The Lab Result Is Not the Cohort: Operational Lessons From FDA’s New Hepatic Impairment Guidance

The denominator in feasibility is usually wrong

Conventional feasibility questions often begin with the number of patients carrying a liver disease diagnosis or meeting selected laboratory thresholds. Those numbers may describe potential reach, but they do not establish an enrollable population.

FDA recommends Child-Pugh classification for participants with cirrhosis due to chronic liver disease. For most oncology studies, the draft allows use of National Cancer Institute criteria, with an important exception for hepatobiliary cancers in which cirrhosis may be the principal risk factor. Child-Pugh classification combines bilirubin, albumin, prothrombin time or INR, ascites, and hepatic encephalopathy. FDA also asks sponsors to collect the etiology of the underlying liver disease. ([fda.gov](https://www.fda.gov/media/71311/download))

Each requirement reduces the usable pool. A diagnosis code does not show whether all classification components are available. A recent laboratory panel does not establish whether the participant is clinically stable. A hepatology practice may have many patients with cirrhosis but relatively few who fit a particular severity cohort and can complete an intensive PK schedule.

A useful feasibility response should therefore separate three populations: patients who appear potentially relevant, patients whose records contain enough information for preliminary classification, and patients who are realistically likely to reach dosing.

• Ask how many potential participants have all required classification components documented within a usable timeframe.
• Separate mild, moderate, and severe impairment estimates instead of reporting one liver disease total.
• Identify which estimates come from searchable records and which depend on physician recall.
• Account for acute illness, recent hospitalization, medication conflicts, transportation, caregiver needs, and the required PK schedule.

Classification is a clinical process, not a screening calculation

The Child-Pugh score includes assessments that can vary between evaluators. Ascites may be absent, controlled, or clinically significant. Encephalopathy grading can be affected by documentation quality, treatment, recent events, and the participant’s condition at the time of assessment.

FDA’s draft asks study reports to include the total Child-Pugh score, the value of every component, and a clear description of how ascites and encephalopathy were evaluated. It also calls for the assessment conditions to be documented, including whether ascites was evaluated with or without diuretic use. ([fda.gov](https://www.fda.gov/media/71311/download))

That level of detail cannot be reconstructed reliably from a total score entered into an electronic case report form. Sites need source worksheets that capture the underlying observations, relevant treatments, evaluator, date, and rationale. Investigator oversight is especially important when the assessment determines cohort assignment or dosing.

Before activation, sponsors should test the classification workflow with realistic cases. If investigators, medical monitors, and data reviewers interpret the same case differently, the study has a design problem that training slides alone will not resolve.

• Define who is qualified to assess ascites and encephalopathy.
• Train staff using borderline and treatment-controlled examples, not only obvious cases.
• Require documentation of every score component rather than a total score alone.
• Establish a rapid medical-monitor escalation process for ambiguous classification.

The screen-to-dose interval is a critical quality factor

FDA recommends that laboratory parameters and clinical status be stable before enrollment. If repeat laboratory results or ascites and encephalopathy assessments change substantially before administration, the draft says assessments should be repeated so the participant can be classified correctly. It gives a change greater than 30 percent as an example and states that the severity group should reflect the assessment closest to pharmacokinetic sampling. ([fda.gov](https://www.fda.gov/media/71311/download))

Operationally, this makes the interval between screening and dosing one of the study’s most vulnerable periods. A participant may qualify at screening but experience an infection, gastrointestinal bleed, medication change, fluid shift, procedure, or hospitalization before dosing. FDA specifically cautions against determining Child-Pugh status during or shortly after acute events because the score may not represent baseline metabolic capacity. ([fda.gov](https://www.fda.gov/media/71311/download))

The protocol and site manual should state which assessments must be repeated, how close to dosing they must occur, who reviews the results, and what happens when a participant moves to another severity category. Without that pathway, the site may face a choice between losing a difficult-to-find participant and making an unsupported classification decision under time pressure.

