropriately linked, this data can confirm eligibility more accurately, reduce screen failures, and ensure that the right patients are being approached, not just the available ones.

This is particularly impactful in rare diseases and complex indications, where finding eligible patients often feels like searching for a needle in a haystack. Real World Data doesn't eliminate recruitment challenges, it reduces guesswork. It shifts patient recruitment from reactive outreach to informed, data-driven targeting. And in an environment where recruitment delays are among the leading causes of trial failure, that shift has a direct bearing on clinical trial success.

One insight that resonated strongly during the session was the value of understanding non-enrolled populations. Screen failures rarely receive the analytical attention they deserve. Yet systematically understanding why patients were excluded, and what distinguishes them from enrolled participants can dramatically improve future protocol design. Fit for purpose Real World Evidence enables this broader lens. By linking data across patients who enrolled, failed screening, or were lost to follow-up, sponsors gain a more realistic picture of real-world disease presentation and treatment patterns. This intelligence feeds directly into evidence generation strategies that are proactive rather than reactive. This shifts trial design from retrospective correction to proactive optimization.

The conversation highlighted a real-world example from a top five pharmaceutical company conducting a late-phase rare disease trial. The study team implemented tokenization through an optional consent embedded within the primary informed consent process. This allowed them to assess overlap between trial participants and multiple Real World Data sources, including claims, electronic health records, labs, and social determinants data.

The results were instructive:

Rather than forcing a one-size-fits-all Real-World Data strategy, the team adopted a hybrid approach, using tokenized Real-World Data where fit existed, and supplementing gaps with targeted medical record retrieval where necessary. This pragmatic, integrated evidence strategy delivered actionable insights without inflating cost or complexity.

Tokenization, medical record retrieval, and data linkage all hinge on informed patient authorization. Experience shows that consent rates are highest when tokenization authorization is introduced at the time of main trial consent, supported by clear, human explanations and not technical abstractions. Sites and study teams play a critical role here. When equipped with simple narratives about why Real-World Evidence matters and how privacy is protected, patients are far more likely to opt in.

As Real-World Evidence strategies grow more sophisticated, sponsors often face a familiar challenge, too many stakeholders, too many data sources, and escalating costs. When teams are unclear about the specific questions they are trying to answer, data strategies expand unnecessarily. Multiple datasets are acquired "just in case," and linkage occurs without a clear analytical purpose, undermining the very efficiency that Real World Data is meant to deliver.

Fit for purpose assessments serve as a corrective force. By defining research questions early and aligning data selection accordingly, sponsors can reduce redundancy, contain costs, and accelerate timelines. Robust evidence generation strategies begin not with data acquisition, but with decision clarity.

Early engagement with therapeutic leads, clinical development teams, and operational partners creates alignment. Equally important is early dialogue with regulatory stakeholders. Health Authorities increasingly expect transparency on how Real-World Data is sourced, linked, and used. Bringing regulators into the conversation early builds confidence, supports integrated evidence strategy development, and reduces downstream friction. Embedding tokenization considerations into protocols, IRB submissions, and consent language from the outset is significantly easier than retrofitting them mid-study.

Machine Learning (ML) already plays a role in abstracting unstructured data, improving patient matching, and accelerating record retrieval. Looking ahead, AI is essential for navigating the sheer volume and fragmentation of Real-World Evidence sources in healthcare. However, while AI amplifies intent, without clear questions and fit for purpose Real World Data strategies, even the most advanced algorithms fall short of supporting meaningful evidence generation. The future is not about more data or more automation, it is about better alignment between data, purpose, and decision-making.

The promise of Real-World Evidence in clinical trials is no longer theoretical. It is practical, operational, and increasingly necessary for clinical trial success. However, realizing that promise requires a shift, from accumulation to intentionality, from siloed datasets to integrated evidence strategies, from transactional data use to sustained evidence generation across the product lifecycle. Fit for purpose Real World Data does not ask sponsors to do everything at once. It asks them to do the right things, in the right order, for the right reasons. And in a clinical landscape where speed, quality, and patient-centricity must coexist, that discipline may be the most powerful accelerator of all.