New Down’s syndrome test more accurate than current screening

“Blood test for Down’s syndrome 'gives better results'," reports BBC News today. The test, which is based on spotting fragments of "rogue DNA", achieved impressive results in a series of trials.

A study of over 15,000 women found that the new blood test more accurately identifies pregnancies with Down's syndrome than the test currently used.

Down's syndrome is caused by having an extra chromosome (the packages of DNA containing information to grow and develop). The new test is able to detect small fragments of DNA from the baby floating about in the mother’s blood, called cell-free DNA (cfDNA).

This blood test measures the number of chromosomes in the mother’s blood, and from that it can see if there are any of these extra chromosomes.

The cfDNA test performed significantly better than the current test across a range of screening test measures for Down’s syndrome, but was not 100% accurate. Importantly, it had a much lower false positive rate than the current test; false positive is where a healthy baby is wrongly identified as having Down’s. A false positive result often leads to an unnecessary further diagnostic test that carries a small risk of causing a miscarriage.

The test is not yet available on the NHS, but it is being reviewed and a decision is expected later this year. It can be accessed privately at a cost of between £400 and £900.

Where did the story come from?

The study was carried out by researchers from the University of California, the Perinatal Diagnostic Center in San Jose, Sahlgrenska University Hospital in Sweden, and several other US institutions. It was funded by Ariosa Diagnostics and the Perinatal Quality Foundation.

The study was published in the peer-reviewed New England Journal of Medicine.

BBC News accurately reported on the study and provided expert opinion from both Great Ormond Street Hospital and the Down's Syndrome Association. Both organisations highlight the need for women to be given clear information about screening, so they can make an informed decision.

What kind of research was this?

This was a diagnostic study, which compared a new antenatal screening test with standard screening for three genetic conditions, including Down’s syndrome.

Normally, people have 23 pairs of chromosomes. However, in these three genetic conditions, there is an extra copy of one of the chromosomes. In Down’s syndrome, there is an extra chromosome 21 (trisomy 21); Edwards' syndrome has an extra chromosome 18 (trisomy 18); and Patau’s syndrome has an extra chromosome 13 (trisomy 13). In most cases, this happens by chance and isn’t inherited from the parents. This is why all mothers-to-be are offered screening to see whether this has happened.

Currently, all pregnant women in the UK are offered screening for these conditions, which involves a two-step process. The test offered depends on how far along the pregnancy is. Women between 11 and 14 weeks pregnant are offered a blood test plus an ultrasound scan, called a combined test. Women between 14 and 20 weeks of pregnancy are offered a different blood test. This is less accurate than the combined test.

If either of these tests indicates an increased risk of having a baby with Down’s, Edwards’ or Patau’s syndromes, the woman will be offered either chorionic villus sampling (CVS) or amniocentesis to find out. Both of these tests involve taking samples from the mother’s abdomen, which can be uncomfortable, although not usually painful. This increases the risk of miscarriage, which occurs in one in 100 women (1%).

The new test detects short fragments of the baby’s DNA floating about in the mother’s blood, called cell-free DNA (cfDNA). By measuring the level of each of the chromosomes, it is possible to see if there are more chromosomes 21, 18 or 13.

The researchers had previously performed proof of principle studies of cfDNA in women at high risk of having a baby with one of these conditions. They now wanted to see how accurate the test was in a large sample of women with any level of risk.

What did the research involve?

The researchers recruited 15,841 pregnant women eligible for screening for Down’s, Edwards’ or Patau’s syndromes. All were tested using the new cfDNA blood test and the standard combined test. The results of the two tests were compared to see which was more accurate at picking up any of the three trisomy conditions.

Women were enrolled in the study between March 2012 and April 2013 from 35 medical centres across the US, Canada and Europe. They were eligible to participate if they were aged 18 or older, and had a singleton pregnancy between weeks 10 and 14.3 at the time of screening.

A blood test for cfDNA was taken at the same time as the standard screening tests. The blood sample was then analysed at a laboratory without the analysts knowing any clinical details about the pregnancy, other than the gestational age and mother’s age (the sample was blinded). The results were not given to the mother or clinician.

The researchers then obtained the outcome of the pregnancy and compared the accuracy of the standard test results with the new cfDNA test. This included any termination of pregnancies and miscarriages if a genetic test had confirmed whether or not they had a trisomy condition.

They originally enrolled 18,955 women, but excluded 3,114, due to:

• them not meeting the inclusion criteria
• withdrawal from the study (either the woman or the investigator)
• sample handling errors
• no standard screening result
• no cfDNA result
• lost to follow-up

What were the basic results?

