Meirluc - I am pretty certain they will announce brief topline results whenever they are available. Look at Amarin as a parallel. They got a late breaking trial talk accepted apparently before they had topline data available (accepted for presentation early September), announced brief topline results in a PR on 9/24, and then presented detailed results at a national conference (AHA) 1 1/2 months later on 11/10. Announcing DCVAX topline in March followed by a detailed ASCO presentation next May is totally plausible.

JT is presenting at the Evercore ISI healthcare conference on Tuesday. I just looked at what this company does, and it is interesting. From their website:

Thanks for the nice summary AVI. This post should be stickied. It is a definitive data-based response. What Nissen is doing (in terms of psychology) is selectively seeking data that supports his argument, ignoring all data to the contrary, and is doing so to protect his ego (huge ego investment in his own competing study). These are well known types of cognitive biases, and they are often unconscious. The bad thing is that in such situations, the person involved is in a closed loop where no competing information can change their minds - arguing with him is pointless.

And why would be they be greedy towards a worthless company? Dilution or not this is a positive signal of something good coming down the pike.

Seems to me that the BOD would not be arguing over money (options) unless there was a lot of money worth arguing over. This is a nearly bankrupt company, so how does this make sense? The BOD would be privy to any business and financial negotiations. Putting 2 and 2 together = ???

Sounds like their own board complained about their stock compensation, which prompted this change to address it. Probably would not matter to them if they expected the company to go belly up soon.

Well, we might lose the battle (fail primary PFS outcome) but win the war (get FDA approval anyway based on long tail). Success/failure of the trial may not really be dichotomous in this case.

Bhatt next publication - On a lark this morning I e-mailed Bhatt to share with him the methods (a publication) and programming language for doing the mediation analyses I posted about yesterday to demonstrate MOA of V. I was surprised to get an almost immediate response from him. He said it was a great idea, and that they plan to do a mediation analysis after they finish the basic biomarker publication (which they are apparently already working on). He seemed excited, using an exclamation point when saying that he looks forward to seeing the results of these additional analyses. He must be like a kid at Christmas right now!

JL - I totally agree. EPA/AA will no doubt predict MACE risk, we know V is linked to EPA/AA, and we also already know TG did not account for the MACE reduction. Even though the links between these seem obvious and support a mechanisms of action, these connections when looked at in separate models (as in JELIS) do not in the technical sense prove the MOA. A researcher like HK could still claim that MOA (as it affects outcomes) has not been proven. You have to think like a researcher and statistician about this to close the case definitively (i.e., you have to prove the common sense interpretation). My statistical post earlier was showing that there is an elegant way to show how all these links fit together in a single statistical model. I think THAT is what would prove the case to even the most picky researcher. If results played out the way you and I would both expect, we could conclusively say "V reduces MACE risk by increasing the EPA/AA ratio and NOT by reducing TGs."

Not sure Raf. I think they collected these biomarkers at some other points as well, but maybe just did not report these yet. Certainly looking at changes between a single set of timepoints is simpler statistically for making a point.

No - In the example I just posted, that at best is only accounting for the link between V status on the left and the path to EPA/AA in the middle. They need to further link it to MACE outcomes( on the right side of the model).

JL - You are right in principle, but I don't think they tested it in a way yet that would satisfy a statistician. Definitely do-able in one model.

Sorry - Should give an example.

Picture an image with V versus Placebo on the left, and MACE on the right. In between the two you have EPA/AA above and TG below. Draw lines between all of them. A mediation analysis (or structural equation model) could test the total direct effect of V on MACE (the direct middle line), and parse out whether the path from V to MACE via TG and/or EPA/AA (the two indirect paths) are significant. Depending on the results, they might conclude that EPA/AA mediates (i.e., mechanism of action) the effect of V on MACE (indirect paths), and likely that there remains a significant direct effect of V (via other untested mechanisms) on MACE. I would think the effect size for EPA/AA would be bigger than for TG (or that the indirect path via TG would be nonsignificant), which in either case, would say that EPA/AA is the most important mechanism of those tested. If the direct effect (from V to MACE) were not significant, this would indicate that EPA/AA is THE mechanism of action. It really is a cool analysis.

Why and How both mean mechanism to me. Take your choice. We know V alters the EPA/AA ratio, but what has not be previously shown is whether this change in turn reduces risk. They can actually test this pathway using mediation analyses or structural equation modelling.

Open label would not be adequate for FDA controlled trial. Blinded placebo controlled is the standard for all FDA studies (devices are an exception).

David - They tried not doing crossover. Patients did not want to participate knowing there might be no benefit to them if they got placebo only.

