Page URL:

The reflex prenatal screening model seems like a step in the wrong direction

20 November 2017
By Catherine Joynson
Assistant Director, Nuffield Council on Bioethics
Appeared in BioNews 927

The new 'reflex' method of antenatal screening for Down's, Edwards' and Patau's syndromes was described last week as a 'transformational advance'. It might be cheaper than other approaches, but it could come at the cost of informed choice for pregnant women and couples.

The authors of the study published in Genetics in Medicine compared their 'reflex' model to the screening test currently offered to women on the NHS – the combined test. They stated that 'no other method of prenatal screening for these disorders has such a high detection rate for such a low false-positive rate'. Further benefits of the reflex model were suggested to be that women would be less likely to become acutely anxious, and cost-savings due to a reduced need for diagnostic and counselling.(1)

The Nuffield Council on Bioethics published a report on the ethics of non-invasive prenatal testing (NIPT) earlier this year.(2) NIPT is a technique for analysing cell-free DNA from the placenta circulating in pregnant women's blood. My involvement with this project leads me to believe that the reflex study, and much of the resulting media coverage, has ignored a swathe of research and public debate that has guided policymakers down a path that is subtly but importantly different to the one advocated by the reflex model.

The reflex model is not a new prenatal test but a different way of offering NIPT. It involves taking the woman's blood at the initial consultation and automatically sending this to be analysed using NIPT if the woman is found to have at least a 1 in 800 chance of having a fetus with Down's, Edwards' or Patau's syndrome following the combined test. The woman is not informed of the result of the combined test before the NIPT stage. The study press release states the reflex model is 'far more accurate and safer' than existing tests – safer in that it results in fewer women having confirmatory diagnostic tests, such as amniocentesis, which carry a small risk of miscarriage.

However, the reflex study is just one of many to show that NIPT is more accurate and leads to fewer false positive results than the combined test. For example, in the NIHR-funded RAPID study that involved 30,000 screened women across eight NHS hospitals, NIPT identified all fetuses with Down's syndrome among women who opted for the test, and resulted in few false positives.(3)

More concerning is the impression that reproductive choice was not given high priority in the reflex study. One of the central recommendations of the Nuffield Council's report is that pregnant women and couples should have access to NIPT but only within an environment that enables them to make autonomous, informed choices. Patient groups and others who contributed to the consultation activities for the Council's report raised concerns that the reflex model may not provide women with enough opportunity for discussion and reflection. As a result, it could compromise their ability to give informed consent to NIPT.(2)

Reflex screening might support informed choice if women and couples were given the opportunity to have an extended discussion with a specially trained professional at the beginning of the screening process. This would allow them to discuss the possibility of their blood sample being sent for NIPT analysis, the implications of different results, the choices that testing may lead to, and what it is like to have a child with the condition being tested for. However, the only reference to the provision of information or support in the study is that women who opted for screening 'received a leaflet describing the reflex DNA approach'.

One of the reasons given by the authors for not returning the combined test results to women before the NIPT stage was to reduce unnecessary anxiety: 'imparting potentially distressing information when this can be completely avoided is self-evidently of benefit'. I would argue that nothing is self-evident in this complicated area of medicine.

Their approach may be better for those women who have false positive results, but it might not be as good for women who have 'true positive' results. We do not currently know. Also, this approach will not necessarily reduce anxiety in the long run and, even if it does, this does not mean women and couples should be denied the opportunity to consider their options in light of a high-chance combined test result. In addition, their approach does not fit with the recent shift in medical practice away from what a reasonable doctor would tell the patient, to what a reasonable patient would want to know.

The reflex model also appears to disregard recent public debate on NIPT for Down's syndrome. Concerns have been reignited that the aim of prenatal screening is to 'screen out' Down's syndrome rather than promote reproductive choice for women. Hence, one thing that almost all contributors to the debate agree on is that NIPT is only ethical if women and couples are given up-to-date, balanced and non-directive information and support during the screening process to enable them to make properly informed choices.

This message has reached policymakers. NIPT will be offered as part of the NHS Fetal Anomaly Screening Programme across England and Wales from 2018. In this model, pregnant women who are found to have at least a 1 in 150 chance of their  fetus having Down's, Edwards' or Patau's syndromes after having the combined test will be given the opportunity to discuss with a specially trained healthcare professional whether they would like to have NIPT, diagnostic testing or no further testing. Public Health England has been developing a programme of training with the input of the Down's, Edwards' and Patau's syndrome communities in preparation for the roll-out next year.

In light of all this, it seems surprising that the researchers responsible for the reflex study are now 'approaching other hospitals in the UK to see if they would be interested in adopting the new test', according to the study press release. Removing an opportunity to give information and choice to pregnant women and couples in the screening process is a move in the opposite direction to that which policy and public attitudes are currently travelling.

