(300) Assessing the feasibility of TREAT-NMD registries for characterising adult patients with SMA receiving risdiplam treatment

Background: Evrysdi (risdiplam) is a disease-modifying therapy (DMT) approved for spinal muscular atrophy (SMA). However, its effectiveness in adult SMA patients has not been formally studied in clinical trials. Aetion, Roche, and TREAT-NMD are collaborating on a real-world registry-based study to assess Evrysdi's effectiveness compared to untreated adults with SMA.

Objectives: To support this study, we conducted a fit-for-purpose (FFP) assessment to evaluate the feasibility of using SMA registries within the TREAT-NMD network.

Methods: The FFP assessment comprised pre-feasibility and feasibility evaluations. These assessments were based on the REQueST tool, EMA guidelines on registry-based studies, and the structured fit-for-purpose data framework (SPIFD).

Results: Initially, we identified and contacted 11 SMA registries within the TREAT-NMD network. Eight were in Europe (Belgium, Spain, UK & Ireland, Czech Republic & Slovakia, Ukraine, Germany & Austria, Switzerland, and Sweden), two in the Pacific region (Australia and New Zealand), and one in North America (Canada). Four registries (Spain, Sweden, Switzerland, and New Zealand) were excluded during the pre-feasibility stage due to low numbers of adult SMA patients, limited Evrysdi usage, or availability issues.

Subsequently, seven registries (Australia, Belgium, Canada, Czech Republic & Slovakia, Germany & Austria, Ukraine, and UK & Ireland) underwent feasibility assessment. Data availability and completeness were evaluated across these registries. The majority (78%) of adult patients had a genetic SMA diagnosis, and 82% had DMT records. However, patient-reported outcomes and motor function measures were inconsistently reported.

Based on registry performance, we categorized them into three tiers: high, medium, and low fit-for-purpose. Belgium and Czech Republic & Slovakia were in the high tier, meeting most data requirements with good completeness. Australia, Canada, Germany & Austria were in the medium tier due to some inadequately collected data. UK & Ireland and Ukraine were in the low tier due to missing or poorly collected key data elements. All except Ukraine were selected to contribute to the study.

Conclusions: Ultimately, six registries within the TREAT-NMD network were chosen for participation. The presence of missing data and variability in data collection emphasize the importance of using multiple registries to effectively address research questions.