Chris - For an Ivy League guy, HK's comments sure make him look stupid. "These results are too good - Don't risk saving millions of lives [which is no longer debatable] until we know exactly why it works." Well, my bet is that within 3-4 months, we have a paper from Bhatt showing why V worked to reduce risk. Everything they need to answer this question is in the R-It study. His comments echo the same concerns about costs to the healthcare system related to the change a few years ago in criteria to broaden statin use. And in that case there were already multiple studies showing that statins work and why they work. I guess this one sided argument (looking only at drug costs and not at economic/health benefits) is a default criticism for many who are not health economists.

Abeta - See this article: Mesenchymal from MRI

Maybe mesenchymal can be retrospectively determined from the MRIs previously obtained?

Senti - Makes sense. I wasn't aware of that.

Raf - My guess would be that they have held it out for a separate publication, maybe in process. You can't publish the same data twice. EPA/AA data may be part of a mechanisms paper, hopefully using mediation analysis to put the mechanism issue to rest.

Which brings up the question in my mind of why nothing seems to have been talked about in terms of mesenchymal in the blinded publication or the recent data update (but maybe I missed it). I thought this was where DCVAX was supposed to shine?

AJ - Cool. Thanks for providing detailed example of this.

Stac - I guess I am thinking they are hedging their bets at this point. Trying to (as cheaply as possible) look like they are GIA, but are open to good BO offers. If for some reason no reasonable offers emerge, I think they would probably hire a lot more reps and potentially have to dilute a bit to enable this. I certainly do not see anything in their behavior that clearly indicates they are only interested in GIA (e.g., ambiguous language used in multiple interviews re: buyout, no sNDA yet, no Europe partner yet, hiring only enough new reps to replace outgoing KOWA reps). All guesses on my part though.

Thanks Float - I had not read that paper in detail yet.

Grth - It is a valid point in terms of science, but is totally irrelevant when it comes to FDA approval. FDA does not care why it works just that it does work on clinical outcomes in a properly designed study.

That is the problem with post hoc data mining, and is why their results are still described as negative in terms of efficacy.

HDG - The sales force staying flat (KOWA co-promotion agreement ends in a month I believe) only makes sense to me if this is a move by Amarin to make it look like they can GIA without having to spend all the money actually needed to succeed at GIA. I am guessing you read this the same way.

Only if it is a journalist

Wow - I finally had a chance to watch the Bloomberg video of JT, and I thought he did an outstanding job. I am very surprised anyone thought otherwise. He is clearly knowledgeable and inspires confidence (at least in me). I thought it was interesting that he did not directly answer the question about more partnerships or BO - he simply sidestepped it by saying "we are positioned to go it alone in the U.S." Very cagey response.

Tilt - If so, then we are golden. That is a very straightforward thing to demonstrate statistically. Crossover issue would be irrelevant.

I have been thinking about the statistical analysis plan they might be developing, which led to some scenarios I thought I would share. One caveat - I am a researcher and do my own statistics, but do not do KM survival analysis (but I know what it is). Although clinician biases in trying to figure out whether a drug works on a patient by patient basis are well know, I think the long history of DCVAX use by a highly experienced clinician (LL) who believes it works based on her clinical practice means that we do not have to consider the scenario where DCVAX proves entirely ineffective.

Scenario 1: PFS (primary outcome) is significant. No idea how likely this is, but if this were the case, that plus the long tail results would IMO be plenty to get DCVAX approved, regardless of whether traditional OS measures are significant.

These other scenarios all assume PFS is NOT significant.

Scenario 2: Significant test of the main effects of DCVAX based on the condition to which patients were assigned at baseline (ITT analysis). This pattern is what we would see if DCVAX works in newly diagnosed GBM, but does not work at all for recurrent GBM. Visualize a graph with a curve steeply sloping downward (placebo regardless of crossover status) and a clearly separate curve sloping downward above it less steeply (DCVAX). This is the most straightforward homerun result we could get for the key prespecified secondary outcome (OS). I don't think the FDA would provide much pushback against this even though it is technically a secondary endpoint, especially in context of the long tail results. Approval likely.

Scenario 3: This could be tested as the interaction between ITT group assignment and timing of DCVAX administration (e.g., placebo coded as 0, crossover coded as 1, and those assigned to DCVAX originally coded as a 2), or ignoring initial group assignment, simply as the main effect of DCVAX timing. A significant analysis like this would indicate that the efficacy of DCVAX depend on when it is given. This would be the most relevant analysis if DCVAX worked well for new GMB and worked somewhat but less well for recurrent GMB. Visualize the steepest downward sloping survival curve for placebos who did not cross over, a separate intermediate downward sloping curve for crossovers, and the shallowest curve for those originally assigned to DCVAX. If significant, the pattern of this effect would be biologically plausible, and because of this, I believe it would support the efficacy of DCVAX, and in context of strong long tail results, would also likely support approval.

