October 2007: Volume 33, Number 10
The Search for the Most Extensive Newborn Screening Panel
What Complexities Will Labs Encounter as Testing Expands?
By Julie McDowell
Today in the U.S., more than 90% of babies will be born in states that require comprehensive screening for at least 21 genetic and congenital conditions, according to the 2007 March of Dimes Newborn Screening Report Card. But uniformity among states continues to be problematic. Only 13 states and the District of Columbia currently adhere to the 2004 recommendation from the American College of Medical Genetics (ACMG) that every baby born in the U.S. be screened for 29 core conditions using tandem mass spectrometry (MS/MS) technology. In addition, laboratories continue to struggle with costs and quality issues associated with implementing an expanded program. Despite these challenges, however, federal healthcare officials are now contemplating adding more conditions to the recommended screening list. In June, the Department of Health and Human Services’ (HHS) Health Resources and Services Administration (HRSA) began accepting nominations for panel additions, forcing the clinical laboratory community to scrutinize the quality of current newborn testing programs while anticipating issues associated with screening for more genetic and congenital conditions.
The recent screening statistics from the March of Dimes represent a dramatic improvement, explained R. Rodney Howell, MD, Chair of the HRSA Advisory Committee on Heritable Disorders and Genetic Diseases in Newborns and Children (ACHDGDNC). In 2005, only 38% of U.S. infants were born in the states that screened for at least 21 of the 29 conditions; in 2006, this figure increased to 64%. Based on current birth calculations, this means that an estimated 3.6 million of the 4 million babies born every year in the U.S. are now screened for the expanded panel of disorders. And these numbers will likely continue to inch upward in 2008. This year, Montana, Kansas, and West Virginia mandated expanded screening programs to be implemented next year.
“Newborn screening is a complex, complicated system and it’s quite distinct because it is a state-based genetic testing program. Therefore, it’s up to the states to fund it,” he explained. “Although there are still states that need to catch up, the bottom line is that most babies in the U.S. are receiving the expanded panel of screenings. Other states are moving forward, some much slower than we would like, but virtually everyone is moving toward a comprehensive panel.” Howell, who is also a Professor of Pediatrics at the University of Miami’s Leonard M. Miller School of Medicine, was chair of the ACMG committee that developed the expanded newborn screening panel recommendations in 2004.
Key Barriers to Statewide Implementation
Many state healthcare officials looking to implement expanded screening say they face challenges relating to funding limitations and lack of support within their organization. In 2004, researchers with the California Department of Health Services’ Genetic Disease Branch and Public Health Institute (Berkeley, Calif.) conducted a national survey of 106 key newborn screening contacts in each state that asked about issues affecting their states’ ability to expand newborn screening to include MS/MS technology. Respondents to the survey, which was published last May in Pediatrics, said that the top lab-related issues limiting expanded screening were costs of equipments and supplies (53%) and the development of population-specific cutoff values (41%). In addition, respondents cited the lack of available accepted protocols and guidelines for MS/MS diagnostic evaluations and adequate educational materials describing the disorders as barriers to implementing expanded screening programs (Pediatrics 2006; 117: S253-S260).
Howell agrees that funding is typically the primary barrier for states wanting to do comprehensive newborn screening. “Without a doubt, money is the most commonly stated reason why states are not implementing the expanded newborn screening panel, along with not having the proper technology and the appropriate people to do this testing,” he said. However, he added, even the poorest states in the U.S. have established expanded programs. “It’s commonly believed that the most fiscally challenged state in the United States is Mississippi. Having said that, they have a complete panel—they test for everything on the expanded newborn panel,” he explained.
Looking at Additional Screening
Even though not every state has mandated screening for the expanded panel of 29 disorders, many in this field are praising HRSA for accepting testing nominations, adding that the time is right to make this list even more comprehensive. HRSA will continue to accept nominations indefinitely; and the ACHDGDNC will appoint an independent expert review group as needed to evaluate the nominations and make recommendations on which disorders should be added to the expanded panel.
