Friday, 3 March 2017

Bad Pharama. It is important that Failed trials are reported

If you are interested to find examples of the consequences of not reporting the results of clinical trials, then look no further than
Ben Goldacre and his book Bad Pharma (2012)

If you saw the recent BBC programme about the clinical trial (anti-CD28 antibody) that went wrong (what a pile of Pants! the programme), then the alarm bells should have been ringing.

This is because doing a similar thing was tried years before in humans and caused the same problem but the academic (not pharma) doing that trial did not report and publish the work.

This could have saved six guys from a lot of problems and one guy would have their fingers and toes still.

Ben Goldacre suggests a number of things to encourage pharma to publish the results within 1 year of completion of the trial,

I say "If the regulators implemented a requirement to publish the trial data on clinical trial.gov or they could not file for another trial. This is a stick to make change as carrots don't seem to work. Would it cause all trials to be published?

There is a UK  website called research fish were each UK government grant has to have output recorded for 5 years after completion of the grant, with the threat of no more funding if you don't complete it...will they have teeth to do this..

Many trials don't complete on time, and then would have the data appearing on clinical trials.com before publication of the trial results. However these aspects could be addressed. Would a stick work 

Probably not, because the (UK) regulators are paid for by pharma and are not going to bite the hand that feeds them.

If the Eli Lilly failed, it is informative, if it made people worse it is informative and then scandelous that it was not reported.

The Merck trial of ataceicept (anti- BAFF & APRIL) was reported as recruiting  in March 2008 and it was terminated, because if made people worse in Sept 2009.

The Lilly trial on talabumab (anti-BAFF) was reported starting April 2009 and recruiting was reported complete in Sept 2010. So it was recruiting a full year after the atacicept trials...yes there were two, were terminated.

Their is data not yet published from the atacicept trials that may allow us to try and understand the apparent failure better, we know this becuase of the protocol.

5 comments:

  1. If the regulators bothered to start clinicaltrials.gov (because too many trial results were swept under the carpet by businesspeople to make drug effects look more pleasing to the eye), then there has to be a system of getting the results out.
    1. Every trial has to be registered if you ever want to submit the result in application to FDA/EMA - that part works
    2. Every result should be reported within max a year of a trial completing recruitment. Good results, sweet nothing results, bad results, reasons for having to stop a trial, who intervened to stop a trial.

    It is all important and all good for learning.

    ReplyDelete
  2. Horrific to learn the history behind the anti CD28,saw it on the news but didn't know that. Six healthy volunteers, awful. And it could have been repeated with tabalumab (on a less horrific scale) wtf didn't Merck make an urgent 'trial terminated as made MS worse' public notice? Ok, I know that's not how it works, just saying what should ideally happen.
    The tabalumab thing has really got to me (in case you haven't noticed). Please, please keep it high up in the blog and give the petition time to gather support. Eg my initial shotgun share on social media yielded a spattering of results, but it will take longer to compose personal targeted messages to contacts working in science, medical professions and the politically minded ethical bunch to get involved. Multiply me up across the blog following, it will happen.

    ProfG is pretty good with the political diplomacy stuff, suspect not your best strength MD? ;-) Feel free to rant on here, I think I might join you ;-)
    Just popping to Amazon (or a more ethical online book store) to get a copy of that book....

    ReplyDelete
    Replies
    1. Correct..diplomacy not my best strength as we will soon see :-)

      The effect seen with anti CD28 was predictable. Mice do not usually succumb to the problem as they are designed to live in bacterial infested environments. Do the same to a guinea pig and you have a dead guinea pig.

      In the scenario it was parixcel that took the blame. These are just a testing company, a German company was behind the bonkers idea.
      In the BBC programme the NHS is the hero (maybe so) but as soon as CD28 was mentioned I would not have been thinking about an infected vial, if they had known what CD28 does. This was never properly explained in the programme and that is one of the reasons I was kicking the television when it was broadcast

      However, you could see the problem with anti CD3 T cell stimulating antibody add CD3 to CD28 and you can grow T cells indefinately.

      We have 5 requests out for data from 4 different companies, two have said no. Two yes and one is thinking about it..it has only been 4 months from the first request.

      Lilly is one of them. We have gone back to the other one if they say no again then I will out them as bad phamra.

      One may feel more obliged to release data after they are outed as hiding information....so we will bring Lilly back into eyesight about why it is important that pharma are open.

      What is it?....Wait and see it will arrive once the embargo is lifted.

      Delete
  3. I tweeted Ben G your blog post yesterday :-)

    ReplyDelete

Please note that all comments are moderated and any personal or marketing-related submissions will not be shown.