Register for free to join our community of investors and share your ideas. You will also get access to streaming quotes, interactive charts, trades, portfolio, live options flow and more tools.
Register for free to join our community of investors and share your ideas. You will also get access to streaming quotes, interactive charts, trades, portfolio, live options flow and more tools.
JL - I think you comment that the composite event rate may be increasing is correct. I think this would be expected given the tightening of study inclusion criteria at about 2 years into the trial, which led to sicker patients being enrolled (who are more likely to have MACE events sooner).
My general take on the recent "no halt" at interim:
1) This was totally expected - the design of the trial made it very hard to stop at 60% for a number of very good reasons that I believe will ultimately benefit shareholders (i.e., good data in subgroups such as diabetics that will permit broader marketing).
2) Assuming the effect size stays the same between the 60% and 80% interim, it is MUCH easier to reach significance at p<.022 than at p<.0076, and moreover, the larger sample of events accruing also makes it easier to meet the stop criterion.
3) The language used in the press release regarding (to paraphrase) wanting more time for the effects to become more "robust" is interesting. I believe "Robust" is the exact wording used in the new SPA describing requirements that secondary endpoint results be strong enough in addition to meeting the primary endpoint stop criterion. Is this coincidence? One possibility is that the DMC told the company that the primary stop criterion has been met but secondary endpoints are not yet ideal. This would be the only scenario in which I can see the company intentionally using the word "robust" in the PR.
Drops in stock price now don't really matter if the above is correct.
That is correct. So if 15% RRR at 60% of sample (events) is not significant, it might be significant at 80% due to more events (larger sample) despite exactly the same effect size.
If you are correct, then the same people will likely drive up prices before the 80% interim, most likely even higher, so nothing will really be lost from where we are now. My hope is that if there is a continuation this month, the price still remains above the recent lows. Since everyone now buying knows an 80% interim will be following in 8 months or so, it seems likely that a sizable portion of recent buyers might just hold the stock rather than miss out on good news. Time will tell.
Stock price movements before results are announced in clinical trials - Interesting study from 2011: http://jnci.oxfordjournals.org/content/early/2011/09/26/jnci.djr338.full
Here is the abstract:
Background Phase III clinical trials and Food and Drug Administration (FDA) regulatory decisions are critical for success of new drugs and can influence a company’s market valuation. Knowledge of trial results before they are made public (ie, “inside information”) can affect the price of a drug company’s stock. We examined the stock prices of companies before and after public announcements regarding experimental anticancer drugs owned by the companies.
Methods We identified drugs that were undergoing evaluation in phase III trials or for regulatory approval by the US FDA from January 2000 to January 2009. Stock prices of companies that owned such drugs were analyzed for 120 trading days before and after the first public announcement of 1) results of clinical trials with positive and negative outcomes and 2) positive and negative regulatory decisions. All statistical tests were two-sided.
Results We identified public announcements from 23 positive trials and 36 negative trials and from 41 positive and nine negative FDA regulatory decisions. The mean stock price for the 120 trading days before a phase III clinical trial announcement increased by 13.7% (95% confidence interval = -2.2% to 29.6%) for companies that reported positive trials and decreased by 0.7% (95% confidence interval = -13.8% to 12.3%) for companies that reported negative trials (P = .09). In a post hoc analysis comparing the stock price averaged over 60 trading days before and after day -60 relative to the clinical trial announcement, the mean stock price increased by 9.4% for companies that reported positive trials and decreased by 4.5% for companies that reported negative trials (P = .03). Changes in company stock prices before FDA regulatory decisions did not differ statistically between companies with positive decision and companies with negative decisions.
Conclusions Trends in company stock prices before the first public announcement differ for companies that report positive vs negative trials. This finding has important legal and ethical implications for investigators, drug companies, and the investment industry.
Sorry - Just realized this was the same ones already posted by Sam on Friday.
