Pneumologie 2015; 69 - A1
DOI: 10.1055/s-0035-1556593

Combining IRT/PAP+SN with DNA analysis for the best CF newborn screening strategy for Germany

O Sommerburg 1, 2, M Stahl 1, 2, M Muckenthaler 2, 3, D Kohlmüller 4, M Happich 3, AE Kulozik 3, GF Hoffmann 2, 4, MA Mall 1, 2, 5
  • 1Division of Pediatric Pulmonology & Allergy and Cystic Fibrosis Center, Department of Pediatrics III, University of Heidelberg, Germany
  • 2Translational Lung Research Center Heidelberg (TLRC), Member of the German Center for Lung Research (DZL), Heidelberg, Germany
  • 3Dept. of Pediatric Oncology, Hematology, Immunology and Pulmonology, University of Heidelberg, Germany
  • 4Division of Metabolic Diseases and Newborn Screening Center, Department of Paediatrics I, Children's Hospital, University of Heidelberg, Germany
  • 5Department of Translational Pulmonology, University of Heidelberg, Germany

Introduction: Evidence from recent European studies suggests that protocols using immunoreactive trypsine (IRT) and pancreatitis associated protein (PAP) for cystic fibrosis (CF) newborn screening (NBS) may be successfully used as a purely biochemical NBS strategy that does not require genetic screening. Pure IRT/PAP protocols can reach acceptable sensitivities and specificities, but this comes at the expense of a relatively low positive predictive value (PPV). The Dutch CHOPIN study (Vernooij-van Langen 2012) has reported that using a three tier strategy (IRT/PAP + extended gene analysis) results in a better PPV.

Objective: Assessment of the performance of different IRT/PAP/DNA CF NBS protocols using different numbers of CFTR mutations included in the DNA test panel.

Methods: Recent data available from the cohort of the Heidelberg CF NBS study (2008 – 2014, 354.722 newborns) were evaluated in a post hoc analysis. The IRT/PAP protocol used in Heidelberg (Sommerburg et al. 2010) used one IRT-independent PAP-cut-off and a safety net (SN, CF-NBS positive, if IRT ≥99.9th IRT percentile) to reach a sufficient sensitivity.

Findings: The combined strategy (IRT/PAP+SN/DNA) resulted in a sensitivity similar to that of IRT/PAP+SN, but a higher specificity and a higher PPV. Nevertheless, the strategy keeps main advantages of IRT/PAP, such as the detection of considerably fewer carriers and fewer newborns with CFSPID when compared with current IRT/DNA screening strategies. Using a mutation panel covering less than 80% allele frequency will lead to a loss of sensitivity when compared to the results of IRT/PAP+SN alone. A sufficient performance (sensitivity and PPV) could already be reached with a mutation panel covering the most common 20 mutations in the German population.

Conclusions: Our results obtained in a cohort of 354.722 newborns support the use of a IRT/PAP+SN/DNA protocol to achieve a better PPV compared to a purely biochemical IRT/PAP+SN protocol.

*Presenting author