LAUSR

The 4-Copy Conundrum in the Treatment of Infants with Spinal Muscular Atrophy Stella Deng, MD, Bo Hoon Lee, MD, Emma Ciafaloni, MD, and Samuel J. Mackenzie, MD, PhD 1,2 Yeo et al. recently reported on the ethical considerations of treating patients with spinal muscular atrophy (SMA), noting that “answers to concerns on [therapy] durability and treatmentmodified phenotypes will only come with the passage of time.” In agreement with this statement, we wish to emphasize the unique questions pertaining to the treatment of pre-symptomatic infants with ≥4 copies of SMN2. Early pre-symptomatic treatment of infants with 2–3 SMN2 copies has been demonstrated to improve outcomes in multiple clinical trials (e.g., ENDEAR, NURTURE, SPR1NT, RAINBOWFISH interim results); however, data for ≥4-copy patients is lacking. In 2020, the Cure SMA working group revised its position statement regarding 4-copy infants, recommending early treatment over individual clinician equipoise, while continuing its recommendation of watchful waiting in the 5-copy population. While we agree that the risk–benefit balance supports early treatment in 4-copy patients, which therapy to recommend is less clear, especially as the long-term durability of episomal gene transfer (onasemnogene abeparvovec-xioi) remains unknown. Individuals with 4 SMN2 copies may rarely develop SMA type 1 or 2, and a significant minority will ultimately be categorized as a 3a phenotype (symptom onset 18 months-3 years). At the other end of the spectrum, individuals with ≥4 SMN2 copies may remain asymptomatic into their 4th and 5th decades. Early data from Germany’s newborn screening program refute a more extreme historical ascertainment bias (i.e., the presence of outright asymptomatic individuals with biallelic loss of SMN1 in the pre-newborn screen era), but the potential for symptom onset in adulthood remains a real possibility for a significant number of 4-copy patients. Without question, administering onasemnogene abeparvovecxioi to newborns fated to develop adult-onset SMA is ethically murky, and the same could be argued for the use of nusinersen and risdiplam on the basis of cost, potential side effects, and burden to patients (regular spinal injections and daily medication administration, respectively). For now, clinicians and families are left with an imperfect set of tools (frequent clinical assessments, electrophysiological measures, precisely distinguishing 4 from >4 SMN2 copies, and assessing other genetic modifiers) to guide them as they weigh the risk of impending motor neuron loss against that of unnecessarily early treatment with drugs lacking long-term safety data. Ultimately, knowing which 4-copy infants to treat and with which modality may “only come with the passage of time.” Author Contributions S.D., B.H.L., E.C., and S.J.M. contributed to the conception and design of the manuscript. S.D. and S.J.M. contributed to drafting the text.