Weighing the benefits of expanded newborn screening

Philippa Brice

12 May 2014

The UK National Screening Committee (NSC) has announced that the current newborn screening programme in England will be expanded to include four new conditions.

Rare metabolic disorders

Blood samples taken from very young babies will now be tested for Maple syrup urine disease (MSUD), Homocystinuria (HCU), Glutaric acidaemia type 1 (GA1) and Isovaleric acidaemia (IVA), along with conditions such as cystic fibrosis and phenylketonuria. These are all rare genetic disorders that interfere with the babies’ ability to metabolise specific substances, resulting in the build-up of toxic waste products in the body. In some cases symptoms may be relatively mild, but in others they can rapidly lead to irreversible brain damage and death. Prompt diagnosis allows medical assessment and care to prevent this; it also avoids what is typically a very drawn out diagnostic process for children in whom disease onset is more gradual.

The evidence for expanded newborn screening

Expansion of the NHS Newborn Blood Spot Screening Programme to include five new inherited metabolic disorders was explored by a pilot scheme. This was underpinned by commissioned research carried out by the PHG Foundation, published in the report Expanded newborn screening, which showed that there was potential for cost-effective prevention of death and severe disability due to these conditions via newborn screening.

Following the pilot, the NSC held a consultation earlier this year on provisional plans to expand newborn screening to include three new genetic disorders (MSUD, HCA and GA1). The PHG Foundation’s response challenged this suggestion, pointing out that omitting two disorders considered on the basis of statistical conclusions from the pilot data was inappropriate, since the numbers involved were so small.

To give some perspective, the one-year pilot screened over 700,000 newborns; 47 were identified as potentially affected by one of the five new disorders under consideration, and diagnostic testing revealed that 20 babies were actually affected; these disorders are very rare indeed, which is one reason why conventional diagnosis can be achingly slow. At the same time, prompt diagnosis can actively save lives, since the onset and devastating impact of the diseases can be incredibly rapid. A few babies developed symptoms so quickly that they were diagnosed before the screening results were returned; however, three of them also died. Whilst this shows that newborn screening will not necessarily benefit all affected children, it also underlines the potential severity of presentation and importance of prompt diagnosis and treatment.

Professor Jim Bonham of the Sheffield Children’s NHS Foundation Trust, who led the pilot, said: “We are delighted with the results because it shows how we can make an enormous difference for these children and their families, in some cases giving them the gift of life”.

Balancing benefits and harms

The NSC were also concerned with the rate of false positive screening results – potential cases identified by screening that on investigation proved not to be affected by the disease. Thus, 27 sets of parents probably experienced anxiety and distress until they heard the results from the diagnostic testing. The question the NSC had to consider was whether or not the expense of expanded screening and follow-up testing, along with the harm caused in terms of unnecessary parental anxiety for the false positive cases, was justified by the health benefits (and cost savings) from the successful identification and treatment of affected children.

The PHG Foundation felt that the pilot evidence, bearing in mind the limitations imposed by the extreme rarity of the diseases and combined with wider international evidence, merited inclusion in screening for all five disorders; the NSC has opted for four out of five.

This highlights what may become an increasingly common dilemma for health commissioners as genomic testing becomes more widespread in health systems. Personalised approaches to disease prevention and treatment offer potential to deliver much greater accuracy, but this will inevitably result in stratification of population and patient groups into smaller sub-populations. The evidence to demonstrate the appropriate path for each of these groups will tend to be less compelling, since it involves fewer people. Therefore, whilst large-scale pilot data will remain a gold standard, there may be a growing need to allow appropriate flexibility within evaluation processes to allow careful consideration of wider evidence too.