To explore the use of genomic sequencing for screening newborns in a public health setting, the National Institutes of Health today awarded up to $25 million in funding over five years to four research teams.
The new program, called Genomic Sequencing and Newborn Screening Disorders, is jointly administered by the Eunice Kennedy Shriver National Institute of Child Health and Human Development and the National Human Genome Research Institute. It resulted from a workshop conducted by NICHD and NHGRI almost two years ago to discuss research into sequencing-based newborn screening.
The awards, totaling $5 million in fiscal year 2013, go to principal investigators Robert Green and Alan Beggs at Brigham and Women's Hospital and Boston Children's Hospital; Stephen Kingsmore at Children's Mercy Hospital in Kansas City; Robert Nussbaum at the University of California, San Francisco; and Cynthia Powell and Jonathan Berg at the University of North Carolina at Chapel Hill. NIH plans to fund these groups with $25 million over five years.
The four pilot projects, which will sequence the genomes or exomes of a couple of thousand newborns in total, will assess whether sequencing provides useful medical information beyond what is already delivered by current newborn screening. Each project will explore technical aspects of genomic sequencing and analysis in newborns, study how the results relate to patient care, and look at ethical, legal, and social implications.
Today, almost all of the four million newborns in the US are already screened for a few dozen inherited disorders using a blood test. Current recommendations call for the screening of 31 disorders, most of them rare genetic diseases with established diagnostic tests and effective therapies. Studies have shown that these existing screening programs, which are administered and regulated by the individual states, prevent more than 12,500 debilitating disorders in children every year and help to save millions of dollars in medical costs.
"Newborn screening is one of the most successful public health programs in our country," said NICHD Director Alan Guttmacher in a teleconference to announce the awards today. However, the 31 diseases currently recommended for screening are only a small fraction of the up to 8,000 inherited disorders, of which 4,000 to 5,000 have known genetic causes that could be detected by genome sequencing today.
Some of the projects will focus on conditions that are already part of the existing newborn screening panel to see whether genome sequencing can yield better answers, but new disorders will be included as well.
"We'd like to see if genomic sequencing can shed light on other disorders that we don't screen for currently, and we'd like to see how genomic data might inform clinical care for newborns more generally," he said. Guttmacher noted that it will remain important to better understand inherited diseases — including those that are already screened for today — in order to develop or improve therapies.
Newborn screening through genome sequencing raises a number of ethical, social, and legal questions that the new program plans to address. These include how to protect a newborn's privacy and what consent process to use; where a newborn's data will be stored and who will have access to it; what conditions will be included in the test; what criteria will be used to determine a positive result; what the follow-up for positive sequencing screening results should be; what results doctors should return to parents; how much newborn screening through sequencing will cost; and what the overall impact on healthcare spending will be.
"These are obviously complicated questions, and we as a society need to consider them thoughtfully before we hope to implement any sort of newborn sequencing program on a wide scale," Guttmacher said.
According to NHGRI Director Eric Green, the program will address whether genome analysis in newborns is an appropriate application of sequencing technologies, and in what situations it works best; what the best ways are to present the information to parents and clinicians; and whether complete genome sequence information might be too much information for them to handle. It will also explore what types of information parents and clinicians want to receive, and study the costs associated with sequencing-based newborn screening.
Current newborn screening costs about $100 per patient, he said, a quarter of which is associated with the test itself, whereas exome sequencing is currently available for less than $1,000, and whole-genome sequencing for $3,000 to $5,000, depending on completeness and quality. One question the program wants to answer is whether the additional information that can be obtained from sequencing will be worth the additional cost.
While sequencing costs continue to drop, the technology "may still be a bit too expensive in the short run for use in population-wide screening programs," Green said. On the other hand, newborn sequencing "could be potentially very powerful, perhaps saving many millions of dollars in healthcare costs." However, the program will not study the economics of implementing sequencing-based newborn screening widely, including whether insurance would pay for it in the future.
Besides the cost of sequencing, the false-positive rate of sequencing-based newborn screening will determine its usefulness. Even with a relatively low false-positive rate, "the cost of follow-up may be such that the whole system becomes economically not feasible," said Guttmacher.
Green acknowledged that the performance of sequencing technologies is still lacking. "They are not quite at a clinical grade," he said. "To truly make genome sequencing part of clinical care, we have to make the accuracy and the fidelity much higher." NHGRI has ongoing programs, he said, to improve the accuracy of sequencing and to develop routines for validating sequencing results.
All of the projects will address the specificity and sensitivity of sequencing-based newborn screening. While some will perform sequencing in a CLIA-certified laboratory, others will confirm positive results by Sanger sequencing.
In order to protect patients in the program, data safety monitoring boards will review protocols, provide advice, and ensure that adverse events are "quickly and appropriately" addressed, Green said. Due to concerns about sharing genome sequence data from newborns, the program will not require grantees to deposit their data in databases such as dbGAP.
Overall, the pilot projects will focus on improving the care of infants, so they will primarily study rare inherited disorders that require immediate intervention. They are "not looking at an infant's risk to develop Alzheimer's disease in their 80s," Green said.
Each group will handle incidental findings in a slightly different way. "Handling incidental findings is one of the questions that these awards are meant to address," said Anastasia Wise, a program director in NHGRI's division of genomic medicine. While some groups will only analyze genes related to the phenotypes of the child, others will return non-actionable incidental findings if parents would like to receive them.
Recommendations for returning incidental findings that were recently published by the American College of Medical Genetics and Genomics (CSN 5/8/2013) should be considered a "version 1.0," Green said. "Any recommendations that exist point to additional questions and additional research studies that we will continue to foster to better inform those practice guidelines."
Each of the four pilot projects will have a slightly different focus. Green and Beggs' team at Brigham and Women's Hospital and Boston Children's Hospital will make genomic data, generated in a CLIA lab, available to parents and doctors throughout infancy and childhood and study the impact and usefulness of the results. It will explore how doctors use the results in healthy as well as sick newborns, compared to current newborn screening results, and how they influence clinical care. In addition, the group will study the response of parents with sick and healthy children, and differences between their response to genomic results and conventional newborn screening results.
Kingsmore's group at Children's Mercy Hospital in Kansas City will study genomic sequencing in newborns in a neonatal intensive care unit, a strategy the team has already explored in a small number of individuals (CSN 7/18/2012). The project aims to reduce the turnaround time to 50 hours to make it comparable to existing newborn screening tests, and to increase the number of diagnoses made. Further, it will study parents' and doctors' perceptions of the risks and benefits of the test, and how they may change over time.
Nussbaum's project at UCSF will focus on the potential of exome sequencing in screening both for disorders that are currently tested for and for others that are not, and to assess the value of the additional information. It will also explore genetic variants that predict the response to drugs frequently used in children, such as codeine and seizure medication. In addition, the team plans to develop a participant protection framework for genomic sequencing during infancy and to explore legal issues related to genome analysis as part of newborn screening.
Powell and Berg's team at the University of North Carolina at Chapel Hill will sequence the exomes of healthy infants as well as those with known genetic conditions, and plans to identify the best ways for returning results to doctors and parents. Their study, which will include multicultural families, will also develop a tool to help parents understand what the results mean and examine challenges that doctors may face in using the new technology.