A Study of valproic acid for patients with spinal muscular atrophy

Introduction

Spinal muscular atrophy (SMA) is an autosomal recessive neuromuscular disorder characterized by degeneration of the anterior horn cells of the spinal cord, resulting in progressive muscular atrophy, and weakness of the limbs and trunk, with the incidence reported to be approximately one in 6000–10 000 live births.1 SMA is classified into five groups; type 1 (Werdnig–Hoffman disease; severe form), type 2 (Dubowitz disease; intermediate form), type 3 (Kugelberg–Welander disease; mild form), type 4 (adult form) and type 0 (prenatal).1-3

The gene responsible for SMA is the survival motor neuron (SMN), which exists as two highly homologous copies within the SMA gene region on chromosome 5q11.2–13.3, namely SMN1 and SMN2.4 The disease is caused by loss of SMN1, with more than 95% of SMA patients showing a homozygous deletion or interruption in SMN1, resulting in a deficiency of the SMN protein.4-6 SMN1 and SMN2 are nearly identical, with the only difference being a single nucleotide change in the coding region, which shows that nucleotide +6 of exon 7 in SMN1 is C and that of SMN2 is T.

Although SMN1 and SMN2 encode the same protein because of a synonymous nucleotide change, SMN2 does not fully compensate for the loss or dysfunction of SMN1. As the C to T change in SMN2 at nucleotide position +6 in exon 7 induces exon skipping, SMN1 and SMN2 show different splicing patterns.7, 8 All SMN1-derived transcripts contain exon 7 and produce full-length SMN (FL-SMN), whereas the majority of SMN2-derived transcripts lack exon 7 (∆7-SMN).

Phenotypic variations in SMA are inversely correlated with SMN2 copy number, and a higher SMN2 copy number ameliorates the clinical phenotype.1, 9, 10 SMN2 might compensate for the loss of SMN1 by modifying disease severity through production of a small amount of full-length SMN protein. Thus, treatment strategies for SMA have focused on increased production of the SMN protein from SMN2.

Valproic acid (VPA) is a histone deacetylase (HDAC) inhibitor as well as an anticonvulsant used for treatment of epileptic patients, as it increases SMN levels in SMA patients through activation of SMN2 transcription and splicing correction of SMN2 exon 7.11, 12 Its effects as a therapeutic agent of SMA are expected.13-20 In the present study, we evaluated the efficacy of VPA in SMA patients.

Methods

The present study was carried out from January 2012 to March 2013.Seven consecutive Japanese SMA patients were recruited, of whom six were type 2 and one was type 3. The type 2 patients were as follows: case A, 34-year-old man; case B, 33-year-old woman; case C, 23-year-old man; case D, 30-year-old woman; case E, 2 years and 10-month-old girl; and case G, 15-year-old girl, whereas the type 3 patient was a 42-year-old man denoted as case F.

None of the participants possessed the SMN1 gene, and had three copies of SMN2 and the neuronal apoptosis inhibitory protein. All except for cases B and E used non-invasive ventilation at night. The demographic features of the patients are summarized in Table 1. All patients underwent physiotherapy, such as range of motion exercises of the extremities and respiration, before, during and after the study. The frequency and contents of physiotherapy differed among the patients, and were dependent on their situation including hospitalization, outpatient status and other factors.

VPA was given daily for 6 months to reach trough levels of 50–100 mg/dL, a dosing level typical of that used in epilepsy patients. L-carnitine was also given. We evaluated using the Modified Hammersmith Functional Motor Scale for SMA (MHFMS),21 and also examined respiratory function and carried out SMN transcript analysis using quantitative real-time polymerase chain reaction (qRT–PCR) measurements with peripheral white blood cell samples22 obtained from the patients before and 1, 3 and 6 months after starting VPA treatment. Blood samples were obtained from all patients in the daytime after fasting.

For SMN transcript analysis, we measured total-SMN, FL-SMN and Δ7-SMN transcript levels using qRT–PCR, with the latter two quantitated from the levels of the products encompassing SMN exons 7 and 8, and exons 5, 6 and 8, respectively. We used glyceraldehyde-3-phosphate dehydrogenase (GAPDH) as an endogenous reference gene, and the levels of SMN are expressed relative to those of GAPDH.22 The detailed methods utilized for qRT–PCR have been described.22 We also evaluated the ratio of FL-SMN to Δ7-SMN transcript (FL/Δ7-SMN). For respiratory function, we assessed vital capacity (VC), maximum insufflation capacity (MIC) and cough peak flow (CPF).23, 24 In addition, we also checked subjective symptoms, side-effects and bodyweight changes in each patient.

Results

Discussion

After a 6-month administration of VPA, many items related to respiratory function were improved in the participants of the present study. Our items used to evaluate respiratory function, such as MIC and CPF, were not utilized in previous reports. Increasing MIC and CPF is important for patients with neuromuscular diseases to maintain good respiratory condition.23 Our results showed that those improved values indicate the effectiveness of VPA administration to maintain a good respiratory condition even in adult SMA patients who show progression. Although Swobota reported improvements in maximum inspiratory pressures, forced vital capacity and forced expiratory volume in 1 s in patients aged over 5 years in an open label study,15 there were no changes in any results of pulmonary function testing carried out in a double-blind trial thereafter.17 Furthermore, no previous studies have reported that respiratory function was clearly improved with VPA administration, except for one that speculated that improved respiratory function might have been a result of growth and development.18

As a next step, our evaluation items of respiratory function, such as MIC and CPF, which have not been used in previous studies, should be evaluated as part of a placebo-controlled randomized trial to confirm the effects of VPA on respiratory condition in SMA patients.

