The Changing Face of Newborn Screening

Thanks largely to genomic advances, an increasing number of advocates are pushing for the use of genome sequencing to improve newborn screening. While this would undoubtedly improve risk assessment for a range of genetic conditions, challenges such as costs and ethical concerns could hinder rollout on a wider scale.

Newborn screening for potentially harmful diseases has existed in some form for around 60 years. Robert Guthrie is widely credited with introducing the first so-called ‘heel prick’ test for newborn babies to test for phenylketonuria (PKU), a serious metabolic condition that can be effectively cured by sticking to a specific diet.

While the number of conditions screened for depends a lot on the country, region and hospital, the introduction of tandem mass spectroscopy helped allow this type of screening to expand to include more than 50 additional, mostly metabolic, conditions, saving 6800 lives a year in the U.S.

Since the human genome project was declared complete in 2003, genetic sequencing technology has changed out of all recognition. Rapid increases in speed and accuracy over the last two decades have occurred alongside dramatic cost reductions making mass sequencing projects such as the UK’s 100,000 Genomes Project, which began in 2013 and ended in 2018, a feasible reality.

Sequencing technology is now also being applied to newborns both to diagnose unknown conditions and to screen healthy babies, albeit still in a research setting. Work done by Rady Children’s Institute for Genomic Medicine in California, led by its president and CEO Stephen Kingsmore, has done much to support sequencing for children with undiagnosed disorders over the last few years. “Thirty-one published studies have shown that genome sequencing of newborns with severe illnesses leads to diagnosis of underpinning genetic diseases in 36%, changes in how those babies are treated in 27%, and changes in outcomes in 18%,” Kingsmore told Inside Precision Medicine.

On the screening side, Robert C. Green, a professor at Harvard Medical School and medical geneticist at Brigham and Women’s Hospital, was ahead of the field when he set up the first BabySeq project in 2015 with co-director Alan Beggs to screen healthy newborns using exome sequencing. After the first phase was completed successfully, seven years down the line the second phase of the project is now underway.

In the U.K., Genomics England is in the first stages of an ambitious pilot project –the Newborn Genomes Programme. This project will start recruiting in 2023 and aims to sequence up to 200,000 healthy newborns to assess the feasibility of a larger roll out across the country’s National Health Service (NHS) in the future.

Despite these positive steps, a number of logistical, financial and ethical challenges remain before largescale genomic sequencing of newborns can be rolled out. The results of these ongoing studies, and the challenges they encounter, will help clarify when and how this type of screening will be adopted in countries around the world.

Rapid sequencing to target rare disease
According to Kingsmore, newborn diagnostic sequencing has the potential to diagnose over 7000 genetic diseases. “However, only about 400 genetic diseases are severe enough and have sufficiently effective treatments today that they should be screened in all newborns today,” he believes. Kingsmore and colleagues proved the validity of such sequencing through Project Baby Bear. Set up in California in 2018, the pilot project was the first state-funded program to use genome sequencing as a diagnostic method for critically ill newborns suspected of having a rare genetic disease.

Sequencing 178 critically ill babies over an 18-month period, the project diagnosed genetic conditions in 76 (43%) cases, 55 of whom were not previously receiving the correct care for their condition. Baby Bear is estimated to have saved $2.5 million in medical costs for the children involved.

Inspired by Project Baby Bear, Caleb Bupp, a clinical geneticist and assistant professor at Michigan State University, and colleagues have set up a sister project in Michigan, also named after their state animal.

“Project Baby Deer is a Michigan wide program…We’ve aimed it at having access for kids, regardless of where they’re born, or what hospital they may be admitted to, which I think was a little bit unique,” Bupp explained in an interview.

Set up in collaboration with Rady, who are managing the rapid sequencing pipeline for the project, Project Baby Deer began roll-out in 2020 and achieved Medicaid coverage in 2021. It includes 8 hospitals across the state and has already enrolled more than 80 children.

Some of the hospitals taking part in the program are smaller and less well resourced than others. Lack of in-depth genetics information or staff with expert knowledge can be a problem, according to Bupp, but he says the pandemic has actually helped solve this issue by boosting telehealth resources and familiarity in the region.

