Background Uptake of cancer trials and in particular prevention trials has been disappointing globally.
Methods Uptake to three randomised chemotherapy breast cancer prevention trials and two dietary prevention trials in women at increased familial risk were assessed and compared with uptake of screening trials across a range of risk categories.
Results Uptake of drug prevention trials remains low at 5.3–13.6%, but is significantly higher in the high (12%) compared to very high risk group (8.4%) for IBIS1 and IBIS2 combined (p=0.004). Recruitment to two dietary prevention studies via mail shot was also disappointingly low at 6.2% and 12.5%. In contrast uptake to two mammography screening trials was >90% in all risk categories.
Conclusions More work must be done to improve recruitment to prevention trials if they are to be seen as viable alternatives to risk reducing surgery.
Impact Trial designs and decision aids need to be developed to improve recruitment.
- Cancer: breast
- genetic epidemiology
Statistics from Altmetric.com
Recently breast cancer (BC) trials' uptake has improved. In 2008/9 14 359 BC patients were recruited to UK trials representing 33% of incident cases, with 9.1% joining randomised trials.1 For the same year 93/428 (22%) BC patients at our institution entered randomised treatment trials, and 70 of 248 (28%) over a more limited time span entered into a randomised study comparing three different diet/exercise programmes after surgery. This represents uptakes of 47%/45% of eligible patients. Prevention trials can improve the statistical power of their studies by recruiting individuals with a higher risk of the specific cancer. Indeed this might be expected to improve recruitment as people at higher risk of a cancer may be more motivated to join a trial. We previously reported a relatively low uptake of the double blind tamoxifen versus placebo trial IBIS1 among women with very high risk of breast cancer (lifetime risk 40%+).2 We have enlarged this study to investigate uptake across a range of risks to a number of breast cancer prevention studies, and report our findings here.
Patients and methods
The Manchester Breast Cancer Family History Clinic has now seen and assessed over 9000 women since 1987. Lifetime risks of breast cancer are calculated using a manual lifetables approach and using the Tyrer–Cuzick model.3 Records are kept of all women approached for prevention studies. We assessed the proportion of women who opted for three chemoprevention studies; two placebo controlled studies of tamoxifen (IBIS1) and anastrazole (IBIS2) and a randomised study using a combination of raloxifene and zoladex (RAZOR) versus placebo as well as two dietary intervention studies; a non-randomised trial to assess the effect of weight loss on markers of breast cancer risk over 12 months (Lifestyle Study) (n=40)4; and a randomised study comparing two different weight loss diets over 6 months (Intermittent Diet Study (n=504).5 Uptake to these trials was compared across lifetime breast cancer risk categories to two screening trials (FH01/FH02) of mammography aged 40–45 and 35–39 years, respectively (table 1). Lifetime risk categories were: very high risk women (lifetime risk 40%+); high risk (lifetime risk 26–39%); and moderate risk (lifetime risk 17–25%). Women were only counted as being approached if eligible. Women who had previously undertaken risk reducing mastectomy were excluded.
The drug and screening trials report the numbers of eligible subjects personally invited to the study by clinical staff within the Family History Clinic. Recruitment to the weight loss studies involved a mail shot.
In total 861 very high risk women (lifetime risk 40%+), 1713 high risk women (lifetime risk 26–39%), and 1785 women at moderate risk (lifetime risk 17–25%) were approached for at least one drug or diet trial. Uptake to drug trials was 5.3–13.6%. Uptake to the IBIS1 study was higher among high but not very high risk groups. This was less evident for IBIS2, but the combined assessment showed a significant difference in uptake of 12% versus 8.4% (p=0.004). RAZOR was initially targeted only at the very high risk group of 601 women. The recruitment rate was only 5.5% and therefore had to be opened out to the other risk categories. Complete records of high/moderate risk women approached were not available, but uptake was estimated to be higher: 11% for high risk and 8% for moderate risk women. In contrast 699/737 (95%) and 372/401 (93%) women eligible for the screening trials (FHO1/FH02) were recruited, with only six women returning their consent forms or stating in clinic they declined entry.
