Broader Eligibility Changes the Operating Model, Not Just the Protocol

FDA is asking for criteria that can be explained

The three guidances address different criteria, but they follow the same logic. Restrictions should reflect the investigational product’s mechanism, pharmacology, toxicity profile, development phase, intended population, and available clinical evidence. Generic language and unexplained numerical thresholds are increasingly difficult to defend operationally or scientifically.

For laboratory criteria, FDA recommends using exclusion thresholds only when they are clearly necessary to mitigate a potential safety concern. Requirements should be customized to the drug, and restrictions inherited from early-phase trials should be revised as pharmacokinetic, safety, exposure-response, organ-impairment, and drug-interaction information becomes available. ([fda.gov](https://www.fda.gov/media/178013/download))

For washout periods, FDA favors relevant clinical and laboratory recovery parameters when they can address the actual safety concern more directly than an arbitrary number of days. Time-based washouts may still be appropriate, but the protocol should explain their scientific basis. Concomitant medications should generally exclude a participant only when a clinically relevant interaction or overlapping toxicity creates a meaningful concern. ([fda.gov](https://www.fda.gov/media/178016/download))

Broader eligibility requires more precise feasibility

A site database count rarely shows whether broader criteria will improve enrollment. The feasibility assessment must separate the theoretical candidate pool from the operationally enrollable population.

Consider a protocol that begins allowing participants with ECOG performance status 2. The site may identify more potential candidates, but some may need caregiver involvement, flexible scheduling, transportation support, longer visits, or decentralized procedures. FDA specifically advises sponsors to consider how inclusion of participants with lower performance status could affect retention and necessary sample size. ([fda.gov](https://www.fda.gov/media/178018/download))

A credible feasibility response should therefore address both access and execution. Sites should estimate how many candidates meet the revised medical criteria, how many can manage the visit schedule, what support is likely to be needed, and whether the study budget and vendor model provide that support.

• Review recent screen failures by criterion rather than relying only on diagnosis counts.
• Identify how many candidates were previously excluded by laboratory thresholds, performance status, washout timing, or medication use.
• Assess transportation, caregiver, mobility, language, and scheduling needs before increasing the enrollment forecast.
• Confirm that referral partners understand the revised criteria and are not applying outdated assumptions.

Screening tools must preserve the nuance in the protocol

Broader criteria should not be translated into a simple instruction to screen more people. The site needs decision tools that preserve the scientific distinctions behind each criterion.

Medication questions are a good example. A prescreener labeled “prohibited medications” may become inaccurate if the protocol permits certain drugs with dose modification, temporary interruption, substitution, or investigator review. The same issue arises when a laboratory result can be repeated, when an abnormal value is acceptable because of the underlying disease, or when eligibility depends on recovery from a clinically significant adverse event rather than a fixed washout date.

The practical response is a structured eligibility matrix maintained under version control. It should identify what is clearly permitted, clearly prohibited, repeatable, correctable, dependent on timing, or subject to investigator or medical-monitor review. This can reduce avoidable screen failures without shifting medical eligibility decisions away from the investigator.

Laboratory inclusivity is also a data-quality issue

FDA’s laboratory guidance calls attention to expected variation across age groups and racial and ethnic populations. It specifically discusses the Duffy null phenotype, which is associated with lower absolute neutrophil counts but not greater infection risk. A 2024 JAMA Network Open analysis found that commonly used neutrophil thresholds could exclude people whose results fall within a Duffy null-associated reference range. ([jamanetwork.com](https://jamanetwork.com/journals/jamanetworkopen/fullarticle/2823538?utm_source=openai))

Operationally, sponsors need to decide how central and local laboratory reference ranges will be handled, whether repeat testing is allowed, and what documentation supports an exception or population-specific interpretation. Sites need clear escalation pathways when a value is technically outside a general reference interval but may not represent the safety risk the criterion was intended to control.

Ambiguity here produces inconsistent screening across sites. One investigator may request medical-monitor review, another may automatically exclude the candidate, and a third may repeat the test without a protocol-defined basis. Broader criteria work only when the decision pathway is reproducible and documented.

Performance status cannot be treated as a checkbox

FDA recommends including adults with ECOG performance status 2, or Karnofsky scores of 60 to 70, unless established safety considerations provide a scientific or clinical rationale for exclusion. The agency also recognizes that performance-status assessment is subjective and does not necessarily identify why a participant has reduced function. ([fda.gov](https://www.fda.gov/media/178018/download))

That distinction matters. Reduced function caused by disease burden may change with effective treatment, while limitations driven by unrelated comorbidities may create different safety and retention considerations. The investigator’s assessment must therefore be clinically grounded, consistently documented, and supported by source information.

FDA also encourages complementary assessments, including patient-reported function, geriatric assessment tools, and potentially digital health technologies. If a protocol adds these measures, sites need training on administration, device support, missing-data follow-up, and the distinction between assessments used for characterization and those used to determine eligibility. ([fda.gov](https://www.fda.gov/media/178018/download))

Broader access can expose weaknesses in retention planning

It is possible to improve eligibility and still fail operationally. A participant may qualify medically but be unable to complete repeated imaging, long infusion visits, intensive pharmacokinetic sampling, or frequent travel. If the study expands eligibility without addressing those burdens, early discontinuations and missing assessments may rise.

Sponsors should model participant support while the protocol and budget are still being developed. Sites should be asked what the target population will realistically require, not simply whether they can recruit it. Depending on the protocol, useful measures may include transportation, caregiver reimbursement, remote follow-up, home health services, local laboratory options, flexible visit windows, and rapid rescheduling pathways.

These measures cannot remove every participation barrier, and decentralized elements must be appropriate for the protocol. They can, however, align the operating model more closely with the population the eligibility criteria are intended to include.

A practical implementation checklist

For active programs, the July guidances provide a reason to review upcoming protocols and amendments before the same criteria are copied into another study. The goal is not maximum inclusivity regardless of risk. It is to remove restrictions that lack a current, treatment-specific justification while building the safeguards and support required for the resulting population.

• For sponsors: document the rationale for each major exclusion criterion and reassess it between development phases.
• For protocol teams: replace vague medication restrictions and automatic washout periods with specific, reviewable rules.
• For CROs: update feasibility templates so sites can report criterion-level candidate losses and participant-support needs.
• For investigators: define how performance status, clinical recovery, laboratory variation, and medication interactions will be assessed and documented.
• For sites: update prescreening scripts, source templates, eligibility checklists, referral education, and escalation workflows before enrollment begins.
• For all parties: monitor screen-failure reasons and early discontinuations by eligibility characteristic to determine whether the broader design is working as intended.