The new test outperformed the current one at detecting Down’s syndrome. Results were similar for Edwards' and Patau’s syndromes, but tended to be less accurate.

One of the most important measures for whether a new screening test is any good is the positive predictive value (PPV). This takes into account the number of correct test results, but also the number of false positives, based on the condition's prevalence.

In rare conditions, like these chromosomal conditions, the false positives are important, because they represent a potentially large group of women who could be sent to have further invasive diagnostic tests they might not need.

The PPV of the new test for Down’s syndrome was 80.9% – significantly higher than the 3.4% scored for the combined test. The PPV difference was lower for women deemed at lower risk of having a baby with Down’s syndrome (76.0% for the new test v 50.0% for the current test).

The detailed results for Down’s syndrome (trisomy 21) were:

• cfDNA screening identified all 38 babies with Down’s syndrome (sensitivity 100%, 95% confidence interval (CI) 90.7 to 100)
• standard screening identified 30 out of 38 babies with Down’s syndrome (sensitivity 78.9%, 95% CI 62.7 to 90.4)
• the cfDNA test was positive in nine pregnancies that did not have Down’s syndrome (false positive rate 0.06%, 95% CI 0.03 to 0.11)
• standard screening was positive in 854 pregnancies that did not have Down’s syndrome (false positive rate 5.4% (95% CI 5.1 to 5.8)

Results for Edwards' syndrome (trisomy 18) were:

• cfDNA identified nine out of 10 cases (sensitivity 90%, 95% CI 55.5 to 99.7)
• standard testing identified eight out of 10 (sensitivity 80%, 95% CI 44.4 to 97.5)
• cfDNA wrongly diagnosed Edwards' syndrome in one case (false positive rate 0.01%, 95% CI 0 to 0.04)
• standard testing was positive in 49 pregnancies that did not have Edwards' syndrome (false positive rate 0.31%, 95% CI 0.23 to 0.41)

The results for Patau’s syndrome (trisomy 13) were:

• cfDNA screening identified both babies (sensitivity 100%,95% confidence interval (CI) 15.8 to 100)
• standard screening identified one out of the two babies (sensitivity 50.0%, 95% CI 1.2 to 98.7)
• the cfDNA test was positive in two pregnancies that did not have Patau’s syndrome (false positive rate 0.02%, 95% CI 0 to 0.06)
• standard screening was positive in 28 pregnancies that did not have Patau’s syndrome (false positive rate 0.25% (95% CI 0.17 to 0.36)

How did the researchers interpret the results?

The researchers concluded that "the performance of cfDNA testing was superior to that of traditional first trimester screening for the detection of trisomy 21". They say that further cost benefit studies are now needed. The researchers also caution that "as emphasised by professional societies, the use of cfDNA testing and other genetic tests requires an explanation of the limitations and benefits of prenatal test choices to the patient".

Conclusion

This large study has shown that the new cfDNA test is better than current standard screening at detecting three trisomy conditions during pregnancy. The confidence in accurately identifying affected pregnancies was strongest for Down’s syndrome. There were much wider confidence intervals for the other two conditions.

The cfDNA test was not 100% accurate, as there were false positive results for each condition, though much fewer than with standard screening.

Around 3% of the cfDNA tests did not produce a result. Careful consideration and further research may be needed to decide the best approach in these cases. Should they all be sent for the next stage of diagnostic tests as a precaution, repeat the test, or be offered the standard test instead? 

The author’s admit that, had they included these "no result" cases in their main analysis, the performance of the cfDNA test would have been lower. How much lower we don’t know, as they don’t appear to have presented an analysis of this scenario.

The potential benefit of the test is that it could reduce the number of women being sent for the CVS or amniocentesis testing, which carry their own risks. As the authors say: "Before cfDNA testing can be widely implemented for general prenatal aneuploidy screening, careful consideration of the screening method and costs is needed."

This test is not yet available on the NHS, though it is being considered under an evaluation project run by Great Ormond Street Hospital. In the evaluation study, which is being carried out on women at low risk, if the results of the test show that a trisomy is highly likely or it is inconclusive, then they are offered the invasive tests to confirm the result. This is because of potential false positive results, which in previous research was found to occur in one in 300 women (0.3%) and false negative results – not picking up the diagnosis in two out of 100 babies.

At present, the test is only offered by private clinics and costs £400 to £900. It takes two weeks to get the result, as the sample is sent to the US. Details of private clinics can be easily found via any internet search engine.