Congress’s 2026 appropriations package for HHS embeds a clear directive: FDA and NIDA are to collaborate on alternative clinical trial endpoints for substance use disorder therapeutics, including alcohol use disorder, moving beyond abstinence-only measures to outcomes such as reduced cravings and reduced disorder severity aligned with NIAAA’s updated recovery definition.
Substantively, this is a pivot in evidentiary expectations that could reshape development programs long constrained by abstinence-centric endpoints. It signals openness to endpoints that reflect functional improvement and symptom reduction, closer to real-world practice and more consistent with select European precedents that accepted reductions in heavy drinking as clinically meaningful. For sponsors pursuing precision or neuromodulatory mechanisms with measurable effects on craving and consumption patterns, the door opens to primary endpoints that better track therapeutic pharmacology. Adial, which is preparing a new Phase 3 for AD04 in genotype-selected heavy drinkers after earlier subgroup signals, is positioned to test this thesis, but the implications extend across the SUD pipeline.
For sponsors and CROs, the operational center of gravity shifts to endpoint selection, validation, and execution. Programs will need to justify instrument choice (e.g., validated craving scales such as PACS or OCDS), define clinically meaningful change thresholds, and harden data integrity for high-variance behavioral measures. Expect protocols to lean into continuous endpoints and responder analyses, with co-primary strategies pairing symptom reduction with functional metrics or reductions in heavy drinking days, complemented by objective biomarkers (e.g., EtG/EtS) to mitigate self-report bias. Powering assumptions, multiplicity control, and rater training will become gating issues, as will standardizing ePRO capture and adherence in ambulatory settings. CDISC harmonization and prespecification of estimands will be under the microscope.
Sites will absorb additional assessment complexity. Visit schedules and remote diaries will be recalibrated to frequent, longitudinal PRO collection; rater drift, placebo response, and retention strategies will require tighter operational controls. ePRO and digital adherence tools can reduce burden but demand upfront training and tech support, particularly across community settings. Broader eligibility tied to improvement over abstinence may expand recruitment pools but complicate stratification; balancing diversity mandates with genotype- or phenotype-based enrichment will challenge feasibility planning.
Regulatory process work now begins. FDA and NIDA will likely convene workshops and issue draft guidance clarifying acceptable instruments, anchors for clinically meaningful change, durability expectations, and when composite or hierarchical endpoints are appropriate. The labeling question looms large: how craving reduction or reduction in heavy drinking translates into claims, and what supportive functional data are required. Payer acceptability is another fulcrum; sponsors should anticipate health economics packages linking symptom reduction to healthcare utilization and productivity. Parallel scientific advice with EMA could reduce transatlantic divergence, but until formal guidance lands, risk remains on endpoint selection and trial interpretability.
For companies, the near-term playbook is tactical. Reassess Phase 2 archives for signals on cravings, severity, or heavy drinking reductions that could underwrite accelerated Phase 3 designs. Update statistical analysis plans to incorporate responder definitions tied to validated anchors. Budget for enhanced rater training, ePRO compliance monitoring, and potential biomarker assays. For AD04 and similar assets, watch for protocol filings that codify non-abstinence primary endpoints and the degree of genetic enrichment required to keep trials feasible. The broader test is whether the field can tame placebo effects and instrument heterogeneity enough to deliver reproducible, clinically persuasive outcomes. If FDA/NIDA can standardize the measurement toolkit, this policy shift could shorten timelines and revive shelved assets; if not, variability and payer skepticism may simply move the bottleneck downstream.
Jon Napitupulu is Director of Media Relations at The Clinical Trial Vanguard. Jon, a computer data scientist, focuses on the latest clinical trial industry news and trends.