“We’ve worked out a fairly good telehealth system for being able to tap into expertise if we need to… I think that’s lowered some of the anxiety and the burden on some of the smaller centers so that they don’t have to feel like they have to figure this out 100% by themselves.”

Newborn sequencing as a public health tool
One of the pioneers in newborn sequencing, Green had previously led several other sequencing projects in adults before getting funding for the first phase of the BabySeq trial in 2015.

The research team managed to recruit 324 babies and their families to take part in the first phase of BabySeq, 67 sick babies admitted to intensive care and 257 healthy babies. The infants were randomly assigned to receive a consultation about family history of genetic disorders only, or the same consultation plus exome sequencing.

The results showed that 11% of the babies sequenced carried dominant monogenic mutations linked to health, 88% had recessive carrier mutations that could impact their parents reproductive planning and 5% carried pharmacogenetic variants that could impact their response to childhood medications.

Green explained that he and his colleagues learnt a lot about how to run these kinds of trials during the study. “We basically went door-to-door to ask the families to participate. And what we found, unsurprisingly, was that immediately after you’ve had a baby, is not necessarily the best time to listen to a complicated pitch about how we’re going to stick your baby for some more blood, and randomize you into a complicated trial… So, it was a struggle to meet our original recruitment goals in that way.” This is just one takeaway that Green and colleagues used when putting together a proposal for the second phase of BabySeq, which has been funded by the NIH as an expansion to 500 families, but has not yet started recruiting.

“We have a stakeholder board now of community members, including individuals from BabySeq 1 who are advising us on better communication strategies. So, we’re bootstrapping, I think, a very much better approach to families,” Green explains.

“We’re also going to do this more in pediatric practices that have been selected for having diverse populations in three separate cities: Boston, New York City and Birmingham, Alabama.”

Sequencing technology has developed at such a pace that what was not possible, or at least difficult, in 2015 is now a reality. Genomics England recently helped launch the NHS Genomic Medicine Service in the U.K. While still in its infancy, according to its website, the new addition to the NHS aims to “be the first national health care system to offer whole genome sequencing as part of routine care.”

As part of the new service, Genomics England is planning an ambitious pilot project to sequence up to 200,000 newborn genomes to test the feasibility of this kind of screening in a research setting. After a public consultation last year that revealed a high level of support for the project, Genomics England plans to begin recruiting babies and their families to take part in the pilot from as early as Spring 2023.

Richard Scott is a Consultant at Great Ormond Street Hospital for Children in London and also Chief Medical Officer at Genomics England. During an interview he explained some of the reasoning behind the new project. “We are at a point where currently the number of conditions that one can screen for in most countries is really quite narrow. And there’s obviously an enormous potential to think about a wider number of conditions that you might be able to screen for using genomics… I think there’s a real impetus to ask whether there are better ways of doing this alongside the heel prick test.”

The idea of running the project within the NHS means that the team running the study can assess how easy it would be to roll out this kind of testing to all newborns and also assess realistic costs and potential savings. The project will also feed data into rare disease research to help develop new therapeutics.

Another consideration that will be considered as part of the project is the potential for sequencing a child’s genome and keeping the data long term to refer back to if they become sick in later life.

“I think the future is definitely one where genomics is entirely embedded in healthcare,” says Scott. “It’s still an open question exactly how that will work. Sequencing technologies have changed a lot in the last 10 or 15 years and they are likely to change a lot again… At the moment, there’s a model to sequence once and go back and look at data whenever is needed, which we’ll be exploring through this programme. But it might also be that we actually store less data, but are able to answer the questions we need to by resequencing.”

Overcoming challenges
Before newborn sequencing can be rolled out on a larger scale, either for diagnostic or screening purposes, there are a number of challenges that need to be overcome.

One is ensuring everyone who needs sequencing for diagnostic purposes can get access to it. Erica Ramos is a genetic counsellor by training and current Vice President, Population Genomics at Genome Medical, a digital health company and specialty medical group.

“It is expensive to do rapid genome sequencing. Currently, it’s only available through these somewhat narrowly focused research efforts and in certain hospitals and health systems… so I think equitable access to that type of program by region is going to be an ongoing concern.”