Many clinical laboratorians are already offering their thoughts on what disorders should be added to the panel. For example, lysosomal storage disorders (LSD) are good candidates for newborn screening, according to Michael J. Bennett, PhD, DABCC, FACB, Professor of Pathology and Laboratory Medicine at the University of Pennsylvania and Director, Metabolic Disease Laboratory at The Children’s Hospital of Philadelphia. Bennett spoke at a full-day symposium titled “Expanded Newborn Screening for Metabolic Diseases by Tandem Mass Spectrometry and Follow-up Testing,” held at July’s AACC Annual Meeting. In addition, he is Chair of the National Academy of Clinical Biochemistry (NACB) committee developing Laboratory Medicine Practice Guidelines (LMPGs) in this area. The draft LMPG, titled Expanded Newborn Screening for Metabolic Diseases by Tandem Mass Spectrometry and Follow-Up Testing, is scheduled to be posted online this month (click here). The draft will be open for public comment until Dec. 31, 2007.
A group of more than 40 diseases related to lysosomal digestive enzyme function, some LSDs cause severe neurodegeneration that is not treatable. However, there is evidence that enzyme replacement therapies might be effective in treating some LSD conditions. “Traditionally, we’ve regarded these diseases as very severe and untreatable, which goes against the paradigm for expanded newborn screening, which is why diagnose a disease that you can’t do anything about,” Bennett explained. “But treatments are becoming available for these diseases, and it’s clear that the earlier we diagnose these conditions, the better chance we are going to have in getting effective treatment. That’s when these conditions start to become better candidates for expanded newborn screening.”
Adding lysosomal disorders to the expanded panel would also require extending the MS/MS technology, as it would necessitate an additional platform. Even with this added platform, it’s still appropriate for whole population screening. This isn’t the case, however, for some disorders detectable in newborns. “The lysosomal methods are appropriate for whole populations, but for conditions such as disorders of purine and pyrimidine metabolism, analysis requires a separation step, which adds time to the process,” said Bennett. “If you have an analysis that takes 30 minutes per sample, you can’t do 4 million samples per year, given the present restraints on U.S. newborn screening programs. So testing for some of the others requires methodological improvements before we can contemplate whole population screening.”
ACHDGDNC’s Howell agrees that there appears to be a change in the current thinking that screening should not be done for disorders that will not be of an immediate benefit to the baby screened. Even though this was one of the criteria that the ACMG and ACHDGDNC used to evaluate the disorders considered for inclusion in the uniform panel in 2004, newborn screening leaders are hearing opposition from parents on this approach. “Right now, we only screen babies for conditions that we can do something about, but the parents of children with certain inherited diseases have spoken out that they don’t think this is the only reason for screening,” said Howell.
Parents often point out the potential benefits of newborn diagnosis of disorders such as Fragile X—the most common form of genetic mental retardation—that could be part of newborn screening. Even though parents and primary care physicians notice developmental delays and other symptoms when the baby is about a year old, a definitive diagnosis often doesn’t come until the child is 30 months old. In the meantime, however, the family has been consulting a host of doctors trying to figure out what is wrong with their child. “Many of these parents feel that they would have benefited if they knew the child had Fragile X because that child could have been put in to an early development program, which has been shown to benefit children,” said Howell. “Because of situations like this, I think there will be an increasing need to think about screening for conditions that will have a broader concept of benefit.”
While some are looking forward to detecting more potential disease in newborns, others in the clinical community are calling for a cautious approach. “Our ability to detect asymptomatic diseases, as well as molecular and genetic abnormalities, is advancing at a lightning pace,” wrote Beth Tarini, MD, MS, in an August 2007 editorial in the Archives of Pediatric and Adolescent Medicine (2007; 161: 767–772). “However, this pace threatens to outstrip our understanding of how to treat disease and our ability to provide appropriate resources to children and families.”
