The MetALD Operations Playbook
From Scientific Theory to Operational Control (ICH E6(R3)-aligned)
MetALD trials are operationally fragile because alcohol exposure and metabolic status fluctuate. This playbook translates MetALD science into implementable Clinical Operations controls: eligibility integrity (“Golden Zone”), dual-stream alcohol governance (eDiary + PEth), safety escalation (“Safety Stop”), and endpoint defensibility (“Shared Pathogenic Feature”).
By Accelsiors CRO — supporting Sponsors and study teams executing hepatology trials.
By Accelsiors CRO — supporting Sponsors and study teams executing hepatology trials.
Quick Answers (MetALD trial operations)
Q1 (H3): What is MetALD (operational definition)?
MetALD is defined by metabolic dysfunction plus moderate alcohol consumption. In trials, it requires a precise alcohol window plus ≥1 cardiometabolic risk factor to avoid phenotype dilution.
Q2 (H3): What is the “Golden Zone” for MetALD enrollment?
- Females: 20–50 g ethanol/day
- Males: 30–60 g ethanol/day
Operational note: convert “drinks/day” into grams/day using region-specific tools to avoid systematic unit error.
Q3 (H3): What biomarker best verifies alcohol exposure in MetALD trials?
PEth (phosphatidylethanol) in whole blood, reflecting cumulative alcohol exposure over ~2–4 weeks.
Q4 (H3): What PEth thresholds are used as operational bands? (illustrative)
- PEth <20 ng/mL: abstinence/very low intake → phenotype drift risk (MASLD-like)
- PEth 35–200 ng/mL: commonly used “moderate use” range supporting MetALD phenotype (when concordant with self-report)
- PEth >200 ng/mL: heavy/harmful drinking → Safety Stop trigger (protocol-defined, medical review)
Q5 (H3): What is a “Discordance Alarm”?
A predefined trigger such as PEth >35 ng/mL AND eDiary = “0 drinks”, prompting automatic reconciliation + supportive re-training (non-punitive) with an auditable trail.
Micro-disclaimer (small text): Thresholds are operational examples; sponsors must define protocol-specific limits and lab manual/SAP handling.
The problem (Why this matters to ClinOps)
Why MetALD trials fail operationally (predictable failure modes)
- Enrolling the wrong patients: “past drinkers” or hidden heavy drinkers dilute signal and create regulatory risk.
- Phenotype drift: participants move toward abstinence (MASLD-like) or harmful intake (ALD-like) mid-study.
- “Alcohol noise”: behavior change can mimic drug effect on enzymes, NITs, and even histology—driving high placebo response.
If alcohol exposure is not governed as a Critical-to-Quality (CtQ) factor, endpoint interpretability can collapse even when the drug works.
The solution (3-Pillar Defense)
The Accelsiors CRO MetALD control framework (QbD “3Pillar Defense”)
Pillar 1 (H3): Definition Rigor (“Golden Zone”)
Hardcode gram/day limits across protocol, screening tools, and IWRS/IRT logic; require active history (e.g., TLFB over 90–180 days), not a single screening question.
Pillar 2 (H3): Active Governance (“Control Loop”)
Pre-specify: Trigger → Action → Evidence for discordance, drift, and safety escalation. Embed logic in DMP/CMP to govern dual streams (eDiary + PEth).
Pillar 3 (H3): Defensible Endpoints (“Shared Pathogenic Feature”)
Use a regulator-ready rationale to adapt MASH endpoint hierarchy to MetALD while treating alcohol fluctuation as a governed covariate (stratification + sensitivity analyses + SAP covariates).
What’s inside the Playbook
What you’ll receive in the gated Playbook (PDF)
- RiskGraded Protocol Audit Matrix (Red/Yellow flags) to stress-test MetALD readiness
- PEth mandate guidance (screening + on-treatment logic) and lab/DMP considerations
- DualStream Reconciliation model (eDiary + PEth) including the Discordance Alarm workflow
- Data Drift trigger library (e.g., “Dry January” regional shifts; relapse signals; flatline reporting)
- Safety governance blueprint including Safety Stop thresholds and escalation ownership
- Biopsy “TwoGate” strategy: NITs for screening/stratification; biopsy for confirmatory endpoints
- Stigma mitigation language for ICF and site scripts (framed as liver safety monitoring)
- Endpoint/SAP defenses for “alcohol noise” (stratification, perprotocol stability, PEth covariates)
- Future-proofing controls for GLP1 / FGF21 confounding (stability run-in, stratification, handgrip/BIA, OCDS + mediation analysis)
Primary offer (Consult) — “Protocol Stress Test”
Book a MetALD Protocol Stress Test (30 minutes)
What we do on the call (bullets):
- Identify single points of failure (e.g., missing PEth mandate, missing Safety Stop, weak drift governance)
- Map your current design to CtQ controls (Trigger → Action → Evidence) aligned with ICH E6(R3) expectations
- Recommend practical amendments that protect interpretability without inflating burden/cost
- Outline the fastest path to implement in your DMP/CMP, site training, and dashboard KRIs
“Targets & triggers” (numbers help answer engines)
Implementation targets (example KPIs & KRIs)
Domain | Target | If you exceed this, re-audit |
Integrity | <10% discordance (PEth vs eDiary) | If >20% triggers Discordance Alarm → review stigma/training + digital tool validation |
Stability | <15% phenotype drift out of Golden Zone | Review eligibility window + monitoring frequency |
Safety | 100% Safety Stop adjudicated within 7 days | Review escalation workflow ownership and triage SLAs |
Efficiency | <30% alcohol-related screen fails, >85% retention | If screen fails >30% → revisit eligibility targeting and site profiling |
