Down's syndrome diagnosis is changing. Clare Sansom looks at the latest techniques
In the UK, about one baby in ten is born with some kind of abnormality. For most the defect is easily curable. But for a substantial minority, such as those suffering from Down's syndrome, the outlook is bleak. People with this condition have three copies of chromosome 21 in their cells. This extra genetic material causes a complex mixture of disabilities. It is possible to diagnose Down's syndrome before birth, but the procedures in use are expensive and not always accurate. This chromosome has now been sequenced and Down's syndrome is one disease to which there may be new approaches as a result. More immediately, clinical trials of new technology indicate that biochemical tests in development can identify women at high risk with significantly greater accuracy. Researchers in the US have also demonstrated that a simple DNA-based test can make a firm diagnosis of Down's syndrome as accurately as the widely accepted procedure of amniocentesis.
Amniocentesis involves taking a small sample of amniotic fluid containing foetal cells and testing it for a variety of abnormalities. It is uncomfortable, and some of the tests can take several weeks. More seriously, there is about a 1 per cent chance that a genetically normal foetus will miscarry after the procedure. For these reasons, as well as cost, amniocentesis is only offered to women at risk of carrying a foetus with a genetic abnormality.
Many abnormalities occur more often in children of older mothers, diabetics, smokers, obese or underweight women, or in particular ethnic groups. The incidence of Down's syndrome increases sharply if a woman is in her late 30s or older. In many areas, only older women will be offered the full range of tests.
The amount of some proteins in the body fluids of pregnant women will be raised or lowered if a foetus is genetically abnormal. The "triple test" for Down's syndrome recommended by the National Health Service Health Technology Assessment programme involves measuring the serum levels of three of these so-called markers. The result is combined with the mother's age and, often, ultrasound measurements, to give a probability that may be as low as one in 50,000 or as high as one in ten. Usually, about 5 per cent of women screened are offered amniocentesis: most of these will be carrying a normal foetus.
Parents in developed countries see ultrasound scans as a routine part of pregnancy, and look forward to them as a chance to "see" their baby for the first time. However, many structural abnormalities are picked up by these scans. The most common ultrasound test for Down's syndrome is the measurement of a temporary swelling of the foetus neck; this is often more pronounced in affected infants. But many structural abnormalities that can be picked up have no, or almost no, clinical consequences. Some babies found to have enlarged kidneys before birth are born with serious renal disease; others are fine.
The accuracy of screening has increased rapidly over the past few years. A test for Down's syndrome developed by the Wolfson Institute in London was recently shown to be up to 85 per cent accurate. It combines ultrasound measurements and two first-trimester biochemical markers with four second trimester markers. There are no plans to make it available on the NHS.
Howard Cuckle of the department of reproductive epidemiology, Leeds University Medical School, emphasises that even the most accurate screening does not give a firm diagnosis. "Diagnosis should be, as near as possible, 100 per cent accurate. Screening tests only aim to identify women as at high enough risk to justify the hazards of a diagnostic procedure. In contrast, there will probably always be some overlap in screening results between affected and unaffected cases."
The policy applied in his department is typical. A woman identified as "screen positive" for Down's syndrome there, and offered amniocentesis, may have an actual risk as low as one in 250.
Researchers at Vysis Inc in Illinois, US have developed a DNA-based diagnostic procedure known as the "Fish test", in which fluorescent markers binding to particular chromosomes are used to identify aneuploidies - anomalies in chromosome number, like the extra copy of chromosome 21 that indicates Down's syndrome. It is faster than amniocentesis and, if introduced into routine practice, should be considerably cheaper. Professor Cuckle says: "If the average costs of diagnosing aneuploidies were to decrease, it should be possible to offer the tests to women at slightly lower risk. This would increase the overall detection rate."
There will always be tension between those who emphasise the public health aspect of Down's syndrome screening (as the majority of affected foetuses will be terminated) and those who emphasise its use in informing choice. The Down's Syndrome Association is opposed to selective termination for the condition. However, the society accepts that screening provides valuable information. It is promoting the new first-trimester tests, as termination, if chosen, is less distressing when performed early in pregnancy. There is also concern that screening out foetuses with serious congenital abnormalities might be a precursor to eugenics, but Professor Cuckle disagrees. "It is very dangerous to extrapolate. Medical and technological advances have always changed the human gene pool."
Professor Cuckle welcomes moves by the government to introduce a national programme for Down's syndrome screening but remains concerned that many women will be denied access simply to save money, and that the advice offered to pregnant women is still inadequate. "Only 15 per cent of the leaflets we analysed contained all the key points identified by the Royal College of Obstetricians and Gynaecologists." It may be that a study by the National Institute of Clinical Excellence could end the "postcode lottery" for women at risk of carrying a Down's syndrome child.