RVX - First I would congratulate them on being so open. It was refreshing (and I suspect you had something to do with it because it was different than their history, at least in the PRs, appears to be).
…
In toto I think the data mining *might* have found something (i.e. combo of 208 with Rosuvastatin in low HDL). Worthy of follow up, albeit with the standard return to the mean problems arising as a result of post hoc analysis. But I would suggest that they are going to have a tough row to hoe to get to approval:
1) Given the repeated, relatively large, percentage of patients with ALT>8ULN (2%) they are going to need a big safety database. And when they get that database there is a higher than normal chance that real Hy's Law liver tox shows up in that database. 10% chance? 30% chance?
1a. Maybe they can target a more seriously ill population as a way to get to approval with less safety risk ala the familial hypercholestemia? HDL<30? (Doesn’t seem likely the FDA would be persuaded – but worth at least a dialog?)
1b. Another option is to dose escalate if the existing data for tox appears to be correlated to magnitude and/or rate of change of ApoA1. Something like that may be what is happening with Kynamaro?
2) Given that they need to convince BP to spend a lot of money on a large outcomes/safety trial based upon post hoc data with all the risks of post hoc (see below for some strengths and weaknesses). Were I them I might try to run another ph ii IVUS just to prove it wasn’t a fluke – or maybe run a 2b/3 study where the 2b is IVUS or non-invasive angiogram (now able to do fairly low dose) and deciding on tweaks to protocol before hitting ph 3.
Strengths and weaknesses:
Strengths:
1) Post Hoc is aligned with their earlier protocol decisions (which statins and what HDL level) which get much less mark down than many post hocs IMO. That said, I’d still expect some regression to the mean vs placebo if they do a repeat.
2) ITT MACE results are pretty good (but the placebo group MACE rate appear pretty high for a non-acute population – so probably better than will show in repeat. Especially ASSURE – although they may be getting more acutes than the entry criteria would lead one to believe because of the need for a baseline IVUS)
3) Concordance of MACE data with IVUS subgroup data. That is worth something IMO – albeit not too much since no matter whether the best IVUS group got that way via treatment or via data mining finding a lucky subset you would still except the same MACE benefits.
Weaknesses
1) Post Hoc is still post hoc.
2) ALT Elevations
3) Smaller items:
3a. The placebo group on Ros and HDL<40 actually did worse IVUS as the Ros dose went up. Not likely to repeat.
3b. Nits:
3bi. They kept talking of great benefit in ¼ the dose of Ros for ¼ of the time. But I think it is pretty clear that for statins most of the benefit comes in the lowest dose and comes initially. Time beyond 6 months is incremental. Dose above ¼ max is incremental (e.g. TNT - 80mg Ator vs 10mg only had HR=0.78).
3bii. They said that they picked Ator and Ros because they were likely to be the generic statins of choice. But if I remember previous PRs they said they chose them because of a subgroup analysis of a previous trial.
Comments welcome as usual.
BTW – Totally separate note: There is any interesting, rarely talked about, newer statin in the US (from a Japanese pharma if I remember correctly). Offhand I cannot remember the name, but it is interesting because it seems to have a very different set of side effects (generally less than the old herd). As I said, not really pertinent to RVX, but an interesting aside. If I had to go on a statin again I’d probably choose that one.
News 
Market Data 
Discover 
