Why precision medicine trials fail on the way from protocol to patient

Precision medicine clinical trials are designed by some of the most capable scientists in drug development. The biomarker hypotheses are well-reasoned, the endpoint selection reflects years of translational work, and the statistical plans are sound. The science holds up. Delivery is where programs break down.

That breakdown shows up in the timelines; the delay accumulates during execution, as a program moves from an elegant design to the work of finding real patients. That work runs across a patchwork of vendors who rarely share systems or data. Recruitment sits with one partner, genetic testing with another, laboratory logistics with a third, and long-term follow-up with a fourth. Each handoff is a place where an eligible patient can stall or disappear.

The argument of this article is straightforward. In precision medicine trials, the greatest risk is concentrated at single points of execution: the moment an eligible patient must be found, genetically identified, consented, tested, and enrolled. When that sequence is spread across disconnected vendors, patients are lost. Unified infrastructure closes the gap.

The execution gap is where precision medicine trials fail

When considering why precision medicine trials fail, one of the critical reasons is downstream of the protocol. A trial can have a validated target, a clean inclusion list, and a well-powered endpoint, and still miss its enrollment timeline by a year. The design holds. The operational chain that turns design into enrolled patients is where programs lose ground.

Precision medicine raises the stakes on that chain. Conventional trial design can often recruit against broad clinical criteria. A biomarker-defined study adds a molecular gate before a patient qualifies, which narrows the eligible population and multiplies the steps between first contact and randomization. Narrow inclusion and exclusion criteria control patient heterogeneity and improve the cleanliness of the signal. They also raise the identification burden, because every added qualifier shrinks the pool and lengthens the path to a confirmed match.

Enrollment planning that treats genetics as a late-stage detail tends to underestimate this burden. Sano has written about how genetics reshapes enrollment planning, because the genetic gate changes both the size of the addressable population and the time it takes to reach each patient. When that reality is priced in during design rather than discovered during execution, the timeline holds up better.

The genetic identification step is where eligible patients are lost

The genetic identification step is the sharpest conversion point in a precision medicine trial, and it is where patient recruitment gets genuinely hard. Matching patients to a targeted therapy is far from guaranteed. A review of precision oncology programs found that the rate of matching patients to targeted drugs ranges from 5% to 49%, clustering around 15% to 20%. Most patients who enter the funnel do not emerge with an actionable match.

The gap between a detected variation and an enrolled patient is wider still. In the NCI-MATCH trial, about 40% of screened patients carried a tumor gene mutation matching a studied drug. A matching mutation is only the first hurdle. Between a positive genetic result and a dosed patient sit consent, confirmatory testing, logistics, and timing, and each of those introduces attrition.

This is why screen failure in genetically defined trials should be read as an operational metric as much as a biological one. Some loss is inherent to the biology of the target. A large share is procedural, produced by delays between testing and result, by results that arrive after a clinical window has closed, or by patients who drop out while waiting for a molecular diagnosis. When genetic testing is treated as an extra service rather than an integrated step, the procedural share of screen failure grows.

Under-diagnosis is an invisible screen failure

Some eligible patients are never lost at screening because they never reach it. They remain undiagnosed. In rare disease, this is the diagnostic odyssey, and it functions as an invisible screen failure that never appears in a single trial's metrics.

The scale is significant. According to Servier, only about one in two people with a rare disease has an accurate diagnosis, and for roughly a quarter of patients, the time to diagnosis can be as long as four years. A patient who fits a trial's molecular profile but carries the wrong label, or no label at all, is invisible to recruitment. The years spent reaching a diagnosis also carry real cost. A peer-reviewed study found the diagnostic odyssey for suspected rare disease is associated with extensive healthcare utilization and high spending, well before any trial enters the picture.

Where diagnosis happens is shifting, which changes where eligible patients surface. Derek Ansel, therapeutic strategy lead for rare disease and oncology at Worldwide Clinical Trials, described on his episode of The Genetics Podcast how, years ago, nephrologists rarely ordered genetic testing, while today they perform the lion's share, moving the diagnostic journey into specialties that once sat outside it. Recruitment strategies built on outdated assumptions about who diagnoses a condition will miss the patients now being identified elsewhere. Genetic literacy compounds the problem, because patients who do not understand why testing matters often drop off before they are ever screened.

Fragmented vendors cannot hand a patient across the line

Even when an eligible, diagnosed patient exists, they still have to cross the line into enrollment, and this is where vendor fragmentation does its damage. Recruitment, genetic testing, consent, and site coordination often live in separate organizations with separate systems. A patient identified by a recruitment vendor becomes a referral, then a test order, then a lab result, then a site appointment, and every transfer between systems is a chance to lose them.

The effect on timelines is well documented. Applied Clinical Trials has documented a long-cited industry benchmark: roughly 20% of investigative sites fail to enroll a single patient and another 30% under-enroll. Fragmentation is a consistent contributor to that pattern, because no single party holds accountability for the patient from first contact to randomization.

Derek Ansel frames the CRO's purpose as giving the sponsor a clear read on whether a drug worked, which means reducing noise and clutter in the data. Fragmented delivery works against that purpose. When a patient's genetic result, consent status, and eligibility live in disconnected systems, the operational noise migrates into the very data the trial exists to interpret. Clean handoffs protect both the timeline and the signal.

Unified infrastructure closes the execution gap

The execution gap closes when recruitment, genetic testing, and long-term follow-up run as one system rather than a relay between vendors. When identification, consent, and testing share a single workflow, the patient stops being handed off and starts moving through a continuous process. Fewer transfers mean fewer points of loss.

This is the model Sano is built around: one platform that unifies recruitment and genetic testing rather than stitching them together after the fact. In programs supported by Sano, participant recruitment and at-home DNA testing for eligibility sit inside the same pathway, so a patient who expresses interest can be genetically screened and confirmed without leaving the process. The genetic gate that fragments most precision trials becomes a step in a single flow rather than a handoff between organizations.

The value of that continuity is structural. It shortens the distance between an eligible patient and an enrolled one, it reduces the procedural share of screen failure, and it keeps genetic results connected to the patients they describe. Long-term follow-up in the same system means the relationship does not reset at each phase. The infrastructure carries the patient across the line the protocol assumed would be easy to cross.

Closing the distance from protocol to patient

Precision medicine trials fail because the distance between a well-designed protocol and an enrolled patient runs through too many disconnected hands. The genetic identification step concentrates that risk, under-diagnosis hides eligible patients before they are ever screened, and vendor fragmentation loses the ones who remain.

Treating recruitment, genetic testing, and follow-up as a single system addresses the problem where it actually lives, at the point of execution. That shift is what turns a strong protocol into a trial that enrolls on time and reads a clean signal. As precision medicine programs take on narrower populations and more demanding molecular criteria, the programs that succeed will be the ones that close the execution gap by design rather than discovering it mid-study.

If you are planning a precision medicine trial and want to close the gap between protocol and patient, get in touch.