Scripts Update for Week Ending 26/08 (taken from Yahoo Board):
V
TRx: 17,444 {vs 17,176; +1.56%} Sector +0.44%
NRx: 6,911 {vs 6,851; +0.88%} Sector -0.50% --- ATH
Ref: 10,533 {vs 10,325; +2.01%} Sector +1.02%
L
TRx: 2,192 {vs 2,093; +4.73%}
NRx: 833 {vs 774; +7.62%}
GenL
TRx: 58,847 {vs 58,867; -0.03%}
NRx: 21,671 {vs 21,938; -1.22%}
V TRx Market Share: 22.23% vs 21.98% --- ATH
V NRx Market Share: 23.49% vs 23.17% --- ATH
V Ref Market Share: 21.47% vs 21.26% --- ATH
Ditto that ditto
Regarding odds of stop at 1st IA, this is a binary event. So, by definition, odds by chance alone are 50% (coin flip). To have a less than 50% chance of stopping you have to have a compelling reason to believe that V does NOT work. In my view, when you factor in the well-known mechanisms of EPA and relevant supporting data from prior trials indicating beneficial EPA effects, the very large sample of R-IT compared to many other trials, the relatively large sample of "events" that have already occurred even at the 60% interim (nearly 1,000), and the relatively conservative effect size assumed when they planned the trial (15% RRR), you come up with an odds of stop that has to be larger than chance. Certainly, 60% probability of stop at 1st IA does not sound at all extreme to me, and even 70% does not sound unreasonable. It is certainly more likely than not that it will stop at first interim. How high the probability is, no one knows for sure, but even at 80% probability it might not stop (1 time out of 5).
The FDA can inspect unblinded data for all clinical trials that will eventually be submitted to support drug approval. This is to monitor and insure patient safety rather than to sneak a peak at efficacy data. Wish your theory was correct though.
The ethics of being on a DMC require that the members not discuss results with anyone. They will be signing confidentiality agreements to that effect. That said, we all know that people like to tell secret information that they have. If results look very good but they decide not to halt yet, it would not be surprising to me if someone on the DMC told another person not bound by confidentiality that "I cannot discuss it but things are looking good," which could be spread and influence stock price later.
The biostatisticians will know, as will the members of the DMC. Probably somewhere around 6 - 10 people will know details of results.
Regarding the DMC, the whole point of having one is to have experts (CV specialists, researchers, biostatisticians) review the unblinded data who are unconnected with the company. The company will not attend the upcoming DMC meeting. The DMC members are chosen by the company, but are kept at arms length. This is to permit decisions to be made as to whether the data meet stop criteria that are not biased in any way by the company. This is a good thing because it prevents others from claiming after the fact that the trial results are biased. Once the decision regarding whether stop criteria are met or not is made by the DMC at their upcoming meeting, then the company is informed, with subsequent actions described as in previous posts depending on what the recommendation is.
P = 0.007 if RRR = 18%, assuming 8,000 patients in the sample. This would meet the cutoff for stopping early. The difficulty in modeling this is that it is a Time to Event analysis, which there is no way (that I am aware of) for us to estimate easily without actual data. I think the 18% RRR goal gives us an idea of what we should hope for at the 60% interim nonetheless.
AVI - I had not seen this meta-analysis previously - thanks for sharing. Despite their conclusions, I think what is interesting is that in Table 3 (specific effects of Omega-3 supplements), the RR for outcomes relevant to MACE indicates 18-22% risk reduction. If this effect size is obtained in R-IT, we will no doubt get significance eventually given the large sample studied. The study was powered to detect an effect smaller than this. Only downside is 15% elevated stroke risk, but if this were a large problem with V, I would have expected that to be apparent in the relatively large ANCHOR trial previously.
Those of you interested in the issues involved in stopping a trial early for benefit, here is an excellent discussion I found that is very readable:
http://circheartfailure.ahajournals.org/content/5/2/294.full
The first example used seems to have many parallels with R-It.
Great point - I was wondering today about that too. DMC might say "you have met your primary endpoint stop criterion and are doing ok on key secondary outcomes." Company might say, "yes, but we want more data on the diabetic subpopulation [for example] so let's continue." In this case, I wonder if we might get a press release saying that they are doing this, which would basically be a positive result but continue anyway. Probably unlikely, but not impossible.
