Making sense of newborn genome sequences
11 January 2019
Since the advent of cheaper whole genome sequencing there has been much discussion and debate on whether we should routinely sequence all babies shortly after birth. Newborn whole-genome sequencing is an attractive prospect for some, as not only does it enable identification of genetic diseases that can manifest in early childhood, but also can provide a stable data set that can be reanalysed over time as the need arises. This data could reveal information about future disease risk or likely drug responses (pharmacogenomics) that could be useful in prevention or treatment decisions, as well as providing a rich resource for ongoing biomedical research
However, this is no simple task given the complexities around the technical aspects of genome sequencing and clinical interpretation of the data it produces, along with ensuring the information derived from sequencing is appropriately communicated. It also gives rise to a myriad of ethical concerns and potential dilemmas, for example how to handle information that has implications at later stages of life for the baby (e.g. BRCA gene status and cancer risk), but may have more immediate or current implications for the parents.
The BabySeq Project is a pilot randomised control trial was initiated in the USA in 2015 that aims to examine some of the issues around newborn genome sequencing. Both healthy and ill newborns were included in this trial and randomised either to receive standard care (newborn screening, family history and genetic counselling) or genetic screening in addition to the standard care. The babies and their families, along with clinicians involved in their care, were followed up to assess if the availability of genomic information impacted on their behaviour. A recent paper provides some of the findings of this project - and highlights some of the challenges of such an endeavour.
Is routine DNA-based newborn screening useful?
Screening of healthy babies is a routine procedure in most developed countries and involves both physical examination as well as routine biochemical blood tests to screen for inherited medical conditions. Early identification of disease in asymptomatic babies can enable appropriate treatments to be initiated early and ameliorate adverse effects of the condition. The exact conditions that are screened for varies between countries and is usually determined by assessment against screening criteria. Some of the factors that impact on inclusion of particular conditions include availability of a validated screening test and agreed treatment pathway, along with consideration of the relative benefits and harms of early identification as well as the costs and ethical implications.
Current biochemical tests can identify many conditions, but are by no means comprehensive. Furthermore, factors such as when exactly the blood sample is collected after birth and the gestational age of the baby can affect test results. DNA-based tests are not affected by such factors, which makes them an appealing complement to the current suite of tests. In addition, they could identify genetic abnormalities for which no biochemical tests currently exist. Genome sequencing of ill newborn and older babies may already be used to seek genetic diseases that may underlie observed symptoms, when a diagnosis can inform management.
Is routine sequencing ready for the clinic?
Many genetic diseases present early in life and can lead to increased mortality and disability if they are not identified and treated at an early stage. Newborn sequencing of healthy or unwell babies can be a useful tool to identify such disease at an early time point. This is much easier when there is certainty linking genetic variants with a specific disease. Currently, however, we are some way from this certainty due to the rare and varied nature of genetic diseases. Although we can obtain genome sequence information, and identify some variants definitively linked with disease, other variants identified that may be involved in disease need careful interpretation in light of previous findings and or clinical indications.
These difficulties in interpretation are illustrated by the paper from the BabySeq pilot trial. 9.4% (15 babies) of the newborns sequenced were identified as’ at-risk‘of a childhood monogenic disease on the basis of finding disease associated variants reported in the literature. Although these variants were considered ’medically actionable’ it was not clear if their identification at this stage was beneficial. This was for a variety of reasons, not least uncertainty as to whether the identified genetic variants would, in fact, ever lead to clinical disease requiring treatment, and identification of milder forms of disease.
The other purported advantage of newborn sequencing is generation of a genomic dataset that can be analysed not at the point of the testing, but as needed in future. The research team also carried out such analysis for a small sub-set of participants who developed disease symptoms, but did not identify variants linked to the symptoms in these cases.
The bottom line
This is an ambitious project that is attempting to untangle some of the practical and ethical questions around sequencing infants. The value of genome sequencing in symptomatic babies, such as those in intensive care, is already recognised. Given the difficulties in interpreting genetic information, we are still some way from demonstrating the benefits of routine sequencing of healthy newborns. Especially when current methods can identify some of those with clinically significant diseases in a quicker and cheaper way.