“Before we roll out additional tests, all players in the newborn screening community—from laboratorians to the specialists to the pediatricians and primary care providers—need to know how to interpret the results appropriately among each other, so that the information can be communicated accurately to the parents,” said Tarini, who is a Researcher and Clinical Lecturer in the Division of General Pediatrics at the University of Michigan Health System in Ann Arbor. It’s vital that laboratorians be involved in this communication loop. “While sometimes laboratorians are not part of this discussion, we need to bring them in because they provide an important perspective,” she explained. “For instance, laboratorians are aware that abnormal levels associated with some disorders are not black and white, and they can explain to the physicians and others involved what the true limitations are of these prognostics.”
The involvement of clinical laboratorians in these discussions will become increasingly important as newborn screening continues to analyze more genetic mutations, Tarini added. “When a mutation is identified, it’s not always clear if it is a disease-causing mutation, so it’s helpful to hear from the laboratory about what this means for the patient, parent, and the diagnosis,” she explained. “In fact, it’s the lab that holds the power to help us understand if the mutations correlate with disease.”
National Children’s Study Now Underway
AACC’s Pediatric Reference Range Committee Will Advise on Specimen Collection
Now that Congress has authorized $69 million to fund the National Children’s Study (NCS), recruitment has begun for this initiative that will examine environmental influences on the health and development of over 100,000 children through infancy, childhood, and early adulthood (CLN, August 2006).
This summer, the National Institute of Child Health and Human Development (NICHD) released a draft of the Research Plan for the first phase of the study—to examine children from birth to 24 months of age—for public comment, which closed at the end of September. NICHD has announced that the project’s Study Protocol is also nearing completion.
To study the effects of environmental factors—including biological, genetic, social, and cultural ones—researchers will be collecting biologic and genetic samples. Access to this collection of samples would enable the clinical laboratory community to establish much needed pediatric reference ranges. “AACC’s Pediatric Reference Range Committee is actually advising the NCS on how they should be collecting samples, storing them, and in the future, how they are going to analyze them,” said Committee Chair Michael J. Bennett, PhD, DABCC, FACB, who is also Director of the Metabolic Disease Laboratory at The Children’s Hospital of Philadelphia. “One of the successes that we’ve had so far is that we’ve convinced the NCS officials that they should collect samples through blood spots. These specimens then would not only be used in newborn screening, but they could be used to measure anything if the analyte is stable, and it uses incredibly small volumes of sample, which is always an issue with pediatrics.”
The NCS was initially authorized by Congress under the Children’s Health Act of 2000. AACC advocated for continued funding of the program, which was needed as part of the Bush Administration’s Fiscal Year 2007 budget in order for the study to move forward. “I’m delighted that they have their funding this year,” said Bennett. “It’s a huge program that is most definitely going to have a big impact on how we view childhood development in years to come. While it’s going to be another decade or so until we see results coming through, I think the results and knowledge we gain from this study will be similar to what we’ve learned from the Framingham Heart Study.”
Increasing False-Positive Rates
Another concern about expanding the core screening panel is the potential increase in false-positive results. In a study published last August in Pediatrics (2006: 448-458), Tarini and her colleagues examined screening data from the National Newborn Screening and Genetics Resource Center (NNSGRC, Austin, Texas). Based on the number of mandated disorders added to state newborn screening panels between 1995 and 2005, they estimated that more than 51,000 infants would have received false-positive results through MS/MS screening in 2005, if the test specificity was 99.9%. When the researchers assumed a specificity of 99.995%, that estimation went down to approximately 2,575 infants who would have received false-positive results.
But these estimations continue to prompt strong reactions from the laboratory community. The statistical model used to estimate these figures is flawed, said Piero Rinaldo, MD, PhD, Co-Director of the Biochemical Genetics Laboratory at the Mayo Clinic (Rochester, Minn.). “According to our actual calculations of 2005 data, they were about 40% off,” he explained. “They predicted a best case scenario with 81 false positives by MS/MS in Minnesota, but we only had 52.”
