Cookies on this website

We use cookies to ensure that we give you the best experience on our website. If you click 'Accept all cookies' we'll assume that you are happy to receive all cookies and you won't see this message again. If you click 'Reject all non-essential cookies' only necessary cookies providing core functionality such as security, network management, and accessibility will be enabled. Click 'Find out more' for information on how to change your cookie settings.

11:00 – 13:00, Cancer Epidemiology Unit, Richard Doll Building, University of Oxford

Present

Professor Julietta Patnick (Chair)

Professor Dame Valerie Beral

Dr Lucy Carpenter Kath Moser

Professor Sir Richard Peto

Professor Amanda Ramirez

Professor Malcolm Reed Professor

Sir Mike Richards Sarah Sellars

Margot Wheaton

Dr Robin Wilson Richard Winder

In attendance

Professor Tom Meade

Dr Gillian Reeves

Documents circulated in advance:

  •  agenda
  • draft protocol (Jan 2011)

Plus for background information:

  • original application for ethical approval
  • patient information leaflet
  • final report of the pilot study

1. Welcome and introductions

2. Background (JP)

The background to the trial was outlined.

Originally randomisation was planned to continue over three years i.e. through one screening round. Now, as announced in January 2011 in Improving Outcomes: A Strategy for Cancer, randomisation will continue for at least six years i.e. over two screening rounds.

Several breast screening units (approx 10% of all units) are not able to participate in the trial:

  1. Gateshead (unsuitable method of batch creation)
  2. Several East Anglian units (unsuitable method of batch creation)
  3. North Devon (private sector unit)
3. Results from the pilot study (KM)

KM tabled a 2-page summary of the pilot study.

The pilot study included 60,000 women across 5 pilot breast screening units. Overall, the pilot was deemed successful and it has provided a firm basis from which to proceed with the main trial. Although women randomised out could request to be screened very few did so. Screening uptake was lower than hoped for. [Pilot study results have subsequently been published: Moser et al. Extending the age range for breast screening in England: pilot study to assess the feasibility and acceptability of randomisation. J Medical Screening 2011;18:96-102]

The following issues were discussed:

a. Women randomised out.

Individual women who were randomised out were not informed of this, nor that they were part of a trial. This approach has ethical approval. Women randomized out can, nevertheless, request screening and their GPs and breast screening units were informed that any woman randomised out could ask to be screened.

b. Digital mammography

Installing digital equipment across the country is proceeding slowly. Although most screening units now have one digital machine this is mostly used for assessment rather than screening. Most women in the trial to date will have been screened using analogue equipment; this is not ideal for the trial. It is expected that the use of digital machines for screening will increase over time and so, with randomisation now continuing over 6 years, a higher proportion of women will be screened with digital equipment in the second screening round than the first. A record needs to be kept of whether screening was performed using digital or analogue equipment.

There was    discussion on the     advantages   and   disadvantages of   digital as compared with analogue mammography.

Action: SS to circulate papers on the benefits and disadvantages of digital vs analogue equipment

c. Uptake rates

Screening uptake was 62% in the pilot study. This is lower than the uptake of 74% for women aged 50-70 invited for routine screening.

There was discussion on how uptake could be improved as the difference between an uptake of 60% or 70% would make a large difference to the power of the main trial:

  1. Monitoring which screening invitation letters did not reach the addresee. What percentage of women do not live at the address the invite goes to? Could look at population movements in/out of area. It is not in the interest of GP consortia to delete ghost patients. This is an issue to revisit at a future meeting.
  2. Offering an ‘out of hours’ service. Research in Manchester and Bristol some years ago indicated that offering an ‘out of hours’ service to previous non-attenders did not improve uptake. This was now being revisited to see if attitudes had changed.
  3. Convenient access and car parking.
  4. Using text messaging/emailing to remind women of appointments had been tested although they did not specifically consider uptake of non-attendees. It was suggested that non-attendees could be randomised to see if these techniques made a difference.
  5. As many people do not work in the same area as they live, offering women screening in areas other than their home address may improve uptake especially for younger women. This approach will be coming into effect soon in some areas.
  6. Past research has established that determinants of uptake of breast cancer screening include post code, car ownership, and previous screening history.
d. ‘false positives’ and terminology

The term ‘negative assessment’. is preferred over ‘false positive’.

Action: Avoid use of ‘false positive’ on trial documents and publications.

