The study compares two groups of patients with spinal muscular atrophy followed in the same centre by the same professionals. In one group the diagnosis was made early in the first weeks of life by neonatal screening, and in the other according to the onset of symptoms. Both groups received treatment, but the evolution of the cases detected early (and therefore treated earlier than in the other group) was much better.  

The work corroborates the advantages of early detection of spinal muscular atrophy by genetic testing before symptoms begin, so that treatment can be initiated. Although some very severe cases may also have very early symptoms, the possibility of very early treatment means that the evolution of these cases is also better than when the manifestations are clearly detectable.   

The detection of cases by genetic screening for spinal muscular atrophy reaches 95 % who show absence of the SMN1 gene responsible for the disease. In the other 5% the gene is present, but with more subtle alterations that cannot be detected with current screening methodology. This is why the inclusion of neonatal screening for spinal muscular atrophy would be an important step forward in the prevention of the disease.

It is a good quality study, but with two cohorts separated in time and not randomised, which increases the risk of bias compared to a study of this type. However, it would be unethical to conduct a randomised study on neonatal screening at this time.  

The study confirms data from clinical trials, which demonstrated much greater efficacy of treatment when patients are treated before the onset of symptoms. It also confirms 'real-life' results from other recent neonatal screening programmes in other countries, which have shown high treatment efficacy when treatments are started before the onset of symptoms. 

As a novelty, this study provides a cohort to compare what happened to patients before screening and suggests that screening improves treatment efficacy, especially in pre-symptomatic children, but also in symptomatic children (something that was not clear in previous studies), because it also allows treatment to be started earlier than when there was no screening.   

In any serious country, neonatal screening should have been introduced by now. Unfortunately, this is not the case in many countries (including Spain). This study provides further evidence on the importance of its implementation.   

In Spain, national neonatal screening has not been implemented. Some ACs have pilot programmes, the level of development and implementation of which varies and I do not know why they have not been made public. Currently, the vast majority of newborns in Spain do not have access to newborn screening. 

The study has the limitations of not being randomised and using two different time cohorts, which are not exactly comparable.  

The biggest limitation is to know if before screening there were other factors that could delay access to treatment. From what I have seen, in Australia nusinersen [a drug used to treat spinal muscular atrophy] was approved in November 2017, although compassionate use access had been available since April 2016. According to this, the comparison cohort starting in August 2016 would have access to nusinersen, but it is unclear whether the ease of access (number of steps, etc.) to compassionate use is the same as when the drug is already approved. In other words, it is possible that factors other than the existence or not of screening may have played a role in delaying the start of treatment before it was implemented. This could influence the results by making screening appear more favourable than it actually is. 

It seems that the patients in the screening cohort were somewhat more severe than those in the other cohort. This could be because prior to screening the patients would have died before arriving at the centre and being able to receive treatment. This could influence the results by making the screening appear less favourable than it actually is. 

The treatments used in the screening cohort and the comparison cohort were different: 33% of the screening cohort were treated with onasemnogene abeparvovec versus 0% of the comparison cohort (all were treated with nusinersen). This could also affect the comparison of screening efficacy.  

These limitations, which are inherent to real-life studies, do not invalidate the conclusions of the study, which demonstrates that the implementation of a neonatal screening programme is an effective measure to initiate treatment early and very significantly improve health outcomes in patients with spinal muscular atrophy. Moreover, a previous study by the same authors had demonstrated its cost-effectiveness.  

Today, there is no doubt that the implementation of neonatal screening for spinal muscular atrophy should be the first measure to be implemented in all countries, such as Spain, where there is access to course-modifying treatments.