I saw the reference to a study design publication in NEJM in an analyst report and on this board. I think that is highly doubtful. NEJM is a top ranked journal that is very highly selective - I think it is unlikely they would publish a trial design with no data. Trial design papers are usually published in specialized journals. Still worthwhile, but publication in NEJM may have to wait until R-It is successful (they absolutely would publish the results if successful).
Agreed - If R-It is successful, this is the kind of paradigm-shifting study that NEJM or JAMA would love. There would be wide press coverage, which would probably help with price movement upward.
The whole point of the DMC is to insulate the company from being accused of biasing the ultimate results. So, having the DMC (if it is done properly) means the company will know nothing about the results after the interim analysis DMC meeting other than "the study met your stop criteria" or "the study did not meet your stop criteria." In the former case, the company would then decide whether they indeed want to stop the study, and if so, the data would become unblinded to them (i.e., they would know exactly how well V worked compared to placebo). I would anticipate a press release shortly thereafter saying the study is being stopped for overwhelming efficacy. This could all happen very quickly after the DMC meets sometime this month.
The company and people who know anything about clinical trials know that a stop at 60% is a high hurdle - the p value criterion they set to stop is much harder to meet than it will be at the 80% interim, and the sample size of MACE events will also be smaller at the 60% interim, which works against meeting stop criteria. The company really does not want to stop the study prematurely, so they are making sure they only stop at 60% if the results are truly overwhelming. I think at the 80% interim it is much easier to meet the stop criteria. While stock price changes are hard to predict, I would not anticipate large and prolonged drops in PPS if we continue at the 60% interim - this really is expected.
Secondary endpoints have always been a focus in the analysis, they just added some extras. I have not seem them specified anywhere, but would expect them to look at efficacy specifically in those with diabetes, look at specific components of their composite MACE endpoint, etc. Regarding your questions:
a) No change from before,
b) FDA will require detailed analysis for the primary endpoint (MACE), with analyses of other secondary endpoints only to the extent the company is seeking approval related to those endpoints (e.g., diabetes specific indication)
c) I don't know.
Doing additional analyses is not that big a deal, regardless. Once the data are cleaned and in the database for analysis, you can whip out hundreds of analyses a day if you wanted to. That is not the roadblock.
Enrolling extra patients in a study to achieve the originally targeted sample size once drop outs are accounted for is not unusual (can be part of the original study design). That may be where the extra patients came from as you suggest.
Sorry - I got that wrong number from someone else's post. Well...down we go (until interim success)
Assuming the most obvious theory is correct, the institutions scheduled to have their notes converted who feel they are overweighted with AMRN now have been selling - this would fit the higher volume today. Once the price hits $2.85 (same as conversion price), there would not seem to be further incentive to sell.
JL - Nicely stated - I agree, and this is a good complement to what I just posted in response to P as well.
P - We agree for once - It is not yet proven that reduction of TG via V or any other drug can affect CV outcomes. ADCOM comments you quote are correct that ANCHOR only proved V reduced a biomarker - that does NOT imply that V didn't or couldn't change MACE outcomes, it just reflects the fact that the ANCHOR study was not designed to test that hypothesis. My impression (FWIW) is that cardiologists as a group still seem to think that TGs are not crucial, however, I also know that scientists are often wrong in what they believe (i.e., there are periodic paradigm shifts). The study I cited does indicate that people who displayed lower TG had reduced MACE events, which provides a plausible causal path for V ==> reduce TG ==> Reduce MACE = R-IT success.
Here is a follow-up to my previous post on a recent in press article in a cohort of nearly 5,000 patients (96% on concomitant statins, 29% diabetic) (http://www.ncbi.nlm.nih.gov/pubmed/27515380):
After actually looking over the full article, results suggest a RRR of 25-26% for higher vs. lower TG levels on MACE events within a 2 year follow-up period. I think this a nice parallel to R-IT and may bode well for outcomes, apart from the benefits of the anti-inflammatory effects of V. Clearly, TG reductions are likely to have significant effects on MACE outcomes. There was significant objective plaque regression as well with lower TG.
