Nusinersen Demonstrates Durability and Safety in SMA in 5-Year NURTURE Data

Newly published results of the open-label, phase 2 NURTURE study (NCT02386553) in the journal Muscle & Nerve suggest a clear demonstration of early treatment with nusinersen (Spinraza; Biogen) in pediatric patients with spinal muscular atrophy (SMA).¹ Notably, over the roughly 5-year treatment period, the investigators pointed to signs of a durable treatment effect and a favorable safety profile among individuals with 2 and 3 copies of the SMN2 gene (2 copies, n = 15; 3 copies, n = 10) who were treated while still asymptomatic.

All told, those with 3 copies of SMN2 achieved all World Health Organization (WHO) motor milestones—and all but a single milestone in 1 child was met within normal developmental timeframes. Those with 2 copies of SMN2 achieved sitting without support at a rate of 100%, with those rates being 93% (n = 14) for walking with assistance, and 86% for walking on their own (n = 13).

At this analysis’s cut off, no child had discontinued the study or the treatment, and all were alive (median age at last visit, 4.9 years; range, 3.8-5.5). Additionally, no additional children were reported to use respiratory intervention since the prior data cut of March 29, 2019 (this publication’s data cut was indicated as February 15, 2021).²

Study investigator Thomas Crawford, MD, codirector of the Muscular Dystrophy Association Clinic and a professor of neurology at Johns Hopkins Medicine, and colleagues indicated that the inclusion and exclusion criteria, as well as the baseline characteristics, should be carefully considered when interpreting these data (SIDEBAR).

“It is particularly important to highlight the differences in baseline characteristics in the NURTURE subgroups and SPR1NT, as this is one of the key reasons that outcomes in this and other trials cannot be compared,” Crawford et al wrote. “For example, while baseline CHOP-INTEND values were higher in NURTURE subgroups 1 and 2 versus all NURTURE children with 2 SMN2 copies, SPR1NT baseline CHOP-INTEND values were intermediate between these groups. Similarly, CMAP [compound muscle action potential] amplitude baseline values in SPR1NT children with 2 SMN2 copies were higher than those in any NURTURE groups. Thus, while our analyses highlight the importance of the different baseline characteristics for trial outcomes, they do not inform on comparative efficacy of these 2 agents.”

SIDEBAR

Key Inclusion Criteria

Age ≤6 weeks at first dose.
Genetic documentation of 5q SMA homozygous gene deletion or mutation or compound heterozygous mutation.
Genetic documentation of 2 or 3 copies of SMN2.
Ulnar compound muscle action potential (CMAP) ≥1 mV at baseline.
Gestational age of 37-42 weeks for singleton births; gestational age of 34-42 weeks for twins.
Meet additional study related criteria.

Key Exclusion Criteria

Hypoxemia (oxygen saturation <96% awake or asleep without any supplemental oxygen or respiratory support).
Any clinical signs or symptoms at Screening or immediately prior to the first dosing (Day 1) that are, in the opinion of the Investigator, strongly suggestive of SMA.
Clinically significant abnormalities in hematology or clinical chemistry parameters.
Treatment with an investigational drug given for the treatment of SMA biological agent, or device. Any history of gene therapy, prior antisense oligonucleotide (ASO) treatment, or cell transplantation.
Meet additional study related criteria.

Note: Other protocol-defined inclusion or exclusion criteria may apply.

“The NURTURE study was 1 of the major 3 studies from Biogen that were used to establish the benefit of nusinersen in the treatment of spinal muscular atrophy,” Crawford told CGTLive^™. “The first 2 were in babies and children, and they are the ones that showed the extraordinary benefit. Babies were rescued from death. My previous career was basically as a hospice doctor, and I went into caring for kids with SMA here at Hopkins and personally held 72 babies that died of this disease. And with the approval of nusinersen, basically, that's not happening anymore. Those babies are being rescued.”

