Background: Pancreatitis-associated protein (PAP) is currently discussed as a marker in newborn screening (NBS) for cystic fibrosis (CF). However, it is not known if PAP concentrations are influenced by sex, gestational age, birth weight, blood transfusion or time of collection and what this would mean for NBS for CF.
Methods: In 2008 all newborns in part of the Netherlands were screened for CF by an IRT/PAP protocol. PAP concentration was determined by the MucoPAP ELISA (DynaBio), which was modified to a Dissociation Enhanced Lanthanide Fluoroimmunoassay (DELFIA) method following a protocol of PerkinElmer.
Results: In healthy newborns, the median PAP concentration was 0.5 μg/l (Interquartile range (IQR 0.3-0.8) whereas this was 3.2 μg/l (IQR 2.0-12.5) in CF infants. PAP concentrations were lower in premature infants 0.94 and 0.91 times for 25 to 31 + 6 weeks GA and 32 to 36 + 6 weeks respectively. A higher PAP concentration was observed in low-birth-weight infants (<2500 gram)(p = 0.001), per 100 gram birth weight gained the PAP concentration decreased with 0.1 %. PAP levels were higher after a blood transfusion, the 95th percentile increased from 1.3 to 3.6 μg/l leading to a higher false-positive rate. The PAP concentration increased when newborn screening was performed more than 168 hours (day 7) after birth (β = 1.63), the 95th percentile increased from 1.3-1.6 μg/l to 4.0 μg/l after 168 hours (72,874 newborns were screened).
Conclusion: Sex, birth weight, and gestational age lead to small differences in PAP concentrations without consequences for the screening algorithm. However, blood transfusion as well as performance of the heel prick after 168 hours (7 days) lead to clinically significant higher PAP levels and to a higher risk on a false-positive screening test result.