Scenario 4: This is the best case scenario for patients but the worst case scenario in terms of interpretability. Visualize the curves for the patients originally assigned to DCVAX and crossover patients overlapping, with both showing a shallow downward sloping curve, with a much steeper downward sloping curve for the non-crossover placebo patients. Clearly DCVAX seems to be doing something (visually) but you would get a nonsignificant main effect of DCVAX in ITT analyses (because there are so many crossovers with long survival who would dilute the treatment effect) and the test in Scenario 3 would also not be significant (because the two curves with the most patients overlap). This is the only scenario in which comparison to SOC controls becomes crucial. There would be no direct statistics to show that DCVAX works, so approval could only be based on showing greater long-tail OS percentages at key milestones (e.g., 2 year, 3 year) than you would expect in SOC. These comparisons can be tested statistically, but how convincing the argument is in terms of approvability depends on whether an SOC number can be quantified for a group similar enough to the DCVAX trial population. While this may be difficult, it would be reasonable to handle this by doing statistical comparisons of milestone survival between DCVAX (initially assigned to DCAVX + crossovers) and prior SOC samples, including all relevant individual difference variables in the model as covariates (in this case, percentage of patients in each study with partial surgical resection, +MGMT status, etc). I think a meta-regression approach (like meta-analysis but looking at the influence of other variables) might be best for comparing the DCVAX trial results to a composite of other similar trials. In theory, if the DCVAX effect in these meta-regression models is significant even after controlling for all relevant sample characteristic differences between studies, then we would be justified in concluding that DCVAX works to increase long-tail survival. Given the orphan stats of GBM and the current regulatory climate at the FDA for orphan disease interventions, I think this has greater than 50/50 odds for approval even in this worst case scenario.

When I worked through these scenarios, I realized that the crossover issue is not necessarily an issue that is impossible to address. No idea what approach they actually plan on using, but I think a good statistician could figure out a way to show DCVAX works if indeed it does work.

Afish - That is way too rational for this board

BB - You have absolutely no idea how things work. Off label prescribing is a totally accepted and legal practice, and Amarin has the right via the 1A amendment (with FDA agreement) to use the R-It publication in promoting the drug to docs. The fact they are doing so means they have FDA agreement to do this. DOJ has nothing to do with it.

I think what is too easy to forget is that we have literally no clue what is going on behind the scenes. My guess is a lot. If there are partnership or buyout negotiations, we will not know ANY of this until after the fact. Just because you do not hear JT singing Amarin's wonders from the rooftops does not mean they are not doing anything. My bet is people here will keep grousing until one day, POP, they announce something amazing and the price takes off. Then all the naysayers will be saying what a great CEO JT is.

Agree and a bean counter is what we need. This is a company with billions of dollars in sales potential and high potential for a huge buyout. Do you want a cheerleader or a someone who knows money and has exactly the expertise needed to maximize value? The mood on this board has resumed the pre 9/24 doom and gloom, BB is in full bipolar tizzy mode, and no one wants to wait a few months for the PPS to go up. What the hell is wrong with people? This is a long term hold not a day trading vehicle.

CBB - I don't think JT mentioning being approached by multiple companies about Europe was a coincidence. He did not have to be that specific (he could have said what he said previously, which is "we are now looking at the possibility of expanding into the European market). His statement was informational, but also could be viewed as sending a signal to other companies that time may be short if they want to BO Amarin.

Mav - Good info. I had never heard of them.

We are aware of these crappy blog posts only because we are searching for them diligently. I have my doubts as to whether any of these people really have an audience. Don't think they amount to a hill of beans, Herper and AF excepted.

Retinadoc - An R21 application might fly....

Hard to say, but most likely is an incremental process. They have almost certainly already fully validated data for all patients who have already died, so nothing more needs to be done for those. For any living patients, I would assume they have validated information up until October, and would need to just obtain final outcome information for those individuals (it sounded like they are continuing to let the data mature while they are getting this process going, which means rechecking everyone). No way to know whether they have put off obtaining certain clinical information until the end of the study (maybe the prior scrubs were incomplete), but if so, that might slow things down some. The validation part of things is what is time consuming. For final study closeout, they have to physically go to each location and see the clinical records to match up for accuracy against the trial records. Hopefully that is incremental only. I think our best guess for time required is to look at past data scrubs. These seemed to be about a 5 month process, and given the incremental issue, I think 4-5 months would be a good ballpark estimate.