EMA workshop on clinical trial data and transparency 22 Nov 2012: a landmark meeting

[Richard Lehman]

Dear All,

Apologies for my relative silence over the last few weeks, which have been momentous for the cause of open data sharing.

Yesterday there was a key meeting which heralded the adoption of an open data policy across Europe by January 2014. The representatives of the European pharmaceutical industry were in a dificult position, as they had come to argue for restriction of data disclosure, whereas the EMA speakers emphasised from the start that this was a meeting to discuss how to release data not whether to release it. 

I attach an incomplete but I hope helpful summary of the main points of the meeting. There were stellar perfomances from three PATH members on the panel, and excellent questions from several others from the floor. If you were at the meeting, please add your recollections and comments. If you were not, it would also be good to hear how you feel we should proceed as a group in the face of such rapid progress. I would particularly urge all members based in Europe to consider taking part in the consultation process at committee level: please keep an eye on the EMA website.

All best,

Richard

[Trish Groves]

Thanks, Richard
Very helpful.
I'm blogging about the meeting for bmj.com today, and we have a news story coming and a nice podcast with Beate Wieseler, Gibnny Barbour, and Ben Goldacre going up on bmj.com later today.

There's one big question I'm not sure we got an answer to yesterday, not least because EMA gave us nothing in writing or verbally beyond this:

Eichler H-G, Abadie E, Breckenridge A, Leufkens H, Rasi G (2012) Open Clinical Trial Data for All? A View from Regulators. PLoS Med 9(4): e1001202. doi:10.1371/journal.pmed.1001202
and this:
http://www.ema.europa.eu/ema/index.jsp?curl=pages/special_topics/general/general_content_000555.jsp&mid=WC0b01ac0580607bfa

The question is:

Will EMA enact this retrospectively, for all drugs they've approved? or only prospectively? They use the term "proactive publication". Hmm.

And, although, this is very encouraging it doesn't tell us whether they'll do this across the board:
Tom Jefferson: Does the EMA have the power to demand release of undisclosed data from Roche about Tamiflu? EMA: yes

I can see that doing it retrospectively would require resources they probably don't have. And starting with prospective disclosure would be a lot better than nothing.
But it's the retrospective data that we and patients need.

bw
Trish

Dr Trish Groves
Deputy editor, BMJ
and Editor-in-chief, BMJ Open

+44 207 383 6018
bmj.com - the open access journal
bmjopen.bmj.com
BMJ Group: http://group.bmj.com
twitter @trished

______________________________________________________________________
This email has been scanned by the Symantec Email Security.cloud service.
For more information please visit http://www.symanteccloud.com
______________________________________________________________________[attachment "EMAdatasharing2012.docx" deleted by Trish Groves/BMJ]

_______________________________________________________________________
The BMJ Group is one of the world's most trusted providers of medical information for doctors, researchers, health care workers and patients group.bmj.com. This email and any attachments are confidential. If you have received this email in error, please delete it and kindly notify us. If the email contains personal views then the BMJ Group accepts no responsibility for these statements. The recipient should check this email and attachments for viruses because the BMJ Group accepts no liability for any damage caused by viruses. Emails sent or received by the BMJ Group may be monitored for size, traffic, distribution and content. BMJ Publishing Group Limited trading as BMJ Group. A private limited company, registered in England and Wales under registration number 03102371. Registered office: BMA House, Tavistock Square, London WC1H 9JR, UK.
_______________________________________________________________________

[Richard Lehman]

Thanks Trish - an extremely important point.
The purpose of all this is to help clinicians reach informed decisions with their patients about interventions which are already available. The elephant in the room is that we cannot do this properly for lack of data which exist but are hidden.
It's the point we need to make time and again. At least the politicians are listening.

Richard

[Ben Goldacre]

i think the idea that this will only be prospective, like all other failed initiatives in publication bias and misreporting, is very concerning. it means that nothing changes until a whole generation of drugs has passed, and the clock keeps getting put back when the initiatives prove to be improperly implemented, which we only discover late, because like with clinicaltrials.gov there is no rolling public audit of compliance.

