A total of 41 patients with SMA (24 females, 58.5%) were included, 33 (80.1%) SMA type 1, seven (17.1%) SMA type 2, and one pre-symptomatic. Thirty-six patients (87.8%) were under previous treatment with nusinersen, and one patient (2.4%) used risdiplam before GT. The time between the last nusinersen injection and GT ranged between 140 to 11 days (mean 76 days). Ten patients (24.4%) continued treatment with nusinersen after infusion of GT due to family and physician expectancy that combined therapy could bring additional benefits.
The mean age at receiving the first treatment was 7.9 (SD = ± 5.5) months (Table 1). The mean age at GT dosing was 18 (SD = ± 6.4) months. Only five patients (12.2%) were dosed before one year, and 31 (75.6%) were dosed up to two years of age. The mean weight at dosing was 10 kg (SD = ± 2.4).
Table 1
Variables | Overall, N = 41 |
SMA type | |
1 | 33 (80.5%) |
2 | 7 (17.1%) |
Pre-symptomatic | 1 (2.4%) |
Number of SMN2 copies | |
2 | 30 (73.2%) |
3 | 11 (26.8%) |
Gender (female) | 24 (58.5%) |
Previous Nusinersen | 36 (87.8%) |
Continued Nusinersen after GT | 10 (24.4%) |
Weight (Kg) | 10.0 (2.4) |
Age at first treatment (months) | 7.9 (5.5) |
Mean age at gene therapy (month) | 18.0 (6.4) |
Age at gene therapy (categories) | |
< 1 year | 5 (12.2%) |
1–2 years | 31 (75.6%) |
> 2 years | 5 (12.2%) |
Total time on prednisolone (days) | 113.7 (81.5) |
Starting dose of prednisolone | |
1mg/Kg | 27 (65.9%) |
2mg/Kg | 14 (34.1%) |
Reported reasons for gene therapy | |
Parental demand, mainly single dose therapy First-line therapy Motor stagnation, seek for new motor milestones No improve in bulbar function | 19 (46.3%) 4 (9.8%) 12 (29.3%) 6 (14.6%) |
Data presented as mean (standard deviation) or n (%).SMN2: survival motor neuron gene 2. |
Forty patients were followed for at least 6 months, and 24 patients (58.5%) for at least twelve months after GT. Baseline characteristics of the cohort and reported reasons for opting for GT are presented in Table 1. One patient died two months after GT due to septic shock. Another patient had a prolonged hospitalization due to thrombotic microangiopathy (detailed later).
Chop-intend And Motor Milestones
At the baseline of GT injection, the CHOP-INTEND scores ranged between 9 and 59 (mean = 33.1, SD = ± 14.8). Mean CHOP-INTEND was 46.1 after 6 months (mean increase of 13 points) and 55.1 after 12 months of GT (mean increase of 22 points from baseline) (Table 2, Fig. 1A). The increase in CHOP-INTEND at six months differed significantly between SMA types 1 and 2 (p = 0.045, Fig. 1B). The linear mixed model showed that an increase in CHOP-INTEND (at least four points) was associated with higher baseline CHOP-INTEND scores (p < 0.001), longer follow-up time (p < 0.001), SMA type 2 (p = 0.016) and younger age at dosing (p = 0.016). No difference in CHOP-INTEND increase at six-month follow-up was found between groups according to previous treatment with nusinersen, SMN2 copy number, gender, or age categories (Fig. 1C).
Table 2
Functional and motor baseline characteristics and after treatment, in overall group and according to SMA type.