Rinaldo added that predictions such as this underscore the need for real newborn screening data, rather than data from predictive statistical models. Since 2004, he has been involved with the Region 4 Genetics Collaborative, a partnership of seven states—Illinois, Indiana, Kentucky, Michigan, Minnesota, Ohio, and Wisconsin—that has been collecting data on newborn screening values. The participants in this collaborative are not limited to the Region 4 states. Currently, there are 38 U.S. state laboratories and 44 screening laboratories in 25 countries contributing monthly data on normal population percentiles, cutoff values, true-positive cases, and performance metrics. The Collaborative is funded by HRSA, and has just been awarded funding through 2012.
“Even though newborn screening has been performed for many, many years, we are bringing some clinical sense to what had evolved into a totally artificial statistical process,” said Rinaldo. “The choice of cutoff values needs to be driven by what happens in real patients.”
An important component of the Collaborative is the monthly feedback e-mailed to participants and posted on the group’s Web site. These include summaries of cutoff ranges, plot analysis, and comparison tools, all of which are anonymous. “By looking at this data and plot analysis, labs are able to analyze if their cutoff values are in danger of missing some cases,” said Rinaldo. “If a lab goes way below the cutoff ranges, then they will have a lot of false positives. But if they go too high above the cutoff target ranges, then they risk missing true positive cases.”
Howell praises the data-gathering efforts of the Region 4 Genetics Collaborative, adding that their initiatives are important in decreasing the false-positive rates. “The Region 4 Project is just fabulous in that they are getting absolute lab data on confirmed, positive children,” he said. “This allows them to make the connection between disorders and values. In addition, they are now refining the cutoff values, which will reduce the false positives tremendously.”
In addition to the state efforts to gather data, Howell also believes similar initiatives are needed at the national level. Many of these conditions—both currently on the expanded panel and under consideration to be added to the panel—are rare, and data on children diagnosed with these disorders need to be collected to refine both testing and treatment options. “What we really need is a system that tracks patients who have rare disorders such as fatty acid oxidation defects in Wyoming, Arizona, or elsewhere in the country,” said Howell, adding that the system should analyze the accuracy of the diagnosis, how the patient was treated, and medical outcome. “For example, many patients with MCAD deficiency are currently treated with carnitine, and some are not, but we really don’t know which is the right treatment. A major research effort that would look at outcomes is needed.” The National Institute of Child Health & Human Development’s Director, Duane Alexander, MD, has expressed an interest in possibly establishing such a system under the agency, according to Howell.
Others in the clinical laboratory community hope that a centralized system would act as a national clearinghouse for all of the data gathered through these various initiatives. “Most of these disorders are so rare that if we don’t have some collaboration to share case-based information, it would be very difficult to develop a clear understanding of the diversity of the disease, the spectrum of clinical presentation, and how that impacts newborn screening,” explained John Sherwin, PhD, Chief of the California Department of Health Services’ Genetic Disease Laboratory. “For example, many assume that a disorder like MCAD deficiency happens in about 1 in 15,000 newborns, but with this data, we might be able to get more accurate statistics. We’ll also be able to gather clinical hallmarks of presentation of these various disorders to help our clinical colleagues understand the spectrum of presentation of these newborn infants, as well as get universal cutoffs and reference ranges for various ethnic groups.”
As newborn screening programs expand in both complexity and breadth in the future, clinical laboratorians will be able to keep pace with the variety of new tools that are coming from these current initiatives—from the Region 4 Collaborative’s data collecting efforts to the NACB’s new LMPGs, said Mayo Clinic’s Rinaldo. “It’s hard to encourage people to both expand and improve their screening programs without telling them how,” he explained. “But now we are developing performance metrics and other tools that are driving quality, which will be one of the many things available to help the laboratory community as newborn screening expands in the near future.”
For More Information
Additional information on the Region 4 Genetics Collaborative can be found online
Information and forms to submit a nomination for the expanded screening panel are available on HRSA’s Web site.