The Importance of Assay Imprecision near the Screen Cutoff for Newborn Screening of Lysosomal Storage Diseases

• Main Authors: Bruce H. Robinson, Michael H. Gelb
• Format: Article
• Language: English
• Published: MDPI AG 2019-03-01
• Series: International Journal of Neonatal Screening
• Subjects: newborn screening; cutoff values; assay imprecision; false positives; false negatives; screening assays
• Online Access: https://www.mdpi.com/2409-515X/5/2/17

For newborn screening (NBS) of lysosomal storage diseases, programs measure enzymatic activities in dried blood spots (DBS) and, in most cases, act on samples where the measurement is below a specific cutoff value. The rate of false positives and negatives in any NBS program is of critical importance. The measured values across a population of newborns are governed by many factors, and in this article we focus on assay imprecision. Assay parameters including the Analytical Range and the Z-Factor have been discussed as a way to compare assay performance for NBS of lysosomal storage diseases. Here we show that these parameters are not rigorously connected to the rate of false positives and negatives. Rather, it is the assay imprecision near the screen cutoff that is the most important parameter that determines the rate of false positives and negatives. We develop the theoretical treatment of assay imprecision and how it is linked to screen performance. What emerges is a useful type of parametric plot that allows for rigorous assessment of the effect of assay imprecision on the rate of false positives and false negatives that is independent of the choice of screen cutoff value. Such plots are useful in choosing cutoff values. They also show that a high assay imprecision cannot be overcome by changing the cutoff value or by use of postanalysis, statistical tools. Given the importance of assay imprecision near the cutoff, we propose that quality control DBS are most useful if they span a range of analyte values near the cutoff. Our treatment is also appropriate for comparing the performance of multiple assay platforms that each measure the same quantity (i.e., the enzymatic activity in DBS). The analysis shows that it is always best to use the assay platform that gives the lowest imprecision near the cutoff.