Skip to Content

How Not to Handle Your Clinical Data

We turn now to Orexigen, one of the small companies trying to make headway in the obesity market. Earlier this year, a patent application from them published, claiming that their drug (Contrave, a sustained release formulation combining the known drugs naltrexone and bupropion) had cardiovascular benefits above and beyond its weight-loss effects. Problem was, they based that claim on the first 25% of the data from an ongoing clinical trial. You’re not supposed to go around doing that. The disclosure led to a public fight with researchers at the Cleveland Clinic (including well-known cardiologist Steve Nissen) about whether the trial had been compromised, and whether Orexigen was misleading its shareholders.
Now, as Matthew Herper reports, the trial has been halted, on the grounds that it has indeed been compromised by that premature data release. The FDA had also stated that they would not accept the trial results for that very reason. Orexigen and their partner, Takeda (who must be wondering how they got into this) announced this yesterday morning, followed a half-hour later by the release of the (half-completed) trial data by the Cleveland Clinic team. (Takeda is also talking breach-of-contract). The cardiovascular benefit that had been noted at the quarter-way mark had slipped below statistical significance, and the best bet is that it was on its way to disappearing entirely, had the trial actually completed.
It’s been ugly. Orexigen released a statement, trying to make their case. Read it at that link and see what you think. I have trouble believing, myself, that they’re really pleased that the study is being terminated, and that it’s yielded important information nonetheless.
The company also seems to be trying (obliquely) to blame the USPTO for disclosing the data, but they’re glossing over the fact that (1) everyone knows the schedule for publication of patent applications, and (2) the PTO is only publishing what you put into your own patent application that you wrote yourself. So this is not even as plausible as blaming the “Reply All” button in an e-mail application. They also seem to be blaming Matt Herper for merely reporting the ways in which the company seems determined to super-glue a clown wig onto its head.
I’ll gladly stand by that opinion, in case Orexigen’s people get around to reading this, and I’ll throw in another one: they would have had trouble paying someone to mess up that trial as throughly as they managed to mess it up themselves. For all the talk in the Orexigen statement about all the tensions and problems with disclosure of clinical data, it’s really not all that hard: you shut up about your interim data, unless you’ve already (from the beginning) planned to disclose something, and the trial has been designed with that in mind. Otherwise, unless the numbers are enough to halt the trial (in either a good or bad way), you just sit tight and see what happens. Because interim data are, well, interim. You run a full trial to see what the real numbers are – until then, they aren’t the real numbers.
Update: more on the data release.

37 comments on “How Not to Handle Your Clinical Data”

  1. Hap says:

    How did the Cylinder Kings get a job running a pharma company?

  2. Anonymous says:

    Trying to point the finger and shift the blame like this just makes them look even more ridiculous and incompetent than they already appear to everyone. Although that’s still probably not as ridiculous and incompetent than they actually are, so keep an eye out for yet more f*** ups.

  3. Anonymous says:

    I don’t know why they would even take the risk of prematurely disclosing CV benefit. It’s a weight loss drug, not CV all they have to do is show efficacy on weight loss and show no CV harm.

  4. Homer Simpson says:


  5. Anonymous says:

    I’d bet this was the consequence of compartmentalization of functions within the company. I can see how the individuals writing and filing the patent might not appreciate the impact that the content of the patent might have on the integrity of the ongoing trial. Those with patent law expertise wouldn’t necessarily be expected to know the potential regulatory pitfalls associated with running clinical trials.

  6. will says:

    @5, I disagree; if you’re a startup pharma company and the IP representation you’ve hired doesn’t have a firm grip on the FDA regulatory process, you’ve hired the wrong patent counsel

  7. Anonymous says:

    Besides, there’s really no excuse for poor communication in a start up – the CEO, Head of R&D and Patent Counsel probably all sit next to each other in the same room.

  8. Magrinho says:

    As per FDA guidance with respect to outcomes, the mantra for obesity drugs became ‘weight loss PLUS…’ Plus could a be a lot of things but generating the data has been ruinous for all comers. Pressure makes people do really stupid shit.
    It’s instructional to go back and read reviews, market forecasts, etc from circa 1997 on the tsumami of money that was to come from obesity drugs.
    But we did get a drug for oily flatulence and the inability to control it. No! Not to prevent oily flatulence. To GIVE you oily flatulence.

