Tuesday November 25 2008
A woman having a mammogram
“Screening women for breast cancer may lead to unnecessary treatment as some cancers will disappear on their own,” reported The Daily Telegraph.
It said that a study found more cancers in women screened every two years than in those screened only once in six years. The newspaper suggests this could mean that some cancers ‘vanish’ and that by detecting them, women could be having unnecessary and painful treatment.
Screening is accepted as a highly effective tool in the detection of breast cancer. Although the changes in breast cancer rates seen in this study suggest that it is possible that some screen-detected cancers may regress, there are also several alternative explanations that warrant further investigation. As it is not known why some cancers might regress and how to identify them, it is likely that most people would rather be treated than not.
Women are advised to continue to attend breast screening programmes.
Where did the story come from?
Dr Per-Henrik Zahl and colleagues from the Norwegian Institute of Public Health and other research centres in Norway and the US conducted this research. The work was funded by the Department of Veterans Affairs in the US. The study was published in the peer-reviewed medical journal Archives of Internal Medicine.
What kind of scientific study was this?
In this cohort study, the researchers compared two screening strategies to see if one detected more cancers than the other. One strategy screened women every two years, while the other screened them once in six years. The researchers were also interested in spontaneous regression (when a cancer shrinks until it is not detectable).
Their theory was that if cancers stayed the same size or got bigger (once they had grown to a detectable size), then both screening approaches would pick up roughly the same amount of cancers overall. The frequent screening would simply pick up the cancers at an earlier stage.
The researchers compared the frequency of breast cancers in four Norwegian counties before and after the introduction of a two-yearly breast cancer screening programme. The screening programme started in 1996, and invited women aged 50 to 69 years old to attend a first mammogram screening in 1996-1997, and every two years after that. In this study, the researchers looked at the women who were aged between 50 and 64 in 1996 and followed them up until 2001 (for six years). These formed their frequent screening group.
The researchers identified a control group of women, who would have been aged 50 to 64 years old in 1992 (those aged 55 to 69 years old at the time of the first screening). They used these women to indicate what would occur if screening was only carried out in this group once every six years (from the period 1992 to 1997).
The researchers used the Norwegian Cancer Registry to identify the women in each group who were diagnosed as having invasive breast cancer over the six-year period, and their ages. They did not include cases of non-invasive breast cancer, such as ductal carcinoma in situ.
The proportion of women diagnosed with invasive breast cancer every year for each age group was then calculated. These were then added up to get the total for the six-year follow-up period. This figure was compared between the control and frequent screening groups, with adjustment for differences in age between the two groups.
To look at whether the effect varied by age, they also compared these figures within four-year overlapping age groups (women aged 50-53 years, 51-54 years, 52-55 years and so on). This method also prevented the women from being counted in both the control and frequent screening groups.
What were the results of the study?
The researchers found that the frequent screening and control groups of women were similar in terms of age, the proportion who had given birth, the age at which they gave birth, and the number of births. Just over three-quarters of women in both groups had attended screening.
After the final screening, at six years, there were 1,909 cases of invasive breast cancer per 100,000 women in the frequent screening group, compared with 1,564 cases per 100,000 women in the control group. This 22% difference was statistically significant. When they looked at women in different age groups, the researchers found that more cancers were detected in the frequent screening group than in the control group in women of all ages, except for the oldest age group (those aged 61-64 years).
What interpretations did the researchers draw from these results?
The researchers concluded that because invasive breast cancers were more common in the frequent screening group than in the control group, it suggests that “some breast cancers detected by repeat mammographic screening would not persist to be detectable by a single mammogram at the end of six years”. They suggest that “this raises the possibility that the natural course of some screen-detected invasive breast cancers is to spontaneously regress”.
What does the NHS Knowledge Service make of this study?
As the authors report, there is little knowledge about what would happen if breast cancers were left alone, as most cases that are detected are treated. This large study provides indirect evidence that some invasive cancers may regress over time. There are some possible alternative explanations of the findings that the authors consider and mostly reject:
- Because the two groups came from different periods(before and after frequent screening was introduced)there may be differences between these periods that account for the differences seen. For example, if there was an improvement in how invasive breast cancer was recorded in the cancer registry, this might lead to an apparent increase in the rates over time. However, the authors say that the records of breast cancer in women below screening age (younger than 50) appear to be relatively stable over this period, which suggests that recording did not change.
- The two groups of women may not have had comparable risk factors for cancer. The authors again suggest that this is unlikely, as the women were similar in a number of ways that affect breast cancer risk (such as childbearing factors). Additionally, although it was from different times in their lives, some women had data that could be used for both the control and frequent screening groups. This should tend to increase similarity between the two groups.
- Mammography may have become more sensitive over time. However, if this were the case, the sizes of the tumours detected should reduce over time, but they did not.
- There may simply have been a natural fluctuation in the incidence of breast cancer over this period. However, the researchers say that before screening was introduced, rates of breast cancer only fluctuated by 1% a year at most. The percentage increase seen during the study (22% over six years) was much higher than this.
- The authors also considered it unlikely that an increased use of hormone replacement therapy (HRT) could be responsible for the increases. Although HRT use did increase over the period, estimation of the effect of this on the overall rate of invasive breast cancer in the population was not as great as the actual increase seen.
The authors say their theory should be tested by analysing data from other screening programmes.
It is important to note that this study did not look at the effect of the screening programme on the number of deaths from breast cancer, therefore no conclusions can be made about whether screening programmes improve survival for women found to have breast cancer.
It is also important to note that these results are from a whole population screening programme, rather than from people specifically at high risk, such as those with familial forms of breast cancer. The relative benefits of screening will be higher in high-risk groups than in the general population.
This study does not conclusively prove that the tumours regressed or indicate which tumours will regress and which will not. For these reasons, women should continue to attend screening programmes and take advice about the likely prognosis of any tumour found, their treatment options, and the benefits and risks of these.