Evaluation of performance of Dutch newborn screening program for cystic fibrosis (2016-2020)

Abstract

Introduction: The Dutch newborn screening (NBS) program for Cystic Fibrosis (CF) started in May 2011 and contains four steps: immunoreactive trypsinogen as general first step and subsequently pancreatitis-associated protein, Innolipa lineblot and extended gene analysis in a selection of cases. Periodic evaluations from 2011-2015 showed a sensitivity below 95% and improvements were implemented in 2016. Our aim was to evaluate performance and costs of the adjusted NBS protocol between 2016 and 2020.

Methods: Data between July 2016 and January 2020 were used for analysis and were obtained from five screening laboratories and seven CF centers in the Netherlands. The dataset was used to calculate sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), number of carriers and subjects with CF screening positive inclusive diagnosis (CF-SPID), and costs per screened newborn. Results were compared with performance of the previous NBS protocol.

Results: 599,137 newborns were included in the Dutch NBS program. A total of 128 children had positive screening for CF, eleven of them were referred based on the new protocol of which five were diagnosed with classic CF. The adjusted program (compared to the previous protocol) had a sensitivity of 96% (91%), specificity 100% (100%), PPV 77% (79%), NPV 100% (100%), number of carriers 23 (17), CF/carrier ratio 4/1 (5/1), number of CF-SPID 8 (8), CF/CF-SPID ratio 12:1 (12:1), costs per screened newborn were €3.19 (€3.02).

Conclusion: The adjusted Dutch NBS program for CF shows a better performance with higher sensitivity and still an acceptable number of carriers and CF-SPID and acceptable increase of costs per screened newborn.