FDA, at best, is inconsistent in their interpretation and guidance. See SRPT's BD addendum:
Jenkins goes on to say: “As always, compromises were not made in our review standards as we considered the applications as efficiently as possible,” Jenkins asserted. “In short, each of these therapies had to demonstrate that it was safe and effective before being approved.
in other words, no flexibility.
Oneway, better go review the data in the WMS presentation, it looks like SRPT is on solid ground wrt a possible imbalance between NH and TX groups. sorry.
1. standard of proof is lower than for full approval - can be approved based on evidence that demonstrates the drug is "reasonably likely to predict clinical benefit." That evidence is what SRPT has to present to Adcomm next week.
2. continued marketing approval is subject to the normal full approval standards. That's why a p-3 (in this case called a "confirmatory study") is still required, should be ongoing at time of accelerated approval, completed timely, and results must confirm drug efficacy and safety.
3. when the confirmatory study is concluded, company must present results to FDA to receive full approval.
Accelerated Approval really is an FDA "thing." You should investigate when it is appropriate and what it entails before taking an uninformed position on Eteplirsin.
it (virtually) rules out chance, but it doesn't rule out population imbalance. fortunately, in SRPT's case, the TX group appears balanced with the NH control.
apparently some other people don't know that FDA is on record as having advised SRPT that they would consider a comparison of SRPT's extended trial results vs Natural History, for purposes of accelerated approval...
without a doubt they WILL be required, per accelerated approval protocol. What's your point?
...while approval will give SRPT a virtual monopoly in the exon-skipping for DMD space, PLUS a $300m+ windfall bonus because of the voucher, making it less likely SRPT will ever need to raise cash in the market again.
BTW, the (extended) trial size can be considered as N=24.
that's what Benny wants you to believe.
I'm guessing the letter says basically the same as what they said at Adcomm, that yes, they (FDA) were very concerned about toxicity, but couldn't formulate a risk / reward proposition because there was also no evidence of effectiveness. There are multiple reasons why Benny wants you to think it is ONLY about effectiveness.
Stat sig is stat sig, but correlation is not causation. So higher N is generally better because it is more likely that randomized treatment and control arm populations are balanced. If they were unbalanced, you might see a stat sig correlation implying a drug effect that was nothing more than an artifact of a population imbalance (e.g, more younger kids got randomized into the TX group). BUT as luck would have it (SRPT's ONLY piece of good luck over the last 4 years) the Natural History studies used as control group in the extended (3+ year) Etep study do appear to be balanced WRT the treatment group, in terms of relevant parameters (age, baseline 6MWT, etc.), as were the 48-week TX pops v pbo.
If you want to speculate on an information-stock-price-action linkage here, the most obvi is that news that BMRN has received CRL from FDA has leaked. Mainstream WS thinking (wrong) is that this is also bad for SRPT. You can't focus on one stock's price without considering the other, or the biotech market in general.