[Trish Groves]

Thanks, Ben and Richard
So neither of you were clear about EMA's intentions for this, either?
Oh dear; I hoped it was just me being dim.
Trish

[Tom Jefferson]

The meeting was about proactive stuff. The retroactive is going out without the policy.

Trish - I never mentioned TF by name in my question. Please edit accordingly.

Tom

Sent from my iPhone

[Ben Goldacre]

afaict they didnt say either way

their decision on this will be an indicator of whether they are keen to fix the problem, or simply repair their reputation after the Ombudsman ruling of maladministration etc. 

[Ben Goldacre]

thank richard

the thing i said (about five times, to my shame) was: industry is claiming that people who analyse their own data do so with immaculate competence, and that secondary analyses may not be so competent. in reality we know that researchers' analyses of their own data are frequently flawed: they break protocols, switch primary outcomes, and fail to public specify analytic strategy. if researchers doing trials, and researchers downloading data from trials, were both expected to post their full protocols etc in advance, on an equal playing field, then we'd help prevent bad analyses in both camps. 

[Richard Lehman]

No, it's my shame that I don't make that sufficiently clear in my report. You were terrific on this. Unfortunately my notes were mostly about stuff that brought up competing or unexpected perspectives.

[Iain Chalmers]

We should be concerned about two categories of evidence about which clarity does not yet exist:

 

Mark Walport clarified with one of the lawyers that there would be access to data available at the EMA about drugs that had not been granted marketing licences as well as about drugs that had been licensed. This was good, but it leaves unaddressed data from those trials which led to a decision to abandon development of a drug, and that were thus never used in support of applications for marketing licences.  The example of a commercial decision not to proceed with lorcainide is one of the most persuasive examples of the contribution that non-publication can make to a major iatrogenic disaster. See: http://www.jameslindlibrary.org/illustrating/records/the-effect-of-lorcainide-on-arrhythmias-and-survival-in-patients/key_passages

 

With regard to ‘unpublic’ data about drugs already in use, I think a first stage is to try to extend (retrospectively) the public registration of the relevant trials, and to add the registration references to the ‘Studies awaiting assessment section of the relevant Cochrane Reviews, and to arrange for there to be a link from the Cover page of the review/protocol as a ‘buyer beware’ warning (Carl Heneghan came up with this idea during our return journey to Oxford).

 

There are already examples of the value of getting aggregate data on unpublished trials of treatments already in use. As people with experience of doing this know only too well, however, obtaining IPData on trials done some years ago presents a challenge which often can’t be met by people who have done trials, whether commercial or non-commercial.  .

 

Finally, I don’t think the discussion on concealing patient identity really bottomed out yesterday. During a brief discussion after the end of proceedings, Mike Clarke and Lesley Stewart challenged Peter Goetzsche’s view that there was no reason for concern.  I don’t know enough about this but I have the impression that Francois had a point that was not explored sufficiently deeply.

 

Best wishes, Iain  

 

---

["Peter C. Gøtzsche"]

Dear Iain,

trials that will never get to EMA because further development was abandoned will not be accessible via EMA, but the idea is that they will become accessible via the new EU portal, through which all drug trials in the EU will need to be registered. I have submitted a critique on the the proposed new Regulation of the Clinical Trials Directive to BMJ and much hope they will accept and publish quickly, as we shall use the arguments in Parliament during the coming months till April in order to improve on the Regulation, which was made by the Commission, which is incredibly lobbied by the drug industry.

Richard, can you send me your minutes from yesterday, I never got them, or your email? Electrons behave unpredictably.

bw

Peter

At 13:49 23-11-2012, Iain Chalmers wrote:

We should be concerned about two categories of evidence about which clarity does not yet exist:
 
Mark Walport clarified with one of the lawyers that there would be access to data available at the EMA about drugs that had not been granted marketing licences as well as about drugs that had been licensed. This was good, but it leaves unaddressed data from those trials which led to a decision to abandon development of a drug, and that were thus never used in support of applications for marketing licences.  The example of a commercial decision not to proceed with lorcainide is one of the most persuasive examples of the contribution that non-publication can make to a major iatrogenic disaster. See: http://www.jameslindlibrary.org/illustrating/records/the-effect-of-lorcainide-on-arrhythmias-and-survival-in-patients/key_passages
 
With regard to ‘unpublic’ data about drugs already in use, I think a first stage is to try to extend (retrospectively) the public registration of the relevant trials, and to add the registration references to the ‘Studies awaiting assessment section of the relevant Cochrane Reviews, and to arrange for there to be a link from the Cover page of the review/protocol as a ‘buyer beware’ warning (Carl Heneghan came up with this idea during our return journey to Oxford).
 