  9. Anonymous says:

    “But we did get a drug for oily flatulence and the inability to control it. No! Not to prevent oily flatulence. To GIVE you oily flatulence.”
    There’s probably a market for that. Somewhere.

  10. Hap says:

    1) Who gets the information from an ongoing clinical trial? Why was it released to anyone else?
    2) I’m not sure why releasing interim results was going to put constructive pressure on the FDA to allow their drug – it seems more likely that someone was trying to cash in on the early results. Probably going to be expensive, because I’m guessing Takeda has some lawyers.
    3) A drug to cause oily flatulence? That sounds like a prop for another low-brow comedy. Thirty or forty of those and it might get its development costs back.

  11. biotech capitalist says:

    this sounds like a great business strategy. I would improve upon it: don’t wait for 25% of your data to come in, just wait until the first adverse event in the placebo group, then you have carte blanche to claim your drug is the cure for cardiovascular disease!

  12. Anonymous says:

    I have a better strategy: Don’t put idiots in charge of your assets and investments.

  13. Philip says:

    @3 Anonymous, they risked it to pump up their stock price and sales. It worked. Their share price went from around $5.75 to $9.37 the day the patent published. Their sales of Contrave have been growing faster than any of their competitors, even though its effectiveness and side effects are not better. Who would not want a 41% lower risk of a CV event?
    @5 Anonymous, there were two data leaks. The first was to just about 100 people in and out of the company. The first leak allowed people in the company to decide to file a patent on 25% data (worse they got the patent). That caused a problem for the FDA. The FDA approved Contrave based in part on the 25% data, but had told the clowns that a new CVOT study would be needed. Then in March the patent published and all hell broke out.
    @6 & 7, I think they only have 47 employees. So a release to 100 means that people outside of the company got the data. What a big no no that was. Total lack of control.
    As a computer geek, I am not an expert on CVOT study design, but I can even find problems in the LIGHT study. Look at the patent and see what happens at 16 weeks. That is when Contrave and placebo started to separate. There was a culling of subjects and the placebo and drug arms changed at that point. The patent number is US 8,969,371 B1 dated March 3, 2015, if you really want to look.
    Bottom line, Orexigen released data twice in what appears to be an effort to raise their share price and sales. I assume that even more lawyers will be suing for an apparent pump and dump. I think they were already facing 17 lawsuits before the 50% data proved the management to be a bunch of clowns.
    I hope the FDA makes an example out of them.

  14. biotechtoreador says:

    Well, at least OREX pays its CEO well: his salary + bonus in 2014 was just north of $1.1 million (more f you add $3.4 mill in options, but those are worth less today).
    OREX shareholders are very kind to pay this much for such amazing idiocy. For that kind of dough, I’m sure Narachi justdives by the BMW dealer en route to the Bentley shop.

  15. Sleepless in SSF says:

    So I have a naive question. The Cleveland Clinic press release says:
    “The FDA also stated that this breach of confidentiality had sufficiently undermined its confidence in the ongoing trial; a new trial would be required to determine whether a 40 percent increase in cardiovascular risk could be ruled out.”
    I understand that relying on such an early interim analysis is risky because the subsequent data may not support your initial conclusions. What I’m not clear on is why the early release invalidates the rest of the trial. Is it due to possible placebo effects if the treatment is claimed to be highly effective? Would that even apply if the study were blinded (though I don’t know if this study was)?

  16. bio_kruncher says:

    I’m not sure I agree that stopping the trial was a good decision by steering committee and FDA regarding protecting patient safety. The net result of stopping study is that more definitive safety info is now pushed much further into the future insuring that many more patients will be prescribed the drug before we know safety with more certainty. I think that the likelihood that knowledge of the interim analysis results would bias the final results of the study is low. The endpoints are fairly objective and likely adjudicated by a blinded committee. As long as the blinding of individual patients would have been maintained for patients, investigators and adjudication, why stop? If too many sponsor/CRO personnel working on the study were unblinded then form a new blinded team to complete the study.
    In the end, it is maybe not so bad for Orexigen. They have to pay for new study but are now certain that they will not have any negative impact to label or approval due to results from this safety study for a few additional years.