Variables | Overall, N = 41 | SMA type 1 N = 33 | SMA type 2 N = 7 |
Two SMN2 copies Three SMN2 copies | 30 (73.2%) 11 (26.8%) | 29 (87.9%) 4 (12.1%) | 0 (0.0%) 7 (100%) |
Age at disease onset (months) | 4.2 (± 3.5) | 3.0 (± 1.8) | 10.2 (3.7) |
Age at first treatment (months) | 7.9 (± 5.5) | 6.3 (± 3.5) | 16.1 (± 5.0) |
Age at gene therapy (months) | 18.0 (± 6.4) | 16.5 (± 5.3) | 25.1 (± 7.2) |
CHOP-INTEND baseline at GT | 33.1 (± 14.8) | 31.2 (± 13.9) | 40.5 (± 13.0) |
CHOP-INTEND 6 months | 46.1 (± 13.4) | 43.9 (± 12.6) | 58.5 (± 11.0) |
CHOP-INTEND 12 months | 55.1 (± 8.1) | 52.3 (± 8.0) | 60.5 (± 7.0) |
Mean change in HINE-2 | 3.2 (± 2.3) | 2.8 (± 1.9) | 5.1 (± 3.4) |
CVS before GT | 18 (43.9%) | 18 (54.5%) | 0 (0.0%) |
CVS after 6 months of GT | 12 (29.3%) | 12 (36.4%) | 0 (0.0%) |
Feeding route (Oral) baseline | 15 (36.6%) | 15 (45.4%) | 7 (100%) |
Feeding route (Oral) after GT | 18 (46.2%) | 18 (54.5%) | 7 (100%) |
Data presented as mean (standard deviation) or number (%). SMN2: survival motor neuron gene 2; GT: gene therapy; CHOP-INTEND: Children's Hospital of Philadelphia Infant Test of Neuromuscular Disorders; HINE-2: section 2 of the Hammersmith Infant Neurological Examination ; CVS: continuous ventilator support; |
The group of patients who continued to treat with nusinersen after receiving GT had no significant difference in CHOP-INTEND increase at 6 months compared to the group that did not continue with nusinersen (p = 0.949, Fig. 1D). Analysis of factors associated with CHOP-INTEND increase was performed separately in the SMA type 1 group and did not bring additional findings.
Thirty-seven patients (30 SMA type 1 and seven SMA type 2) were evaluated according to the motor milestone acquired at the end of the follow-up, and 31 (83.7%) acquired at least one new motor milestone according to HINE-2, and 27 (70.3%) according to WHO criteria. Among SMA type 1 patients, the mean change in HINE-2 was 2.8 points (± 1.9) (Table 2), with four patients (13.3%) reaching head control and three (10%) sitting with support. According to WHO criteria, fourteen patients (46.6%) reached the ability to sit alone, three (10%) were able to stand with support, and one (3,3%) reached the ability to walk with support after GT (Fig. 2A). SMA type 2 patients presented a mean change in HINE-2 of 5.1 points (± 3.4) (Table 2), two patients (28.6%) reached the ability to walk with support, and three (42.8%) acquired the ability to walk alone according to WHO motor milestones (Fig. 2B). Linear regression showed an association between an increase in HINE-2 and three SMN2 copies (p = 0.015). There was no association between an increase in motor milestones and previous treatment with nusinersen, gender, or age at dosing. Linear correlation was found between gain in motor milestones measured by HINE-2 and baseline CHOP-INTEND (Spearman’s correlation coefficient = 0.408, p = 0.016) (Fig. 3).
Respiratory Support And Bulbar Function
Among the 41 patients, twelve (29.3%) did not use any ventilatory support, eleven (26.8%) were in noninvasive ventilation (NIV < 16 hours/day), and 18 (43.9%) were in continuous ventilatory support (CVS > 16 hours/day) at the baseline (GT injection). In the last evaluation, among those on CVS at baseline, the daily time of ventilatory support was reduced in six (33.3%), and they were using NIV < 16 hours/day. Another four patients (22.2%) presented a reduction in the daily ventilation time but were still on CVS, and another eight (44.4%) presented no change in the daily ventilator support time.
SMA type 2 patients did not present moderate or severe dysphagia and did not require any nutritional support (gastrostomy) at baseline and after GT. Among the 34 SMA type 1 patients, only 15 (44.1%) were fed orally, and the other 19 (55.9%) exclusively used gastrostomy at the baseline. In the last evaluation, the number of children who used the oral feeding route increased to 18 (52.9%). Among the other children who still needed a gastrostomy, seven (36.8%) showed improvement in swallowing of saliva and reduction in procedures for sialorrhea.
