Understanding the gap between feasibility estimates and patient availability
Patient availability is often overestimated during trial planning. This usually starts at feasibility, where sites are asked to report how many patients they have with a given disease. Those numbers are then used to model enrollment timelines and site selection.
The reality is that site-reported patient counts rarely reflect current, reachable, or trial-eligible populations. Sites may include patients they have not contacted in years, patients whose records are outdated, or patients who are no longer actively receiving care at that institution. The number reflects a presence at some point in the past rather than present availability. Moreover, the number doesn’t account for disease education or willingness to participate in a trial.
Nevertheless, sponsors often aggregate patient estimates across multiple sites when planning enrollment, but the cohort that ultimately enrolls is typically far smaller than those projections suggest. This blog outlines the structural reasons behind this gap, what sponsors tend to miss during site selection, and how patient-level data can support more accurate enrollment planning.
Site selection in crowded indications
Site selection decisions often prioritize reported patient volume and prior trial experience. In rare disease and gene therapy, this approach breaks down quickly because many sponsors pursue the same indications at the same time (a phenomenon commonly referred to as “herding”).
These therapeutic areas are highly congested. Multiple trials compete for overlapping patient populations, often with similar inclusion criteria. Even when sites do have patients on record, those patients may already be enrolled in another study, be in screening elsewhere, or have exhausted their willingness to participate in research.
As a result, patient availability becomes constrained not by disease prevalence, but by competition and patient fatigue. This is rarely visible during feasibility.
What enrollment looks like in practice
An analysis of over 2,000 clinical trials revealed that less than half of trials met their planned enrollment targets, with a median enrollment that was 31% lower than originally specified.
These shortfalls are not apparent at the outset; enrollment challenges often emerge weeks or months after activation, once outreach begins and eligibility is assessed in practice. Without ongoing feedback from sites and patient-facing teams, sponsors can lose valuable time before recognizing that assumptions were incorrect.
Continuous feedback throughout the trial is essential. Enrollment should be monitored as a dynamic signal, not a static forecast created during feasibility.
Data that supports more realistic planning
One way to improve accuracy is for sponsors to rely less on site-reported counts and more on patient-level data they control or directly assess. Purpose-built patient registries or detailed databases can provide a more grounded view of the population.
High-quality datasets allow sponsors to filter by genetic variants, years since diagnosis, disease progression, and recency of patient contact. This level of detail is particularly valuable for site selection, as it reflects not just who exists in records, but who may be reachable and relevant to the protocol.
That said, patient presence in a registry does not equate to enrollment readiness. Many patients have limited education about their disease or genetics. Others may not be aware of clinical trial options or may be unwilling to participate due to burden, logistics, or prior experiences. Patient data improves planning accuracy, but it does not remove the need for engagement and education.
Planning for real availability
Overestimation of patient availability is caused by outdated site records, overlapping trials competing for the same patients, and limited insight into whether patients are reachable, eligible, and willing to participate at the time recruitment begins. As a result, experienced teams tend to recalibrate how they interpret feasibility data. In practice, site-reported numbers are often treated as an upper bound rather than an expected outcome, with conservative adjustments applied systematically, particularly in congested indications where patient competition and site burden are high.
Sponsors should also plan for enrollment as a dynamic process rather than a fixed forecast. Early signals from outreach, screening, and site feedback are often more informative than initial feasibility estimates. Building structured feedback loops into trial execution allows sponsors to identify underperforming sites quickly and adjust recruitment strategies before timelines slip.
Incorporating patient-level data into feasibility and site selection can further improve accuracy. While this data does not guarantee participation, it reduces reliance on assumptions and surfaces constraints earlier in planning.
How Sano supports this approach
Sano enables this planning model by providing patient data infrastructure designed for precision patient finding and sustained engagement. Rather than relying on site-reported estimates, sponsors work with a patient population that has been identified against protocol-level criteria and engaged directly.
Because patients are educated about their disease, genetics, and research options upfront, estimates reflect individuals who are trial-ready, engaged, and reachable, not inferred from historical records. This reduces uncertainty during feasibility and lowers drop-off during enrollment.
Sano maintains this patient population over time by delivering personalized educational materials and updates about relevant studies, allowing sponsors to continue engaging the same cohort as programs move through phases, expand sites, or enter new regions. This continuity helps sponsors plan enrollment with greater confidence across the development lifecycle.
To learn more, explore our Virtual Waiting Room page.