On mean Hammersmith Functional Motor Scale Expanded (HFMSE) total scores, patients showed continued improvement over the course of the study. The predicted mean slopes in HFMSE total score over time, measured as the point change per year, were similar between the 14 patients with 2 SMN2 copies (+6.30; 95% CI, 4.70–7.90) and the 10 patients with 3 SMN2 copies (+6.99; 95% CI, 4.98-9.00). Of note, these predicted mean total score and slopes of improvement over time were higher in children in NURTURE with either 2 or 3 SMN2 copies compared with those from the 34 participants in the phase 3 CHERISH study (NCT02292537) with symptomatic later-onset SMA who were treated with nusinersen (+5.22; 95% CI, 4.51-5.94) or the 24 treated with sham procedure (+0.138; 95% CI, −0.79 to 1.06).

Since the prior publication of data,² an additional 2 children with 2 SMN2 copies achieved the maximum Children's Hospital of Philadelphia Infant Test of Neuromuscular Disorders (CHOP INTEND) score at age 3.8 years (Study Day 1375) and 4.8 years (Study Day 1755). All told, 22 of the study children (88%; 2 SMN2 copies, n = 12; 3 SMN2 copies, n = 10) achieved the maximum CHOP INTEND score. Mean CHOP INTEND scores increased steadily from baseline prior to stabilization near the maximum score, and mean scores were higher throughout the study and at the most recent assessment in all NURTURE participants than in ENDEAR study of participants with symptomatic infantile-onset SMA—though Crawford et al noted that scores did increase in nusinersen-treated ENDEAR participants.

Additionally, in NURTURE, 11 children (2 SMN2 copies, n = 5; 3 SMN2 copies, n = 6) had at least 1 valid 6-Minute Walk Test (6MWT) assessment, totaling distances of 150 m to 325 m and 238 m to 444 m, respectively, indicating their capability to sustain independent walking.

The patient subgroups with 2 SMN2 copies, minimum baseline compound muscle action potential amplitude ≥2 mV without baseline areflexia (Subgroup 1, n = 8; Subgroup 2, n = 11) had better motor and nonmotor outcomes versus all children with 2 SMN2 copies. No children in Subgroup 1 received respiratory intervention or gastrostomy tube placement, compared with 4 and 5 children, respectively, among all NURTURE children with 2 SMN2 copies. A single child in this subgroup (13%; Participant 10) met the protocol-specified criteria for SMA symptoms compared with 8 of 15 (53%) among all NURTURE children with 2 SMN2 copies.

“Treatment following onset of early neurodegeneration before definitive symptoms creates an opportunity to examine previously obscured features,” Crawford and colleagues wrote. “The experience suggests new assessment opportunities (eg, assessing 2 SMN2 copy children with HFMSE and 6MWT) and previously unknown limitations.”

Crawford and colleagues additionally wrote that with survival beyond 2 years of age, some infant-specific items on CHOP-INTEND can no longer be reliably tested, and “progressive deficits in older children are also less well-assessed by infantile metrics due to confounding factors, for example, obesity, scoliosis, contractures.” They noted that ordinal scales from the pretreatment era may be insensitive to treatment effects and limitations—specifically, subtle features such as fasciculations, tremor, and muscle tone might be detected by patient-reported assessments or physical exam, rather than these scales.

As for safety, no new concerns were identified with the 2 additional years of follow-up from this analysis. No adverse events (AEs) were deemed to be study-drug related, with 10 participants (40%) reporting an AE that was considered possibly related to study drug. All of these AEs resolved despite continued treatment, save for one case of proteinuria in 1 child and a case of clonus in another. In total, 12 patients (48%) had 1 or more serious AEs, though none were considered by investigators to be study-drug related.

There were no discontinuations of treatment or withdrawals because of an AE, and the lumbar puncture procedure—which required sedation for some participants—was deemed generally well tolerated. No cases of meningitis, hydrocephalus, or renal/liver failure were reported, and additionally, no clinically relevant trends related to nusinersen in hematology, blood chemistry, urinalysis, coagulation, vital signs, or electrocardiograms were observed by Crawford et al.

NURTURE’s limitations included participant cooperation and ability to adhere to test instructions, thus making HMFSE and 6MWT likely to be more reliable in older children; additionally, COVID-19 pandemic-related constraints resulted in treatment delays of up to 2 months—though the investigators noted that “rapid restoration of nusinersen cerebrospinal fluid levels would be expected following administration of subsequent doses at regularly scheduled visits.”