Safety Data And Corticosteroids Use
One patient presented signs of anaphylaxis (hypotension and tachycardia) during the GT infusion. The infusion was then stopped, and the condition was reversed with volume, an intravenous corticosteroid, an antihistamine, and adrenaline. After complete resolution, the infusion was restarted less than 2 hours later. Thirty patients (73,1%) had a fever, pyrexia, or anorexia in the first week after infusion. Nausea or vomiting occurred in 20 patients (48,8%). However, all were using some gastric protector or anti-emetic, and this event was more common in the group of patients using the highest dose of corticosteroid.
Liver transaminases elevation at least two times to the upper limit of normal value occurred in 29 (70.7%) patients, and in four (9.7%), the elevation was greater than 20 times the upper limit of normal value (maximum elevation of ALT 2,200 U/L in the first week after dosing). No patient showed changes in liver function (elevated bilirubin or prolonged prothrombin time). In the first three months of follow-up, consecutive elevations of transaminases (2 times of upper limit of normal value) occurred 8.2 weeks (± 4.3) on average, with an initial peak in the first week and a later peak ranging from the third to sixth week after infusion (Fig. 4). Weight, age, or previous use of nusinersen did not correlate with consecutive elevations of transaminases in the first months. They starting dose of prednisolone at 2mg/kg/day correlated with a lower consecutive mean elevation of transaminases in the first months (5.3 weeks ± 4.3, p = 0.003) concerning the group of patients who started prednisolone at the usual dose of 1mg/kg/day (9.4 weeks ± 3.2), detailed in Fig. 4.
The median time of corticosteroid use was 113.7 days (± 81.5), ranging from 60 to 450 days. Only twelve patients (29.2%) used corticosteroids for 60 days recommended by the label. Twenty patients (48.8%) used corticosteroids beyond the initial 3-month period due to a late increase in transaminases when trying to wean off prednisolone. Total time of prednisolone use was associated with age, with patients younger than one year using it for a median period of 62.2 days (± 8.6) and older patients having longer use (p = 0.021, Fig. 5). Gender, previous use of nusinersen, weight, or even initiation of a higher dose of corticosteroids did not correlate with a shorter duration of steroid use.
Thrombocytopenia with platelet levels less than or equal to 120,000/microL occurred in thirteen (31.7%) patients, moderate thrombocytopenia (< 100,000/microL) in eleven (26.2%) patients, and severe thrombocytopenia (< 50,000/microL) in three (7.3%) patients. All but one of the cases occurred within the first week and were resolved entirely within the second week after treatment. In this cohort, thrombocytopenia correlated with greater weight (OR = 2.22, CI = 1.30–4.77, p = 0.014) but did not correlate with previous use of nusinersen (p = 0.223).
One case of thrombotic microangiopathy occurred in a female patient with SMA type 1, dosed at 15 months of age and weighing 8.2 kg. This patient had an advanced stage of disease (baseline CHOP-INTEND of 12 points) and previous treatment with nusinersen, receiving the last dose 58 days before gene therapy. The subject was tracheostomized and used invasive ventilation 24 hours/day. The patient had had recurrent respiratory infections and had been hospitalized for months before the gene therapy was administered. Thrombotic microangiopathy was treated with plasma transfusion and dialysis (the patient did not tolerate plasmapheresis for hemodynamic reasons). The condition resolved after 30 days with residual renal dysfunction but no need for permanent renal replacement therapy.
A slight increase in troponin I to baseline values occurred in three (7.3%) patients without any repercussions on cardiac function. In these cases, it did not lead to an increase in prednisolone dose. Electrocardiogram and echocardiogram were normal in those patients who underwent this evaluation, as it is not a test routinely performed after therapy.
One death occurred approximately two months after GT dosing, in a 2-year-old patient, with severe impairment in baseline motor function (CHOP-INTEND of 09 points), under CVS and who was in the final period of corticosteroid use (0.2 mg/kg/day). The patient had diarrhea for two days, and at hospital admission, he showed signs of septic shock, necrosis of intestinal loops, and an E. coli agent was isolated in the blood culture. The patient died after three days of hospitalization at the intensive care unit.