Here is the actual quote from the article relevant to RRR:
"Figure VA in the online-only Data Supplement illustrates
Kaplan–Meier curves assessing MACE across
patients stratified according to on-treatment TG levels <
versus ≥ median, illustrating that cumulative incidence of
first MACE at 24 months was significantly greater in those
with ≥ median TG levels (21.2% versus 15.9%; log-rank
P<0.001). Figure VB in the online-only Data Supplement
stratifies the population according to on-treatment TG levels
< versus ≥200 mg/dL, illustrating significantly higher
cumulative incidence of first MACE within patients with
TG ≥200 mg/dL compared with lower TG levels (23.7%
versus 17.5%; log-rank P<0.001)."
Related to these posts, maybe as has been suggested by another poster these note holders were aware this would happen and have been selling some extra shares for the past couples of weeks or so? There has been fairly high volume with generally downward pressure at a time when everything else might say we would be expecting upward drifts (the first IA readout). I realize that this may be like trying to explain why the wind is blowing a certain direction at a given time...
P - Funny - Lovaza is "just fish oil" as well, and GSK seems to have done quite well with it. Does not matter whether it is a natural substance or not, it can be sold in purified form and in regulated fashion to earn money. Penicillin is purified mold, morphine is purified opium, etc.
Not sure if Zum or anyone posted this before -
This study came out online recently. Combined past trials with nearly 5,000 patients. Found that lower TG had significant beneficial effect on MACE events (as hypothesized in R-IT) and reduced objective arterial plaque progression. Supports V effects on R-IT outcomes via TG mechanisms (ignored effects via anti-inflammatory properties).
http://www.ncbi.nlm.nih.gov/pubmed/27515380
OBJECTIVES:
Non-high-density lipoprotein cholesterol (non-HDLC) levels reflect the full burden of cholesterol transported in atherogenic lipoproteins. Genetic studies suggest a causal association between elevated triglycerides (TGs)-rich lipoproteins and atherosclerosis. We evaluated associations between achieved non-HDLC and TG levels on changes in coronary atheroma volume.
APPROACH AND RESULTS:
Data were analyzed from 9 clinical trials involving 4957 patients with coronary disease undergoing serial intravascular ultrasonography to assess changes in percent atheroma volume (?PAV) and were evaluated against on-treatment non-HDLC and TG levels. The effects of lower (<100 mg/dL) versus higher (≥100 mg/dL) achieved non-HDLC levels and lower (<200 mg/dL) versus higher (≥200 mg/dL) achieved TG levels were evaluated in populations with variable on-treatment low-density lipoprotein cholesterol (LDLC) </≥70 mg/dL and C-reactive protein </≥2 mg/L and in patients with or without diabetes mellitus. On-treatment non-HDLC levels linearly associated with ?PAV. Overt PAV progression (?PAV>0) was associated with achieved TG levels >200 mg/dL, respectively. Lower on-treatment non-HDLC and TG levels associated with significant PAV regression compared with higher non-HDLC and TG levels across all levels of LDLC and C-reactive protein and irrespective of diabetic status (P<0.001 across all comparisons). ?PAV were more strongly influenced by changes in non-HDLC (ß=0.62; P<0.001) compared with changes in LDLC (ß=0.51; P<0.001). Kaplan-Meier sensitivity analyses demonstrated significantly greater major adverse cardiovascular event rates in those with higher versus lower non-HDLC and TG levels, with an earlier separation of the non-HDLC compared with the LDLC curve.
CONCLUSIONS:
Achieved non-HDLC levels seem more closely associated with coronary atheroma progression than LDLC. Plaque progression associates with achieved TGs, but only above levels of 200 mg/dL. These observations support a more prominent role for non-HDLC (and possibly TG) lowering in combating residual cardiovascular risk.
Since the note holders were forced to take the stock instead of principal + 3.5% interest, it would seem they would have incentive to hold for possible upside after getting shares cheaply rather than dump all shares as PPS is moving down. Maybe as stated earlier they will short the stock as protection rather than outright selling shares? Options in next couple of weeks might be an indicator?
