Is Your Clinical Trial Enrolling the Right Patients? Q&A with Jen Lamppa

One of FDA’s main focuses at the moment is streamlining the drug approval process. In June of this year, the agency announced a new pilot program designed specifically to shorten the early-stage drug trial timelines.

A key factor of this is the growing acceptance of RWD in trial design. The agency hopes that by allowing RWD collected by qualified research institutions, drug developers will not only be able to move through early stages faster, but also reduce the need for late-stage trials as well.

Pharmaceutical Executive spoke with Jen Lamppa, SVP of commercial strategy, how RWD impacts trial design, specifically during the patient eligibility screening phase. The addition of this data is expected to impact recruitment timelines, although it may have other unintended impacts as well. As such, pharma and biotech companies will likely need to adjust their recruitment models to properly take advantage of the new environment.

Click here for the video version of this interview!

Pharmaceutical Executive: What are the costly ways that RWD can impact clinical trial recruitment?
Jen Lamppa: Two things stand out when it comes to self-reported data in clinical trial enrollment. The first is the sheer volume of self-reported data that needs to be collected. The second — and by far the more costly — is the accuracy of that data.

When a patient walks into a doctor's office and is asked about their medical history, the honest truth is that it is often easier to ask the clinician, "What did I say last time? What medications are on my chart?" There is a significant burden placed on patients at the moment of enrollment: they are expected to accurately recall their current health status, their medical history, and what has happened to them, often under time pressure and in an unfamiliar clinical environment.

Real-world data tells the true story. It contains the documented record of what actually happened to a patient historically — and that record is essential for validating whether a patient coming into a study is who they say they are and whether they fit the intended study population.

The consequences of inaccurate self-reported data can be extremely costly. You can end up enrolling a patient who should not have been in the study — someone who doesn't fit the right patient profile, or who carries undue risks that the protocol was designed to exclude. There are documented examples of trials where the wrong patients were enrolled or where inaccurate information was provided at baseline.

In a best-case scenario, the consequence is a time-consuming audit and the potential exclusion of that patient from the analysis. In a worst-case scenario, there are safety implications — and those can be completely detrimental, in some cases becoming a showstopper for the entire study.

PE: How do clinical data layers impact how patient availability is determined?
Lamppa: Clinical and administrative data layers are a powerful way to validate what a patient is self-reporting — and to reduce the burden placed on the patient in the first place. Using real-time, patient-specific data feeds, you can get an immediate read on diagnoses, current medications, health status, and prior health history before or at the point of enrollment.

That capability improves the accuracy of the information being collected and strengthens the validation process. It adds a layer of refinement not just to who you are reaching out to, but to what information you choose to ingest on a patient and how you verify it — creating a more reliable foundation for the entire enrollment process.

PE: What recent FDA actions are impacting sponsors?
Lamppa: The FDA's recent guidance encouraging sponsors to leverage real-world data is a wholly positive signal. That guidance largely focuses on the use of real-world data in the evidence package — specifically, the ability of sponsors to use real-world data in some cases without providing the identified patient record. That is a fair limitation of many real-world data sources that have been aggregated for registry and research use, and the FDA's increasing tolerance and acceptance of de-identified real-world data as part of the evidence package is, in effect, an invitation for sponsors to explore how to also use that data upstream to streamline trial starts and patient enrollment.

If sponsors are investing in real-world data and making it part of critical study workflows, the most effective approach is to start by profiling patients and designing studies with a higher likelihood of completion — enrolling diverse and representative patient populations without the undue restrictions that legacy study designs have historically imposed. From there, that investment carries through to finding where those patients are, pre-screening them, and providing supplemental information to validate that a patient belongs in that study. All of that ultimately feeds into the eventual collection of patient-specific data for the study data package.

The FDA's movement to encourage real-world data use through downstream acceptance is, in effect, a call to action for sponsors to begin exploring — if they haven't already — how to use it upstream as well. Doing so will make the entire process more efficient and generate greater value from an investment that is already being made.

PE: How is AI being used to prevent clinical trial mismatches?
Lamppa: AI can be applied very effectively in this space — particularly to accelerate analysis of unstructured clinical data. In administrative data such as claims, the bulk of information is in structured formats: diagnosis codes, procedure codes, medication codes. But in clinical data, where much of the real value sits — the value most adjacent to what is collected in a clinical study — a significant portion is unstructured. It lives in clinical notes, radiology reports, and images. AI is being applied broadly to extract insights from that unstructured layer, and then to validate that those insights are accurate.

The ability to harness AI to accelerate specific analytics — including modeling and sensitivity analyses — is critical to actually putting real-world data to use for its intended purposes. The traditional approach has always been manual adjudication and validation, which does not scale. That limitation has been widely recognized for some time, and there will always be a place for manual review in data management and validation. But the application of AI to reduce the manual burden, compress timelines, and ultimately improve the accuracy of what is being collected on patients is foundational to the broader use of real-world data — particularly complex clinical real-world data.