- Concerns over age extension trial of mammography screening - part 1 (26 May 2014)
- New Scientist article on breast cancer screening age extension trial (10 Nov 2014)
- Concerns over age extension trial of mammography screening - part 2 (14 Jan 2015)
- Concerns over age extension trial of mammography screening - part 4 (15 June 2015)
Our intensive lobbying has achieved substantial results, albeit still incomplete. Main points:
- The proposed new patient information sheet will now actually tell women that they are entering a clinical trial.
- The Harrow Research Ethics Committee approved* a substantial amendment to the protocol, on 14th October 2014 This amendment is incorporated in the new protocol, version 3.
- The protocol now runs to 11 pages, instead of eight originally. It still falls far short of Good Clinical Practice (GCP) compliance; for example there is nothing about quality assurance or publication policy, and in general it doesn’t come close to following the ICH format. There is now a little detail on data processing, and there is a statistical power calculation (although based on contentious assumptions). The sponsor has refused permission for HealthWatch to publish this version of the protocol. This is strange in view of the current moves towards transparency of commercially sponsored trials.
- At the REC meeting, the investigators stated that full roll-out of the age extension will be dependent on the results of this trial. However the trial website still states that “The age extension will proceed regardless of whether this study goes ahead or not, and therefore regardless of whether the phasing-in is randomised or not.”
* We are using the informal terms `approved’ and `approval’ for clarity and brevity. The official term is `favourable opinion’.
We have obtained the following documents:
- Protocol version 3 (redacted)
- Data Monitoring and Ethics Committee (DMEC) minutes dated 8th November 2011, 9th January 2013, and 28th January 2014
- Minutes of the REC meeting of 14/10/2014.
- Revised patient information sheet dated May 2014
We have appealed the REC’s decision, on the following main grounds:
Abuse of process
As the trial team now says that full roll-out of screening in the extended age range will wait for the trial results, it’s clear that they are now in equipoise. The rationale for the original protocol was that screening was accepted as beneficial, ie they were not in equipoise. This is a major change, and the REC should have reviewed the original approval. If the rationale has changed, the original hypothesis was flawed. We have other concerns about processes.
The characterisation of the research
The REC minutes refer to an “epidemiological” study. It is not. It is a clinical trial. In addition, the researchers claim that the scientific rationale is unchanged, but see `abuse of process’ above.
Failure to comply with Good Clinical Practice
The chief executive of Public Health England, Duncan Selbie, has stated that “we…will ensure that we continue to operate within the requirements for clinical trials as they apply in England”. We believe he has thus committed PHE to compliance with Good Clinical Practice as defined by the International Conference on Harmonisation of Technical Requirements for Registration of Pharmaceuticals for Human Use (ICH). The EU has adopted ICH GCP and member states have in turn transposed EU law into national law.
The protocol remains flawed
The new protocol is now 11 pages long with 11 references (virtually entirely self-citation). It still does not appear to have sought or received the benefit of peer review. The background is descriptive, rather than making the case for a trial and contains an odd ‘opportunistic’ justification. There is no research question, just an assertion under ‘Aim’ that “The cluster-randomised Age Extension Trial will assess reliably the risks and benefits of offering an extra screening invitation”.
The protocol design and implied consent
We do not accept that the case for continuing with implied consent was made well enough (indeed it was not made at all in the ‘new, improved’ 2014 protocol).
The outcome measures
There is no proper primary outcome measure. It is extraordinary, given the planned size, that the researchers have not chosen all-cause death as an endpoint, nor any other measures of harm. If matters are uncertain, surely the chance of causing more harm than good would be missed by such a biased approach? All-cause death is not mentioned in the Data Collection section although it is in the summary.
Misleading information given to the REC
If the chief investigator stated verbally to the REC (as per minutes) that “1 life is saved for every 200 women screened”, then the committee was misled.
Inadequate reassurance of oversight by the sponsor
We are worried that the REC did not explore the role of the sponsor who is responsible for implementing and maintaining quality assurance and quality control systems with written SOPs to ensure that trials are conducted and data are generated, documented (recorded), and reported in compliance with the protocol and applicable external requirements. Indeed the sponsor does not seem to have a clear idea of how many women have been enrolled; the DMEC minutes in 2014 estimated that 0.8m women are already enrolled, the REC minutes state 0.5m, and yet the patient information sheet says 1.5m.
Inadequate oversight by Data Monitoring and Ethics Committee
There is no description of this in the protocol, and there is no ethical oversight regarding the change in rationale. What are the monitoring systems for looking for adverse events? No stopping rules have been set up, either for futility or for harm.
The information women receive remains unclear and misleading
The proposed PIS is somewhat more honest and informative than before. It should be a ‘tailored’ invitation leaflet for the trial. It is wrong to post it out along with the accompanying NHS leaflet. The screening programme and the trial are being presented to women together, when they are actually separate.
The question of the chief investigator being a fit and appropriate person was inadequately addressed
ICH GCP requires that all trial personnel must be qualified by education, training and experience to carry out their assigned duties. The REC minutes show that the CI was not asked about these. The REC simply asked her to define her job, which actually has nothing to do with the clinical research trial.
The REC gave approval for the substantial amendment, but without seeing the new protocol version that would embody it. Nor did it give any deadline for issuing the new protocol. Meanwhile the existing protocol – by virtue of the present REC decision, accepted as flawed – was allowed to remain in force.
We also provided a detailed critique of the trial according to ICH GCP requirements. Our appeal letter was sent on 17th December 2014, and has been acknowledged.
Overall, it’s good to see that some of the flaws in this trial have been addressed, though it is an indictment of current ethical oversight that none of this would have happened without HealthWatch persistently pressuring all parties. The chief investigator in particular has not been willing to engage with us, in stark contrast to the Health Research Authority which has been consistently helpful. Responses from Public Health England have been very slow and patchy, and we have had to engage at several different levels. The sponsor, the University of Oxford, has been cordial but as we have reported here, this looks like more of an administrative position, without the resources normally expected of a real sponsor.
We are grateful for your feedback and interest and will keep HealthWatch members (“for science and integrity in medicine”) informed of further progress as it happens.