The lower uptake rate for dietary studies reflects the general poor response to mail shots. Only 12.5% (40/340) entered the lifestyle study and 6.3% (50/797) entered the intermittent study. There was no consistent pattern of uptake to these studies according to risk categories. This most likely reflects that motivation to join a weight loss study reflects a general desire to lose weight and is not motivated by risk reduction. Uptake to drug prevention studies thus remains low. In the highest risk group of unaffected BRCA1/2 mutations carriers only 22/256 (8%) have entered a prevention trial, whereas risk reducing surgery uptake was >50% after 7 years.6 This also compared to 319/1713 (18.6%) high risk women (p=0.001) entering a prevention trial.
We have shown that uptake to drug prevention trials remains disappointingly low at 5.3–13.6%. In contrast the great majority of women from all three risk categories have entered a non-randomised screening trial. Uptake of screening is already known to be much higher than uptake into prevention studies, with the majority of highest risk women entering MRI screening.2 The reasons for the poor uptake to prevention trials need to be explored. Anecdotally our high risk women report not feeling comfortable with placebo studies as they will wish to be in an active treatment arm. It is likely that trials aimed at this very high risk group will need to consider dropping a placebo arm in favour of two proven preventive agents before recruitment is likely to rise. For entry into the STAR trial which involved randomisation to either tamoxifen or raloxifene, but did not have a placebo, uptake was higher. A total of 184 460 women were screened using the modified Gail model to determine their breast cancer risk. Of these, 96 368 had a 5 year risk of at least 1.66%.7 From this group, 20 616 (21.4%) agreed to be screened to determine full eligibility for the trial based on the medical criteria defined below; 20 168 were found to meet all eligibility criteria of the study. Of this latter group, 19 747 women expressed a desire to go forward with participation in the trial, signed consent forms, and were randomised to receive either tamoxifen or raloxifene. Participants were screened and enrolled through nearly 200 clinical centres throughout North America. This multicentre trial represents an adjusted uptake of 21%.
It is certainly possible that women at very high risk do not feel that the potential risk reductions of 35–70% are sufficient, given their high absolute risk of breast cancer. Recent introduction of MRI screening in the UK might have improved uptake of prevention studies with the possibility of reliable early detection. However, if anything, uptake of prevention trials appears to have reduced in more recent cohorts and uptake of risk reducing mastectomy has not dropped.6 Perhaps this is linked to widespread press coverage of adverse effects of certain cancer prevention drugs8 and hormone treatments.9 Improving risk communication may have some benefit in uptake according to research on uptake of screening.10
Recruitment to clinical trials is likely to be dependent on a number of factors such as having a dedicated trial manager, being a cancer or drug trial, and having interventions only available inside the trial.11 Training of study staff has also been highlighted as an important area,11–13 and providing them with a best practices tool box that in the future would include successful mechanisms for eligibility screening and recruitment.12 Development of complex interventions to target poorly recruiting centres or individuals that include training has also shown some success.13
We are not aware of any other groups that have assessed uptake across a series of prevention trials, or to specify this by risk category. Uptake of tamoxifen in North America, where it is licensed for prevention (this is not the case in Europe), has shown only 0.2% of the female population aged 40–79 years took the drug in the year 2000.14 However, this study was not able to assess uptake in those strictly eligible on risk grounds or after being approached with prevention as an option. Amongst BRCA1/2 mutation carriers a Canadian group reported 11% uptake (29/270) of either tamoxifen or raloxifene in women who had not undergone risk reducing mastectomy.15 This figure is rather similar to our 8% uptake in mutation carriers for any drug prevention trial. It is likely that more must be done before we will get sufficient individuals into cancer prevention trials to move forward on an alternative to surgery in the highest risk groups.
We acknowledge the support of the Biomedical Research Centre at Central Manchester Foundation Trust, NIHR and the Genesis Breast Cancer Prevention Appeal.
Competing interests None.
Ethics approval Ethics committee approval for the studies described was obtained from the Central Manchester Ethics Committee.
Provenance and peer review Not commissioned; externally peer reviewed.
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.