One thing I find interesting about the dilution is that it seems that when earnings are negative (there is a loss per share), averaging over a larger number of shares will reduce the loss per share from what it would have been before. This is the converse of the negative influence of dilution on profits. This may give us a "better" loss per share on the next earnings statement(appear closer to profitability) despite no fundamental change other than more shares. Is this interpretation correct? Might this have been a perceived advantage of the conversion from the company perspective?
Chas - You stated:
one of the factors influencing P is that higher risk subjects are more likely to have an event earlier, leaving a lower-risk pool of subjects eligible for a primary event (slightly lowering observed P over time).
P - Your post may be just your usual "turd in the punchbowl" comments. However, I do feel I have to respond. Your post stating <5% probability of success at ANY time point are as unjustified as those claiming 100% likely stop at first interim. This is a multiyear study costing more $100 million, and a company does not invest that much money on a <5% chance of positive findings. Company management stand to make a lot of money on their stock if PPS increases so obviously the have a vested interest in designing R-IT to succeed if possible. The company employs scientist consultants with expertise in this area when designing the study. The scientists involved in developing the study design are competent (PI is at Harvard and they do not give out faculty positions easily). There are just as many, or more, studies by numerous researchers highlighting multiple beneficial effects of EPA than there are studies suggesting there is no effect such as those you repeatedly cite. The R-IT sample size of more than 8000 patients was chosen based on conservative effect sizes derived from those actually observed in the most similar previous studies, the desired level of power to detect an effect if there was one (90%), and their willingness to have an erroneous positive finding (e.g., p<.05 = 5%). They have also enriched the study for sicker people most likely to benefit both from direct TG-related and antiinflammatory-related effects of EPA, with an increase in "disease severity" as time went on. This increases the odds of significant effects. None of us can predict what the actual findings will be, but the likelihood of success is certainly >50% at the 2nd IA and final analysis, if not before, and may be substantially higher odds than this.
That makes sense. If so, it would seem like shifting from debt to equity and adding corresponding number of shares should have zero effect on the total enterprise value. No idea what if any impact this "non-change in value" will/should have on PPS.
Can someone confirm whether the language below in the press release is standard in an exchange like this? These are ordinary ADR shares correct?:
"The ADSs issuable upon exchange of the 2014 Notes and 2015 Notes will not be registered under the Securities Act of 1933, as amended (the “Securities Act”), or the securities laws of any other jurisdiction, and may not be offered or sold in the United States absent registration under the Securities Act or an applicable exemption from registration requirements."
Looking at the 8-K and what HD wrote re: names of the note holders, looks like Amarin and its subsidiary have both said "we don't want to owe any of you guys any more or pay you interest anymore. Now that our stock price is back up, here's your money (stock) to do with what you want." A big question for stock price would then seem to be whether the former note holders want to keep the stock because they believe it will increase in PPS (neutral to PPS?) or sell now to get what they can (negative to PPS, at least short term). Seems that, if anything, the former note holders might be motivated to promote the stock to increase PPS and build on their earnings from this exchange of notes for stock. Only direct advantage to the company seems to be less debt on the books, and possible benefit when there are negative earnings per share (based on something I read, at least - I guess the losses get diluted?).
I have to say that with a big stock dilution and the dumping of biotech stocks due to news this week, I am a little surprised (but happy) the stock has remained as high as it has been. I just hope it continues an upward trajectory.
Raf - I look forward to that. I can only hope I do as well as you did.
Looks to me like the dilution and this note conversion are related:
From a finance website:
Q: Does a convertible note holder have a choice about converting a note in an equity financing round?
A The terms of conversion are usually listed in the convertible note. Typically, conversion to equity is automatic at the next equity raise, but may be conditional on a round meeting certain thresholds such as a minimum round size.
BB - I would love good news re: Anchor SPA. Good luck. However, I don't think the company or anyone else really expects a change in this from the FDA so it does not seem like there is any real downside risk specifically from this, only upside.
I am either insane or brilliant. I bought another 5,000 shares at 3.02 today, near bottom.