  17. No Surprise says:

    #5, absolutely not compartmentalization. I have worked with the CEO before, and I’d bet the glue for the clown wig comment of Derek’s was self-applied.
    This is about ego and marketing determining scientific and IP strategy because, well, things must fit to a strategy.

  18. Hap says:

    If FDA approved the drug based on an interim analysis that the subsequent data doesn’t fit, couldn’t they pull the approval? At that point, the negative impact for the company would be immediate with a potential (highly uncertain) benefit in the future.

  19. Chrispy says:

    I don’t understand the scientific rationale for discontinuing the trial, but perhaps there isn’t one. If you piss off the FDA badly enough then they will make you regret it, science be damned. It really is just stupid. Trying to prevent the publication of the 50% data set is even more idiotic, especially when a guy like Nissen is one of the clinicians. (Recall that he was one of the main people who got Vioxx yanked from the market — he’s no stranger to standing up to corporations.) Now I wonder, even if you have your $200 million, who is going to run your trial after you’ve acted in such bad faith on the first one? This kind of crap is a big part of the reason people hate our industry. Frankly, I do, too.

  20. bio_kruncher says:

    More stuff recently posted on Herper’s site. Must read.
    Orexigen saying that the study is not part of formal post marketing commitment so they are not required to copmlete/conduct new study. Also implying Nissen and Cleveland Clinics release of 50% results not based on final adjudicated data?

  21. Anonymous says:

    Interesting product review from Orexigen’s CEO:

  22. Biotech Capitalist says:

    If I were the FDA, I would discontinue the trial. The purpose of the double blind randomized controlled trial is to get unbiased statistical understanding of the effect of a drug on an endpoint. Interim analysis done in isolation by a committee is there to ensure the trial is running well and there is no change to equipoise either by great efficacy or bad safety. If there are leaks, if there is fundamental disagreement between those running the trial (ie Nissen and the sponsor), if the company botched its data handling, I would absolutely shut it down. If they messed this up so badly in the interim, how can the scientific community and the FDA trust the results that come out of it?

  23. Philip says:

    @18, Hap: You are correct that the 50% did not fit with the 25% reported data that showed a 41% cardio benefit. But, the 50% did not show an increased risk over placebo. Who knows what would have happened if the study went to completion. My guess is that Contrave would be a small negative for cardio health.
    One big question is what will the FDA do. The LIGHT study should have been completed in 2017. The new study, if it is ever done, will not be completed until 2022. So the FDA expected results in two to three years, now it will be five or more. Would it be fair to pull Contrave until the new study is within three years of completion? I think so, but I do not the FDA to do what I think is correct.
    I still think the LIGHT study is really flawed and could not show a benefit because of design.

  24. bio_kruncher says:

    @22 If the blind is intact for patients, investigators and adjudicators, then change in equipoise or other possible bias should equally effect both Contrave and placebo patients. To ensure patient safety, I would rather have LIGHT results in 1-2 yrs(guessing?) with very slight chance of bias than to wait until 2022 or later to get results. Yes, 2022 is target date for completion of new study.

  25. Hap says:

    Considering what’s happened so far [trial breached at least twice, broadly, and changes in trial conditions noted above (@13)], a slight chance of bias is highly optimistic.

  26. Biotech Capitalist says:

    @24 Patients volunteer their health and lives to take our cooked up experimental medicines. They deserve far better than even “very slight chance of bias.” And like Hap @25 said, disastrous trial breaches are more than slight.

  27. Anonymous says:

    Releasing your data prematurely to influence investor sentiment is already a big no-no. These are commercial entities but this is also scientists doing science and standards have to be upheld. You can’t just have people releasing tranches of data in post-hoc fashion doing whatever they like and making inferences so their stock prices increase.

  28. DN says:

    “The purpose of the double blind randomized controlled trial is to get unbiased statistical understanding of the effect of a drug on an endpoint.”

    Double-blinding is so that the nurse does not deliver a bottle of placebo and start crying at the poor doomed patient. It is to keep the patients and raters from being affected by knowledge about which group they are in.

    The FDA blew the decision to cancel the study. It’s not just that the scientific value of the study was barely affected, it’s that they panic-cancelled out of spite. The correct scientific approach would have been to ask if the trial would have been approved with a plan to release partial data. The answer was almost certainly yes.