In contrast, there was no relationship between VPA blood level and change in FL-SMN transcription products. Also, the level of VPA in blood in cases with improved MHFMS was less than an optimal level. Thus, improvement at the study end-point was not necessarily associated with VPA blood level or FL-SMN transcription product.

Past reports of VPA administration in SMA patients are summarized in Table 4, with most of those cases being SMA type 2 and 3.13-20 The VPA dose in each of those reports is assumed to have been in accordance with the dose or blood level when used as an anticonvulsant. In four reports, carnitine administration was combined.15, 17-19 Two of those were designed as a double-blind study, whereas the others were open label.17, 20 In six reports, either SMN transcription level or SMN protein level was evaluated.14-18, 20 As an evaluation of motor function, the original Hammersmith Functional Motor Scale (HFMS) or MHFMS was used in four reports,15, 17-19 whereas a muscle strength test was also used in five reports.13, 17-20 One study found increased levels of SMN mRNA in association with VPA administration,14 and another noted increased levels of SMN protein with VPA administration.16 However, in three reports, there was no evident change in SMN transcription level with VPA administration15, 17, 18 In the report by Weihl, motor function efficacy was noted in SMA type 3 and 4 patients,13 whereas Darbar reported that HFMS improvement in SMA type 3 was not observed.19 Kissel reported no statistically significant differences regarding changes in maximum voluntary isometric contraction in ambulatory SMA adults.20 Also, Swoboda reported that children aged under 5 years,15 furthermore, those aged 2–3 years with SMA type 2 showed MHFMS improvement with VPA administration.17

So the effects of VPA on SMA patients are controversial. Based on the present results, we expect that respiratory function in adult patients with progression, as well as motor function in younger children, has a possibility to improve after VPA administration. However, our open study was limited by the number of cases analyzed, and establishment of a control was difficult. We cannot conclude that the change in end-point after VPA administration is exclusively related to VPA administration. In particular, in the MHFMS of case E, growth development could have influenced our evaluation of clinical manifestations. The effects of growth development on motor functional evaluation should be evaluated in a placebo-controlled randomized trial.

As for the effects of VPA on SMN, promotion of SMN2 gene transcription by activation of the SMN2 gene promoter (production increase of full length type SMN2 mRNA and Δ7-SMN2 mRNA), splice progress of the SMN2 gene exon 7 by gene activation to encode a splicing related protein and a combination of these two mechanisms have been considered.11, 12 An increase in FL- and Δ7-SMN transcription product quantity is expected to occur with VPA administration. Furthermore, VPA is a multifunctional drug that is expected to have a neuroprotective effect.25 Therefore, it is also speculated that VPA blood level, FL-SMN transcription product quantity and improvement in outcome are not necessarily linked.

Regarding the change in quantity of the SMN transcription product after VPA administration, we considered the effects of fluctuations in the system of measurement. Whether an increase in SMN protein in peripheral blood leukocytes reflects an increase in that in ventral horn cells remains unknown. It is also not clear if an increase in SMN protein in the ventral horn cells is directly associated with clinical manifestation improvement. In addition, if an imperceptible change in motor function occurs, it might not be possible to detect the difference using the method of evaluation utilized in the present study. Thus, subtle changes in clinical signs and symptoms might not be detected by the present evaluation method.

It was also difficult to evaluate the effects of physiotherapy on motor and respiratory functions in a comprehensive manner because of variations in each patient. In the present cases, uniform physiotherapy was not possible because of functional differences among our patients, the therapeutic environment (hospitalization, outpatient status and so on) and ethical reasons. Thus, variations in physiotherapy should be minimized to better evaluate the effectiveness of VPA in future studies.

We found no serious side-effects caused by VPA administration. However, many of our patients gained bodyweight as compared with pretreatment, which was induced by VPA. Excessive weight gain has negative effects on motor and respiratory conditions in such patients, thus careful administration of VPA is required. As for carnitine administration, we should recognize side-effects including a precordial sense of incongruity and a complaint of belching, which have not been reported in previous reports.

VPA is expected to show good effects as a therapeutic drug for SMA in younger patients for motor functional improvement and even in adult patients for respiratory improvement. To fully elucidate its effectiveness and efficacy in SMA patients, development of an evaluation method to better determine minimal changes in clinical manifestations including respiratory function items, such as MIC and CPF, which were not used in previous reports, as well as introduction of a new biomarker that can be easily evaluated and is able to differentiate responders to VPA treatment from non-responders are required. Although improvements in MHFMS in the young girl and respiratory function improvements in the older subjects were observed in the present study, our results are difficult to interpret because of the open label nature and small scale. The effects of growth development on motor functional evaluation in child cases, respiratory function using MIC and CPF, and efficacy of VPA for SMA should be evaluated using a placebo-controlled randomized trial protocol.