So I spotted graphs for n=73 (Mar 2018), 91, 113 (Aug 2018) and they all look pretty consistent to me. There's enough wibble in the data between n=73 and n=113 to look like statistics, I'm not surprised 91 and 113 fit to a very similar curve and it's also good to see the fit settle slightly lower from 73 to 113. The greatest variation is at week 1 which makes sense to me.
If we infer from the information we have, 12 weeks of standard placebo care corresponds to 3-4 week of treatment care. To me this looks consistent with earlier datasets (eg as described June 2017). Now back in June 2017 they did a cost comparison of $1200 for the treatment vs $250 per week for standard care so if the results are firm the cost benefit should be well past breakeven.
While I'd like to see a final version too, I don't know that it's a good look now that the trial is finished and soon to be unblinded.
While I think the evidence shows that the odds are good the trial will be a success, imagine if it wasn't.
I think it's pretty standard practice for a company to maintain radio silence at this time.
FTT Price at posting:
7.8¢ Sentiment: None Disclosure: Held
(20min delay)