Thursday June 26 2008
The researchers say the test could be a simple mouth swab
“All women could soon be offered a genetic test which would tell them whether they were likely or unlikely to contract breast cancer”, The Guardian reported today. The test could possibly be a simple mouth swab, carried out at around the age of 30. Women whose results showed them to be at high risk could be screened at a younger age while those at low risk could choose to delay it until they were “55 or older”.
The Independent also covered the story and said that because of genetic inheritance, the odds of getting breast cancer could “vary more than sixfold” and the current breast screening programme fails to target those at highest risk.
The stories are based on a study that used estimates of the increased risk of breast cancer due to seven common genetic variations. Individually, each variation slightly increases risk, but when they occur together, they are expected to more than double the risk of a woman developing breast cancer over a lifetime.
Although there are a number of assumptions behind this study - that are based on the risks estimated in other genetic studies - the research is reliable and highlights the potential of this sort of ‘pre-screening’. The researchers hope that this sort of testing might improve the accuracy of traditional methods of detecting breast cancer. However, more research into the harms associated with such pre-screening, along with the costs of the testing, are required before it is possible to be sure that this test is suitable to be offered to women.
Where did the story come from?
Dr Paul Pharoah and colleagues from the Departments of Oncology, and Public Health and Primary Care at the University of Cambridge and the Cancer Research UK Cambridge Research Institute carried out the research. Funding sources and potential conflicts of interest were not reported. The study was published in the peer-reviewed medical journal: The New England Journal of Medicine.
What kind of scientific study was this?
In this modelling study, the researchers used data from genetic studies to estimate the theoretical chances of getting breast cancer. This chance of breast cancer (or expected risk) was estimated for each of the 2,187 possible combinations of seven specific gene variations.
The researchers explain that women’s chances of getting breast cancer can vary more than sixfold because of their genetic inheritance. A genetic test based on these seven gene sites could potentially help determine individual cancer risk or could focus screening to concentrate on those at higher risk.
At present, women as young as 30 years of age, who have a very strong family history of breast cancer with high-risk genes (known as BRCA1 and BRCA2), are offered screening by MRI scan. Over a lifetime, these genes increase the chances of a woman developing cancer from around 9% to 80%. However, they are also rare - about three in a thousand women are carriers, meaning that few women benefit from such gene testing.
The other more common genetic variations that the researchers were interested in, occur more frequently. For example, one such site of variation is known as rs2981582. The high risk variation at this site occurs in 38% of the general population, with the other 62% of women carrying the low-risk variation. The researchers say that the risk due to this common variation (or allele) accounts for about 2% of the genetic risk of breast cancer – a small amount when looked at in isolation. They chose seven such sites of variation known as SNPs (single nucleotide pairs). At each site, one of the two variations is high risk and the other is low risk. If the high risk variation is present it will increase the risk of cancer incrementally. For this study the researchers extracted the data on risk from four different genome-wide association studies that had reported their results relating to the seven SNPs found on six different chromosomes.
The researchers were also interested in estimating the increase in the risk of breast cancer if all the high-risk variations occurred together in the same woman. For this, they assumed that the relative risks due to each of the seven variations (or susceptibility alleles) could be multiplied together. This allowed them to estimate the ‘discrimination’ provided by the test, (i.e. how accurate the gene tests were at identifying the women most likely to develop cancer from those that were unlikely to). If such testing had good discrimination for a woman individually, then women who had all the risky variations might belong to a suitable group in which to offer personalised screening and prevention programmes.
The researchers said that dividing people into high and low risk groups like this might also be useful as a pre-screening procedure in a population and they wanted to test this by adding it to the current sequence of tests for healthy women. This would indicate if the tests had potential to improve the national breast cancer screening programme.
What were the results of the study?
The researchers said that 56 out of every 10 million women (about 3,300 women in the UK) carry two copies of all the low-risk variations at each of the seven gene sites, and these women have less than half the risk of breast cancer of the general population – a 4.2% lifetime risk compared with a 9.4% risk.
At the other end of the scale, researchers say around seven in 10 million women have high-risk variations at each of the seven sites, giving them a 23% risk of developing breast cancer over a lifetime – two and a half times that of the population as a whole.
What interpretations did the researchers draw from these results?
The researchers concluded that their analysis suggests the “risk profile generated by the known, common, moderate-risk alleles does not provide sufficient discrimination to warrant individualized prevention”. However, they also say that it may be possible to use risk stratification in the context of programmes for disease prevention in the general population. This means that although the test was not useful for telling an individual whether they might develop breast cancer, it might provide useful information for saying who should go on to the next stage of a screening programme, such as mammography or MRI screening.
What does the NHS Knowledge Service make of this study?
This study demonstrates the potential for genetic tests in disease prevention. It illustrates how genetic studies can provide information that could refine the sequence of testing in screening programmes. The authors suggest that there are a few outstanding questions and barriers “to be overcome before such potential is realised”, including:
- Their simple model is based on several assumptions that may not be robust. For example, they assume that the risks attributable to single allele variations can be multiplied and that the benefits of mammography are related to the reduction in risk alone. It is likely that there may be some interaction between the function of single alleles found in the genome and also that the benefits of mammography might depend on both the risk reduction and the age of the patient (a factor that is known to determine the sensitivity of the screening procedure).
- Any added benefit of this sort of testing, if it is confirmed in further trials, will need to be balanced against the added complexity and costs of adjusting population screening programmes to take into account individual genetic profiles. This added complexity might, the authors say, reduce the programmes efficiency.
- Screening at younger ages may be better performed by MRI-screening, which is more expensive than the traditional mammography X-ray. The authors say that screening for breast cancer by means of MRI would be prohibitively expensive unless targeted at those at risk.
- The best way to communicate personal risk and the contribution that such discussions make to informed decision making is not yet clear. It is likely that considerable public and professional education will be required before the general understanding of the complexities of genetic testing for common variations such as these is sufficient for honest, informed consent.
More research into the harms associated with this sort of pre-screening in addition to the costs involved will be required before it is possible to be sure that this test is good enough to be offered to women.
Sir Muir Gray adds...
It looks like this long promised test will arrive.