Changes in electrophysiological findings of spinal muscular atrophy type I after the administration of nusinersen and onasemnogene abeparvovec: two case reports

Spinal muscular atrophy (SMA) is an autosomal recessive neurodegenerative disorder caused by deletions or variants in the survival motor neuron 1 (SMN1) gene, and approximately 95% of all patients with this condition have a homozygous SMN1 deletion [1, 2]. It causes motor neuron degeneration and progressive muscle weakness and atrophy, resulting in significant motor disability, respiratory and swallowing dysfunction. It is classified into types 1 through 4 based on the age of onset, symptoms, and maximum motor milestones achieved [3]. Patients diagnosed with SMA type 1, known as Werdnig–Hoffmann disease, display symptoms in the first 6 months of infancy, and never achieve the ability to sit independently. The neighboring SMN2 can, in part, compensate for nonfunctional SMN1; hence, higher SMN2 copy numbers in SMA are associated with later onset and milder phenotypes [4, 5].

Recently, there have been significant advances in the therapeutic field, and three different drugs have been approved [6]. Nusinersen, an antisense oligonucleotide drug, and risdiplam, a small-molecule drug, modify the pre-mRNA splicing of SMN2 to increase the production of functional SMN, albeit by different mechanisms [7, 8]. The other therapy known as onasemnogene abeparvovec (OA) is a form of adeno-associated virus serotype 9 vector-based gene therapy delivering a transgene encoding human SMN proteins [9, 10]. Although clinical improvement in patients with SMA after these treatments has been reported [7‐10], changes in electrophysiological findings have rarely been reported. Compound muscle action potential (CMAP) amplitudes in nerve conduction study (NCS) of patients with SMA type I treated with nusinersen or OA have been reported to increase over time compared to nontreated patients [7, 9]. However, changes in needle electromyography (EMG) findings after these treatments have not yet been reported. In this paper, we report changes in NCS and needle EMG findings over time in two cases with SMA type I after treatment with nusinersen and OA.

In NCS, motor nerve responses were evoked under intravenous sodium thiopental-induced sleep, while the skin temperature was kept higher than 31 °C. In needle EMG, a needle was inserted into deltoid and rectus femoris muscles during intravenous sodium thiopental-induced sleep. After recording at rest, we awakened the patient and recorded the examination of voluntary contraction. The CMAP amplitudes of the median and tibial motor nerves at 2 months, 1 year, and 2 years of age are shown in Table 1. Although the CMAP amplitudes of both nerves were very low before nusinersen started, they increased over time after the treatment, and that of the tibial nerve reached the normal range 18 months after OA administration. The MCVs of these nerves were always in the normal range. The F-wave evaluation was not performed because she was sedated. The needle EMG showed fibrillation potentials at rest in both the deltoid and rectus femoris muscles at 2 months, 1 year, and 2 years of age. During voluntary weak contractions, the amplitudes of motor unit potentials (MUPs) were mostly around 0.5 mV at 2 months of age; however, the MUPs at 1 year and 2 years of age showed high amplitudes that reached 7–8 mV (Fig. 1a).

NCS and needle EMG were performed in the same manner as in patient 1. The CMAP amplitudes of the median and tibial nerves at 6 months, 1 year, and 2 years of age are shown in Table 1. The MCVs of these nerves were always in the normal range. The F-wave evaluation was not done because she was sedated. Needle EMG revealed her fibrillation potential at rest in both the deltoid and rectus femoris muscles at 6 months and 1 year of age. During voluntary weak contractions, the amplitudes of the MUPs were mostly around 0.5 mV at 6 months; however, the MUPs at 1 year revealed a high amplitude that reached 6–7 mV (Fig. 1b).

Informed consent was obtained from the parents of each patient for the use of their child’s data.

We report changes in NCS and needle EMG findings over time in two cases with SMA type I treated with nusinersen and OA in this paper. NCS in patients with SMA typically shows decreased CMAP amplitudes reflecting chronic motor axonal degeneration, and CMAP amplitudes correlated with clinical severity and age [2, 11, 12]. The peroneal CMAP amplitude was reported to increase in infantile-onset SMA treated with nusinersen compared to the nontreated SMA [7]; the same has been reported for the ulnar CMAP in SMA treated with OA [9]. In our patients, the CMAP amplitudes of the median and tibial motor nerves were significantly lower than those of healthy children of the same age before the treatment [13]. However, they increased over time after the treatment and some of them reached the normal range. Previous papers have reported that patients with SMA gain better motor development, the earlier they are treated [7], and that patients with symptomatic SMA who received nusinersen showed a significant increase in CMAP and motor unit number estimation was more increased in patients with shorter disease durations [14]. The treatment was initiated earlier in patient 1 (2 months old) than in patient 2 (6 months old), so patient 1 probably had better acquisition of motor development and more elevated CMAP amplitude of the tibial nerve. In patient 1, the CMAP amplitude of the tibial nerve had better recovery compared to the median nerve. However, clinical improvement was greater in the upper limbs, so the difference in the degree of improvement in CMAP between upper and lower limbs may not necessarily reflect the degree of clinical improvement.

In contrast, the changes in needle EMG findings after these therapeutic interventions have not yet been reported. Needle EMG in patients with SMA reveals abnormal spontaneous activity, including fibrillation potentials at rest reflecting active denervation and long durations and high amplitudes of MUPs during voluntary contraction reflecting chronic compensatory changes of reinnervation. However, in severely affected patients like SMA type I, MUPs may reveal reduced amplitudes and durations because of severe active denervation [2]. McDonald described that MUPs were lower in infants with amplitudes ranging from 150 µV to approximately 2000 µV, and generally MUPs more than 1000 µV in 0–3-year-old children are rare [15]. In our patients, needle EMG after the treatment showed high-amplitude MUPs that reached 6–8 mV during voluntary weak contractions, which were not seen before the treatment. It is known that high-amplitude MUPs are seen when nerve reinnervation occurs after denervation [16]; so, it is suggested that peripheral nerve reinnervation occurs after these treatments, but it does not necessarily prove recovery of motor neurons. In fact, fibrillation potentials suggestive of denervation were still seen in both patients. There are no reports on the changes in needle EMG findings in patients with SMA after the treatment; so, we need to follow how these reinnervation and denervation findings change. The accumulation of these findings will be important for evaluating the effectiveness of treatment for SMA in the future.