Risdiplam (1) – Evrysdi®

• Clinical trials
  • The FIREFISH trial is an open-label, single-arm trial that included patients with genetic evidence of 5q-associated SMA and an onset of clinical signs or symptoms from 28 days of age up to ≤ 3 months, aged between 1 month (28 days) and ≤ 7 months (210 days) at the time of enrolment, and who have two copies of the SMN2 gene.
  • The ENDEAR study is a double-blind RCT that enrolled patients with genetically confirmed 5q-associated SMA who were aged ≤ 7 months at the start of the study, with an age at symptom onset of ≤ 6 months, and who had 2 copies of the SMN2 gene.
  • The second part of the SUNFISH study is a double-blind RCT that included non-ambulatory patients with a genetically confirmed diagnosis of 5q-associated SMA, aged between 2 and 25 years at screening, and presenting with clinical symptoms of type 2 or 3 SMA.

a) Patients aged 2 months or older with 5q-associated spinal muscular atrophy (5q-SMA) type 1

• Hint for a non-quantifiable additional benefit
• In summary, for risdiplam in the treatment of patients aged 2 months and older with 5q-SMA type 1, taking into account the , an expected advantage of oral administration over intrathecal injection, and taking particular account of the severity of the condition, there is evidence of a hint of an additional benefit compared with nusinersen; however, due to the limited available evidence, the extent of this benefit is non-quantifiable.
• Morbidity – death or permanent ventilation
  • For the composite endpoint of death or long-term ventilation, as well as for the individual component of long-term ventilation, a clear, statistically significant advantage in favour of risdiplam compared with nusinersen is evident based on the comparison of individual arms from the FIREFISH and ENDEAR studies.
  • For the individual component of ‘permanent ventilation’ within the combined endpoint, a clear, statistically significant advantage in favour of risdiplam compared with nusinersen is evident based on the ‘naive’ comparison of individual arms of the FIREFISH and ENDEAR studies.
  • Table 1: Comparison of individual arms of different studies (unadjusted): Risdiplam (FIREFISH) vs. Nusinersen (SHINE-ENDEAR)
• Overall assessment
  • The ‘naive’ comparison presented is subject to very high levels of uncertainty.
  • The ‘naive’ comparison of individual arms of the FIREFISH and ENDEAR studies, which compared risdiplam and nusinersen, shows a clear, statistically significant advantage for risdiplam in terms of the endpoints of death or long-term ventilation, as well as for the individual component of long-term ventilation.
  • However, based on this ‘naive’ comparison, it cannot be reliably ruled out that these effects arise solely from systematic bias due to confounding factors, as it is reasonable to assume, particularly in light of the exclusion criteria for the FIREFISH and ENDEAR studies, that the study population in the ENDEAR trial has a poorer prognosis with regard to respiratory events.
  • However, the observed differences in the time to permanent ventilation and the combined endpoint of time to death or permanent ventilation suggest that risdiplam is at least not inferior to nusinersen.
  • Furthermore, it is assumed that oral administration of risdiplam offers a noticeable advantage over intrathecal administration of nusinersen, particularly in younger children.

b) Patients aged 2 months and over with 5q-SMA type 2

• The additional benefit is not proven
• In summary, based on the identified studies, an adjusted indirect comparison of risdiplam versus nusinersen for patients with SMA type 2 is not possible.

c1) Patients aged 2 months and older with 5q-SMA type 3, for whom intrathecal administration of nusinersen is an option

• An additional benefit is not proven
• The SUNFISH study is therefore not suitable for drawing conclusions regarding the additional benefit of risdiplam compared with the appropriate comparator therapy for this patient population with SMA type 3.

c2) Patients aged 2 months and over with 5q-SMA type 3 for whom intrathecal administration of nusinersen is not an option

• Hint for a non-quantifiable additional benefit
• In summary, for risdiplam in the treatment of patients aged 2 months and older with 5q-SMA type 3, for whom intrathecal administration of nusinersen is not an option, taking into account the advantage demonstrated in the SUNFISH study regarding motor function of the upper limbs (RULM), there is a hint of an additional benefit compared with the appropriate comparator therapy (BSC), the extent of which is non-quantifiable due to the considerable uncertainties in the study data used.
• mortality
  • In the patient population with SMA Type 3, no deaths occurred up to month 12.
• Morbidity – MFM-32 (gross and fine motor skills)
  • There was no statistically significant difference between the treatment groups in the mean change at month 12.
• Morbidity – Hammersmith Functional Motor Scale Expanded (HFMSE)
  • There was no statistically significant difference in the mean change between the treatment groups at month 12.
• Morbidity – Revised Upper Limb Module – RULM
  • For the mean change, there is a statistically significant advantage in favour of risdiplam + BSC compared with placebo + BSC.
  • The 95% confidence interval lies entirely above the non-significance threshold of 0.2. This is interpreted as a significant effect.
• Morbidity – Health status (EQ-5D VAS)
  • For the mean change at month 12, there is no statistically significant difference between the treatment groups.
• Side effects
  • No discontinuations due to adverse events occurred up to month 12.
  • For the specific AE ‘skin and subcutaneous tissue disorders’ (SOC, AEs), a statistically significant disadvantage was observed at month 12 compared with best standard care (BSC).
  • Overall, in the category of side effects, neither an advantage nor a disadvantage for risdiplam compared with BSC can be inferred.
• Overall assessment / Conclusion
  • For patients aged 2 months and over with 5q-SMA type 3, for whom intrathecal administration of nusinersen is not an option, the results from the SUNFISH study for patients with clinically diagnosed SMA type 3 are used.
  • No significant differences were observed between the treatment arms in terms of mortality and the MFM-32, HFMSE and EQ-5D VAS endpoints in the morbidity category.
  • For the RULM endpoint in the morbidity category, a statistically significant advantage was observed in favour of risdiplam compared with BSC, which is interpreted as a clinically relevant effect.
  • Health-related quality of life was not assessed.
  • In the ‘Side Effects’ category, the results for SUEs are not taken into account, as no analyses are available that exclude events attributable to the underlying condition.
  • The available data are subject to very high levels of uncertainty, as results for the overall population of patients aged 2 months and over with 5q-SMA type 3 from the SUNFISH study are provisionally used for the comparatively small group of patients aged 2 months and over with 5q-SMA type 3, for whom intrathecal administration of nusinersen is not an option.

d1) presymptomatic patients aged 2 months and over with 5q-SMA and up to three copies of the SMN2 gene

• An additional benefit is not proven
• No data are available for the assessment of the additional benefit of risdiplam compared with the appropriate comparator therapy, nusinersen, in presymptomatic patients aged 2 months and older with 5q-SMA and up to three copies of the SMN2 gene.

d2) presymptomatic patients aged 2 months and over with 5q-SMA and four copies of the SMN2 gene

• An additional benefit is not proven
• For the assessment of the additional benefit of risdiplam compared with the appropriate comparator therapy ‘treatment as medically indicated, with a choice of nusinersen or best standard care’ in presymptomatic patients aged 2 months and older with 5q-SMA and four copies of the SMN2 gene, no data are available.

Courtesy translation only, please refer to the German original.