Precious SMA natural history data

Classic chromosome 5q spinal muscular atrophy (SMA) is one of the most common neuromuscular conditions in childhood and the most common fatal genetic disease in infants. Byers and Banker¹ first described the variability of this disease in terms of clinical onset and phenotypic severity of SMA and introduced us to the clinical spectrum of “infantile SMA.” They suggested that SMA was a progressive disease. Others, however, favored a static clinical course after an initial period of deterioration. This debate has continued to the present. Most accept that a continual gradient in phenotypic severity exists, meaning every patient with SMA is truly unique. The modern classification of SMA, established in 1992, is therefore based on age at symptom onset and best motor performance achieved.² SMA type I, also known as Werdnig-Hoffmann disease, exhibits subtypes of differing severity and, therefore, has been further divided into IA, IB, and IC, based on age at clinical onset. Dubowitz³ proposed a decimal classification system based on a continual rather than a discrete variable (e.g., SMA type 1.1, 1.5, and 1.9) to better capture the graded severity of the clinical phenotypes.

A large number of natural history studies, before 1995 and after the discovery of the survival of motor neuron 1 (SMN1) gene and SMN protein, have been published for type I SMA (“nonsitters”), and also for type II (“sitters”) and type III (“walkers”).⁴ SMA is caused by decreased levels rather than complete loss of the SMN protein, leading to loss or selective dysfunction of motor neurons in the spinal cord and brainstem.²

In this issue of Neurology®, Pane et al.⁵ report an observational study of functional abilities in infants, children, and adults with SMA type I. This is an important natural history study because it includes information largely on older patients, which has been lacking in the literature with only a few exceptions of non–molecularly confirmed studies. Pane et al. conclude that, after age 2 years, most patients with SMA type I survive only if they have nutritional support (gastrostomy) and respiratory support (tracheostomy or noninvasive ventilation for more than 16 hours per day). These patients exhibit low scores on functional scales. Some patients, generally with the mildest phenotypes (1.9 or IC), did not require nutritional or respiratory support after age 2 years. This report, however, is not a prospective longitudinal study and does not indicate the proportion of patients in the mild 1.9 cohort who did not survive.

In a large series of type I patients reported by Zerres and Rudnik-Schoneborn,⁶ the survival probabilities at 2, 4, 10, and 20 years of age were reported as 32%, 18%, 8%, and 0%, respectively, showing that the survival rate continues to decline after age 2 years. Clearly, this decline is multifactorial and compounded by the increasing complication rate in this fragile population. The type and extent of nutritional and respiratory support were not specified in their report. Pane et al. also provide information on SMN2 copy number and suggest that survival beyond age 2 years is not related to a higher SMN2 copy number only, as a small number of long-term survivors (after age 10 years without tracheostomy or more than 16 hours per day of ventilation) had 2 SMN2 copies. Admittedly, this is a limited and slightly biased view of the entire population.

The present authors caution against classifying patients only on the basis of symptom onset—prudent advice in our opinion. Clinical onset is difficult to ascertain precisely, either because of lack of parental awareness of early symptoms and signs in their child or because pediatricians (and sometimes pediatric neurology specialists) may overlook subtle early signs such as delayed head control, mild hypotonia, or areflexia. Therefore, classifying patients on the basis of symptom onset has inherent flaws. An objective biomarker will be useful in future studies.

Regarding nutrition, the number of gastrostomy tube (G-tube) placements increased over time in patients classified as 1.9. But even after age 4 years, less than 50% of patients underwent G-tube placement. Yet, G-tube insertion is a reversible procedure, and it may prevent overt or silent aspiration. It also facilitates adequate nourishment, which has a positive effect on overall outcome of these patients. Because G-tube placement can prevent undernutrition, we suggest that more patients with SMA type I may benefit from G-tube feeding.

The authors also report that 34 patients (28%) acquired comprehensible speech, implying normal intelligence. Thirty of the 34 were in the mildest 1.9 group. Although studies are limited, most surviving type I infants, even when nonverbal, have normal IQs if measured using nonverbal means.⁷

Not surprisingly, patients with 3 SMN2 copies had a better outcome than infants with 2 copies, but perhaps surprisingly, 2 patients with 4 SMN2 copies did not exhibit a milder phenotype. This observation has relevance to the use of newborn screening algorithms based on SMN2 copy number when making decisions regarding the treatment of newborns diagnosed with genetic SMA.⁸ With the exception of symptomatic newborns with 1 SMN2 copy, for whom palliative care might be more appropriate, presymptomatic newborns with 1, 2, 3, or 4 SMN2 copies should be candidates for immediate treatment, with nusinersen and with other promising treatments currently in development. Most patients with 4 SMN2 copies will develop SMA type III and approximately 7% to 10% will develop type II disease. But some, as pointed out by the authors, may develop type I disease. Even if the SMA phenotype cannot be totally rescued, early treatment can improve survival and quality of life.

The SMA natural history data presented in this report are precious for many of the reasons stated above. Robust natural history data on naive patient populations will become less available in the future once effective treatments are approved, and we will become more dependent on the historical record for comparisons. Pane et al. have clearly documented for us the modern natural history of the severe form of SMA before the advent of effective disease-modifying treatment. Their data will serve as a clear benchmark to measure future treatment successes.

• © 2018 American Academy of Neurology