Primary Endpoints for Phase 3 Huntington's Disease Trials
Selecting a primary endpoint for a Phase 3 Huntington's disease trial is one of the most consequential decisions a clinical development team will make. The choice directly shapes regulatory acceptability, statistical powering, trial duration, and—ultimately—the product label. With a small patient population, limited approved therapies, and a disease spectrum spanning symptomatic chorea management to putative disease modification, endpoint selection in HD carries outsized risk if misaligned with agency expectations.
This analysis examines the primary endpoints that have been accepted by FDA and EMA in approved HD chorea trials, and surveys the endpoint strategies currently registered in Phase 3 disease-modifying programs. It covers the clinical outcome assessments used, the measurement instruments regulators have credited, and how endpoint choice differs between symptomatic and disease-modifying development programs.
Want to ask Rhizome your own regulatory questions? Try it for free.
Primary endpoints for Phase 3 Huntington's disease trials: what regulators have accepted
The core distinction: what is the trial trying to show?
There is no single "best" primary endpoint for HD. The defensible choice depends on the therapeutic claim, and the regulatory record splits cleanly into two archetypes:
- Symptomatic trials targeting chorea (VMAT2 inhibitors). Here the established, FDA- and EMA-accepted primary endpoint is the UHDRS Total Maximal Chorea (TMC) score.
- Disease-modifying trials (antisense oligonucleotides, small molecules aiming to slow progression). Here sponsors have used Total Functional Capacity (TFC), Total Motor Score (TMS), or the composite UHDRS (cUHDRS) as the primary measure of global clinical decline.
Symptomatic (chorea) trials: UHDRS Total Maximal Chorea
The TMC score is the through-line across all three approved chorea drugs. It is UHDRS Part I, item 12 (a–g), rating chorea in seven body regions from 0 to 4 each, for a total range of 0 to 28 122225. The primary analysis in each pivotal trial was the change from baseline to a "maintenance"/end-of-treatment window defined as the average of two late visits, which dampens visit-to-visit variability.
- Tetrabenazine (Xenazine), TETRA-HD (Protocol TBZ 103,004): primary endpoint was change in UHDRS TMC from baseline to the maintenance phase, defined as the average of Week 9 and Week 12 scores 12611. Key secondary endpoints were the Clinical Global Impression (CGI), UHDRS Total Motor Score, the UHDRS Functional Assessment Checklist (Part IV), and a UHDRS gait item 1119.
- Deutetrabenazine (Austedo), First-HD (SD-809-C-15): primary endpoint was change in UHDRS TMC from baseline to maintenance, with maintenance defined as the mean of the Week 9 and Week 12 scores 282935. Secondary endpoints were Patient Global Impression of Change (PGIC) and Clinical Global Impression of Change (CGIC) at Week 12, SF-36 Physical Functioning, and the Berg Balance Test 282935. EMA's assessment of the same program identified the identical UHDRS-TMC primary endpoint and maintenance-window definition 39.
- Valbenazine (Ingrezza), KINECT-HD: primary endpoint was change in UHDRS TMC from baseline to end of treatment, with end of treatment defined as the average of Week 10 and Week 12 2225. Key secondary endpoints were CGI-C and PGI-C responder/impression analyses 127128.
Takeaway for a chorea program: UHDRS TMC (baseline to an averaged end-of-treatment window) is the endpoint with the deepest regulatory precedent, and clinician- and patient-rated global impression scales (CGI-C, PGI-C/PGIC) are the conventional key secondaries that support the clinical meaningfulness of a TMC change.
Disease-modifying trials: TFC, TMS, and the composite UHDRS
For interventions claiming to slow progression rather than suppress chorea, single-symptom scales are too narrow, and sponsors have anchored on measures of global functional/clinical decline. The registry record shows three approaches in Phase 3:
- Total Functional Capacity (TFC): change from baseline in TFC has been used as a Phase 3 primary endpoint in early HD, with timepoints seen at 52 weeks in the EU register 120121122. TFC captures global daily independence (work, chores, finances, activities of daily living) 888992.
- Total Motor Score (TMS): used as a primary or co-primary motor measure, with a 26-week timeframe noted in one EU Phase 3 record before extension to TFC over 52 weeks 121122. TMS spans oculomotor function, dysarthria, chorea, dystonia, gait, and postural stability 828889.
- Composite UHDRS (cUHDRS): used as the Phase 3 primary in the tominersen program, specified as change from baseline in cUHDRS (with TFC as the primary for U.S. participants) at 16 months 78; a related EU record specifies change from baseline to Week 65 in cUHDRS total score in the mITT population 120123.
