Roche shares SMA data ahead of scrap with Biogen, Novartis

Roche has presented more of the data that could enable it to challenge Biogen and Novartis for the spinal muscular atrophy (SMA) market. The readout shows 29% of infants with the most severe form of the disease could briefly sit without support after taking risdiplam, adding to evidence that Roche has a viable challenger for the market.

Competition in the SMA market has intensified significantly in recent years, with Biogen’s Spinraza and Novartis’ Zolgensma providing treatment options for a previously underserved population. The availability of a blockbuster antisense oligonucleotide and emerging gene therapy challenger have raised the bar for success in SMA, meaning Roche’s small-molecule SMN2 splicing modifier needs to excel if it is to live up to annual sales forecasts that run as high as $2 billion. 

Roche shared more clinical data that will shape whether risdiplam can live up to those expectations Tuesday. The data come from the second part of a pivotal trial, which Roche revealed had met its primary endpoint in January.

The latest update shows the primary endpoint success was achieved after 12 of the 41 infants in the study were able to sit without support for five seconds after 12 months of daily dosing with Roche’s oral drug. Roche’s trial lacked a control arm, but, based on natural history data, the Swiss pharma wouldn’t have expected any of the Type 1 SMA patients enrolled in the trial to sit without support in the absence of treatment.

Roche also reported that 18 of the infants could hold their head upright and 13 were able to roll to the side. Two of the infants were able to stand with support. All bar four of the patients experienced a four-point or greater increase on the Children’s Hospital of Philadelphia Infant Test of Neuromuscular Disorders (CHOP INTEND) scale. The median increase was 20 points. In the absence of treatment, the CHOP INTEND scores of infants with Type 1 SMA decline over time. 

The improvements may be translating into reduced mortality. Three of the patients enrolled in the trial had died as of the data cutoff, but the median age of the other 38 infants was 20.7 months and the vast majority of participants were able to feed orally. Feeding support is often necessary in Type 1 SMA patients beyond 12 months of age.

Almost one-third of patients suffered bad cases of pneumonia, making it by far the most common severe adverse event, but the trial was free from new safety signals. The lack of safety red flags puts Roche on track to win FDA approval for risdiplam in August, barring any late surprises. 

What slice of the SMA market Roche will claim remains uncertain, though. The designs of the studies of risdiplam, Spinraza and Zolgensma differed in key regards, making cross-trial comparisons, which can be unreliable at the best of times, particularly problematic. 

For example, the data shared by Roche today come from a trial of patients who developed symptoms attributable to Type 1 SMA by three months of age. Biogen and Novartis, in contrast, won approval on the basis of studies that enrolled subjects that showed symptoms of infantile-onset SMA before six months of age. The clinical trials also used different primary endpoints.