Kingsmore and the Rady team have spent years setting up a fast and efficient sequencing pipeline, but this is not an easy thing to set up overnight. “Rapid genome sequencing is not the same ballgame as whole genome sequencing, to be able to have the workflows and all of the tools and technologies that enable that fast evaluation and confident evaluation takes a big team and a long time to develop,” explains Ramos.

Stanford professor and clinical geneticist Euan Ashley and his team recently hit the news with a new world record for sequencing a genome in just over 5 hours and completing a genetic diagnosis for a patient in under 8 hours. He says that the type of sequencing machine used can make a difference to how affordable and accessible this type of service can be.

Their record used Oxford Nanopore’s PromethION 48 sequencer. While their reasoning for using this machine, rather than the more common Illumina machines found in most hospital laboratories, was its speed and long-read sequencing ability, Ashley says the different approach could actually be a more affordable option for smaller centers.

“With some sequencing machines, your hospital or the healthcare system has to come up with half a million dollars or more just to buy the machine. The way Oxford Nanopore operates is you pay for the flow cells, but you don’t really pay for the machine. Getting hold of one of the machines comes down to whether you are going to order enough flow cells.” he explained.

Simply sequencing faster can also save money, particularly if the patient is critically ill. “There is a big difference between eight hours and 24–48 hours,” says Ashley. “For these patients it could be life and death. It’s also money, because it’s $15,000 a day to stay in a critical care unit. And that could be even higher, depending on the support you have.”

Another potential problem for smaller centers wanting to rollout newborn sequencing programs is that the advanced genetics expertise needed for sequence analysis can be lacking. This is something that Fabric Genomics is tackling with its artificial intelligence (AI) driven genome analysis tools, which are already being used by Kingsmore’s team at Rady. “They use our AI algorithm, which is called GEM. And that does a very good job at very quickly distilling an entire genome down into a very small number of candidates to look at,” explained Jeanette McCarthy, VP of Precision Medicine at Fabric. But she adds that the company also helps centers with less available expertise. “We offer a whole package from when the data comes off the sequencer through generating a physician ready clinical report. A laboratory might purchase our software, but then they can’t find anybody to hire to analyze it. There’s a limited pool of genetics experts that can handle the whole genome sequence data analysis, so we offer those services to our customers.”

In terms of newborn screening in healthy babies, a big point of current debate is what findings should be returned to parents about their children. While many err on the side of returning only ‘medically actionable’ findings, in other words, genetic variants linked to diseases that are treatable and start in childhood, there are others who think a greater range of information could be shared.

“If you tell a parent that their infant is carrying a mutation for something that could eventually be deadly, that’s pretty scary. But we have now done this in several hundred infants, found such mutations and told their parents. We demonstrated that these parents are resilient and understand and want that information,” emphasizes Green.

“We never want to be cavalier about the potential for distress. But the narrative of widespread catastrophic distress in the face of genetic risk information is simply not true.”

A future of genetic medicine
We may still be a few years off from a future where everyone has their genome sequenced and accessible in their medical records in case of future need, but the experts seem to agree that we are moving slowly in this direction.

Sequencing newborns to help treat undiagnosed conditions is growing in acceptance and is, arguably, the first step on the road to wider rollout of early sequencing for screening purposes.

One thing that is hard to predict is how far genomic technology, and indeed technology in general, will have advanced in 10-20 years. At the moment storing even one genome takes a lot of computing power and the chance of being able to access genomic data easily, such as through electronic health records, seems remote. However, technology changes quickly and it’s possible that there is a breakthrough that will allow this just around the corner.

From the consultations carried out by both the BabySeq and Genomics England researchers there seems to be demand from the public for sequencing-based newborn screening. There is still debate among experts as to what findings should be returned to parents and their children, but most are advocating for the sharing of information on childhood onset diseases that are currently treatable, at least as a first step.

“After we’ve done this for 5 or 10 years, people may feel differently, because the general populations attitude toward technology changes over time,” says David Bick, an experienced medical geneticist and clinical advisor for Genomics England on the newborn sequencing project.

“There’s just so many different what ifs that I think it’s really important that we continue to have these conversations about what the ethics and morals of these activities are, and also how they apply in different scenarios. Not just in the early adopter, wealthy, White, employed and English-speaking population that has really led a lot of these types of efforts so far,” adds Ramos.