There are already examples of the value of getting aggregate data on unpublished trials of treatments already in use. As people with experience of doing this know only too well, however, obtaining IPData on trials done some years ago presents a challenge which often can’t be met by people who have done trials, whether commercial or non-commercial.  .
 
Finally, I don’t think the discussion on concealing patient identity really bottomed out yesterday. During a brief discussion after the end of proceedings, Mike Clarke and Lesley Stewart challenged Peter Goetzsche’s view that there was no reason for concern.  I don’t know enough about this but I have the impression that Francois had a point that was not explored sufficiently deeply.
 
Best wishes, Iain 
 

Dear All,

Apologies for my relative silence over the last few weeks, which have been momentous for the cause of open data sharing.

Yesterday there was a key meeting which heralded the adoption of an open data policy across Europe by January 2014. The representatives of the European pharmaceutical industry were in a dificult position, as they had come to argue for restriction of data disclosure, whereas the EMA speakers emphasised from the start that this was a meeting to discuss how to release data not whether to release it. 

I attach an incomplete but I hope helpful summary of the main points of the meeting. There were stellar perfomances from three PATH members on the panel, and excellent questions from several others from the floor. If you were at the meeting, please add your recollections and comments. If you were not, it would also be good to hear how you feel we should proceed as a group in the face of such rapid progress. I would particularly urge all members based in Europe to consider taking part in the consultation process at committee level: please keep an eye on the EMA website.

All best,

Richard

______________________________________________________________________
This email has been scanned by the Symantec Email Security.cloud service.
For more information please visit http://www.symanteccloud.com
______________________________________________________________________[attachment "EMAdatasharing2012.docx" deleted by Trish Groves/BMJ]

_______________________________________________________________________
The BMJ Group is one of the world's most trusted providers of medical information for doctors, researchers, health care workers and patients group.bmj.com. This email and any attachments are confidential. If you have received this email in error, please delete it and kindly notify us. If the email contains personal views then the BMJ Group accepts no responsibility for these statements. The recipient should check this email and attachments for viruses because the BMJ Group accepts no liability for any damage caused by viruses. Emails sent or received by the BMJ Group may be monitored for size, traffic, distribution and content. BMJ Publishing Group Limited trading as BMJ Group. A private limited company, registered in England and Wales under registration number 03102371. Registered office: BMA House, Tavistock Square, London WC1H 9JR, UK.
_______________________________________________________________________

Peter C. Gøtzsche, Professor, Director, MD, DrMedSci, MSc
Nordic Cochrane Centre, Rigshospitalet, Dept. 7811
e-mail: p...@cochrane.dk, Tel: +45 35 45 71 12, Fax: +45 35 45 70 07
____________________________

NEW BOOK:
Mammography screening: truth, lies and controversy.
London: Radcliffe Publishing; 2012.
See www.cochrane.dk
Featured, for example, in The Guardian
JAMA (16 May 2012) and BMJ (17 May 2012)
__________________________

Visiting address: Tagensvej 22, 1st floor

["Peter C. Gøtzsche"]

Dear all,

dont be so negative! At the meeting, I explained that I could get access to the clinical documentation for three SSRIs from the Swedish Drug Agency, unredacted, but also that it took up 70 m of paper that were deposited in a mountain cave (you usually put explosives and nuclear waste in caves, so it makes sense to me to put the explosive data there too).