Registry text also shows cognitive and secondary measures (SDMT, Stroop Word Reading, TMS) tracked alongside the primary in the tominersen Phase 3 at 16 months 78, and a separate Phase 3 metformin study using a composite UHDRS cognitive score (built from SDMT, verbal fluency, color naming, word naming, and Stroop interference) as its primary endpoint 80. This illustrates that "the primary endpoint" in disease-modifying HD trials is a design choice among validated global or domain-specific composites, not a fixed regulatory standard.
The composite UHDRS (cUHDRS) in detail
Because the cUHDRS is increasingly the default global endpoint for disease-modifying trials, its construction matters. It is an equally weighted composite of four UHDRS-based measures: TFC, TMS, SDMT (Symbol Digit Modalities Test), and Stroop Word Reading. Each component is converted to a z-score and the z-scores are summed 878990. The components cover:
- TFC – global daily functioning/independence 888992
- TMS – motor impairment (oculomotor, dysarthria, chorea, dystonia, gait, postural stability) 828889
- SDMT – cognitive processing speed, attention, working memory, psychomotor speed 888992
- Stroop Word Reading – attention, processing speed, mental flexibility/psychomotor speed 828889
The rationale for using it as a primary endpoint is that it captures multidomain clinical decline (motor, cognitive, functional) rather than a single domain, tracks underlying progressive brain changes, and relates to change in daily functional ability 89. It is also sensitive to decline; a change of at least 1.2 points has been characterized as a meaningful worsening 8790.
Endpoint selection summary
| Trial objective | Primary endpoint | Scale / range | Typical timepoint | Key secondaries | Precedent |
|---|---|---|---|---|---|
| Suppress chorea (symptomatic) | UHDRS Total Maximal Chorea (TMC) | 7 regions x 0–4, total 0–28 122 | Baseline to averaged end-of-treatment window (Wk 9/12 or Wk 10/12) 12922 | CGI-C, PGI-C/PGIC; TMS, functional checklist 1128127 | Xenazine, Austedo, Ingrezza (FDA + EMA) 2353925 |
| Slow motor progression | UHDRS Total Motor Score (TMS) | Motor items, multidomain 8288 | ~26 weeks (EU record) 121 | TFC, cognitive measures | Disease-modifying Phase 3 121122 |
| Slow functional decline | Total Functional Capacity (TFC) | Global daily function 8892 | 52 weeks (EU record) 120121 | TMS, cUHDRS components | Disease-modifying Phase 3 120122 |
| Slow overall clinical decline | Composite UHDRS (cUHDRS) | z-score sum of TFC+TMS+SDMT+SWR 8789 | 16 months / Week 65 78120123 | Individual cUHDRS components (TMS, SDMT, Stroop) 78 | Tominersen Phase 3 78 |
| Slow cognitive decline | Composite UHDRS cognitive score | SDMT, fluency, color/word naming, Stroop interference 80 | Not stated in registry row 80 | Motor/functional measures | Metformin Phase 3 80 |
Practical considerations for a Phase 3 design
- Match the endpoint to the mechanism and label you want. A VMAT2-type chorea suppressant has a clear, precedented path with UHDRS TMC and global-impression secondaries. A disease-modifying agent needs a global or multidomain measure (TFC or cUHDRS) and a longer readout (roughly 12 to 16 months) to detect a slope difference.
- Use averaged end-of-treatment windows for chorea. Every approved chorea drug averaged two late visits for the TMC primary, reducing measurement noise; regulators have accepted this consistently 12922.
- Pre-specify how clinical meaningfulness is supported. CGI-C and PGI-C/PGIC have served as the anchor secondaries that translate a TMC point change into a patient-relevant effect 28127128. For cUHDRS, a pre-specified meaningful-change threshold (for example, ≥1.2) helps interpretation 8790.
- Cognitive scales are usually components or secondaries, not standalone primaries, except where cognition is the explicit therapeutic target (as in the metformin cognitive-composite design) 7880.
Limitations
The chorea (symptomatic) endpoints are drawn from FDA review/label records and the EMA EPAR and are well established. The disease-modifying endpoints are drawn from trial-registry protocol text (ClinicalTrials.gov, CTIS, EudraCT); registry excerpts sometimes state the scales and visit schedule without the full primary-endpoint wording, so exact estimands (analysis population, comparison timepoint) should be confirmed against each trial's statistical analysis plan 5678120. A PubMed search intended to add measurement-property references failed on repeated attempts and is not reflected here. For any specific program, a targeted follow-up on FDA/EMA scientific advice for that mechanism, and on the estimand and multiplicity strategy, is worth asking Rhizome directly.