New Tool Improves Newborn Screening Accuracy

Ricki Lewis, PhD

February 16, 2012

February 16, 2012 — A new approach to analyzing metabolite levels in newborn blood samples decreases the likelihood of false-negatives and false-positives, according to results from a new study by Gregg Marquardt, MSS, from the Department of Laboratory Medicine and Pathology, Mayo Clinic College of Medicine, Rochester, Minnesota, and colleagues. The study was published online February 16 in Genetics in Medicine.

The US Centers for Disease Control and Prevention has named newborn screening for metabolic disorders 1 of the 10 great public health achievements of the last decade. Tandem mass spectrometry is used to screen samples for up to 60 conditions for abnormal levels of markers. However, screening using analyte cutoffs can result in errors.

To improve the accuracy of newborn screening, researchers in 154 laboratories in 49 countries collaborated to apply multivariate pattern-recognition software for several analytes per disease, converting analyte cutoffs to composite scores that reflect the degree of overlap between a population without a particular metabolic disorder and a diagnosed population with that disorder. The researchers define "clinical relevance" as a marker reaching the median of the disease range while remaining outside the percentile limits of the healthy population.

The analysis generates a single score, which also considers penetration within the disease range, differences between diseases, and weighted correction factors that reflect the degree of overlap between the normal and disease population for each informative marker. The correction is unique to each condition, the researchers write.

The article offers examples of how the new technique corrects false-negatives and false-positives. The false-negative example is the urea cycle disorder argininosuccinic acid (ASA) lyase deficiency. Using the cutoff for the marker citrulline, the screening case was deemed uninformative. However, the new approach designated the score "most likely" indicative of ASA lyase deficiency.

The new tool greatly improves sensitivity. The database includes 86 cases (0.7% of the total) that were reported as normal on conventional newborn screening but were diagnosed, with any of 23 conditions, following clinical presentation. The 86 cases fell to 67 by excluding conditions with uninformative markers, and of those cases, 59 (88%) would have been identified using the new tool.

Four disorders illustrate the common scenario of a heterozygote testing falsely positive because of an intermediate concentration of a metabolite. The new approach is more accurate than simply elevating the cutoff for diagnosis to exclude heterozygotes.

At publication time, the database included 767,464 results from 12,721 cases representing the 60 diseases that make up the uniform panel that the Secretary of Health and Human Services adopted in 2010. Altogether, the researchers developed 90 tools that either diagnose a particular condition or distinguish between conditions. Most of the analytes are amino acids or acylcarnitines.

The researchers conclude, "The impact of this objective, evidence-driven approach to the interpretation of laboratory results could be substantial."

In commenting on the study, James P. Evans, MD, PhD, Bryson Distinguished Professor of Genetics and Medicine at the University of North Carolina at Chapel Hill, agreed that this approach could be influential. "This effort represents a collaborative tour de force involving over 150 centers from 49 countries. The investigators implemented an innovative approach to the analysis of newborn screening results, and in so doing achieved significant reductions in false positive and false negative rates. Adoption of such approaches promises to significantly enhance the efficacy of newborn screening throughout the world," he told Medscape Medical News.

This work was supported by a grant to the Region 4 Genetics Collaborative from the Health Resources and Service Administration of the Maternal and Child Health Bureau Cooperative Agreement; by contracts from the Eunice Kennedy Shriver National Institute of Child Health and Human Development, National Institutes of Health, Department of Health and Human Services, and the Newborn Screening Translational Research Network; and by the T. Denny Sanford Professorship fund, Mayo Clinic College of Medicine. A full list of author disclosures is available on the journal's Web site. Dr. Evans has disclosed no relevant financial relationships.

Gen Med. Published online February 16, 2012. Abstract


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.