It is impossible for the EMA to retroactively make many many many many tons of paper avaible for the thousands of drugs we have on the market. Let's be happy that we can get access to all this if we ask for it!

bw

Peter

At 13:23 23-11-2012, Ben Goldacre wrote:

afaict they didnt say either way

their decision on this will be an indicator of whether they are keen to fix the problem, or simply repair their reputation after the Ombudsman ruling of maladministration etc.Â

Thanks, Ben and Richard

Trish

Thanks, Richard

Very helpful.

and this:

http://www.ema.europa.eu/ema/index.jsp?curl=pages/special_topics/general/general_content_000555.jsp&mid=WC0b01ac0580607bfa

The question is:

bw

Trish

Dr Trish Groves

Deputy editor, BMJ

BMJ Group: http://group.bmj.com

Sent by:       Â

pathtodata...@googlegroups.com

Dear All,

Apologies for my relative silence over the last few weeks, which have been momentous for the cause of open data sharing.

Yesterday there was a key meeting which heralded the adoption of an open data policy across Europe by January 2014. The representatives of the European pharmaceutical industry were in a dificult position, as they had come to argue for restriction of data disclosure, whereas the EMA speakers emphasised from the start that this was a meeting to discuss how to release data not whether to release it. Â

I attach an incomplete but I hope helpful summary of the main points of the meeting. There were stellar perfomances from three PATH members on the panel, and excellent questions from several others from the floor. If you were at the meeting, please add your recollections and comments. If you were not, it would also be good to hear how you feel we should proceed as a group in the face of such rapid progress. I would particularly urge all members based in Europe to consider taking part in the consultation process at committee level: please keep an eye on the EMA website.

All best,

Richard

______________________________________________________________________

This email has been scanned by the Symantec Email Security.cloud service.

For more information please visit http://www.symanteccloud.com

______________________________________________________________________[attachment "EMAdatasharing2012.docx" deleted by Trish Groves/BMJ]

_______________________________________________________________________

The BMJ Group is one of the world's most trusted providers of medical information for doctors, researchers, health care workers and patients group.bmj.com . This email and any attachments are confidential. If you have received this email in error, please delete it and kindly notify us. If the email contains personal views then the BMJ Group accepts no responsibility for these statements. The recipient should check this email and attachments for viruses because the BMJ Group accepts no liability for any damage caused by viruses. Emails sent or received by the BMJ Group may be monitored for size, traffic, distribution and content. BMJ Publishing Group Limited trading as BMJ Group. A private limited company, registered in England and Wales under registration number 03102371. Registered office: BMA House, Tavistock Square, London WC1H 9JR, UK.

_______________________________________________________________________

______________________________________________________________________

This email has been scanned by the Symantec Email Security.cloud service.

For more information please visit http://www.symanteccloud.com

______________________________________________________________________

_______________________________________________________________________

The BMJ Group is one of the world's most trusted providers of medical information for doctors, researchers, health care workers and patients group.bmj.com. This email and any attachments are confidential. If you have received this email in error, please delete it and kindly notify us. If the email contains personal views then the BMJ Group accepts no responsibility for these statements. The recipient should check this email and attachments for viruses because the BMJ Group accepts no liability for any damage caused by viruses. Emails sent or received by the BMJ Group may be monitored for size, traffic, distribution and content. BMJ Publishing Group Limited trading as BMJ Group. A private limited company, registered in England and Wales under registration number 03102371. Registered office: BMA House, Tavistock Square, London WC1H 9JR, UK.

_______________________________________________________________________

[Iain Chalmers]

Thanks, Peter. Iain

 

---

[Ben Goldacre]

I agree, documenting the gaps is extremely important. I've always been surprised that Cochrane reviews, while producing funnel plots etc, don't also routinely contain a list of trials they know exist but cannot access.

As I understand it, the EU portal will not contain information on trials conducted outside the EU, eg in BRIC nations. 

From all discussions on this issue, I'm struck by how many gaps there are in this non-overlapping patchwork of legislation, and legislative wishes. 

To add another patch: as a minimum, I think all nations should demand, as an intermediate measure to full disclosure of results and data, that all companies provide a list of all trials conducted for all uses (licensed and unlicensed) of all drugs they sell, before they are allowed to sell them. But this is still only companies. The levers for non-industry academics will be different.