The preliminary results of the pilot indicate that women aged 47-49 had a higher ‘negative assessment’ rate than women aged 71-73. It would be interesting to examine whether having a negative assessment discourages women from subsequent attendance at breast screening.

4. The main trial

a)    Additional ethics approval (JP)

Since obtaining ethical approval for the trial in February 2010, additional approval has been obtained to:

  1. extend the trial to cover two screening rounds i.e. 6 years
  2. include data from the pilot study in the main trial

b)   protocol (JP)

Since making the original application for ethical approval, it has become apparent that a more refined analysis is required. A revised protocol had therefore been drafted for discussion at the meeting. This included several differences form the original ethics application - specifically linkage with:

  • historic screening records
  • HES data
  • other, as yet unspecified, NHS data

Although it is known that attending previous screening invitation is strongly related to attending current invitation, it may be hard to justify looking at attending cervical screening as a predictor of attending breast screening as cervical screening is so much more invasive than breast screening. It was suggested we look at data on historic screening and other determinants of screening uptake, to see what is shown, and then decide how to proceed with main trial analysis.

Action: KM to do exploratory analysis looking at predictors of screening uptake including historical screening acceptance, postcode, car ownership etc. It may be possible to use data from the pilot study for this analysis.

Action: KM to amend protocol

  1. to say the analysis will be done firstly including all women, and subsequently restricted to women most likely to attend screening. The way in which the latter group of women is defined will depend on what the exploratory analyses come up with in terms of predictors of taking up screening invitation.
  2. to make clear that it is batches that are randomised and not individual women.
  3. to make clearer that all women randomised in will be invited.

c) progress to date (KM)

KM tabled a 1-page summary of progress of the trial to end January 2011.

19 breast screening units have started randomising. The total number of study participants by the end of January 2011 was 128, 871. By the end of March 2011, it is estimated that there will be over 155, 000 study participants.

Before starting randomisation, sites need to:

  • have at least one digital machine
  • be keeping to NHSBSP targets
  • have local R&D approval

Obtaining digital equipment is proving the biggest hurdle to starting randomisation due to equipment cost, and also maintenance costs which are very high for digital machines.

Action: SS to compile a document listing all 73 units and identifying which of the above criteria screening units have fulfilled/not yet fulfilled. MR suggested this could be reviewed at the Cancer Programme Board.

Action: KM to calculate numbers that will be randomised in six years and add to protocol

Clarification was requested as to what the endpoint of the trial was. It was clarified that the primary endpoint is breast cancer mortality. For the 47-49 age group this is defined as the cumulative risk of breast cancer mortality by 60 and for the 71-73 year age group it is defined as the cumulative risk of breast cancer mortality by age 80.

4. Discussion and adoption of terms of reference

The terms of reference were reviewed. It was noted that the terms of reference should include a reference to patient safety which is built into NHSBSP procedures.

5. Any other business
  • Patient information leaflet

Leaflet needs updating to cover extension of randomisation to 2 screening rounds. In addition several suggested amendments to the current information leaflet were proposed.

Action: Suggested changes to information leaflet to be discussed with JP. KM to coordinate update and changes to leaflet, including clearance with ethics committee

  • Extension of trial over 2 screening rounds

Sites that have already obtained local R&D approval will need to inform their R&D department that the trial is now running over two screening rounds (rather than one as previously planned); sites in the process of obtaining approval will need to ensure their application is for 72 months and the appropriate number of women.

Action: SS to inform all sites (including previous pilot sites) to proceed as

above, and to send sites the letter from ethics committee giving approval for the trial to extend over 2 screening rounds

Pilot sites will have completed two screening rounds by summer 2015. As things stand currently they will at that point start inviting all women in the extended age groups.

  • A potential third round and setting completion dates

The idea of extending the trial for a third screening round was raised; this would increase the power of the trial and enable longer term outcomes to be examined. In case this becomes a possibility at a future date, it was stressed that it was important to say in the protocol that the trial will continue until ‘at least’ 2016.

The possibility of setting completion dates was raised so as to synchronise completion of rounds. This issue should be discussed at a further meeting.

6. Date of next meeting

The Data Monitoring Committee meeting due to be held in the afternoon was postponed. It will now meet as a teleconference.

It was agreed that the Age Extension Trial Management Group should meet annually face-to-face, with the next meeting in January/ February 2012. It was felt that the next meeting should be for at least 3hrs with a break in the middle.