The recent Cure SMA 2026 meeting featured presentations on evolving treatment pathways for spinal muscular atrophy (SMA). Real-world findings and gene transfer optimization trials demonstrated durable motor outcomes, stable safety profiles, and the ongoing refinement of genetically targeted therapies.

Following these presentations, featured expert Kapil Arya, MD, was interviewed by Conference Reporter Associate Editor-in-Chief Sandra Ramadani, PhD. Clinical perspectives from Dr Arya on these findings are presented here.

Currently, we have 4 genetically targeted SMA treatments, resulting in an evolving clinical practice that is now focused on multiple treatment trajectories: switching treatments; bridging treatments, such as initially starting risdiplam and moving on to gene replacement therapy (GRT); or using add-on therapies, such as adding risdiplam to GRT. So far, the evidence for these clinical practices is largely real world open label, but there were some studies in this space presented at Cure SMA 2026 that are worth noting.

At this year’s Cure SMA meeting, Claudia A. Chiriboga, MD, presented the final results from the JEWELFISH study, which investigated risdiplam therapy in non–treatment-naive patients with SMA (abstract 20). JEWELFISH demonstrated that risdiplam is safe, well tolerated, and effective in previously treated patients with SMA. A key finding from this study is that treatment with risdiplam resulted in a rapid increase in SMN protein in the blood—an almost 2-fold increase from baseline after 4 weeks—that was sustained over 5 years of long-term monitoring. Motor assessments also showed an overall stabilization of motor function across diverse age groups, with almost one-third of patients achieving a score that was greater than or equal to their baseline score. The safety profile was consistent with that observed in treatment-naive patients. Adverse events and serious adverse events actually declined by almost 50% between the first and second years and then remained stable. These are very impressive results.

When it comes to GRT, long-term follow-up studies investigating onasemnogene abeparvovec provide us with concrete insights into GRT’s impact on both symptomatic and presymptomatic patients. At Cure SMA 2026, Megan A. Waldrop, MD, presented the AVXS-101-LT-002 long-term follow-up analysis of intravenous onasemnogene abeparvovec GRT, focusing on presymptomatic infants with SMA (abstract 23). The infants in this trial had 2 or 3 copies of the SMN2 gene and were able to sit independently. Impressively, all patients achieved event-free survival, and no patient required permanent ventilation or a gastrostomy. Swallowing function was also maintained in all patients, allowing for oral feeding. This shows that a one-time infusion can result in a robust expression of the SMN protein, even many years later, without any significant late-onset motor function decline.

Real-world evidence on intravenous onasemnogene abeparvovec in presymptomatic infants shows some acute issues with the liver and platelets, but no significant long-term issues. The bigger concern is with the intrathecal administration of onasemnogene abeparvovec. Will there be any toxicity to the sensory ganglion? Will it cause damage to the transmission of sensory information cranially, perhaps causing ataxia? We currently do not have much long-term data on this, so it is all within the realm of conjecture.

In another presentation at Cure SMA 2026, Crystal M. Proud, MD, reported the extended 64-week outcomes from the phase 3 STEER study, which evaluated intrathecal onasemnogene abeparvovec (OAV101) in patients with SMA (abstract 24). At week 64, there was a small but statistically significant increase of 2 or 3 points from baseline on the Hammersmith Functional Motor Scale Expanded (HFMSE) compared with sham controls. The safety profile was consistent with that from earlier trials. The most serious adverse effects were upper respiratory tract infection and fever, but nothing very worrisome.

Richard S. Finkel, MD, reported the real-world findings from the RESTORE Registry, which is an ongoing assessment that I am currently involved in that looks at long-term outcomes for patients with SMA in the context of treatment advancements (abstract 16). For patients receiving intravenous onasemnogene abeparvovec monotherapy, close to 90% achieved or maintained their motor function scores as measured by the Children’s Hospital of Philadelphia Infant Test of Neuromuscular Disorders (CHOP INTEND), with no significant adverse effects other than what we already know based on published data. For me, the key takeaway from this presentation is that patients with 2 or more copies of SMN2 show consistent and meaningful motor function improvement after receiving early one-time treatment. These are very promising data.

Although not directly related to drug efficacy, some interesting qualitative data were presented during a Cure SMA 2026 clinical care session moderated by Pooja Mohan Rao, MD. For this, a group of psychologists and physical therapists studied what it means to patients when we say “meaningful change” (ie, whether a 2- or 3-point HFMSE improvement is meaningful to them or not). Very poignantly, patient and caregiver testimony suggests that even a single-point increase is a meaningful change—even a stabilization of motor function scales is considered a meaningful improvement. Even if it is not captured by the scales, patients’ stamina improves and they are able to do small things, such as comb their hair or adjust their clothes. This is clinically meaningful to them.

Indeed, at Cure SMA 2026, Dr Finkel also presented an integrated analysis of the DEVOTE Part C and ONWARD studies looking at high-dose nusinersen in nusinersen-experienced patients that showed that, despite minimal improvements in HFMSE scores, patients reported feeling more energetic and having an improvement in their stamina (abstract 22). They did not experience the dip in performance that they experienced previously with lower-dose nusinersen just before their next infusion.