2.3 Scaling up
They may look at dozens of alleles, and involve thousands of people, but existing screening programmes have been concerned with individual genes. But the technologies now being developed will soon permit the recording of hundreds of genes at a time. So-called gene chips combine the skills of microchip designers with DNA sequence information to offer rapid, easy-to-read results for an individual covering hundreds of genetic variants. A gene chip is a thin slice of glass about the size of a postage stamp. Stuck to the surface is a grid, each line the width of a human hair and each containing a small sequence of single-stranded DNA, a gene variant. The total DNA can represent either hundreds of gene variants for just one gene, for a number of genes or even a gene set for the entire genome. When a patient's DNA is added to the chip, pieces stick to matching sequences and the rest are washed away. The results are read by means of an electronic scanner and analysed by a computer software program, which identifies the matches within a matter of hours. Gene chips are already in use; for example, the one used for detecting variants in the breast cancer gene BRCA1 can detect any nucleotide change in any position in that gene.
If such a DNA chip was developed for cystic fibrosis, what would be the consequences for screening for this disease?
Since at least 900 variants are known, this technology could identify all variants or alleles of the gene and not just the common ones, making screening much more exact, and eventually foolproof.
When devices like this come into wider medical use, potentially as desktop boxes in the doctor's surgery, then more general DNA profiling of individuals, i.e. identifying variants for many hundreds of genes simultaneously, becomes possible. In 2005, the UK Human Genetics Commission recognized that DNA profiling is likely to become feasible in less than 20 years time. There will be no problem handling all the data, according to the forecasters: ‘assuming that there are about 1000 clinically relevant genotypic markers [variants] per person, then genotyping one billion people would result in about 10 terabytes of data, an amount that would fit on a mere 1000 DVD optical discs’, says a writer in the journal Science. But what will we do with all this information?
In general, it will not be the kind of information derived from earlier screening programmes, which has been used to choose specific treatments for one known disease, or to offer advice about the chance of a pregnancy producing a child with a particular, usually not very common, disorder. Instead, it will add up to a catalogue of individual susceptibility factors, or genetic risk factors, or alleles, for the most common, multifactorial diseases – like cancer, heart disease, stroke, diabetes, hypertension or Alzheimer's disease. (The term ‘genetic risk factor’ refers to a susceptible disease allele and should not be confused with ‘risk factor’, meaning an ‘environmental factor’. Both genetic risk factors and risk factors increase the risk of an individual developing a particular disease.)
And while it will help doctors make predictions, they will be statistical predictions, telling people they are more likely to develop this disease, less likely to develop that, than the population at large. This will pose quite different problems of counselling and decision-making than those seen with the single-gene disorders – like CF or HD, for example – which make up the bulk of our experience of genetic testing so far.
The complexities of hereditary influences on breast cancer hint at some of these problems. Spend a few minutes thinking about what might be some of the main problems of counselling and decision-making posed by the discovery of the two genes – BRCA1 and BRCA2 – strongly associated with breast cancer, before reading on.
There are a number of features of the link between certain alleles of the two genes and breast cancer that may be hard to convey clearly, to health professionals as well as patients. They include:
Most breast cancer is not associated with either of these two genes – each of which accounts for perhaps 2.5 per cent of the total incidence of the disease.
Although certain alleles of the genes concerned are associated with an enhanced risk of breast cancer, testing for them does not give a clear-cut result. A positive test does not mean that a woman will definitely get breast cancer. A negative test does not mean that she will definitely not. This is not related to inaccuracies in the test – false positive or false negative – it is simply a property of the genetic information. The development of cancer is a multi-step process and involves mutations in at least five or six genes. An individual who inherits a recessive, mutant allele in either the BRCA1 or BRCA2 gene is one step nearer to developing cancer than an individual who inherits two normal alleles of both of these genes. The absence of a clear-cut result contrasts testing for a single gene disorder such as HD, where a positive test for a mutant allele means a certainty of developing the disease.
If a test result is positive, there is no certain route to prevention. Options include so-called prophylactic mastectomy, or breast amputation, along with hysterectomy of womb and ovaries to reduce the risk of ovarian cancer associated with the same genes. Less drastically, regular monitoring may be recommended, but if this is done using X-ray mammography, it may itself increase cancer risks (X-rays are mutagenics).
Cancer is a source of particular anxiety to many, so it would probably be wrong to suggest that testing positive increases the numbers of the ‘worried well’. They may already have been concerned about breast cancer, especially as many will have seen a high incidence of the disease in their families. What is clear is that publicity about ‘cancer genes’ increases demand for testing (in contrast with the experience of HD), and that responsible management of the tests demands a great deal of explanation and counselling. One result has been that some doctors in cancer clinics have had to become genetic counsellors.
This begins to suggest some of the demands that would be produced by more widespread genetic screening that yielded information about health risks and disease probabilities. Another difficulty is that, although more health workers can be trained to help people to deal with information like this, not too much is known about how their customers will respond. We know a certain amount about how people understand probabilities (not too well, on the whole), but much less about how they may react to specific predictions.
Can you think of different ways in which people might deal with a genetically-based prediction that they were at high risk, or genetically susceptible, for heart disease in middle age?
Research suggests there are two broad classes of reaction: activism and fatalism. Activists try and take control, and strive to minimise their risk by diet, exercise or drugs, and by avoiding smoking. Fatalists, on the other hand, hear the prediction as something they can do little about, and decide they will indulge freely in all the things health-educators say are bad for you – because they are going to get sick anyway.
The trouble is, it is hard to know who will react which way. There are similar uncertainties about other common disorders that are often suggested as candidates for susceptibility, or genetic risk, prediction. If a gene of major effect in the development of schizophrenia is ever identified, for example, some families may feel they are relieved of blame or guilt for the occurrence of the disorder, while others may interpret a test that shows the presence of the gene as showing that they are to blame after all. Again, we do not really know which is more likely.
Click below to view the video sequence.
Now would be a good time to view this video sequence, which requires you to consider some medical benefits arising from knowledge about our genes.