P.190Feasibility, reliability and convergent validity for digital biomarkers captured via a smartphone application (app) to assess motor behaviors in individuals with spinal muscular atrophy (SMA) in the JEWELFISH trial

Abstract

Frequent visits to the clinic to complete assessments impose a large burden on individuals with spinal muscular atrophy (SMA) and their caregivers. New tools may complement or improve regular assessment of symptom severity and progression. Smartphones enable remote patient monitoring through widely accepted platforms and have previously demonstrated value in disorders such as Parkinson's disease. We developed a smartphone application (app) comprising nine tests for the assessment of finger, hand, arm and lung functions via tapping, tracing, squeezing, sustained-phonation and phone-turning games. Using the app, we are assessing the feasibility, reliability and convergent validity of potential digital biomarkers captured via smartphone in patients with Type 2 or 3 SMA from the JEWELFISH trial (NCT03032172)—an ongoing, multicenter, open-label study evaluating the safety, tolerability and PK/PD relationship of risdiplam (RG7916/RO7034067) in non-naïve patients with SMA (who have previously received olesoxime, AVXS-101 or therapies targeting SMN2-splicing). Risdiplam is an investigational, orally administered, centrally and peripherally distributed small molecule that modulates SMN2 pre-mRNA splicing to increase survival of motor neuron (SMN) protein levels. Patients perform daily active tests at home and data features are computed from sensor data. We are using the aggregated sensor data features over all tests to quantify convergent validity relative to clinical assessment of disease severity (determined by motor function measurement and forced vital capacity). Smartphones may enable long-term, frequent, reliable, low-burden and disease-relevant monitoring for individuals with type 2 or 3 SMA. We will present data from JEWELFISH exploring the validity of this approach.