Log In

Forgot Password?
Create New Account

Loading... please wait

Abstract Details

Discovering Dominant Features Associated with Disease Progression in Huntington’s Disease (HD): A Data-driven Approach Using the Enroll-HD Database
Movement Disorders
P6 - Poster Session 6 (12:00 PM-1:00 PM)
3-007
Knowledge of prognostic factors can aid clinical decision making and should be taken into account when analysing HD data. To date, few variables have been identified as strong independent predictors of HD progression.
To systematically test and rank the statistical importance of potential prognostic factors on different aspects of disease progression in manifest HD (Stages I–V).
Enroll-HD is a global HD registry, from which participants with adult-onset HD, aged 25–65 years, Independence Scale (IS) ≥70 at baseline and ≥2 annual visits were identified (N=1,608). In all, 103 potentially prognostic features were considered including baseline demographics, clinical characteristics, comorbidities, symptoms, pharmacological and non-pharmacological treatments and IS. A random forest-based algorithm was used to rank features according to ability to predict outcomes (using the permutation importance accuracy). Outcome measures were 2-year changes in composite Unified Huntington’s Disease Rating Scale (cUHDRS), Total Motor Score (TMS), Symbol Digit Modalities Test (SDMT), Stroop Word Reading (SWR) and Total Functional Capacity (TFC).
Cytosine adenine guanine (CAG)-age product (CAP) was the most predictive feature across all outcomes, and CAG was second for all except SWR, where IS was second and CAG third. IS ranked third for SDMT, TMS and cUHDRS and eighth for TFC. Other features associated with faster progression were being accompanied to visits (all outcomes), history of cognitive impairment (all but SWR/TFC), tetrabenazine use (all but SWR), antipsychotic use (all but TMS) and apathy symptoms (all but SDMT/SWR). 

Results suggest that in addition to CAG repeat length, CAP and IS, there are other important predictors in HD. These include factors such as attending visits accompanied, history of cognitive impairment and tetrabenazine or antipsychotic use. 


Though causality of associations should be further explored, results suggest important predictors that are candidates for statistical control in further analyses of HD data.


Authors/Disclosures

PRESENTER
No disclosure on file
No disclosure on file
No disclosure on file
Scott Schobel Scott Schobel has received personal compensation for serving as an employee of F. Hoffman-La Roche Ltd.
No disclosure on file
Jeffrey D. Long Jeffrey Long has received personal compensation in the range of $5,000-$9,999 for serving as a Consultant for Triplet. Jeffrey Long has received personal compensation in the range of $10,000-$49,999 for serving as a Consultant for Vacinnex. Jeffrey Long has received personal compensation in the range of $500-$4,999 for serving as a Consultant for Wave. Jeffrey Long has received personal compensation in the range of $0-$499 for serving as a Consultant for PTC. Jeffrey Long has received personal compensation in the range of $500-$4,999 for serving as a Consultant for Genentech. Jeffrey Long has received personal compensation in the range of $10,000-$49,999 for serving on a Scientific Advisory or Data Safety Monitoring board for Roche. Jeffrey Long has received personal compensation in the range of $500-$4,999 for serving on a Scientific Advisory or Data Safety Monitoring board for uniQure. Jeffrey Long has received research support from NIH DSMB.