Life Sciences

EMA Policy 0070: Teams and Responsibility

Share:
Life Sciences

"Policy 0070 has put the frighteners on Regulatory Affairs teams, especially those charged with data transparency, because they’ll be to blame if a patient is identified from a public-facing clinical study report," warns Cathal Gallagher at d-Wise.


If you’re being paid £35,000, (or $44,000US), a year to do the job you’ve always done in Regulatory Affairs, you’re unlikely to welcome the knowledge that you’re all that stands between your employer and a fine worth 4-6 percent of the company’s global income.

But this is the situation potentially facing many internal regulatory personnel - particularly those with responsibility for data transparency/data sharing - as life sciences businesses prepare their public-facing clinical study reports (CSRs) to comply with new EMA guidance under Policy 0070[1], which took effect in October.clinicaltransparencysecurity.png

Because straightforward redaction is now discouraged in favour of more systematic ‘anonymisation’ of clinical trial participants, companies must tread a fine line between disguising subjects sufficiently, and maximising the CSR’s value to external parties. If firms fail in the first requirement - their duty to protect an individual’s right to anonymity – the penalties could run into the millions.

Previous EMA requirements permitted redaction. In addition, associated data transparency roles were chiefly concerned with responding to specific requests for clinical trial information from internal users or other researchers (Policy 0043). This meant they were covered by agreed legal limits around information use, safeguarding subjects’ identities. With Policy 0070, all CSRs should be made available, up front - in support of any future public request - so the risks are much higher.

Without trace?

This is a big deal for those responsible, and the whole task (of risk avoidance) is much more complicated than it sounds. One of the requirements will be to look for risks of subject identification across the board – in anything produced by medical writers, including content already printed in publications. Teams will need to consider what’s already been shared with other researchers - what’s already in the public domain.

As they bolster their resources, pharma companies will need to lay down very strict guidelines about what’s expected, because of the higher risks involved in data transparency roles. But regulatory skills are hard enough to come by and this isn’t going to make recruitment any easier – or cheaper.

clinical-1.pngIt’s a lot to take on board, especially when teams and firms put the new requirements into context. The biggest clinical study data request site, Clinical Study Data Request, has received just 254 requests for CSR documents under the existing EMA data-sharing initiative. When I asked a biotech researcher why this might be, he pointed to (a) a lack of awareness (i.e. of the chance to access other firms’ clinical trial content), and (b) to scepticism about the content’s likely value (at the end of what could be a very protracted requisition process).

Policy 0070 is a bolder proposition, however. The EMA’s intentions are to increase public trust in life sciences, whatever it takes. That’s why it is insisting on anonymisation rather than blacking out whole swathes of documents, in order to retain the integrity of the original research, while still offering patients reassurance of anonymity.

Companies must get on top of this.  Equally, the EMA needs to be clearer and bolder in its demands. Policy 0043 was made law, and it may be that the Agency’s only option is to take Policy 0070 the same way, if it is serious about taking transparency to the next level.

Avoidance isn’t the answer

In the meantime, there are other risk factors to consider, not least that companies could inadvertently give away trade secrets, disclosing company confidential information (CCI) about how drugs are made, as they publish more information externally. This further increases the pressure on responsible parties and requires that a broader range of business stakeholders, including legal professionals, get involved – pinpointing potential areas of vulnerability.

Resistance is futile: there is little point putting off the inevitable. The only way firms and their teams can protect themselves is to implement strict processes and guidelines, and manage all of this systematically, so that nothing is left to chance. When a lot of the required content is narrative in nature, this won’t be easy -  not without heavy reliance on ‘comment banks’ whose treatment (with de-identification rules) could be automated. But it is better to start from the ground up.

The alternative is that medical writers will find themselves producing content for two audiences – doubling their workload and their stress levels, and requiring at the very least that they are re-trained.

There’s going to be a big bill to pay, either way.

 

[1] Policy 0070 is the latest transparency initiative of the EMA to publish clinical reports assessed as part of a Marketing Authorisation Application, with a view to improving the transparency of assessment and aiding the development of new knowledge. First publications were due from October 2016 (http://www.ema.europa.eu/docs/en_GB/document_library/Presentation/2016/06/WC500209454.pdf)  

 

blur-screens.png

Clinical Trial De-identification

To see advance highlights of Blur’s latest functionality, please visit: http://www.d-wise.com/blur-clinical-trial-data-de-identification