KINGSTON, Mass. (
) -- While there seems to be a debate related to the dystrophin being produced by eteplirsen and whether or not it has proven to be correlated with a clinical benefit, I believe the FDA guidance on accelerated approval requires dystrophin production only be "reasonably likely to predict" clinical benefit.
All of the arguments related to the size of the eteplirsen study, the statistical analysis or the debate about the natural history of the disease ignore the data. Eteplirsen is producing dystrophin in all patients after 48 weeks of treatment and has led to a stabilization of all ambulatory patients in the study over 74 weeks in the early treatment arm, and now over 38 weeks in the placebo patients. Importantly, the placebo patients were on a steep decline before eteplirsen produced dystrophin in their muscles.
To those who do not have to live with the personal devastation of Duchenne muscular dystrophy (DMD), as my family has with my son Jett, I'm here to tell you we cannot wait any longer for drugs like eteplirsen to be approved. Those not living with DMD can afford to debate the need for a more precise and definitive correlation of dystrophin and stabilization of the disease. My son cannot. The fact that eteplirsen has proven to be safe in all of these patients, including those who have now received high doses of the drug over a 74-week period, puts an exclamation point on the need for this drug today. DMD is a rapid, progressive and irreversible disorder, which makes the risk-benefit ratio of eteplirsen extremely favorable.
"The Correlation Between Dystrophin Levels And Clinical Benefit Is Poor"
-- Cowen research report, April 22, 2013.
I find it interesting that this "correlation" concern is gaining traction now,
eteplirsen has demonstrated the ability to produce dystrophin at statistically significant levels. Eteplirsen was designed to produce dystrophin in DMD patients, and the data show the drug is doing exactly that. The FDA's regulations for subpart H accelerated approval don't require a surrogate endpoint to show a "proven and confirmed correlation," only that surrogate is "reasonably likely to predict" a clinical benefit. What does "reasonably likely" mean? Does it mean a surrogate endpoint has to mirror or correlate 100%? No, it means what it says: "The drug product has an effect on a surrogate endpoint that is reasonably likely, based on epidemiologic, therapeutic, pathophysiologic, or other evidence, to predict clinical benefit or on the basis of an effect on a clinical endpoint other than survival or irreversible morbidity."
In the eteplirsen study conducted by
, all 12 of the patients showed positive dystrophin fibers in their biopsies. All ambulatory participants demonstrated stabilization of their walking in the six-minute walk test after dystrophin was produced by the drug. The two twin boys enrolled in the study who rapidly lost ambulation before dystrophin was produced have shown stability on pulmonary function and muscle strength. This means all patients demonstrate positive effects associated with eteplirsen after dystrophin is produced. Not some patients, all patients.
Is it reasonable to believe there was variability in quantifying all 12 of the muscle biopsies? Is it reasonable to believe there is a sampling error in all 12 of the muscle biopsies? Even though there might be variation in the amount of dystrophin in the individual patient, the data shows that all dystrophin increased over time. Despite the limitations of dystrophin variability, it is quite encouraging that all the boys displayed dystrophin production. This bolsters confidence that eteplirsen is working. Data presented at the Muscular Dystrophy Association conference this week show stabilization in the six-minute walk test along with an increase in dystrophin positive fibers, around the same time.