1) Prenatal reflex DNA screening for trisomies 21, 18, and 13
Genetics in Medicine |  9 November 2017
2) Non-invasive prenatal testing: ethical issues
Nuffield Council on Bioethics |  2017
3) Uptake, outcomes, and costs of implementing non-invasive prenatal testing for Down’s syndrome into NHS maternity care
BMJ |  4 July 2016
1 June 2021 - by Michaela Chen 
A long-running court case over non-invasive prenatal testing patents has been settled...
27 November 2017 - by Jonathan Bestwick 
Antenatal screening for Down's, Edwards' and Patau's syndromes with the reflex DNA method was shown in our recent study to have higher screening performance compared with the combined test alone, and compared with the proposed recall DNA method...
13 November 2017 - by Dr Barbara Kramarz 
A new prenatal screening method for Down's syndrome, Edwards syndrome and Patau syndromes is safer, more reliable and cheaper than existing approaches, a study has found...
6 November 2017 - by Ruth Retassie 
Genetic testing for cystic fibrosis, fragile X syndrome and spinal muscular atrophy is recommended for all would-be parents by a study in Australia...
25 September 2017 - by Dr Rosie Morley 
Parents pass on more new genetic mutations to their children with age, and fathers pass on more than mothers, according to a study...
The evidence shows that prenatal reflex DNA screening is a step in the right direction ( - 23/11/2017)
Joynson, in her “comment” article gives unfounded general opinions and ignores the improvement in antenatal screening performance afforded by the reflex DNA method compared to the recall method in which women are recalled for a second screening test (DNA) following an initial positive test (Combined test). The reflex DNA method has a higher detection rate (95% v 81%) and a lower false-positive rate (0.02% v 2.3%).(1) Although most of the false-positives in the recall method are reclassified negative by the DNA test, this does not avoid the distress associated with being recalled on account of a positive screening test. Joynson seems to overlook the substantial advantages of the reflex DNA method.

My response to Joynson’s specific comments are as follows:
1    There is no lack of informed choice. Women are informed of what is available and decide on the basis of the screening test regarded as a whole. Obtaining consent for reflex DNA screening is no different from obtaining consent for other established screening tests. The assertion that “reproductive choice was not given high priority” is incorrect. Joynson raises the question as to whether reflex DNA screening is a test or a method. It is a test as well as a method because the women having the test are given a single test result. In contrast, with the recall method, women are given two test results. Reflex DNA screening was not presented as the invention of a new marker but as the application of a known maker in a way that achieves a more accurate and safer method of screening than existing tests.

2    The accusation that the authors of the reflex project (1) have ignored a “swathe of research and public debate” with reference to the Nuffield Council report titled “Non-invasive prenatal testing: ethical issues” (2) is unjustified. The reflex method is only referred to in one paragraph (4.25) of the 169 page Nuffield Council report, which does so incorrectly in a chapter titled “NIPT in the private sector”, the reflex method was and is provided through the NHS. Policy makers have failed to recognise that measurement of DNA in screening or diagnosis is not conceptually different from the measurement of proteins, steroids or ultrasound markers; the use of the term NIPT disguises this fact and fails to recognise that the test is a screening test (not a diagnostic test) and the term conceptually isolates the test as non-invasive when all screening tests are non-invasive. NIPT is a term that is best avoided. In practice the question is which of the markers, or combination of markers, achieves the best screening performance at an affordable cost.

3    Joynson argues that the RAPID study (2) using the recall method is better than the Combined test alone, as is the reflex method, and implies that the recall method and the reflex method are similar in screening performance. This implication is incorrect as is apparent from the observation that of the 30,790 women screened in the RAPID study there were 118 false-positive results . By comparison in the reflex DNA project there were 4 false-positive results out of the 22,812 women screened, a rate over 20 times lower. Furthermore the recall screening method leads to women having unnecessary invasive diagnostic tests. In the RAPID study about two thirds of women who opted for an invasive diagnostic test in the 1 in 150 risk category (a category in which they had a choice of a DNA test or a diagnostic test) had an unaffected pregnancy, reflecting the anxiety caused by the initial positive Combined test screening result. In the circumstances the women understandably wanted a definitive diagnostic result instead of another screening result.

4    Joynson suggests that an “extended discussion” (ie counselling) is needed before a woman decides whether to be screened with the reflex DNA test. There is no reason why this process of obtaining consent need be more or less extended than with other tests. Women offered all screening tests are informed of the test in question and choose whether to be screened. They are not counselled at this stage. Counselling is needed once a woman has been identified as having a positive screening result.

5    In connection with the professional duty to “reduce unnecessary anxiety” Joynson argues that contrary to what is said in our paper nothing is self-evident in this complicated area of medicine. My colleagues and I disagree. Regardless of whether the area of medicine is simple or complicated, causing unnecessary harm is always wrong and unethical. It is surprising that Dr Joynson does not recognise this.

6    Joynson suggests that reflex DNA screening may not be as good for women with true positive results as it is for those with false-positive results. This is a surprising proposition which has no basis in fact.

7    Joynson raises an issue that does not specifically affect reflex DNA screening, namely that the aim of prenatal screening for Down’s syndrome may be to “screen out” the disorder implying eugenics. There is no way in which an antenatal screening service has anything to do with eugenics. It provides women and their partners with a choice, without pressure to either accept or reject screening. The fact that such screening services are available does not mean that individuals with Down’s syndrome are different from other members of society.

I believe that the points made by Joynson are neither correct nor relevant. What is of concern is that she, the Assistant Director of the Nuffield Council  on Bioethics, is supporting a poorer method of screening when a better one could be made available at the same or similar cost. I believe that the Nuffield Council should review its position in this area. It should recommend the most effective and safe method of screening for those who consent to screening rather than support an inferior method and not seek consent in respect of separate components of a single screening test.

Jonathan Bestwick
Wolfson Institute of Preventive Medicine
Barts and the London School of Medicine and Dentistry
Queen Mary University of London

1.    Wald NJ, Huttly WJ, Bestwick JP et al. Prenatal reflex DNA screening for trisomies 21, 18 and 13. Genet Med 2017 doi:10.1038/gim.2017.188. Epub ahead of print
2.    Non-invasive prenatal testing: ethical issues. Nuffield Council on Bioethics. 2017
3.    Chitty LS, Wright D, Hill M et al. Uptake, outcomes, and costs of implementing non-invasive prenatal testing for Down’s syndrome into NHS maternity care: prospective cohort study in eight diverse maternity units. BMJ 2016;354:i3426
to add a Comment.

By posting a comment you agree to abide by the BioNews terms and conditions

Syndicate this story - click here to enquire about using this story.