SIOG 2014
Alternatives to population-based screening in elderly breast cancer patients
Dr Nienke De Glas - Leiden University Medical Center, Leiden, The Netherlands
In the Netherlands there’s a large population-based screening that did use an upper age limit of 70 years until 1998. But in 1998 the upper age limit of screening was extended to 75 years, despite the fact that previous randomised trials barely included any women over the ages of 68.
Why was it extended?
There have been some observational studies that have shown that mortality in women who had screen detected breast cancer was better but these observational studies are highly biased due to several reasons. But still, in the Netherlands, that was reason enough to extend the age limit.
What we did was we looked at the incidence rates of early stage tumours and advanced stage tumours in the Netherlands after the implementation of screening in the age group of 70-75. What you would expect if a screening programme is successful is that the incidence rates of early stage tumours increases while the incidence rates of advanced stage tumours decreases as a result of the earlier detection. However, what we did find was we found an enormous increase in the incidence of early stage tumours but we found almost no decrease in advanced stage tumours. So we calculated that for every advanced stage tumour that is prevented, almost twenty women were over-diagnosed, so received unnecessary treatment for cancers that otherwise probably would have been indolent or not have led to mortality.
What were the consequences of over-diagnosis?
We only looked at the incidence rate but you can extrapolate that there’s a large group of women that receive unnecessary treatment. Especially in this age group this is important because older women are at increased risk of possible complications, of adverse events of endocrine treatment and chemotherapy. So especially in this age group, over-diagnosis and over-treatment is a big risk and has complications. We could conclude that the implementation of screening in this age group has led to a large proportion of over-diagnosis, unnecessary treatment, a large amount of costs for the screening, but also for unnecessary treatment. We did not look into the cost aspect but I think you could calculate that it must be an enormous amount of money, yes. I think that in this age group we should not offer a population-based screening but of course there is a group of women who could benefit from screening. So I think we should move towards a more personalised way of screening women based, of course, on the patient’s preference but also life expectancy. Women with a large remaining life expectancy could benefit from breast cancer screening. Also we should look at breast cancer risk and there are many tools that can calculate breast cancer risk of women in a population. So I think we should move towards a more personalised way of screening women instead of large population-based screening programmes.
What would be your message to cancer doctors?
I would mainly want to make a statement to the policy-makers because I think that population-based screening is not the way to go, at least for older women. Of course there is an on-going debate for the whole breast cancer screening but in older women it has really large risks because of the risk of complications of treatment in the older population. For the doctors it’s very difficult because if there is a cancer detected of course you want to give the patient the best treatment. I think there are things that we don’t want to know in certain patients, or certain women, actually not even patients,
Could the Dutch government stop doing this?
Of course it’s a very political discussion with a large pro and con people in the field. So it’s a very difficult and on-going debate. I’m not sure what will happen next but I think this is a very strong message that it can actually harm people. I think do no harm is one of the main things we should think about as doctors and we can do harm by screening people. These are quite strong data and a very large population-based study where there’s a high coverage of the screening programme, an attendance rate of almost 80% also in the older women. So, yes, these data can add to the debate of screening, especially in the older age group, yes.