• Build prespecified reassessment checkpoints into the schedule.
• Minimize avoidable delays between final eligibility review and dosing.
• Track hospitalizations, procedures, infections, transfusions, and medication changes during the interval.
• Confirm cohort capacity before reassigning a participant whose status changes.

Medication reconciliation is part of the scientific design

Participants with chronic liver disease commonly have comorbidities and complex treatment regimens. FDA notes that anticoagulants may affect INR, while other conditions or interventions may alter bilirubin or albumin. The draft also calls for careful assessment of concomitant medications that could influence the investigational drug’s pharmacokinetics or pharmacodynamics. ([fda.gov](https://www.fda.gov/media/71311/download))

For the site, this means medication reconciliation cannot be treated as a clerical list copied from the medical record. Staff need the actual drug, dose, route, frequency, indication, start date, recent changes, adherence pattern, and use of nonprescription products. The investigator must then distinguish therapies that affect eligibility classification from those that may interact with the investigational product.

Peer-reviewed operational guidance similarly identifies polypharmacy and medication interactions as major considerations in hepatic impairment studies. It recommends designing restrictions carefully so that medically stable participants are not excluded unnecessarily when an interaction can be evaluated and managed appropriately. ([pmc.ncbi.nlm.nih.gov](https://pmc.ncbi.nlm.nih.gov/articles/PMC9841300/?utm_source=openai))

• Complete medication review early enough to resolve discrepancies before dosing.
• Do not assume the electronic health record contains outside prescriptions or nonprescription products.
• Define whether prohibited medications can be stopped safely and whether a washout is scientifically justified.
• Reconcile the medication list again immediately before administration.

PK execution should influence site selection

The new draft recommends sampling of sufficient frequency and duration to characterize the parent drug and active metabolites. Depending on the drug’s extraction and protein-binding properties, protocols may also require measurement of both total and unbound concentrations. FDA recommends retaining sufficient samples for possible future binding analyses in certain circumstances. ([fda.gov](https://www.fda.gov/media/71311/download))

These requirements affect staffing, processing times, freezer capacity, backup equipment, courier arrangements, meal timing, and the participant’s ability to remain at the site. A center with strong hepatology access but inconsistent PK execution is not a low-risk choice. Neither is an experienced phase 1 unit without reliable access to appropriately classified participants.

The best model may be a coordinated network in which hepatology practices identify potential participants, research staff perform structured prescreening, and a qualified unit conducts dosing and intensive sampling. Responsibilities for safety communication, source transfer, medication review, and post-dose follow-up must be settled before recruitment begins.

• Review actual processing and storage requirements during feasibility, not after selection.
• Verify backup centrifuges, freezers, temperature monitoring, and after-hours coverage.
• Map every time-sensitive handoff from collection through shipment.
• Assess whether visit length and transportation plans are realistic for medically complex participants.

Early evidence can support broader participation later

FDA’s broader participation guidance recognizes that people with liver impairment are often excluded early because dose and safety information are not yet available. It also states that exclusions should decrease as information on metabolism, excretion, interactions, and appropriate dose adjustment becomes available. ([fda.gov](https://www.fda.gov/media/190162/download))

The September draft connects dedicated hepatic impairment work to that larger development objective. FDA recommends assessing hepatic impairment early enough to guide later eligibility criteria and dosage selection. It also notes that patients with hepatic impairment have often been excluded even when they are part of the intended treatment population. ([fda.gov](https://www.fda.gov/media/71311/download))

That makes the operational quality of these studies consequential beyond a small clinical pharmacology protocol. Reliable classification and interpretable PK data can help sponsors replace broad organ-function exclusions with evidence-based eligibility and monitoring plans.

The practical lesson is straightforward: do not select sites for these studies by counting diagnoses. Select them by testing the full path from referral to classification, stability confirmation, dosing, and final sample. That is the path that determines whether the study is genuinely feasible.