Newborn screening for spinal muscular atrophy in Australia: a non-randomised cohort study.

Background: In light of a new therapeutic era for spinal muscular atrophy (SMA), newborn screening has been proposed as a gateway to facilitate expedient diagnosis and access to therapeutics. However, there is paucity of evidence on health outcomes outside the homogenous populations in clinical trials to justify broader implementation of newborn screening for SMA. In this real-world study, we aimed to investigate the effectiveness of newborn screening coupled with access to disease-modifying therapeutics, as an intervention for SMA.

Methods: In this prospective, non-randomised cohort study done at Sydney Children's Hospital Network (NSW, Australia), we included children younger than 16 years with homozygous exon 7 deletions of survival motor neuron 1 gene (SMN1) mutations, non-selectively assigned to a screening group (incident population diagnosed by newborn screening) from Aug 1, 2018, to Aug 1, 2020, or a comparator group (incident population diagnosed by clinical referral) from Aug 1, 2016, to July 31, 2018. We excluded infants with compound heterozygous SMN1 mutations and those participating in ongoing and unpublished clinical trials. Effectiveness of newborn screening for SMA was compared using motor development milestone attainment defined by WHO Multicentre Growth Reference Study at 2 years post diagnosis. Secondary outcome measures included mortality and change in Hammersmith Infant Neurological Examination-2 (HINE-2) score, ventilation requirements, and enteral requirements 2 years from the time of diagnosis.

Findings: 34 children met the study inclusion criteria, but 33 children were included in the study population after one neonate was excluded due to participation in an ongoing unpublished clinical trial. 15 children were included in the screening group (seven [47%] male and eight [53%] female; median age 2·1 weeks [IQR 1·9-2·7]) and 18 children (nine [50%] male and nine [50%] female) were included in the comparator group (median age 47·8 weeks [13·0-99·9]). The 2-year survival rate was 93% (14 of 15 children) in the screening group and 89% (16 of 18) in the comparator group. Among survivors, 11 (79%) of 14 walked independently or with assistance in the screening group, compared with one (6%) of 16 children in the comparator group (χ=16·27; p<0·0001). A significantly greater change in motor function was observed in the screening group compared with the comparator group over 2 years (HINE-2 score group difference, 12·32; p<0·0001). The requirement for non-intensive ventilation or feeding support at follow-up was higher in the comparator group than in the screening group (odds ratio 7·1 [95% CI 0·7-70·2]). Significant predictors of functional motor outcomes as determined by HINE-2 score at 2 years post diagnosis were HINE-2 score (p=0·0022), CHOP-INTEND (p=0·0001), compound muscle action potential (CMAP; p=0·0006), and disease status (p=0·023) at diagnosis.

Interpretation: Newborn screening for SMA, coupled with early access to disease-modifying therapies, effectively ameliorates the functional burden and associated comorbidities for affected children. For children diagnosed through newborn screening, motor score, CMAP, and disease status at diagnosis has clinical utility to determine functional independence.

Funding: Brain Foundation and National Health and Medical Research Council.