  29. Matthew Herper says:

    @28 — Not without spending alpha, it wouldn’t. And the alpha spend at 25% would have been large, no?

  30. Philip says:

    @28 DN, I would like to add a bit to @29 Matthew Herper’s comment.
    There are rules to running an FDA study. These rules have provided a level playing field and reliable data for many years. Orexigen broke the rules. Not just once, but twice. The rule breaking helped Orexigen’s stock price go from around $5.75 to $9.37 in one day and probably contributed to an increase in sales for Contrave. So it is clear that breaking the rules was of finance benefit to Orexigen.
    Trusting Orexigen with future interim data is obviously out of the question. Contrave’s approval was based on the FDA getting their data by 2017, now it will be 2022 at the earliest.
    What do you think should happen to Orexigen? What would be fair?

  31. matt says:

    In the post, it indicates that perhaps Orexigen was blaming USPTO for early release of data. I don’t think that’s the case. Instead, they are ignoring their prior release of the 25% data (with some handwaving that of course they said it was preliminary data but look at the patent we got based on that preliminary BIG BENEFIT TO CV HEALTH), and trying to blame Steve Nissen for releasing the 50% data without full analysis. Now that results don’t really favor them, they require very thorough and detailed analysis before release, and perhaps the public never needs to worry its pretty little head about it.
    Despite the remarks about the CEO gluing the clown hat to his head, the press release is a rather glorious piece of bullshit indeed. Somebody polished that turd so bright you could use it for a telescope mirror. So well, I think, that Matthew Herper had to include statements from Takeda and several background facts to help pierce its powerful reality-distortion field.

  32. Hap says:

    Perhaps a good analogy might be to a mistrial: if you do things you’re not supposed to in a court case, you have to start over if you want to prosecute (assuming the case is not dismissed outright). The trial doesn’t get to continue just because you’re likely to have won – you get to do it over because you messed up, and perhaps next time you’ll do it right.
    Oh, if it’s not obvious, I am not a lawyer.

  33. Kevin says:

    The FDA’s decision is based on what’s good for the entire clinical trial ecosystem, and not solely this particular drug or subset of patients. If they let Orexigen get away with these sorts of shenanigans on a wink and a nod this time around, then every company with mildly positive incomplete study data is going to be leaking like crazy.
    So the FDA had two options:
    1) Delay the (potential) approval of another mildly interesting weight-loss drug (that probably doesn’t have any important CV effects one way or the other) for a few years to ensure that Orexigen handles their study data properly; or
    2) Incentivize every small company to leak loudly and irresponsibly whenever they have the opportunity, to…um…encourage the interest of patients, physicians, and the stock market.
    I can see the FDA might have taken a different tack if this were a first-in-class drug for something deeply deadly and untreatable–but this isn’t that. If a company is going to be this seemingly reckless and irresponsible with their preliminary trial data, yes, it’s the FDA’s role to smack them down hard.

  34. Humble Scrivener says:

    @33 Preach it, Kevin! In the whole clinical trial ecosystem, the FDA will get what it tolerates. Tolerating a non-prespecified data disclosure that materially affects equipoise could encourage other applicants to push this and other envelopes. Since ultimately the patients will get what the FDA tolerates, I think the FDA was justified.
    To say nothing of the ethics of publication by investor meeting; it is conceivable that a journal might regard the data in the eventual primary manuscript as predisclosed without peer review under these circumstances.

  35. Joshua Belanger says:

    This is just bizzare. I never expected to see those company names or those drug names again. I recognized the company names Orexigen and Takeda, and the names of the component drugs of Contrave immediately because about three years ago (2012) I, via a temp agency, worked for Academic Network (a subsidiary, as I recall, of Stericycle, though that’s not terribly relevant here), which was the company doing the pre-screening for the now aborted clinical trial mentioned here. I remember spending a lot of time asking callers if they were currently taking any opioid medications. (Naltrexone reacts badly, to put it mildly, with opioid analgesics,) In any case, it’s astonishing to be even that remotely connected to a screw-up this big.

  36. Sean McDonough says:

    bio_kruncher: Wait, isn’t it illegal to make the drug available to prescribe before full trials have been completed?

Comments are closed.